{"collection":"philosophy-ethics","full_text_available":true,"graph_url":"https://unchartedknowledge.5gfusion.net/v2/api/page/gutenberg_47848/graph","html_url":"https://unchartedknowledge.5gfusion.net/v2/page/gutenberg_47848","id":"gutenberg_47848","license":"Public Domain (US)","rights_basis":"gutenberg_metadata_copyright_false","source":"gutenberg","source_url":"https://www.gutenberg.org/ebooks/47848","summary":null,"tags":["Butterflies","Category: Science - Biology","Heredity","Insects","Insects -- Evolution","Lepidoptera","Mechanism (Philosophy)","Zoology"],"text":"Transcriber\u2019s notes: Italics is represented by _underscores_; boldface\nis represented by =equals signs=.\n\nPages 1-400, Plates I-II, and some referenced footnotes are in Volume I.\n\n\n\n\n  STUDIES IN THE THEORY\n  OF DESCENT\n\n  BY\n\n  DR. AUGUST WEISMANN\n  PROFESSOR IN THE UNIVERSITY OF FREIBURG\n\n  _WITH NOTES AND ADDITIONS BY THE AUTHOR_\n\n  TRANSLATED AND EDITED, WITH NOTES, BY\n\n  RAPHAEL MELDOLA, F.C.S.\n  LATE VICE-PRESIDENT OF THE ENTOMOLOGICAL SOCIETY OF LONDON\n\n  WITH A PREFATORY NOTICE BY\n\n  CHARLES DARWIN, LL.D., F.R.S.\n  _Author of \u201cThe Origin of Species,\u201d &c._\n\n  IN TWO VOLUMES\n  VOL. II.\n\n  WITH EIGHT COLOURED PLATES\n\n  London:\n\n  SAMPSON LOW, MARSTON, SEARLE, & RIVINGTON\n  CROWN BUILDINGS, 188, FLEET STREET\n\n  1882\n\n  [_All rights reserved_]\n\n\n\n\nI.\n\nLARVA AND IMAGO VARY IN STRUCTURE INDEPENDENTLY OF EACH OTHER.\n\n\nIt would be meaningless to assert that the two stages above mentioned\nwere _completely_ independent of one another. It is obvious that the\namount of organic and living matter contained in the caterpillar\ndetermines the size of the butterfly, and that the quantity of organic\nmatter in the egg must determine the size of the emergent larva.\nThe assertion in the above heading refers only to the structure;\nbut even for this it cannot be taken as signifying an absolute, but\nonly a relative independence, which, however, certainly obtains in\na very high degree. Although it is conceivable that every change of\nstructure in the imago may entail a correlative change of structure\nin the larva, no such cases have as yet been proved; on the contrary,\nall facts indicate an almost complete independence of the two stages.\nIt is quite different with cases of _indirect_ dependence, such, for\nexample, as are brought about by \u2018nurse-breeding.\u2019 This phenomenon is\nalmost completely absent in Lepidoptera, but is found in Diptera, and\nespecially in Hymenoptera in every degree. The larv\u00e6 of ichneumons\nwhich live in other insects, require (not always, but in most\ninstances) that the female imago should possess a sharp ovipositor,\nso that in this case also the structure and mode of life of the larva\ninfluences the perfect insect. This does not depend, however, on\ninherent laws of growth (correlation), but on the action of external\ninfluences, to which the organism endeavours to adapt itself by natural\nselection.\n\nI will now let the facts speak for themselves.\n\nIt is shown by those species in which only one stage is di- or\npolymorphic that not every change in the one stage entails a\ncorresponding change in the other. Thus, in all seasonally dimorphic\nspecies we find that the caterpillars of butterflies which are often\nwidely different in the colour and marking of their successive\ngenerations are absolutely identical. On the other hand, many species\ncan be adduced of which the larv\u00e6 are dimorphic whilst the imagines\noccur only in one form (compare the first and second essays in this\nvolume).\n\nThere are however facts which directly prove that any one stage can\nchange independently of the others; I refer to the circumstance that\nany one stage may become independently variable--that the property of\ngreater variability or of greater constancy by no means always occurs\nin an equal degree in all the three stages of larva, pupa, and imago,\nbut that sometimes the caterpillar is very variable and the pupa and\nimago quite constant. On the other hand, all three stages may be\nequally variable or equally constant, although this seldom occurs.\n\nIf variability is to be understood as indicating the period of\nre-modelling of a living form, whether in its totality or only in\nsingle characters or groups of characters, from the simple fact of the\nheterochronic variability of the ontogenetic stages, it follows that\nthe latter can be modified individually, and that the re-modelling\nof one stage by no means necessarily entails that of the others. It\ncannot however be doubted that variability, from whatever cause it may\nhave arisen, is in all cases competent to produce a new form. From the\ncontinued crossing of variable individuals alone, an equalization of\ndifferences must at length take place, and with this a new, although\nnot always a widely deviating, constant form must arise.\n\nThat the different stages of development of a species may actually be\npartly variable and partly constant, and that the variable or constant\ncharacter of one stage has no influence on the other stages, is shown\nby the following cases, which are, at the same time, well adapted to\nthrow light on the causes of variability, and are thus calculated to\ncontribute towards the solution of the main problem with which this\ninvestigation is concerned.\n\nWhen, in the following pages, I speak of _variability_, I do not\nrefer to the occurrence of local varieties, or to variations which\noccur in the course of time, but I mean a high degree of individual\nvariability--a considerable fluctuation of characters in the\nindividuals of one and the same district or of the same brood. I\nconsider a species to be constant, on the other hand, when the\nindividuals from a small or large district differ from one another only\nto a very slight extent. Constant forms are likewise generally, but not\ninvariably, such as are poor in local varieties, whilst variable forms\nare those which are rich in such variations. Since the terms \u201cvariable\u201d\nand \u201cconstant\u201d are but relative, I will confine myself to the most\nextreme cases, those in which the individual peculiarities fluctuate\nwithin very wide or very narrow limits.\n\nAs no observations upon the degree of variability shown by a species in\nthe different stages of its development were available, I was obliged\nto fall back upon my own, at least so far as relates to the larval and\npupal stages, whilst for the imaginal stage the wide experience of my\nesteemed friend Dr. Staudinger has been of essential service to me.\n\nLet us in the first place confine our attention to the three chief\nforms which every Lepidopteron presents, viz. larva, pupa, and imago.\nWith respect to the constancy or variability of these three forms, we\nactually find in nature all the combinations which are theoretically\nconceivable.\n\n(1.) There are species which possess a high degree of constancy in\nall three stages, such, for example, as _Limenitis Camilla_, _Pieris\nBrassic\u00e6_,[171] _Sphinx Ligustri_, and _Euchelia Jacob\u00e6\u00e6_.\n\n(2.) There are species showing a high degree of variability in all\nthree stages. This case must be of rare occurrence, as I am only able\nto adduce _Araschnia Prorsa-Levana_, a fact which arises from the\ncircumstance that the pupal stage is, as a rule, but seldom variable.\n\n(3.) There are species which are variable in two stages and constant\nin the third. To this class, for example, belongs _Smerinthus Tili\u00e6_,\nof which the larva and imago are very variable, whilst the pupa is\nquite constant. The same is the case with _Lasiocampa Pini_, the\nwell-known fir moth. Many butterflies show this same phenomenon in\nother combinations, such, for instance, as _Vanessa Urtic\u00e6_ and\n_Polychloros_, in which the larva and pupa are very variable, and the\nimago very constant. In a less degree the same is also the case with\n_Vanessa Atalanta_, whilst in _Pieris Napi_ the pupa and imago are\nvariable, and the caterpillar remarkably constant, this likewise being\nthe case with the local form _Bryoni\u00e6_, which, according to my theory,\nis to be regarded as the parent form of _Napi_ (See Part I. of the\npresent volume).\n\n(4.) There are species which are constant in two stages, and variable\nonly in the third. Thus, a few species can be found in which the\nlarva and pupa are constant and the imago variable. This is the case\nwith _Saturnia Yamamai_, the imago of which is well known to present\nnumberless shades of colour, varying from light yellow to greyish\nblack, whilst the green caterpillar shows only slight individual\ndifferences of marking, and scarcely any differences of colour. The\npupa of this species is quite constant. _Arctia Caja_ and _Hebe_, and\n_Chelonia Plantaginis_ belong to this same category.\n\nThere are a very large number of species which possess very constant\nimagines and pup\u00e6, but extremely variable larv\u00e6. The following are\nthe cases known to me:--_Macroglossa Stellatarum_, _Fuciformis_\nand _Bombyliformis_; _Ch\u00e6rocampa Elpenor_, _Celerio_, and _Nerii_;\n_Deilephila Galii_, _Livornica_, H\u00fcbn., _Hippopha\u00ebs_, _Vespertilio_,\nand _Zygophylli_; _Sphinx Convolvuli_; _Acherontia Atropos_;\n_Smerinthus Ocellatus_ and _Tili\u00e6_; _Callimorpha Hera_; _Cucullia\nVerbasci_ and _Scrophulari\u00e6_.\n\nCases in which the variability depends entirely upon the pupa, while\nthe larva and imago are extremely constant, are of great rarity.\n_Vanessa Io_ is a case in point, the pupa being light or dark brown, or\nbright golden green, whilst in the two other stages scarcely any light\nshades of colour or variations in the very complicated marking are to\nbe met with.\n\nThe facts thus justify the above view that the individual stages of\ndevelopment change independently--that a change occurring in one stage\nis without influence on the preceding and succeeding stages. Were this\nnot the case no one stage could possibly become variable without all\nthe other stages becoming so. Did there exist a correlation between\nlarv\u00e6, pup\u00e6, and imagines of such a nature that every change in the\nlarva entailed a corresponding change in the imago, as soon as a large\nnumber of larval characters became fluctuating (_i.e._ as soon as this\nstage became variable), a large number of imaginal characters would\nnecessarily also become fluctuating (_i.e._ this stage would also\nbecome correspondingly variable).\n\nThere is one other interpretation which might perhaps be attempted from\nthe point of view of the old doctrine of species. It might be said that\nit is a special property of certain larval or imaginal markings to be\nvariable whilst others are constant, and since the larval and imaginal\nmarkings of a species are generally quite distinct, it may easily\nhappen that a butterfly possessing markings having the property of\nconstancy may belong to a caterpillar having variable markings.\n\nThere is a soul of truth underlying this objection, since it is\ntrue that the various forms of markings which occur in Lepidoptera\napparently reach different degrees of constancy. If we speak of the\nconstancy or variability of a species, a different meaning is attached\nto these expressions according as we are dealing _e.g._ with a species\nof _Sphinx_ or a species of _Arctia_. That which in the latter would\nbe estimated as a high degree of constancy, in the former would be\ntaken as a considerable amount of variability. It is of interest, in\nconnection with the question as to the causes of constancy, to note\nthat the power of any form of marking to attain to a high degree of\nconstancy is by no means inversely proportional to the complication of\nthe marking, as would have been expected _\u00e0 priori_.\n\nThus, the species of _Sphinx_ and of allied genera possess on their\nfore-wings, which are mostly coloured with a mixture of dull grey,\nwhite and black, an exceedingly complicated arrangement of lines\nwhich, in constant species, show a high degree of uniformity: on the\nother hand, the checquered fore-wings of our _Arctiid\u00e6_, which are far\nmore coarsely marked, always show, even in the most constant species,\nwell-marked individual differences. The different types of marking must\ntherefore be measured by different standards.\n\nBut in granting this, we decidedly refute the statement that constancy\nand variability are inherent properties of certain forms of marking.\n\nThis reasoning is based on the simple fact that a given type of marking\ncomprises both species of great constancy and of (relatively) great\nvariability.\n\nThus, the fore-wings of _Sphinx Ligustri_ and _S. Convolvuli_ are\nextremely constant, whilst the very similarly marked _Anceryx\n(Hyloicus) Pinastri_ is exceedingly variable. Similarly _Deilephila\nEuphorbi\u00e6_ is known by its great variability of colouring and marking,\nwhilst _D. Galii_, which resembles this species so closely as to be\nsometimes confounded with it, possesses a high degree of constancy,\nand further, the Corsican and Sardinian _D. Dahlii_ is very variable.\nAmong the family _Arctiid\u00e6_, _Callimorpha Hera_ and the Alpine _Arctia\nFlavia_ are cases of constancy, whilst _A. Caja_, which is so similar\nto the last species, is so generally variable that two perfectly\nidentical specimens can scarcely be found together.\n\nThe same can be shown to hold good for the markings of caterpillars.\nThus, the larva of _D. Dahlii_ shows very considerable variability,\nwhilst that of _D. Galii_ is very constant in marking (disregarding the\nground-colour). So also the larva of _Vanessa Urtic\u00e6_ is very variable\nand that of _V. Antiopa_ very constant, &c.\n\nThe great differences with respect to constancy or variability which\nare displayed by the different stages of one and the same species,\nmust therefore find their explanation elsewhere than in the type of\nthe marking itself. The explanation must be found in the circumstance\nthat each stage changes independently of the others, and at different\nperiods can enter a new phase of variability.\n\nWe are here led in anticipation to the main question:--Are changes\nproduced by internal or external causes? is it the physical nature of\nthe organism which is compelled to become remoulded spontaneously after\nthe lapse of a certain period of time? or does such modification only\noccur when produced directly or indirectly by the external conditions\nof life?\n\nIn the cases before us the facts undoubtedly indicate a complete\ndependence of the transformations upon external conditions of life.\n\nThe independent appearance of variability in the separate stages of the\nmetamorphosis might, however, be regarded as only apparent. It might\nstill be attempted to attribute the changes to a purely inherent cause,\n_i.e._, to a phyletic vital force, by assuming that the latter acts\nperiodically in such a manner that at first one and then the following\nstage becomes variable, until finally the entire species is transformed.\n\nThere is but little to be said in reply to this if we once take refuge\nin entirely unknown forces, the operation of which can be arbitrarily\nconceived to be either constant or periodic.\n\nBut granting that such a transforming power exists and acts\nperiodically, the variability must always pass over the different\nstages in a fixed direction, like a wave over the surface of\nwater--imago, pupa, and larva, or larva, pupa, and imago, must\n_successively_ become variable. Cases like that of _Araschnia\nProrsa_, in which all three stages are variable, may certainly be\nthus explained, but those instances in which the larva and imago\nare extremely variable, and the pupa quite constant, are entirely\ninexplicable from this point of view.\n\nThe latter can, however, be very simply explained if we suppose the\nchanges to be dependent upon external influences. From this standpoint\nwe not only see how it is possible that an intermediate stage should\nremain uninfluenced by the changes which affect the two other stages,\nbut we can also understand why it should just be the pupal stage that\nplays this part so frequently. If we ask why most pup\u00e6 are constant\nand are relatively but very slightly variable, the answer will be\nfound in the facts that all pup\u00e6 which remain concealed in the earth\nor inside plants (_Sesiid\u00e6_), or which are protected by stout cocoons,\nshow complete constancy, whilst any considerable amount of variability\noccurs only in those pup\u00e6 which are suspended or openly exposed. This\nis closely connected with a fact to which I have called attention on a\nformer occasion,[172] viz., that dimorphism occurs in certain pup\u00e6,\nbut only in those which are openly exposed and which are therefore\nvisible to their foes. I am only acquainted with such cases among the\npup\u00e6 of butterflies, and it is likewise only among these that I have\nfound any considerable amount of variability.\n\nFacts of this kind indicate that Nature does not uselessly sport with\nforms, but that at any rate changes of this sort result from external\ninfluences. The greater frequency of variability among larv\u00e6 and its\ncomparative rarity in imagines is also undoubtedly in favour of this\nview.\n\nIt has already been shown that species with variable larv\u00e6 and constant\nimagines are extremely common, but that those with constant larv\u00e6\nand variable imagines are very rare. This confirms the conclusions,\nalready drawn above, first, that the variability of the imago cannot\nowe its existence to the variability of the larv\u00e6, and secondly, that\nthe causes which produce variability affect the larval condition more\ncommonly than that of the imago.\n\nWhere can these causes be otherwise sought than in the external\nconditions of life, which are so widely different in the two stages,\nand which are much more variable for the larva than for the imago?\n\nLet us take the species of one genus, _e.g._ those of _Deilephila_.\nThe imagines of our European species--as far as we know--all live in\nprecisely the same manner; they all fly at twilight,[173] showing a\npreference for the same flowers and very often frequenting the same\nspots, so that in the haunts of one species the others are almost\nalways to be met with, supposing them to occur in the same locality.\nThey conceal themselves by day in similar places, and are attacked by\nsimilar foes.\n\nIt is quite different with the caterpillars. These, even in the case of\nthe most closely allied species, live under different conditions, as\nappears from the fact that they feed on different plants. The latter\ncan, however, produce changes both directly and indirectly. The larv\u00e6\nmay acquire adaptive colours and markings, and these would vary in\naccordance with the colour and structure of the food-plant; or they may\nbecome brightly coloured as a sign of distastefulness in cases where\nthey are inedible. Then again the colour of the soil on which the larv\u00e6\nlive would act upon their colours making these adaptive. Certain habits\nof the caterpillars may also be dependent upon the nature of their\nfood-plants. Thus, _e.g._ _Deilephila Hippopha\u00ebs_ feeds only at night,\nand conceals itself by day under moss and among the leaves at the base\nof the food-plant; but _D. Euphorbi\u00e6_ could not acquire such a habit,\nbecause _Euphorbia Cyparissias_ generally grows on arid soil which is\npoor in vegetation, and which therefore affords no concealment, and\nfurthermore, because a caterpillar, as long as it continues to feed,\ncannot, and as a matter of fact does not, ever wander far from its\nfood-plant. A habit of concealment by burying in the earth also, such\nfor example as occurs in _Acherontia Atropos_, could not be acquired by\n_D. Euphorbi\u00e6_, because its food-plant generally grows on hard, dry,\nand stony ground.\n\nIn addition to these considerations, the foes would be different\naccording as the caterpillar lived on plants which formed dense\nthickets covering large extents of the shore (_Hippophae_) or grew\nisolated on dry hillocks and declivities where the herbage was scanty\nor altogether absent; or again, according as the insect, in conjunction\nwith such local differences, fed by day or had acquired the habit\nof feeding only by night. It must in fact be admitted that new and\nimproved adaptations, or, in more general terms, that inducements to\nchange, when depending on the environment, must be more frequently\ndissimilar for larv\u00e6 than for the imagines. We must accordingly expect\nto find actual change, or that condition of variability which may be\nregarded as initiative to change, occurring more commonly in larv\u00e6 than\nin perfect insects.\n\nSince facts are in complete accordance with the results of these _\u00e0\npriori_ considerations we may also venture to conclude that the basis\nof the considerations is likewise correct, viz., the supposition that\nthe changes of colour and marking in caterpillars, pup\u00e6, and imagines\nresult from external influences only.\n\nThis must not be taken as signifying that the single stages of the\nlarval development are also only able to change through the action of\nexternal influences. The larval stages are correlated with each other,\nas has already been shown (see the previous essay): new characters\narise in the adult caterpillar at the last stage and are then gradually\ntransferred back to the younger stages quite independently of external\ninfluences, this recession being entirely brought about by the laws of\ncorrelation. Natural selection here only exerts a secondary action,\nsince it can accelerate or retard this transference, according as the\nnew characters are advantageous or disadvantageous to the younger\nstages.\n\nNow as considerable individual differences appear in the first\nacquisition of a new character with respect to the rapidity and\ncompleteness with which the individuals acquire such a character, the\nsame must obtain for the transference of an improvement acquired in\nthe last stage to the next younger stage. The new character would be\nacquired by different individuals in different degrees and at different\nrates--it would have, to a certain extent, to struggle with the older\ncharacters of the stage; in brief, the younger stage would become\nvariable.\n\nVariability of this kind might well be designated as _secondary_,\nin contradistinction to _primary_ variability; the latter (primary)\ndepends upon an unequal reaction of the individual organisms to\nexternal influences, the former (secondary) results from the unequal\nstrength and rate of the action of the innate laws of growth governing\nthe organism. In both cases alike exceeding variability may occur, but\nthe causes producing this variability are dissimilar.\n\nThe different stages of larval development would thus frequently\ndisplay independent variability in a manner similar to the pupal or\nimaginal stages, since they can show individual variability while\nthe other stages of development remain constant. This appearance\nof independent variability in the different stages of the larval\ndevelopment, however, is in truth deceptive--we have here in fact\na kind of wave of variability, which passes downwards through the\ndevelopmental stages, becoming gradually weaker, and finally dying out\ncompletely.\n\nIn accordance with this, we very frequently find that only the last or\ntwo last stages are variable, while the younger stages are constant.\nThus in _Macroglossa Stellatarum_, the larv\u00e6 are constant in the first,\nsecond, and third stages, but become variable in the fourth, and in\nthe fifth stage first show that high degree of variability which has\nalready been described in detail (See. Pl. III., Figs. 3-12). The\nlarv\u00e6, of _Vanessa Cardui_ also, according to my notes, are extremely\nconstant in the first four stages in spite of their complicated\nmarking, but become variable in the fifth stage, although to no very\ngreat extent.\n\nIn _Smerinthus Tili\u00e6_, _Ocellatus_ and _Populi_ also, the greatest\nlarval variability is shown only in the last stage, the preceding\nstages being very constant. These cases by no means depend upon the\nmarking of the young stages being simpler and therefore being less\ncapable of varying. The reverse case also occurs. In a somewhat similar\nmanner as the young of the tapir and wild hog are striped, while the\nadult animals are plainly coloured, the young caterpillars of _Saturnia\nYamamai_ possess longitudinal black lines on a yellow ground, while as\nearly as in the second stage a simple green colour appears in the place\nof this complicated but perfectly constant marking. If the young stages\nare so frequently constant, this rather depends upon the fact that the\ntransference of a new character to these stages not only takes place\ngradually, but also with continually diminishing energy, in a manner\nsomewhat similar to physical motion, which continually diminishes in\nspeed by the action of resistance till it is completely arrested. This\nconstancy of the younger stages may further be due to the circumstance\nthat the characters would only be transferred when they had become\nfixed in the last stage, and were consequently no longer variable. The\ntransferred characters may thus have acquired a greater regularity,\n_i.e._ a less degree of variability, than they possessed at their first\norigination. Extensive investigations in this special direction must\nbe made if the precise laws, in accordance with which the backward\ntransference of new characters takes place, are to be discovered. By\nsuch researches only should we arrive with certainty at the causes\nwhich determine the lesser variability of the young larval stages.\n\nIt may also occur that the early stages are variable, whilst the\nlater stages are constant, although this case appears to happen less\nfrequently. Thus, the caterpillars of _Gastropacha Quercifolia_ vary\nconsiderably in the second stage but are constant at a later period,\nand the same is the case with _Spilosoma Urtic\u00e6_, which in the second\nstage may be almost considered to be dimorphic, but which subsequently\nbecomes constant.\n\nCases in which the first stage is variable appear to be of the least\nfrequent occurrence. I know of only one such instance, viz., _Anceryx\nPinastri_, of which the newly hatched larv\u00e6 (Pl. VI., Fig. 53) show\nconsiderable differences in the brownish-black crescentic spots. The\nsecond (Fig. 54), third, and fourth stages are then tolerably constant,\nwhile the fifth stage again is very variable.\n\nAn instance of this kind can be easily explained by two waves of\nvariation, the first of which now affects only the first stage,\nwhile the second has just commenced to affect the fifth stage. Such\na supposition is not opposed to any theoretical considerations, but\nrather has much probability in its favour, since we know that species\nare from time to time subject to be remodelled; and further, that the\ncoalescence of several stages of phyletic development in the ontogeny\nof one and the same species (see p. 226, development of the genus\n_Deilephila_) shows that during the backward transference of one\ncharacter, new characters may appear in the last stage of the ontogeny,\nand indeed very frequently at a time when the next youngest character\nhas not been transferred back so far as to the first stage.\n\nThat this secondary variability is to a certain extent brought about by\nthe conflict between the old and new characters, the latter striving to\nsuppress the former, is shown by the caterpillar of _Saturnia Carpini_\nwhich I have observed for many years from this point of view, and than\nwhich I do not know a more beautiful illustration.\n\nWhen these larv\u00e6 leave the egg they are black, but in the adult state\nare almost bright green--this at least being the case in a local form\nwhich, from the district in the vicinity of Genoa where it is found,\nI will designate as the var. _Ligurica_. Now whilst these two extreme\nstages of development are relatively constant, the intermediate stages\nshow a variability which becomes greater the nearer the last stage\nis approached, this variation in the marking depending simply on the\nstruggle between the green colour and the more anciently inherited\nblack. In this manner there arises, especially in the fourth stage\nof the German local form, an incredible mixture of the most diverse\nmarkings, all of which can, however, be very easily explained from the\nforegoing point of view.\n\nThe simpler and, as I am inclined to believe, the older form of the\ntransformation is presented to us in the local variety _Ligurica_. In\nthe last stage, when 7.5 centimeters long, this form is of a beautiful\nbright green colour without any trace of black marking[174] (Pl. VIII.,\nFig. 77). The colour of the six orange warts which are situated on\neach segment is also similar in all specimens, so that this stage is\nperfectly constant.\n\nOur German _S. Carpini_ shows different characters in the fifth stage.\nIt is true that individual specimens occur which are entirely green\nwithout any black, but these are rare; the majority possess a more or\nless broad black ring encircling the middle of each segment (Pl. VIII.,\nFigs. 78 and 79). Those specimens in which the black ring has become\nbroken up into large or small spots surrounding the base of the warts\nconstitute intermediate forms (Fig. 80). The last stage of the German\nlocal form, unlike that of the Genoese local form, is therefore very\nvariable.\n\nThe two forms, moreover, do not simply differ in being more or less\nadvanced in phyletic development, but also in several other points. As\nit is of great theoretical interest to show that a species can develop\nlocal differences only in the stage of larva, I will here subjoin the\nplain facts.\n\nThe differences consist in that the Genoese local form goes through\nfive moults whilst the German local form, like most caterpillars,\nhas only four moults. Further, in the Genoese form the light green,\nwhich is also possessed by the German form in the fourth stage, when\nit once appears, is retained to the end of the larval development,\nwhilst in the fifth stage of the German form this colour is replaced\nby a dull greyish-green (compare Figs. 77 and 78). There is further\na very considerable difference in the earlier stages which shows\nthat the phyletic transforming process has taken a quite independent\ncourse in the two forms. Since the struggle between the green and\nblack--retaining this idea--appears to be quite finished in the last\nstage of the Genoese form, we should expect that the new colour, green,\nwould now also have encroached further upon the younger stages than\nin the German form. Nevertheless, this is not the case, but quite\nthe reverse happens, the black maintaining its ground longer in the\nItalian than in the German form.\n\nIn the Genoese form the two first stages are completely black, and\nin the third stage an orange-yellow lateral stripe first appears. In\nthe German form this stripe appears in the second stage, and there\nis not subsequently added, at least on the middle segments, a yellow\nborder surrounding some of the warts of the median series. In the\nthird stage, however, the yellow (which is but the precursor of the\nlater green colour) becomes further extended, so that the caterpillars\noften appear of an orange colour, some or all of the warts and certain\nspots and stripes only being black (Figs. 66 and 68). The warts are\nalso often yellow while the ground remains in most part black--in\nbrief, the bright colour is in full struggle with the black, and an\nendless series of variations is the result of this conflict, whilst in\nthe corresponding stage of the Genoese form almost complete constancy\nprevails.\n\nThis constancy remains also in the following (fourth) stage, the\ncaterpillar still being deep black, only the yellow (sulphur-coloured)\nlateral stripe, which has now become brighter, indicating the impending\nchange (Fig. 67). This takes place in the fifth stage, in which the\nground-colour suddenly becomes bright green, the black remaining at\nmost only in traces on the anterior edges of the segments.\n\nThis is the same marking as is shown by the fourth stage of the German\nform, only in this case individuals quite destitute of black do not\noccur. In many specimens indeed black forms the ground-colour, the\ngreen only appearing in certain spots (Figs. 71 to 75); in others the\ngreen predominates, and these two extremes are connected by innumerable\nintermediate forms, so that this stage must be regarded as the most\nvariable of all.\n\nThe sixth stage of the Genoese and the fifth of the German form have\nalready been compared together. The results may be thus tabulated:--\n\n  _A. German form._                  _B. Genoese form._\n\n  STAGE I. 9 days.                 9 days.\n    Black; constant.                   Black; constant.\n\n  STAGE II. 8 days.                11 days.\n    Black, with orange-yellow          Black; constant.\n      lateral stripe; variable.\n\n  STAGE III. 5 days (in some       12 days.\n    cases as much as 16 days).\n    Black, with yellow; very           Black, with orange-yellow\n      variable.                          lateral stripes; constant.\n\n  STAGE IV. 16 days (in some       6 days.\n    cases only 5 days).\n    Bright green and black,            Black, with bright yellowish\n      mixed; very variable.              lateral stripe; constant.\n\n  STAGE V. 6 days (frequently      6 days.\n    longer).\n    Dark green, with or without        Bright green, small traces\n      black bands; variable.             of black; variable.\n\n  STAGE VI. Pupation.              18 days.\n                                       Bright green, without any\n                                         black; constant.\n\n                                   STAGE VII. Pupation.\n\nFrom this comparison we perceive that the process of transformation\nhas at least become preliminarily concluded in the Genoese form. Why\nthe backward transference of the newly-acquired character to the young\nstages has not yet occurred, or, at least, why it is not in progress,\ndoes not appear; neither can it be stated whether this will take place\nlater, although we may venture to suppose that such will be the case.\nAt first sight but a relatively short time appears necessary for the\nsingle stage V., which is still in a state of fluctuation (variable),\nto become constant by continued crossing, like all the other stages.\n\nThat the transformation is still in full progress in the German form,\nis shown by the fact that in this case all the stages are variable with\nthe exception of the first--the second stage being only variable to a\nsmall extent, the third to a much greater extent, and the fourth to the\nhighest degree conceivable, whilst the fifth and last stage is again\nless variable--so that the greatest struggle between the old and new\ncharacters takes place in the fourth stage.\n\nAmong the innumerable variations presented by this last stage a\ncomplete series of transitional forms can be arranged so as to show the\ngradual conquest of the black by the green, and thus indicating, step\nby step, the course which the latter colour has taken.\n\nIn the blackest specimens there is nothing green but the lateral\n(infra-spiracular) line which was yellow in the preceding stage, and a\ncrescent-shaped streak at the base of the middle warts together with\na still smaller crescent at the base of the upper warts (Figs. 71 and\n81). These spots become extended in lighter specimens and approximate\nso as to leave only narrow black bridges, a third spot being added at\nthe posterior edge of the warts (Figs. 72 and 82). The three spots\nthen extend on all sides, still leaving for a long period narrow\nblack lines at the boundaries where their growth has caused them to\nabut. In this manner there frequently arises on the green ground a\ntrue hieroglyphic-like marking (Figs. 85 and 86). Finally the black\ndisappears from the anterior edge and diminishes on the middle line of\nthe back where it still partly remains as a T-shaped figure (Figs. 73\nand 74), although generally replaced elsewhere by the green with the\nexception of small residues.\n\nOne point remained for a long time inexplicable to me, viz., the change\nof the light green into dark grey-green which appeared in the last\nstage in connection with a total change of the black marking.\n\nSupposing that new characters are actually acquired only in the last\nstage, and that from this they are transferred to the younger stages,\nwe should expect to find completely developed in the last stage the\nsame colouring and markings as are possessed more or less incompletely\nin the fourth stage. Now since the developmental tendency to the\nremoval of black and to the predominance of green--if we may thus\nventure to express it--is obvious in the fourth stage, we may expect to\nfind in the fifth stage a bright green ground-colour, either without\nany mixture of black or with such black spots and streaks as were\nretained in the fourth stage as residues of the original ground-colour.\nBut instead of this the fifth stage shows a dark green colour, and a\nmore or less developed black marking which cannot in any way be derived\nfrom that of the fourth stage.\n\nThe Genoese local form observed last year first gave me an explanation\nto the extent that in this form the last stage is actually only the\npotential penultimate stage, or, more correctly expressed, that\nthe same characters which at present distinguish the last stage of\nthis form, are already more or less completely transferred to the\npenultimate stage.\n\nThe apparently paradoxical behaviour of the German form can be\nexplained by supposing that before the pure bright green had become\ncompletely transferred to the penultimate stage a further change\nappeared in the last stage, the green ground-colour becoming darker,\nand black transverse bands being formed. The marking of the last\nstage would then be regarded as the reverse of that of the preceding\nstage; the absence of black would be the older, simple black spots at\nthe base of the warts the next in succession, and a connected black\ntransverse band the most advanced state of the development.\n\nWhether this explanation is correct, and if so, what causes have\nproduced the second change, may perhaps be learnt at some future\ntime by a comparison with the ontogeny of other _Saturniid\u00e6_; in the\nmeantime this explanation receives support from another side by the\nbehaviour of the Genoese local form. If the last stage of the German\nform has actually commenced to be again re-modelled, then this variety\nis further advanced in phyletic development than the Genoese form; and\nthis corresponds entirely with the theory that in the former the light\ncolour (the orange considered as preliminary to the transformation into\ngreen) has already been carried down into the second stage, whilst in\nthe Genoese variety even in the fourth stage only the first rudiments\nof the colour-transformation show themselves.\n\nThe Genoese form is to a certain extent intermediate between the\nGerman form of _Saturnia Carpini_ and the nearly related _S. Spini_,\na species inhabiting East Germany. In this latter the larv\u00e6, even in\nthe adult state, are completely black with yellow warts. This form of\ncaterpillar must therefore be regarded as phyletically the oldest, and\nthis very well agrees with the character of the moth, which differs\nessentially from _S. Carpini_ only in not being sexually dimorphic. In\n_Carpini_ the male possesses a far more brilliant colouring than the\nfemale, the latter agreeing so completely with the female of _Spini_\nthat it can hardly be distinguished therefrom, especially in the case\nof the somewhat larger South European specimens of the last species.\nNow as the more simple colouring of the female must in any case be\nregarded as the original form, we must consider _Spini_, both sexes\nof which possess this colouring, to be phyletically the older form,\nand _Carpini_, the male of which has become differently coloured, must\nbe considered as the younger type. This completely accords with the\ncharacters of the larv\u00e6.\n\nI must here mention that I have also asked myself the question whether\nthe variations of the different larval stages are connected together as\ncause and effect--whether the lightest specimens of the fifth stage may\nperhaps not also have been the lightest individuals of the third and\nfourth stages.\n\nSuch relationship is only apparent between the third and fourth stages;\nthe darkest larv\u00e6 of the third stage become the darker varieties of the\nfourth stage, although it is true that the lighter forms of the third\nsometimes also become dark varieties in the fourth stage. Between the\nfourth and fifth stages there is scarcely any connection of this kind\nto be recognized. Thus, the darkest varieties of the fourth stage\nsometimes become the lightest forms of the fifth stage, whilst in other\ncases from the lightest individuals of the fourth stage there arise\nall the possible modifications of the fifth stage. Further details may\nbe omitted: the negative result cannot cause any surprise, as it is a\nnecessary consequence of the continued crossing that must take place.\n\nWe thus see that the three chief stages of development (larva, pupa,\nand imago) actually change in colour independently of each other,\nthe single stages of the larval development being however in greater\ndependence upon one another, and being connected indeed in such a\nmanner that a new character cannot be added to the last stage without\nbeing transferred in the course of time to the preceding stage, and at\na later period from this again even to the youngest stage, supposing\nit not to be previously delayed in the course of its transference by\nunknown opposing forces. On this last point, however, the facts at\npresent available do not admit of any certain decision.\n\nBut why do the individual larval stages behave in this respect so\nvery differently to the chief stages of the whole development? why\nare the former so exactly correlated whilst the latter are not? If\nnew characters have a general tendency to become transferred to the\nyounger ontogenetic stages, why are not new imaginal characters first\ntransferred to the pupa, and finally to the larva?\n\nThe answer to these questions is not far to find. The wider two stages\nof a species differ in structure, the less does correlation become\npossible; the nearer the two stages are morphologically related, the\nmore powerful does the action of correlation become. It is readily\nconceivable that the more widely two succeeding stages deviate\nin structure and mode of life, the less possible does it become\nfor characters to be transferred from one to the other. How is it\npossible, for example, that a new character in the proboscis or on the\nwings of a butterfly can be transferred to the caterpillar? If such\ncorrelation existed it could only manifest itself by some other part\nof the caterpillar changing in correspondence with the change of the\nproboscis or wings of the butterfly. That this is not the case has, in\nmy opinion, been conclusively shown by all the foregoing considerations\nrespecting the independent variability of the chief stages of the\nmetamorphosis.\n\nThere are, moreover, an endless number of facts which prove the\nindependence of the individual stages of development--I refer to\nthe multitudinous phenomena presented by metamorphosis itself.\nThe existence of that form of development which we designate as\nmetamorphosis is alone sufficient to prove incontestably that the\nsingle stages are able to change independently of one another to a most\nremarkable extent.\n\nIf we now ask the question: how has the so-called \u201ccomplete\u201d\nmetamorphosis of insects arisen? the answer can only be: through the\ngradual adaptation of the different stages of development to conditions\nof life which have continually deviated more and more widely from each\nother.[175]\n\nBut if individual stages of the post-embryonic development can finally\nattain to such complete diversity of structure as that of the larva and\nimago through gradual adaptations to continually diverging conditions\nof life, this shows that the characters acquired by the single stages\nare always only transferred to the same stages of the following\ngeneration, whilst the other stages remain uninfluenced thereby. This\ndepends upon that form of heredity designated by Darwin \u201cinheritance at\ncorresponding periods of life,\u201d and by Haeckel \u201chomochronic heredity.\u201d\n\n\n\n\nII.\n\nDOES THE FORM-RELATIONSHIP OF THE LARVA COINCIDE WITH THAT OF THE IMAGO?\n\n\nHaving thus established the independence in the variability of\nthe individual stages of metamorphosis, I will now turn to the\nconsideration of the question as to how far a parallelism is displayed\nin the phyletic development of these stages. Is there a complete\ncongruence of form-relationship between larv\u00e6 on the one hand and\nimagines on the other? does the classification founded on the\nmorphology of the imagines agree with that based on the morphology of\nthe larv\u00e6 or not?\n\nIf, according to Claus,[176] we divide the order Lepidoptera into six\ngreat groups of families, it is at once seen that these groups, which\nwere originally founded exclusively on imaginal characters, cannot by\nany means be so clearly and sharply defined by the larval characters.\n\nThis is certainly the case with the _Geometr\u00e6_, of which the larv\u00e6\npossess only ten legs, and on this account progress with that peculiar\n\u201clooping\u201d movement which strikes even the uninitiated. This group,\nwhich is very small, is however the only one which can be founded on\nthe morphology of the larv\u00e6; it comprises only two nearly related\nfamilies (_Phytometrid\u00e6_ and _Dendrometrid\u00e6_), and it is not yet\ndecided whether these should not be united into one group comprising\nthe family characters of the whole of the \u201cloopers.\u201d\n\nNeither the group of Micro-lepidoptera, nor those of the _Noctuina_,\n_Bombycina_, _Sphingina_, and _Rhopalocera_, can be based\nsystematically on larval characters. Several of these groups are indeed\nbut indistinctly defined, and even the imagines present no common\ncharacteristics by which the groups can be sharply distinguished.\n\nThis is well shown by the _Rhopalocera_ or butterflies. These insects,\nin their large and generally brilliantly coloured wings, which are\nusually held erect when at rest, and in their clubbed antenn\u00e6, possess\ncharacters which are nowhere else found associated together, and\nwhich thus serve to constitute them a sharply defined group.[177]\nThe caterpillars, however, show a quite different state of affairs.\nAlthough the larval structure is so characteristic in the individual\nfamilies of butterflies, these \u201clarval-families\u201d cannot be united\ninto a larger group by any common characters, and the \u201c_Rhopalocera_\u201d\nwould never have been established if only the larv\u00e6 had been known.\nIt is true that they all have sixteen legs, that they never possess a\nSphinx-like horn, and that they are seldom hairy, as is the case with\nmany _Bombycid\u00e6_,[178] but these common _negative_ characters occur\nalso in quite distinct groups.\n\nIn the butterflies, therefore, a perfect congruence of\nform-relationship does not exist, inasmuch as the imagines constitute\none large group of higher order whilst the larv\u00e6 can only be formed\ninto families. If it be admitted that the common characters of\nbutterflies depend on their derivation from a common ancestor, the\nimagines must have retained certain common characters which enable them\nto be recognized as allies, whilst the larv\u00e6 have preserved no such\ncharacters from the period at which the families diverged.\n\nWithout going at present into the causes of these phenomena I will pass\non to the consideration of further facts, and will now proceed to\ninvestigate both the form-relationships within the families. Here there\ncan be no doubt that in an overwhelmingly large majority of cases the\nphyletic development has proceeded with very close parallelism in both\nstages; larval and imaginal families agree almost completely.\n\nThus, under the group _Rhopalocera_ there is a series of families\nwhich equally well permit of their being founded on the structure of\nthe larva or on that of the imago, and in which the larv\u00e6 and imagines\ntherefore deviate from one another to the same extent. This is the\ncase, for instance, with the families of the _Pierid\u00e6_, _Papilionid\u00e6_,\n_Danaid\u00e6_, and _Lyc\u00e6nid\u00e6_.\n\nBut there are also families of which the limits would be very different\nif the larv\u00e6 were made the basis of the classification instead of the\nbutterflies as heretofore. To this category belongs the sub-family\n_Nymphalin\u00e6_. Here also a very characteristic form of caterpillar\nindeed prevails, but it does not occur in all the genera, being\nreplaced in some by a quite different form of larva.\n\nIn the latest catalogue of Diurnal Lepidoptera, that of Kirby (1871),\n112 genera are comprised under this family. Of these most of the larv\u00e6\npossess one or several rows of spines on most or on all the segments, a\ncharacter which, as thus disposed, is not met with in any other family.\n\nThis character is noticeable in genera 1 to 90, if, from those\ngenera of which the larv\u00e6 are known, we may draw a conclusion with\nreference to their allies. I am acquainted with larv\u00e6 of genus 2,\n_Agraulis_, Boisd. (_Dione_, H\u00fcbn.); of genus 3, _Cethosia_, Fabr.; 10,\n_Atella_, Doubl.; 12, _Argynnis_, Fabr.; 13, _Melit\u00e6a_,[179] Fabr.;\n19, _Araschnia_, H\u00fcbn.; 22, _Vanessa_, Fabr.; 23, _Pyrameis_, H\u00fcbn.;\n24, _Junonia_, H\u00fcbn.; 31, _Ergolis_, Boisd.; 65, _Hypolimnas_, H\u00fcbn.\n(_Diadema_, Boisd.); 77, _Limenitis_, Fabr.; 81, _Neptis_, Fabr.; 82,\n_Athyma_, Westw.; and finally with those of genus 90, _Euthalia_,\nH\u00fcbn.--which, according to Horsfield\u2019s figures, possess only two rows\nof spines, these being remarkably long and curved, and fringing both\nsides. It may be safely assumed that the intermediate genera would\nagree in possessing this important character of the Nymphalideous\nlarv\u00e6, viz., spines.\n\nAfter the genus 90 there are 22 more genera, and these are spineless,\nat least in the case of the two chief genera, 93, _Apatura_, and\n104, _Nymphalis_. Of the remainder I know neither figures nor\ndescriptions.[180] In the two genera named the larv\u00e6 are provided with\ntwo or more spine-like tentacles on the head, and the last segment\nends in a fork-like process directed backwards. The body is otherwise\nsmooth, and differs also in form from that of the larv\u00e6 of the other\n_Nymphalin\u00e6_, being thickest in the middle, and tapering anteriorly and\nposteriorly; neither is the form cylindrical, but somewhat flattened\nand slug-shaped. If therefore we were to arrange these butterflies by\nthe larv\u00e6 instead of by the imagines, these two genera and their allies\nwould form a distinct family, and could not remain associated with the\n90 other Nymphalideous genera.\n\nWe have here a case of _incongruence_; the imagines of the genera 1-90\nand 91-112 are more closely allied than their larv\u00e6.\n\nFrom still another side there arises a similar disagreement. The larv\u00e6\nof the genera _Apatura_ and _Nymphalis_ agree very closely in their\nbodily form and in their forked caudal appendage with the caterpillars\nof another sub-family of butterflies, the _Satyrin\u00e6_, whilst their\nimagines differ chiefly from those of the latter sub-family in the\nabsence of an enlargement of certain veins of the fore-wings, an\nessential character of the _Satyrin\u00e6_.\n\nThis double disagreement has also been noticed by those systematists\nwho have taken the form of the caterpillar into consideration.\nThus, Morris[181] attempted to incorporate the genera _Apatura_ and\n_Nymphalis_ into the family _Libytheid\u00e6_, placing the latter as\ntransitional from the _Nymphalid\u00e6_ to the _Satyrid\u00e6_. But although the\nimagines of the genera _Apatura_, _Nymphalis_, and _Libythea_ may be\nmost closely related--as I believe they actually are--the larv\u00e6 are\nwidely different, being at least as different as are those of _Apatura_\nand _Nymphalis_ from the remaining _Nymphalin\u00e6_.\n\nNow if we could safely raise _Apatura_ and _Nymphalis_ into a distinct\nfamily--an arrangement which in the estimation of Staudinger[182] is\ncorrect--and if this were interpolated between the _Satyrid\u00e6_ and\n_Nymphalid\u00e6_, such an arrangement could only be based on the larval\nstructure, and that of the imagines would thus remain unconsidered,\nsince no other common characters can be found for these two genera than\nthose which they possess in common with the other Nymphalideous genera.\n\nThe emperor-butterflies (_Apatura_), by the ocelli of their fore-wings\ncertainly put us somewhat in mind of the _Satyrin\u00e6_, in which such\nspots are always present; but this character does not occur in the\ngenus _Nymphalis_, and is likewise absent in most of the other genera\nof this group. The genus _Apatura_ shows in addition a most striking\nsimilarity in the markings of the wings to the purely Nymphalideous\ngenus _Limenitis_, and it is therefore placed, by those systematists\nwho leave this genus in the same family, in the closest proximity to\n_Limenitis_. This resemblance cannot depend upon mimicry, since not\nonly one or another but _all_ the species of the two genera possess a\nsimilar marking; and further, because similarity of marking alone does\nnot constitute mimicry, but a resemblance in colour must also be added.\nThe genus _Limenitis_ actually contains a case of imitation, but in\nquite another direction; this will be treated of subsequently.\n\nIt cannot therefore be well denied that in this case the larv\u00e6 show\ndifferent relationships to the imagines.\n\nIf the \u201cnatural\u201d system is the expression of the genetic relationship\nof living forms, the question arises in this and in similar cases\nas to whether the more credence is to be attached to the larv\u00e6\nor to the imagines--or, in more scientific phraseology, which\nof the two inherited classes of characters have been the most\ndistinctly and completely preserved, and which of these, through its\nform-relationship, admits of the most distinct recognition of the\nblood-relationship, or, inversely, which has diverged the most widely\nfrom the ancestral form? The decision in single instances cannot but be\ndifficult, and appears indeed at first sight impossible; nevertheless\nthis will be arrived at in most cases as soon as the ontogeny of the\nlarv\u00e6, and therewith a portion of the phylogeny of this stage, can be\naccurately ascertained.\n\nAs in the _Rhopalocera_ most of the families show a complete congruence\nin the form-relationship of the caterpillars and perfect insects, so\na similar congruence is also found in the majority of the families\nbelonging to other groups. Thus, the two allied families of the group\n_Sphingina_ can also be very well characterized by their larv\u00e6;[183]\nboth the _Sphingid\u00e6_ and the _Sesiid\u00e6_ possess throughout a\ncharacteristic form of larva.\n\nOf the group _Bombycina_ the family of the _Saturniid\u00e6_ possess\nthick cylindrical caterpillars, of which the segments are beset with\na certain number of knob-like warts. It is true that two genera of\nthis family (_Endromis_ and _Aglia_) are without these characteristic\nwarts, but the imagines of these genera also show extensive and common\ndifferences from those of the other genera. A distinct family has in\nfact already been based on these genera (_Endromid\u00e6_, Boisd.). Thus the\ncongruence is not thereby disturbed.\n\nSo also the families _Liparid\u00e6_, _Euprepiid\u00e6_, and _Lithosiid\u00e6_ appear\nsharply defined in both forms; and similar families occur likewise\nunder the _Noctuina_, although in this group the erection of families\npresents great difficulties owing to the near relationship of the\ngenera, and is always to some extent arbitrary. It is important,\nhowever, that it is precisely the transitional families which present\nintermediate forms both as larv\u00e6 and as imagines.\n\nSuch an instance is offered by the _Acronyctid\u00e6_, a family belonging\nto the group _Noctuina_. The imagines here show in certain points an\napproximation to the group _Bombycina_; and their larv\u00e6, which are\nthickly covered with hairs, likewise possess the characteristics of\nmany of the caterpillars of this group.[184]\n\nA second illustration is furnished by the family _Ophiusid\u00e6_, which is\nstill placed by all systematists under the _Noctuina_, its affinity to\nthe _Geometrina_, however, being represented by its being located at\nthe end of the _Noctuina_. The broad wings and narrow bodies of these\nmoths remind us in fact of the appearance of the \u201cgeometers;\u201d and the\nlarv\u00e6, like the imagines, show a striking resemblance to those of the\n_Geometrina_ in the absence of the anterior abdominal legs. For this\nreason H\u00fcbner in his work on caterpillars has termed the species of\nthis family \u201c_Semi-Geometr\u00e6_.\u201d\n\nAll these cases show a complete congruence in the two kinds of\nform-relationship; but exceptions are not wanting. Thus, the family\n_Bombycid\u00e6_ would certainly never have been formed if the larval\nstructure only had been taken into consideration, since, whilst the\ngenera _Gastropacha_, _Clisiocampa_, _Lasiocampa_, _Odonestis_, and\ntheir allies, are thickly covered with short silky hairs disposed in\na very characteristic manner, the caterpillars of the genus _Bombyx_,\nto which the common silkworm, _B. Mori_, belongs, are quite naked and\nsimilar to many Sphinx-caterpillars (_Ch\u00e6rocampa_). Are the imagines of\nthe genera united under this family, at any rate morphologically, as\nunequally related as their larv\u00e6? Whether it is correct to combine them\ninto one family is a question that does not belong here; we are now\nonly concerned with the fact that the two stages are related in form in\nvery different degrees.\n\nAn especially striking case of incongruence is offered by the family\n_Notodontid\u00e6_, under which Boisduval, depending only on imaginal\ncharacters, united genera of which the larv\u00e6 differed to a very great\nextent. In O. Wilde\u2019s work on caterpillars this family is on this\naccount quite correctly characterized as follows:--\u201cLarv\u00e6 of various\nforms, naked or with thin hairs, sixteen or fourteen legs.\u201d[185]\nIn fact in the whole order Lepidoptera there can scarcely be found\nassociated together such diverse larv\u00e6 as are here placed in one\nimago-family; on one side the short cylindrical caterpillars of\nthe genus _Cnethocampa_, Steph. (_C. Processionea_, _Pithyocampa_,\n&c.), which are covered with fine, brittle, hooked hairs, and are\nvery similar to the larv\u00e6 of _Gastropacha_ with which they were\nformerly united; and on the other side there are the naked, humped,\nand flat-headed larv\u00e6 of the genus _Harpyia_, Ochs., with their two\nlong forked appendages replacing the hindmost pair of legs, and the\ngrotesquely formed caterpillars of the genera _Stauropus_, Germ.,\n_Hybocampa_, Linn., and _Notodonta_, Ochs.\n\nThe morphological congruence between larv\u00e6 and imagines declares itself\nmost sharply in genera, where it is the rule almost without exception.\nIn this case we can indeed be sure that a genus or sub-genus founded\non the imagines only will, in accordance with correct principles,\npresent a corresponding difference in the larv\u00e6. Had the latter been\nknown first we should have been led to construct the same genera as\nthose which are now established on the structure of the imagines, and\nthese, through other circumstances, would have stood in the same degree\nof morphological relationship as the genera founded on the imagines.\nThere is therefore a congruence in a double sense; in the first place\nthe differences between the larv\u00e6 and imagines of any two genera are\nequally great, and, in the next place, the common characters possessed\nby these two stages combined cause them to form precisely the same\ngroups defined with equal sharpness; the genera coincide completely.\n\nSo also the butterflies of the sub-family _Nymphalin\u00e6_ can well be\nseparated into genera by the characters of the larv\u00e6, and these, as\nfar as I am able to judge, would agree with the genera founded on the\nimagines.\n\nThe genus _Melit\u00e6a_, for example, can be characterized by the\npossession of 7-9 fleshy tubercles bearing hairy spines; the genus\n_Argynnis_ may be distinguished by always having six hairy unbranched\nspines on each segment, and the genus _Cethosia_ by two similar spines\non each segment; the genus _Vanessa_ shows sometimes as many as\nseven branched spines; and the genus _Limenitis_ never more than two\nbranched blunt spines on each segment, and so forth. If we go further\ninto details it will be seen that the most closely related imagines,\nas might indeed have been expected, likewise possess the most nearly\nallied larv\u00e6, whilst very small differences between the imagines are\nalso generally represented by corresponding differences in the larv\u00e6.\nThus, for instance, the genus _Vanessa_ of Fabricius has been divided\ninto several genera by later authors. Of these sub-genera, _Grapta_,\nDoubl. (containing the European _C.-album_, the American _Fabricii_,\n_Interrogationis_, _Faunus_, _Comma_, &c.), is distinguished by the\nfact that the larv\u00e6 not only possess branched spines on all the\nsegments with the exception of the prothorax, but these spines are\nalso present on the head; in the genus _Vanessa_ (_sens\u00fb strictiori_),\nDoubl., the head and prothorax are spineless (_e.g._ _V. Urtic\u00e6_); in\nthe tropical genus _Junonia_, H\u00fcbn., which was also formerly (Godart,\n1819[186]) united with _Vanessa_, the larv\u00e6 bear branched spines on all\nthe segments, the head and prothorax included.\n\nIt is possible to go still further and to separate two species of\n_Vanessa_ as two new genera, although they have hitherto been preserved\nfrom this fate even by the systematists most given to \u201csplitting.\u201d\nThis decision is certainly justifiable, simply because these species\nat present stand quite alone, and the practical necessity of forming a\ndistinct genus does not make itself felt, and this practical necessity\nmoreover frequently comes into conflict with scientific claims: science\nerects a new genus based on the amount of morphological difference, it\nbeing quite immaterial whether one or many species make up this genus;\nsuch an excessive subdivision is, however, a hindrance to practical\nrequirements, as the cumbrous array of names thereby becomes still\nfurther augmented.\n\nThe two species which I might separate from _Vanessa_ on the ground of\ntheir greater divergence, are the very common and widely distributed\n_V. Io_ and _Antiopa_, the Peacock Butterfly and the Camberwell Beauty.\nIn the very remarkable pattern of their wings, both show most marked\ncharacteristics; _Io_ possesses a large ocellus on each wing, and\n_Antiopa_ has a broad light yellow border which is not found in any\nother species of _Vanessa_. There can be no doubt but that each of\nthese would have been long ago raised into a genus if similarly marked\nspecies of _Vanessa_ occurred in other parts of the world, as is the\ncase with the other species of the genus. Thus, it is well known that\nthere is a whole series of species resembling our _V. Cardui_, and\nanother series resembling our _V. C.-album_, the two series possessing\nthe same respective types of marking; indeed on these grounds the\nsub-genera _Pyrameis_ and _Grapta_ have been erected.[187]\n\nI should not have considered it worth while to have made these remarks\nif it had not been for the fact that the caterpillars of _V. Io_ and\n_V. Antiopa_ differ in small particulars from one another and from the\nother species of the genus. These differences relate to the number and\nposition of the spines, as can be seen from the following table:--\n\nSPECIES OF THE GENUS VANESSA, FABR.\n\n  +-------------------+-----------------------------------------------+\n  |                   |   Number of Spines on the head and segments   |\n  |                   |                 of the larva.                 |\n  |                   +-----+-----+-----+-----+-----+-----+-----+-----+\n  |                   |Head.|Segm.|Segm.|Segm.|Segm.|Segm.|Segm.|Segm.|\n  |                   |     |  I. | II. | III.| IV. |  V. |VI.- | XII.|\n  |                   |     |     |     |     |     |     | XI. |     |\n  |                   +-----+-----+-----+-----+-----+-----+-----+-----+\n  |V. Io              |  0  |  0  |  2  |  2  |  4  |  6  |  6  |  4  |\n  |V. Antiopa         |  0  |  0  |  4  |  4  |  6  |  6  |  7  |  4  |\n  |V. Urtic\u00e6          |  0  |  0  |  4  |  4  |  7  |  7  |  7  |  4  |\n  |V. Polychloros     |  0  |  0  |  4  |  4  |  7  |  7  |  7  |  4  |\n  |V. Ichnusa         |  0  |  0  |  4  |  4  |  7  |  7  |  7  |  4  |\n  |V. Atalanta        |  0  |  0  |  4  |  4  |  7  |  7  |  7  |  4  |\n  |V. C.-album        |  2  |  0  |  4  |  4  |  7  |  7  |  7  |  4  |\n  |V. Interrogationis |  2  |  0  |  4  |  4  |  7  |  7  |  7  |  4  |\n  |V. Levana          |  2  |  0  |  4  |  4  |  7  |  7  |  7  |  4  |\n  +-------------------+-----+-----+-----+-----+-----+-----+-----+-----+\n\nThis character of the number of spines will not be considered as too\nunimportant when we observe how perfectly constant it remains in the\nnearly allied species. This is the case in the three consecutive\nforms, _Urtic\u00e6_, _Polychloros_, and _Ichnusa_. Now when we see\nthat two species which differ in their imaginal characters present\ncorrespondingly small differences in their larv\u00e6, this exact systematic\ncongruence indicates a completely parallel phyletic development.\n\nExceptions are, however, to be met with here. Thus, H\u00fcbner has united\none group of the species of _Vanessa_ into the genus _Pyrameis_ just\nmentioned, on account of certain characteristic distinctions of the\nbutterflies. I do not know, however, how this genus admits of being\ngrounded on the structure of the larv\u00e6; the latter, as appears from the\nabove table, agree exactly in the number and position of the spines\nwith the caterpillars of _Vanessa_ (_sens\u00fb strictiori_), nor can any\ncommon form of marking be detected which would enable them to be\nseparated from _Vanessa_.\n\nStill more striking is the incongruence in the genus _Araschnia_,\nH\u00fcbn. (_A. Prorsa-Levana_), which, like the genus _Pyrameis_, is\nentirely based on imaginal characters. This is distinguished from all\nthe other sub-genera of the old genus _Vanessa_ by a small difference\nin the venation of the wings (the discoidal cell of the hind-wings is\nopen instead of closed). Now it is well-known that in butterflies the\nwing-venation, as most correctly shown by Herrich-Sch\u00e4ffer, is the\nsafest criterion of \u201crelationship.\u201d It thus happens that this genus,\ntypified by the common _Levana_, is in Kirby\u2019s Catalogue separated\nfrom _Vanessa_ by two genera, and according to Herrich-Sch\u00e4ffer[188]\nby forty genera! Nevertheless, the larv\u00e6 agree so exactly in their\nspinal formula with _Grapta_ that we should have no hesitation in\nregarding them as a species of this sub-genus. It appears to me very\nprobable that in this case the form-relationship of the caterpillar\ngives more correct information as to the blood-relationship of the\nspecies than that of the imago--in any case the larv\u00e6 show a different\nform-relationship to the imagines.\n\nJust as in the case of butterflies there are many genera of _Sphingid\u00e6_\nwhich can be based on the structure of the larv\u00e6, and which agree with\nthose founded on the imagines.\n\nThus, the genus _Macroglossa_ is characterized by a straight anal horn,\na spherical head, and by a marking composed of longitudinal stripes,\nthese characters not occurring elsewhere in this combination. The\nnearly allied genus _Pterogon_, on the other hand, cannot be based\non the larv\u00e6 only, since not only is the marking of the adult larva\nvery distinct in the different species, but the anal horn is present\nin two species, whilst in a third (_P. \u0152nother\u00e6_) it is replaced by\na knob-like eye-spot. The genus _Sphinx_ (_sens\u00fb strictiori_) is\ndistinguished by the simple, curved caudal horn, the smooth, egg-shaped\nhead and smooth skin, and by a marking mainly composed of seven oblique\nstripes. The genus _Deilephila_ is distinguished from the preceding by\na dorsal plate, situated on the prothorax and interrupting the marking,\nas well as by the pattern, which here consists of a subdorsal line\nwith ring-spots more or less numerous and developed; the skin also\nis rough, \u201cshagreened,\u201d although it must be admitted that there are\nexceptions (_Vespertilio_). The genus _Ch\u00e6rocampa_ admits also of being\nbased on the form-relationship of its caterpillars, although this is\ncertainly only possible by disregarding the marking and taking alone\ninto consideration the peculiar pig-like form of the larv\u00e6. The genus\n_Acherontia_, so nearly related to _Sphinx_, possesses in the doubly\ncurved caudal horn a character common to the genus (three species\nknown[189]). Finally may be mentioned the genus _Smerinthus_, of which\nthe larv\u00e6, by their anteriorly tapering form, their shagreened skin and\nalmost triangular head with the apex upwards, their simply curved anal\nhorn, and by their seven oblique stripes on each side, constitute a\ngenus as sharply defined as that formed by the moths.\n\nAlthough in all the systematic divisions hitherto treated of there are\ncases where the form-relationship of the larva does not completely\ncoincide with that of the imago, such incongruences are of far more\nfrequent occurrence in the smallest systematic group, viz. species.\n\nThe larv\u00e6 of two species have very frequently a much nearer\nform-relationship than their imagines. Thus, the caterpillars of\n_Smerinthus Ocellatus_ and _S. Populi_ are closely allied in\nstructure, marking, and colouring, whilst the moths in these two last\ncharacters and in the form of the wings are widely separated.[190]\nJudging from the larv\u00e6 we should expect to obtain two very similar\nmoths, but in fact both _Populi_ and _Ocellatus_ have many near allies,\nand these closely related species sometimes possess larv\u00e6 which differ\nmore widely than those of more distantly related species of imagines.\n\nThus, in Amur-land and North America there occur species of\n_Smerinthus_ which closely resemble our _Ocellatus_ in colour, marking,\nand form of wing, and which possess the characteristic large blue\nocellus on the hind-wings. _S. Exc\u00e6catus_ is quite correctly regarded\nas the representative American form of our _Ocellatus_, but its\ncaterpillar, instead of being leaf-green, is of a chrome-yellow, and\npossesses dark green instead of white oblique stripes, and has moreover\na number of red spots, and a red band on the head--in brief, in the\nvery characters (colour and certain of the markings) in which the\nimagines completely agree it is widely different from _Ocellatus_. It\nappears also to be covered with short bristles, judging from Abbot and\nSmith\u2019s figure.[191]\n\nJust in the same way that the species having the nearest conceivable\nform-relationship to _Ocellatus_ possesses a relatively strongly\ndiverging larva, so does the nearest form-relation of _Populi_ (imago)\noffer a parallel case. This species, which is also North American,\nlives on _Juglans Alba_. The imago of _Smerinthus Juglandis_ differs\nconsiderably from _S. Populi_ in the form of the wings, but it\nresembles the European species so closely in marking and colouring that\nno doubt can exist as to the near relationship of the two forms. The\ncaterpillar of _S. Juglandis_,[192] however, differs to a great extent\nfrom that of _Populi_ in colour--it is not possible to confound these\ntwo larv\u00e6; but those of _Populi_ and _Ocellatus_ are not only easily\nmistaken for one another, but are distinguished with difficulty even by\nexperts.\n\nIn this same family of the _Sphingid\u00e6_ cases are not wanting in which,\non the other hand, the moths are far more closely allied than the\nlarv\u00e6. This is especially striking in the genus _Deilephila_, eight\nspecies of which are allied in the imaginal state in a remarkable\ndegree, whilst the larv\u00e6 differ greatly from one another in colour,\nand to as great an extent in marking. These eight species are _D.\nNic\u00e6a_, _Euphorbi\u00e6_, _Dahlii_, _Galii_, _Livornica_, _Lineata_,\n_Zygophylli_, and _Hippopha\u00ebs_. Of these, _Nic\u00e6a_, _Euphorbi\u00e6_,\n_Dahlii_, _Zygophylli_, and _Hippopha\u00ebs_ are so much alike in their\nwhole structure, in the form of the wings, and in marking, that\nfew entomologists can correctly identify them off-hand without\ncomparison. The larv\u00e6 of these four species, however, are of very\ndifferent appearances. Those of _Euphorbi\u00e6_ and _Dahlii_ are most\nalike, both being distinguished by the possession of a double row of\nlarge ring-spots. _Zygophylli_ (see Fig. 50, Pl. VI.) possesses only\nfaint indications of ring-spots on a white subdorsal line; and in\n_Hippopha\u00ebs_ there is only an orange-red spot on the eleventh segment,\nthe entire marking consisting of a subdorsal line on which, in some\nindividuals, there are situated more or less developed ring-spots (see\nFigs. 59 and 60, Pl. VII.). If we only compare the larv\u00e6 and imagines\nof _D. Euphorbi\u00e6_ and _Hippopha\u00ebs_, we cannot but be struck with\nastonishment at the great difference of form-relationship in the two\nstages of development.\n\nIn the case of _D. Euphorbi\u00e6_ and _Nic\u00e6a_ this difference is almost\ngreater. Whilst these larv\u00e6 show great differences in colour, marking,\nand in the roughness or smoothness of the skin (compare Fig. 51, Pl.\nVI. with Figs. 43 and 44, Pl. V.), the moths cannot be distinguished\nwith certainty. As has already been stated, the imago of the rare\n_D. Nic\u00e6a_ is for this reason wanting in most collections; it cannot\nbe detected whether a specimen is genuine, _i.e._ whether it may not\nperhaps be a somewhat large example of _D. Euphorbi\u00e6_.\n\nAn especially striking instance of incongruence is offered by the\ntwo species of _Ch\u00e6rocampa_ most common with us, viz., _Elpenor_ and\n_Porcellus_, the large and small Elephant Hawk-moths. The larv\u00e6 are\nso similar, even in the smallest details of marking, that they could\nscarcely be identified with certainty were it not that one species\n(_Elpenor_) is considerably larger and possesses a less curved caudal\nhorn than the other. The moths of these two species much resemble\none another in their dull green and red colours, but differ in the\narrangement of these colours, _i.e._ in marking, and also in the form\nof their wings, to such an extent that _Porcellus_ has been referred to\nthe genus _Pergesa_[193] of Walker. If systemy, as is admitted on many\nsides, has only to indicate the morphological relationship, this author\nis not to blame--but in this case a special larval classification must\nlikewise be admitted, in a manner somewhat similar to that at present\nadopted provisionally in text-books of zoology for the Hydroid Polypes\nand inferior Medus\u00e6. This case of _Porcellus_, however, shows that\nthose are correct who maintain that systemy claims to express, although\nincompletely, the blood-relationship, and that systematists have always\nunconsciously formed their groups as though they intended to express\nthe genetic connection of the forms. Only on this supposition can it\nappear incorrect to us to thus separate two species of which the larv\u00e6\nagree so completely.\n\nI cannot conclude this review of the various systematic groups without\ntaking a glance at the groups comprised within species, viz. varieties.\nWhilst in species incongruence is of frequent occurrence, in varieties\nthis is the rule, for which reason it admits in this case of being more\nsharply defined, since we are not concerned with a double difference\nbut only with the question whether in the one stage a difference or an\nabsolute similarity is observable. By far the majority of varieties are\neither simply imaginal or merely larval varieties--only the one stage\ndiverges, the other is quite constant.\n\nThus, as has already been shown, in all the seasonally dimorphic\nbutterflies known to me the caterpillars of the two generations of\nimagines, which are often so widely different, are exactly alike; and\nthe same obtains for the majority of purely climatic varieties of\nbutterflies. Unfortunately there are as yet no connected observations\non this point. The only certain instance that I can here mention is\nthat of the Alpine and Polar form of _Pteris Napi_. This variety,\n_Bryoni\u00e6_, the female of which differs so greatly in marking and\ncolouring, possesses larv\u00e6 which cannot be distinguished from those of\nthe ordinary form of _Napi_.(See part I. appendix I. p. 124.)\n\nThat caterpillars can also vary locally without thereby affecting the\nimagines is shown by the frequently mentioned and closely investigated\ncases of di- and polymorphism in the larv\u00e6 of a number of _Sphingid\u00e6_\n(_M. Stellatarum_, _A. Atropos_, _S. Convolvuli_, _C. Elpenor_, and\n_Porcellus_, &c.). The same thing is still more clearly shown by those\ninstances in which there are not several but only one distinct larval\nform occurring in each of two different localities.\n\nTo this class belongs the above-mentioned case of _Ch\u00e6rocampa Celerio_\n(p. 197), supposing our information concerning this species to be\ncorrect; likewise the recently-mentioned case of the Ligurian variety\nof the caterpillar of _Saturnia Carpini_; and finally the case of\n_Eriogaster Lanestris_, so well known to lepidopterists. This insect\ninhabits the plains of Germany, and in the Alps extends to an elevation\nof 7000 feet, where it possesses a larva differently marked and\ncoloured (_E. Arbuscul\u00e6_) to those of the lowlands whilst the moths are\nsmaller, but do not differ in other respects from those of the plains.\n\nAmong the Alpine species many other such cases may occur, but these\ncould only be discovered by making investigations having special\nreference to this point. Of the Alpine butterflies, for example,\nnot a single species can have been reared from the caterpillar;\nfor this reason but few observations have on the whole been given\nby entomologists respecting the Alpine larv\u00e6, which are not known\nsufficiently well to enable such a question to be decided.\n\nThe investigation of the form-relationships existing between larv\u00e6 on\nthe one hand and imagines on the other has thus led to the following\nresults:--\n\nWe learn on comparison that incongruences or inequalities of\nform-relationship occur in all systematic groups from varieties to\nfamilies. These incongruences are of two kinds, in some cases being\ndisclosed by the fact that the larv\u00e6 of two systematic groups, _e.g._\ntwo species, are more closely related in form than their imagines (or\ninversely), whilst in other cases the larv\u00e6 form different systematic\ngroups to those formed by the imagines.\n\nThe results of the investigation into the occurrence of incongruences\namong the various systematic groups may be thus briefly summarised:--\n\nIncongruences appear to occur most frequently among varieties, since\nit very frequently happens that it is only the larva or only the\nimago which has diverged into a variety, the other stage remaining\nmonomorphic. The systematic division of varieties is thus very often\none-sided.\n\nAmong species also incongruences are of frequent occurrence. Sometimes\nthe imagines are much more nearly related in form than the larv\u00e6, and\nat others the reverse happens; whilst again the case appears also\nto occur in which only the one stage (larva) diverges to the extent\nof specific difference, the other stage remaining monomorphic (_D.\nEuphorbi\u00e6_ and _Nic\u00e6a_).\n\nThe agreement in form-relationship appears to be most complete in\ngenera. In the greater number of cases the larval and imaginal genera\ncoincide, not only in the sharpness of their limits, but also--as far\nas one can judge--in the weight of their distinctive characters, and\ntherefore in the amount of their divergence. Of all the systematic\ngroups, genera show the greatest congruence.\n\nIn families there is again an increase of irregularity. Although larval\nand imaginal families generally agree, there are so many exceptions\nthat the groups would be smaller if they were based exclusively\non the larval structure than if founded entirely on the imagines\n(_Nymphalid\u00e6_, _Bombycid\u00e6_).\n\nIf we turn to the groups of families we find a considerably increased\nincongruence; complete agreement is here again rather the exception,\nand it further happens in these cases that it is always the larv\u00e6\nwhich, to a certain extent, remain at a lower grade, and which form\nwell defined families; but these can seldom be associated into groups\nof a higher order having a common character, as in the case of the\nimagines (_Rhopalocera_).\n\nAfter having thus collected (so far as I am able) the facts, we have\nnow to attempt their interpretation, and from the observed congruence\nand incongruence of form-relationship of the two stages to endeavour to\ndraw a conclusion as to the underlying causes of the transformations.\n\nIt is clear at starting that all cases of incongruence can only be the\nexpression or the consequence of a phyletic development which has not\nbeen exactly parallel in the two stages of larva and imago--that one\nstage must have changed either more rapidly or more slowly than the\nother. An \u201cunequal phyletic development\u201d is thus the immediate cause of\nincongruence.\n\nThus, the occurrence of different larv\u00e6 in species of which the\nimagines have remained alike may be simply understood as cases in\nwhich the imago only has experienced a change--has taken a forward\nstep in phyletic development, whilst the larv\u00e6 have remained behind.\nIf we conceive this one-sided development to be repeated several\ntimes, there would arise two larval forms as widely different as those\nof _Deilephila Nic\u00e6a_, and _Euphorbi\u00e6_, whilst the imagines, as is\nactually the case in these species, would remain the same.\n\nThe more commonly occurring case in which one stage has a greater\nform-divergence than the other, is explicable by the one stage having\nchanged more frequently or more strongly than the other.\n\nThe explanation of the phenomena thus far lies on the surface, and it\nis scarcely possible to advance any other; but why should one stage\nbecome changed more frequently or to a greater extent than the other?\nwhy should one portion be induced to change more frequently or more\nstrongly than another? whence come these inducements to change? These\nquestions bring us to the main point of inquiry:--Are the causes which\ngive rise to these changes internal or external? Are the latter the\nresult of a phyletic vital force, or are they only due to the action of\nthe external conditions of life?\n\nAlthough an answer to this question will be found in the preceding\nessay, I will not support myself on the results there obtained, but\nwill endeavour to give another solution of the problem on fresh\ngrounds. The answer will indeed be the same as before:--A phyletic\nforce must be discountenanced, since in the first place it does not\nexplain the phenomena, and in the second place the phenomena can be\nwell explained without its assumption.\n\nThe admission of a phyletic vital force does not explain the phenomena.\nThe assumption that there is a transforming power innate in the\norganism indeed agrees quite well with the phenomenon of congruence,\nbut not with that of incongruence. Since a large number of cases of\nthe latter depend upon the fact that the larv\u00e6 are more frequently\ninfluenced by causes of change than their imagines, or _vice vers\u00e2_,\nhow can this be reconciled with such an internal force? On this\nassumption would not each stage of a species be compelled to change, if\nnot contemporaneously at least successively, with the same frequency\nand intensity, by the action of an innate force? and how by means of\nthe latter can there ever result a greater form-divergence in the larv\u00e6\nthan in the imagines?\n\nIt is delusive to believe that these unequal deviations can be\nexplained by assuming that the phyletic force acts periodically.\nGranting that it does so, and that the internal power successively\ncompels the imago, pupa, and finally the larva to change, there would\nthen pass a kind of wave of transformation over the different stages\nof the species, as was actually shown above to be the case in the\nsingle larval stages. The only possible way of explaining the unequal\ndistances between larv\u00e6 and imagines would therefore be to assume that\ntwo allied groups, _e.g._ species, were not contemporaneously affected\nby the wave, so that at a certain period of time the imago alone of one\nspecies had become changed, whilst in the other species the wave of\ntransformation had also reached the larva. In this case the imagines of\nthe two species would thus appear to be more nearly related than their\nlarv\u00e6.\n\nNow this strained explanation is eminently inapplicable to varieties,\nstill less to species, and least of all to higher systematic groups,\nfor the simple reason that every wave of transformation may be\nassumed to be at the most of such strength as to produce a deviation\nof form equal to that of a variety. Were the change resulting from\na single disturbance greater, we should not only find one-sided\nvarieties, _i.e._ those belonging to _one_ stage, but we should\nalso meet as frequently with one-sided species. If, however, a wave\nof transformation can only produce a variety even in the case of\ngreatest form-divergence, the above hypothetical uncontemporaneous\naction of such a wave in two species could only give rise to such\nsmall differences in the two stages that we could but designate them\nas varieties. An accumulation of the results of the action of several\nsuccessive waves passing over the same species could not happen,\nbecause the distance from a neighbouring species would always become\nthe same in two stages as soon as _one_ wave had ended its course. In\nthis manner there could therefore only arise divergences of the value\nof varieties, and incongruences in systematic groups of a higher rank\ncould not thus be explained.\n\nAll explanations of the second form of incongruence from the point of\nview of a phyletic force can also be shown to be absurd. How can the\nfact be explained that larval and imaginal families by no means always\ncoincide; or that the larv\u00e6 can only be formed into families whilst\nthe imagines partly form sharply defined groups of a higher order? How\ncan an internal directive force within the same organism urge in two\nquite distinct directions? If the evolution of a definite system were\ndesigned, and the admission of such a continually acting power rendered\nnecessary, why such an incomplete, uncertain, and confused performance?\n\nI must leave others to answer these questions; to me a vital force\nappears to be inadmissible, not only because we cannot understand the\nphenomena by its aid, but above all because it is superfluous for their\nexplanation. In accordance with general principles the assumption of an\nunknown force can, however, only be made when it is indispensable to\nthe comprehension of the phenomena.\n\nI believe that the phenomena can be quite well understood without any\nsuch assumption--both the phenomena of congruence and incongruence, in\ntheir two forms of unequal divergence and unequal group-formation.\n\nLet us in the first place admit that there is no directive force in\nthe organism inciting periodic change, but that every change is always\nthe consequence of external conditions, being ultimately nothing but\nthe reaction--the response of the organism to some of the influences\nproceeding from the environment; every living form would in this case\nremain constant so long as it was not compelled to change by inciting\ncauses. Such transforming factors can act directly or indirectly,\n_i.e._ they can produce new changes immediately, or can bring about\na remodelling by the combination, accumulation, or suppression\nof individual variations already present (adaptation by natural\nselection). Both forms of this action of external influences have long\nbeen shown to be in actual operation, so that no new assumption will be\nmade, but only an attempt to explain the phenomena in question by the\nsole action of these known factors of species formation.\n\nIf, in the first instance, we fix our attention upon that form of\nincongruence which manifests itself through unequal divergence of\nform-relationship, it will appear prominently that this bears precise\nrelations to the different systematic groups. This form of incongruence\nconstitutes the rule in varieties of the order Lepidoptera, it is of\nvery frequent occurrence in species, but disappears almost completely\nin genera, and entirely in the case of families and the higher groups.\nOn the whole, therefore, as we turn to more and more comprehensive\ngroups, the incongruence diminishes whilst the congruence increases,\nuntil finally the latter becomes the rule.\n\nNow if congruence presupposes an equal number of transforming\nimpulses, we perceive that the number of the impulses which have\naffected larv\u00e6 and imagines agree with one another the more closely the\nlarger the systematic groups which are compared together. How can this\nbe otherwise? The larger the systematic group the longer the period of\ntime which must have been necessary for its formation, and the more\nnumerous the transforming impulses which must have acted upon it before\nits formation was completed.\n\nBut if the supposition that the impulse to change always comes from\nthe environment in no way favours the idea that such impulses always\naffect both stages contemporaneously, and are equal in number during\nthe same period of time, there is not, on the other hand, the least\nground for assuming that throughout long periods the larv\u00e6 or the\nimagines only would have been affected by such transforming influences.\nThis could have been inferred from the fact that varieties frequently\ndepend only upon one stage, whilst specific differences in larv\u00e6 only\nalso occur occasionally, the imagines remaining alike; but no single\ngenus is known of which all the species possess similar larv\u00e6. Within\nthe period of time during which genera can be formed the transforming\nimpulses therefore never actually affect the one stage only, but always\ninfluence both.\n\nBut if this is the case--if within the period of time which is\nsufficient for the production of species, the one stage only is but\nseldom and quite exceptionally influenced by transforming impulses,\nwhilst both stages are as a rule affected, although not with the same\nfrequency, it must necessarily follow that on the whole, as the period\nof time increases, the difference in the number of these impulses which\naffect the larva and of those which affect the imago must continually\ndecrease, and with this difference the magnitude of the morphological\ndifferences resulting from the transforming influences must at the same\ntime also diminish. With the number of the successively increasing\nchanges the difference in the magnitude of the change in the two stages\nwould always relatively diminish until it had quite vanished from our\nperception; just in the same manner as we can distinguish a group of\nthree grains of corn from one composed of six, but not a heap of 103\ngrains from one containing 106 grains.\n\nThat the small systematic groups must have required a short period and\nthe large groups a long period of time for their formation requires no\nspecial proof, but results immediately from the theory of descent.\n\nAll the foregoing considerations would, however, only hold good if\nthe transforming impulses were equal in strength, or, not to speak\nfiguratively, if the changes only occurred in equivalent portions of\nthe body, _i.e._ in such portions as those in which the changes are\nof the same physiological and morphological importance to the whole\norganism.\n\nNow in the lower systematic groups this is always the case. Varieties,\nspecies, and genera are always distinguished by only relatively small\ndifferences; deep-seated distinctions do not here occur, as is implied\nin the conception of these categories. The true cause of this is, I\nbelieve, to be found in the circumstance that all changes take place\nonly by the smallest steps, so that greater differences can only arise\nin the course of longer periods of time, within which a great number of\ntypes (species) can, however, come into existence, and these would be\nrelated by blood and in form in different degrees, and would therefore\nform a systematic group of a higher rank.\n\nThe short periods necessary for the production of inferior groups, such\nas genera, would not result in incongruences if only _un_typical parts\nof the larv\u00e6, such as marking or spines, underwent change, whilst in\nthe imagines typical parts--wings and legs--became transformed. The\nchanges which could have occurred in the wings, &c., during this period\nof time would have been much too small to produce any considerable\ninfluence on the other parts of the body by correlation; and two\nspecies of which the larv\u00e6 and imagines, had changed with the same\nfrequency, would show a similar amount of divergence between the larv\u00e6\nand between the imagines, although on the one side only _un_typical\nparts--_i.e._ those of no importance to the whole organization--and\non the other side typical parts, were affected. The _number_ of the\nchanges would here alone determine whether congruence or incongruence\noccurred between the two stages.\n\nThe case would be quite different if, throughout a long period of time,\nin the one stage only typical and in the other only untypical parts\nwere subjected to change. In the first case a complete transformation\nof the whole structure would occur, since not only would the typical\nparts, such as the wings, undergo a much further and increasing\ntransformation in the same direction, but these changes would also lead\nto secondary alterations.\n\nIn this manner, I believe, must be explained the fact that in the\nhigher groups still greater form-divergences of the two stages occur;\nand if this explanation is correct, the cause of this striking\nphenomenon, viz., that incongruence diminishes from varieties to\ngenera, in which latter it occurs but exceptionally, whilst in families\nand in the higher groups it again continually increases, is likewise\nrevealed. Up to genera the incongruence depends entirely upon the one\nstage having become changed _more frequently_ than the other; but\nin families and groups of families, and in the orders Diptera and\nHymenoptera, as will be shown subsequently, in sub-orders and tribes,\nit depends upon the _importance of the part of the body_ affected by\nthe predominant change. In the latter case the number of changes is\nof no importance, because these are so numerous that the difference\nvanishes from our perception; but an equal number of changes, even\nwhen very great, may now produce a much greater or a much smaller\ntransformation in the entire bodily structure according as they affect\ntypical or untypical portions, or according as they keep in the same\ndirection throughout a long period of time, or change their direction\nfrequently.\n\nThose unequal form-divergences which occur in the higher systematic\ngroups a re always associated with a different formation of groups--the\nlarv\u00e6 form different systematic groups to the imagines, so that one\nof these stages constitutes a higher or a lower group; or else the\ngroups are of equal importance in the two stages, but are of unequal\nmagnitude--they do not coincide, but the one overlaps the other.\n\nIncongruences of this last kind appear in certain cases within families\n(_Nymphalid\u00e6_), but I will not now subject these to closer analysis,\nbecause their causes will appear more clearly when subsequently\nconsidering the orders Hymenoptera and Diptera. Incongruences of the\nfirst kind, however, admit of a clear explanation in the case of\nbutterflies. They appear most distinctly in the groups composed of\nfamilies.\n\nNobody has as yet been able to establish the group _Rhopalocera_ by\nmeans of any single character common to the larv\u00e6; nevertheless, this\ngroup in the imagines is the sharpest and best defined of the whole\norder. If we inform the merest tyro that clubbed antenn\u00e6 are the chief\ncharacter of the butterflies, he will never hesitate in assigning\none of these insects to its correct group. Such a typical character,\ncommon to all families, is, however, absent in the larv\u00e6; and it might\nbe correctly said that there were no Rhopalocerous larv\u00e6, or rather\nthat there were only larv\u00e6 of _Equites_, _Nymphales_, and _Heliconii_.\nThe larv\u00e6 of the various families can be readily separated by means\nof characteristic distinctions, and it would not be difficult for an\nadept to distinguish to this extent in single cases a Rhopalocerous\ncaterpillar as such; but these larv\u00e6 possess only _family_ characters,\nand not those of a higher order.\n\nThis incongruence partly depends upon the circumstance that the\nform-divergence between a Rhopalocerous and a Heterocerous family is\nmuch greater on the side of the imagines than on that of the larv\u00e6.\nWere there but a single family of butterflies in existence, such as\nthe _Equites_, we should be obliged to elevate this to the rank of\na sub-order on the side of the imagines, but not on that of the\nlarv\u00e6. Such cases actually occur, and an instance of this kind will be\nmentioned later in connection with the Diptera. But this alone does not\nexplain why, on the side of the imagines, a whole series of families\nshow the same amount of morphological divergence from the families\nof other groups. There are two things, therefore, which must here be\nexplained:--First, why is the form-divergence between the imagines of\nthe Rhopalocera and Heterocera greater than that between their larv\u00e6?\nand, secondly, why can the imagines of the Rhopalocera be formed into\none large group by means of common characters whilst the larv\u00e6 cannot?\n\nThe answers to both these questions can easily be given from our\npresent standpoint. As far as the first question is concerned, this\nfinds its solution in the fact that the form-divergence always\ncorresponds exactly with the divergence of function, _i.e._ with the\ndivergence in the mode of life.\n\nIf we compare a butterfly with a moth there can be no doubt that the\ndifference in the conditions of life is far greater on the side of the\nimagines than on that of the larv\u00e6. The differences in the mode of life\nof the larv\u00e6 are on the whole but very small. They are all vegetable\nfeeders, requiring large quantities of food, and can only cease\nfeeding during a short time, for which reason they never leave their\nfood-plants for long, and it is of more importance for them to remain\nfirmly attached than to be able to run rapidly. It is unnecessary for\nthem to seek long for their food, as they generally find themselves\namidst an abundance, and upon this depends the small development of\ntheir eyes and other organs of sense. On the whole caterpillars live\nunder very uniform conditions, although these may vary in manifold\ndetails.\n\nThe greatest difference in the mode of life which occurs amongst\nLepidopterous larv\u00e6 is shown by wood feeders. But even these, which\nby their constant exclusion from light, the hardness of their food,\ntheir confinement within narrow hard-walled galleries, and by the\npeculiar kind of movement necessitated by these galleries, are so\ndifferently situated in many particulars to those larv\u00e6 which live\nopenly on plants, have not experienced any general change in the\ntypical conformation of the body by adaptation to these conditions of\nlife. These larv\u00e6, which, as has already been mentioned, belong to\nthe most diverse families, are more or less colourless and flattened,\nand have very strong jaws and small feet; but in none of them do we\nfind a smaller number of segments, or any disappearance, or important\ntransformation of the typical limbs; they all without exception possess\nsixteen legs, like the other larv\u00e6 excepting the _Geometr\u00e6_.\n\nNow if even under the most widely diverging conditions of life\nadaptation of form is produced by relatively small, and to a\ncertain extent superficial, changes, we should expect less typical\ntransformations in the great majority of caterpillars which live\non the exterior of plants or in their softer parts (most of the\nMicro-lepidoptera). The great diversity in the forms of caterpillars\ndepends essentially upon a different formation of the skin and its\nunderlying portions. The skin is sometimes naked, and can then acquire\nthe most diverse colours, either protective or conspicuous, or it may\ndevelop offensive or defensive markings; in other cases it may be\ncovered with hairs which sting, or with spines which prick; certain\nof its glands may develop to an enormous size, and acquire brilliant\ncolours and the power of emitting stinking secretions (the tentacles of\nthe _Papilionid\u00e6_ and Cuspidate larv\u00e6); by the development of warts,\nangles, humps, &c., any species of caterpillar may be invested with the\nmost grotesque shape, the significance of which with respect to the\nlife of the insect is as yet in most cases by no means clear: _typical\nportions_ are not, however, essentially influenced by these manifold\nvariations. At most only the form of the individual segments of the\nbody, and with these the shape of the whole insect, become changed\n(onisciform larv\u00e6 of _Lyc\u00e6nid\u00e6_), but a segment is never suppressed,\nand even any considerable lengthening of the legs occurs but very\nseldom (_Stauropus Fagi_).[194]\n\nWe may therefore fairly assert that the structure of larv\u00e6 is on the\nwhole remarkably uniform, in consequence of the uniformity in the\nconditions of life. Notwithstanding the great variety of external\naspects, the general structure of caterpillars does not become\nchanged--it is only their outward garb which varies, sometimes in one\ndirection, and sometimes in another, and which, starting from inherited\ncharacters, becomes adapted to the various special conditions of life\nin the best possible manner.\n\nAll this is quite different in the case of the imagines, where we\nmeet with very important differences in the conditions of life. The\nbutterflies, which live under the influence of direct sunlight and a\nmuch higher temperature, and which are on the wing for a much longer\nperiod during the day, must evidently be differently equipped to the\nmoths in their motor organs (wings), degree of hairiness, and in the\ndevelopment of their eyes and other organs of sense. It is true that we\nare not at present in a condition to furnish special proofs that the\nindividual organs of butterflies are exactly adapted to a diurnal life,\nbut we may safely draw this general conclusion from the circumstance\nthat no butterfly is of nocturnal habits.[195] It cannot be stated\nin objection that there are many moths which fly by day. It certainly\nappears that no great structural change is necessary to confer upon a\nLepidopteron organized for nocturnal life the power of also flying by\nday; but this proves nothing against the view that the structure of\nthe butterflies depends upon adaptation to a diurnal life. Analogous\ncases are known to occur in many other groups of animals. Thus, the\ndecapodous Crustacea are obviously organized for an aquatic life; but\nthere are some crabs which take long journeys by land. Fish appear\nno less to be exclusively adapted to live in water; nevertheless the\n\u201cclimbing-perch\u201d (_Anabas_) can live for hours on land.\n\nIt is not the circumstance that some of the moths fly by day which is\nextraordinary and demands a special explanation, but the reverse fact\njust mentioned, that no known butterfly flies by night. We may conclude\nfrom this that the organization of the latter is not adapted to a\nnocturnal life.\n\nIf we assume[196] that the Lepidopterous family adapted to a diurnal\nlife gives rise in the course of time to a nocturnal family, there\ncan be no doubt but that the transformation of structure would be far\ngreater on the part of the imagines than on that of the larv\u00e6. The\nlatter would not remain quite unchanged--not because their imagines\nhad taken to a nocturnal life which for the larva would be quite\nimmaterial, but because this change could only occur very gradually\nin the course of a large number of generations, and during this long\nperiod the conditions of life would necessarily often change with\nrespect to the larv\u00e6. It has been shown above that within the period of\ntime necessary for the formation of a new species impulses to change\noccur on both sides; how much more numerous therefore must these be\nin the case of a group of much higher rank, for the establishment of\nwhich a considerably longer period is required. In the case assumed,\ntherefore, the larv\u00e6 would also change, but they would suffer much\nsmaller transformations than the imagines. Whilst in the latter almost\nall the typical portions of the body would undergo deep changes in\nconsequence of the entirely different conditions of life, the larv\u00e6\nwould perhaps only change in marking, hairs, bristles, or other\nexternal characters, the typical parts experiencing only unimportant\nmodifications.\n\nIn this manner it can easily be understood why the larv\u00e6 of a family\nof _Noctu\u00e6_ do not differ to a greater extent from those of a family of\nbutterflies than do the latter from some other Rhopalocerous family, or\nwhy the imagines of a Rhopalocerous and a Heterocerous family present\nmuch greater form-divergences than their larv\u00e6. At the same time is\ntherefore explained the unequal value that must be attributed to any\nsingle family of butterflies in its larv\u00e6 and in its imagines. The\nunequal form-divergences coincide exactly with the inequalities in the\nconditions of life.\n\nWhen whole families of butterflies show the same structure in their\ntypical parts (antenn\u00e6, wings, &c.), and, what is of more importance,\ncan be separated as a systematic group of a higher order (_i.e._ as a\nsection or sub-order) from the other Lepidoptera whilst their larval\nfamilies do not appear to be connected by any common character, the\ncause of this incongruence lies simply in the circumstance that the\nimagines live under some peculiar conditions which are common to them\nall, but which do not recur in other Lepidopterous groups. Their larv\u00e6\nlive in precisely the same manner as those of all the other families of\nLepidoptera--they do not differ in their mode of life from those of the\nHeterocerous families to a greater extent than they do from one another.\n\nWe therefore see here a community of form within the same compass as\nthat in which there is community in the conditions of life. In all\nbutterflies such community is found in their diurnal habits, and in\naccordance with this we find that these only, and not their larv\u00e6, can\nbe formed into a group having common characters.\n\nIn the larv\u00e6 also we only find agreement in the conditions of life\nwithin a much wider compass, viz. within the whole order. Between\nthe limits of the order Lepidoptera the conditions of life in the\ncaterpillars are, as has just been shown, on the whole very uniform,\nand the structure of the larv\u00e6 accordingly agrees almost exactly in all\nLepidopterous families in every essential, _i.e._ typical, part.\n\nIn this way is explained the hitherto incomprehensible phenomenon that\nthe sub-ordinal group _Rhopalocera_ cannot be based on the larv\u00e6, but\nthat Lepidopterous caterpillars can as a whole be associated into a\nhigher group (order); they constitute altogether families and an order,\nbut not the intermediate group of a sub-order. By this means we at the\nsame time reply to an objection that may be raised, viz. that larval\nforms cannot be formed into high systematic groups because of their\n\u201clow and undeveloped\u201d organization.\n\nTo this form of incongruence, viz. to the formation of systematic\ngroups of unequal value and magnitude, I must attach the greatest\nweight with respect to theoretical considerations. I maintain that\nthis, as I have already briefly indicated above, is wholly incompatible\nwith the admission of a phyletic force. How is it conceivable that such\na power could work in the same organism in two entirely different\ndirections--that it should in the same species lead to the constitution\nof quite different systems for the larv\u00e6 and for the imagines, or that\nit should lead only to the formation of families in the larv\u00e6 and to\nsub-orders in the imagines? If an internal force existed which had\na tendency to call into existence certain groups of animal forms of\nsuch a nature that these constituted one harmonious whole of which\nthe components bore to one another fixed morphological relationships,\nit would certainly have been an easy matter for such a power to have\ngiven to the larv\u00e6 of butterflies some small character which would\nhave distinguished them as such, and which would in some measure have\nimpressed them with the stamp of \u201c_Rhopalocera_.\u201d Of such a character\nwe find no trace however; on the contrary, everything goes to show that\nthe transformations of the organic world result entirely from external\ninfluences.\n\n\n\n\nIII.\n\nINCONGRUENCES IN OTHER ORDERS OF INSECTS.\n\n\nAlthough the order Lepidoptera is for many reasons especially\nfavourable for an investigation such as that undertaken in the previous\nsection, it will nevertheless be advantageous to inquire into the\nform-relationships of the two chief stages in some other orders of\nmetamorphic insects, and to investigate whether in these cases the\nformation of systematic groups also coincides with common conditions of\nlife.\n\n\nHYMENOPTERA.\n\nIn this order there cannot be the least doubt as to the\nform-relationship of the imagines. The characteristic combination of\nthe pro- and meso-thorax, the number and venation of the wings, and the\nmouth-organs formed for biting and licking, are found throughout the\nwhole order, and leave no doubt that the Hymenoptera are well based on\ntheir imaginal characters.\n\nBut it is quite different with the larv\u00e6. It may be boldly\nasserted that the order would never have been founded if the larv\u00e6\nonly had been known. Two distinct larval types here occur, the\none--caterpillar-like--possessing a distinct horny head provided with\nthe typical masticatory organs of insects, and a body having thirteen\nsegments, to which, in addition to a variable number of abdominal legs,\nthere are always attached three pairs of horny thoracic legs: the\nother type is maggot-shaped, without the horny head, and is entirely\ndestitute of mouth-organs, or at least of the three pairs of typical\ninsect jaws, and is also without abdominal and thoracic legs. The\nnumber of segments is extremely variable; the larv\u00e6 of the saw-flies\nhave thirteen besides the head, the maggot-shaped larv\u00e6 of bees possess\nfourteen segments altogether, and the gall-flies and ichneumons only\ntwelve or ten. We should be much mistaken also if we expected to find\nconnecting characters in the internal organs. The intestine is quite\ndifferent in the two types of larv\u00e6, the posterior opening being absent\nin the maggot-like grubs; at most only the tracheal and nervous systems\nshow a certain agreement, but this is not complete.\n\nThe order Hymenoptera, precisely speaking and conceived only\nmorphologically, exists therefore but in the imagines; in the larv\u00e6\nthere exist only the caterpillar- and maggot-formed groups. The former\nshows a great resemblance to Lepidopterous larv\u00e6, and in the absence\nof all knowledge of the further development it might be attempted to\nunite them with these into one group. The two certainly differ in\ncertain details of structure in the mouth-organs and in the number of\nsegments, abdominal legs, &c., to a sufficient extent to warrant their\nbeing considered as two sub-orders of one larval order; but they would\nin any case be regarded as much more nearly related in form than the\ncaterpillar- and maggot-like types of the Hymenopterous larv\u00e6.\n\nIs it not conceivable, however, that the imagines of the\nHymenoptera--that ichneumons and wasps may be only accidentally alike,\nand that they have in fact arisen from quite distinct ancestral forms,\nthe one having proceeded with the Lepidopterous caterpillars from one\nroot, and the other with the grub-like Dipterous larv\u00e6 from another\nroot?\n\nThis is certainly not the case; the common characters are too\ndeep-seated to allow the supposition that the resemblance is here\nonly superficial. From the structure of the imagines alone the common\norigin of all the Hymenoptera may be inferred with great probability.\nThis would be raised into a certainty if we could demonstrate the\nphyletic development of the maggot-formed out of the caterpillar-formed\nHymenopterous larv\u00e6 by means of the ontogeny of the former. From the\nbeautiful investigations of B\u00fctschli on the embryonic development of\nbees[197] we know that the embryo of the grub possesses a complete\nhead, consisting of four segments and provided with the three typical\npairs of jaws. These head segments do not subsequently become formed\ninto a true horny head, but shrivel up; whilst the jaws disappear with\nthe exception of the first pair, which are retained in the form of soft\nprocesses with small horny points. We know also that from the three\nforemost segments of the embryo the three typical pairs of legs are\ndeveloped in the form of round buds, just as they first appear in all\ninsects.[198] These rudimentary limbs undergo complete degeneration\nbefore the birth of the larva, as also do those of the whole[199] of\nthe remaining segments, which, even in this primitive condition, show a\nsmall difference to the three foremost rudimentary legs.\n\nThe grub-like larv\u00e6 of the Hymenoptera have therefore descended from\nforms which possessed a horny head with antenn\u00e6 and three pairs\nof gnathites and a 13-segmented body, of which the three foremost\nsegments were provided with legs differing somewhat from those of the\nother segments; that is to say, they have descended from larv\u00e6 which\npossessed a structure generally similar to that of the existing saw-fly\nlarv\u00e6. The common derivation of all the Hymenoptera from one source is\nthus established with certainty.[200]\n\nBut upon what does this great inequality in the form-relationship of\nthe larv\u00e6 and imagines depend? The existing maggot-like grubs are\nwithout doubt much further removed from the active caterpillar-like\nlarv\u00e6 than are the saw-flies from the Aculeate Hymenoptera. Whilst\nthese two groups differ only through various modifications of the\ntypical parts (limbs, &c.), their larv\u00e6 are separable by much\ndeeper-seated distinctions; limbs of typical importance entirely\nvanish in the one group, but in the other attain to complete\ndevelopment.\n\nIn the Hymenoptera there exists therefore a very considerable\nincongruence in the systems based morphologically, _i.e._ on the pure\nform-relationships of the larv\u00e6 and of the imagines. The reason of this\nis not difficult to find: _the conditions of life differ much less in\nthe case of the imagines than in that of the larv\u00e6_. In the former the\nconditions of life are similar in their broad features. Hymenoptera\nlive chiefly in the air and fly by day, and in their mode of obtaining\nfood do not present any considerable differences. Their larv\u00e6, on the\nother hand, live under almost diametrically opposite conditions. Those\nof the saw-flies live after the manner of caterpillars upon or in\nplants, in both cases their peculiar locomotion being adapted for the\nacquisition and their masticatory organs for the reduction of food.\nThe larv\u00e6 of the other Hymenoptera, however, do not as a rule require\nany means of locomotion for reaching nor any organs of mastication for\nswallowing their food, since they are fed in cells, like the bees and\nwasps, or grow up in plant galls of which they suck the juice, or are\nparasitic on other insects by whose blood they are nourished. We can\nreadily comprehend that in the whole of this last group the legs should\ndisappear, that the jaws should likewise vanish or should become\ndiminished to one pair retained in a much reduced condition, that the\nhorny casing of the head, the surface of attachment of the muscles of\nthe jaws, should consequently be lost, and that even the segments of\nthe head itself should become more or less shrivelled up as the organs\nof sense therein located became suppressed.\n\nThe incongruence manifests itself however in yet another manner than\nby the relatively greater morphological divergence of the larv\u00e6: a\ndifferent grouping is possible for the larv\u00e6 and for the imagines. If\nwe divide the Hymenoptera simply according to the form-relationships of\nthe imagines, the old division into the two sub-orders _Terebrantia_\nor _Ditrocha_ and _Aculeata_ or _Monotrocha_ will be the most correct.\nThe distinguishing characters of a sting or ovipositor and a one- or\ntwo-jointed trochanter are still of the greatest value. But these\ntwo sub-orders do not by any means correspond with the two types of\nlarv\u00e6 since, in the _Terebrantia_, there occur families with both\ncaterpillar-formed and maggot-formed larv\u00e6.\n\nThe cause is to be found in that a portion of these families possess\nlarv\u00e6 which are parasitic in other insects or in galls, their bodily\nstructure having by these means become transformed in a quite different\ndirection. The mode of life of the imagines is, on the other hand,\nessentially the same.\n\nWe have here therefore another case like that which we met with among\nthe Rhopalocerous Lepidoptera, in which the imagines appear to be\ncapable of being formed into a higher group than the larv\u00e6, because the\nformer live under conditions of life which are on the whole similar\nwhilst the latter live under very divergent conditions.\n\nThe old division of the Hymenoptera into two sub-orders has certainly\nbeen abandoned in the later zoological text-books; they are now divided\ninto three:--saw-flies, parasitic, and aculeate Hymenoptera; but even\nthis arrangement has been adopted with reference to the different\nstructure of the larv\u00e6. Whether this system is better than the older,\n_i.e._ whether it better expresses the _genealogical_ relationship, I\nwill not now stop to investigate.[201]\n\n\nDIPTERA.\n\nThe imagines of the Diptera (_genuina_), with the exception of the\n_Aphaniptera_ and _Pupipara_, agree in all their chief characters,\nsuch as the number and structure of the wings, the number and joints\nof the legs, the peculiar formation of the thorax (fusion of the three\nsegments);[202] and even the structure of the mouth organs varies only\nwithin narrow limits. This is in accordance with their mode of life,\nwhich is very uniform in its main features: all the true Diptera live\nin the light, moving chiefly by means of flight, but having also the\npower of running; all those which take food in the imago condition\nfeed upon fluids. Their larv\u00e6, on the other hand, are formed on two\nessentially different types, the one--which I shall designate as the\ngnat-type--possessing a horny head with eyes, three pairs of jaws,\nand long or short antenn\u00e6, together with a 12- or 13-segmented body,\nwhich is never provided with the three typical pairs of thoracic legs,\nbut frequently has the so-called abdominal legs on the first and last\nsegments. The other Dipterous larv\u00e6 are maggot-shaped and without a\nhorny head, or in fact without any head, since the first segment, the\nhomologue of the head, can in no case be distinguished through its\nbeing larger than the others; it is on the contrary much smaller. The\ntypical insect mouth-parts are entirely absent, being replaced by a\nvariously formed and quite peculiar arrangement of hooks situated on\nthe mouth and capable of protrusion. Never more than eleven segments\nare present besides the first, which is destitute of eyes; neither are\nabdominal legs ever developed.\n\nThe mode of life differs very considerably in the two groups of larv\u00e6.\nAlthough the Dipterous maggots are not as a rule quite incapable of\nlocomotion like the grubs of the Hymenoptera (bees, ichneumons), the\nmajority are nevertheless possessed of but little power of movement\nin the food-substance on which they were deposited as eggs. They do\nnot go in search of food, either because they are parasitic in other\ninsects in the same manner as the ichneumons (_Tachina_), or else they\nlive on decaying animal or vegetable substances or amidst large swarms\nof their prey, like the larv\u00e6 of the _Syrphid\u00e6_ amongst _Aphides_.\nThey generally undergo pupation in the same place as that which they\ninhabit as larv\u00e6 and indeed in their larval skin which hardens into an\noval pupa-case. Some few leave their feeding place and pupate after\ntraversing a short distance (_Eristalis_).\n\nAs in the case of the Hymenoptera the structure of the larv\u00e6 can here\nalso be explained by peculiarities in their mode of life. Creatures\nwhich live in a mass of food neither require special organs of\nlocomotion nor specially developed organs of sense (eyes). They have\nno use for the three pairs of jaws since they only feed on liquid\nsubstances, and the hooks within the mouth do not serve for the\nreduction of food but only for fastening the whole body. With the jaws\nand their muscular system there likewise disappears the necessity for a\nhard surface of attachment, _i.e._ a corneous head.\n\nThe mode of life of the larv\u00e6 of the gnat-type is quite different\nin most points. The majority, and indeed the most typically formed\nof these, have to go in search of their food, whether they are\npredaceous, such as the _Culicid\u00e6_ and many of the other _Nemocera_\n(_Corethra_, _Simulium_), or whether they feed on plants, which they\nin some cases weave into a protective dwelling tube (certain species\nof _Chironomus_). Many live in water and move with great rapidity;\nothers bury in the earth or in vegetable substances; and even those\nspecies which live on fungi sometimes wander great distances, as in\nthe well-known case of the \u201carmy worm\u201d where thousands of the larv\u00e6 of\n_Sciara Thom\u00e6_ thus migrate.\n\nNow the two types of larv\u00e6 correspond generally with the two large\ngroups into which, as it appears to me correctly, the Diptera\n(_genuina_) are as a rule divided. In this respect there is therefore\nan equality of form-relationship--the grouping is the same, and\nthe incongruence depends only upon the form-divergence between the\ntwo kinds of larv\u00e6 being greater than between the two kinds of\nimagines.[203]\n\nThat the form-divergence is greater in the larv\u00e6 than in the imagines\ncannot be doubted; that this distant form-relationship cannot,\nhowever, be referred to a very remote common origin, _i.e._ to a very\nremote blood-relationship, not only appears from the existence of\ntransition-forms between the two sub-orders, but can be demonstrated\nhere, as in the case of the Hymenoptera, by the embryonic development\nof the maggot-like larv\u00e6.\n\nSeventeen years ago I showed[204] that the grub-formed larv\u00e6 of the\n_Muscid\u00e6_ in the embryonic state possessed a well-developed head with\nantenn\u00e6 and three pairs of jaws, but that later in the course of the\nembryonic development a marked reduction and transformation of these\nparts takes place, so that finally the four head segments appear as a\nsingle small ring formed from the coalesced pairs of maxill\u00e6, whilst\nthe so-called \u201cfore-head\u201d (the first head segment), together with the\nmandibles, becomes transformed into a suctorial-head armed with hooks\nand lying within the body. At the time of writing I drew no conclusion\nfrom these facts with reference to the phyletic development of these\nlarval forms; nor did B\u00fctschli, six years later, in the precisely\nanalogous case of the larv\u00e6 of the bees. The inference is, however, so\nobvious that it is astonishing that it should not have been drawn till\nthe present time.[205]\n\nThere can be no doubt that the maggot-like larv\u00e6 of insects are not by\nany means ancient forms, but are, on the contrary, quite recent, as\nfirst pointed out by Fritz M\u00fcller,[206] and afterwards by Packard[207]\nand Brauer,[208] and as is maintained in the latest work by Paul\nMayer[209] on the phylogeny of insects.\n\nThe Dipterous maggots have evidently descended from a larval form\nwhich possessed a horny head with antenn\u00e6 and three pairs of jaws, but\nwhich had no appendages to the abdominal segments; they are therefore\nordinary Dipterous larv\u00e6 of the gnat-type which have become modified\nin a quite peculiar manner and adapted to a new mode of life, just as\nthe grubs of the Hymenoptera are larv\u00e6 of the saw-fly type, which have\nbecome similarly transformed, although by no means in the same manner.\nThe resemblance between the two types of larva is to a great extent\npurely external, and depends upon the process designated \u201cconvergence\u201d\nby Oscar Schmidt, _i.e._ upon the adaptation of heterogeneous animal\nforms to similar conditions of life. By adaptation to a life within\na mass of fluid nutriment, the caterpillar-formed larv\u00e6 of the\nHymenoptera and the _Tipula_-like larv\u00e6 of the Diptera have acquired\na similar external appearance, and many similarities in internal\nstructure, or, in brief, have attained to a considerable degree of\nform-relationship, which would certainly have tended to conceal the\nwide divergence in blood-relationship did not the embryological forms\non the one side and the imagines on the other provide us with an\nexplanation.\n\nIt is certainly of great interest that in another order of insects--the\nColeoptera--grub-formed larv\u00e6 occur quite irregularly, and their origin\ncan be here traced to precisely the same conditions of life as those\nwhich have produced the grubs of bees. I refer to the honey-devouring\nlarv\u00e6 of the _Melo\u00efd\u00e6_ (_Melo\u00eb_, _Sitaris_, _Cantharis_). The case is\nthe more instructive, inasmuch that the six-legged larval form is not\nyet relegated to the development within the egg, but is retained in the\nfirst larval stage. In the _second larval stage_ the maggot-form is\nfirst assumed, although this is certainly not so well pronounced as in\nthe Diptera or Hymenoptera, as neither the head nor the thoracic legs\nare so completely suppressed as in these orders. Nevertheless, these\nparts have made a great advance in the process of transformation.\n\nThe grub-like larv\u00e6 of the Hymenoptera and Diptera appear to me\nespecially instructive with reference to the main question of the\ncauses of transformation. The reply to the questions: what gives the\nimpetus to change? is this impetus internal or external? can scarcely\nbe given with greater clearness than here. If these larv\u00e6 have\nabandoned their ancestral form and have acquired a widely divergent\nstructure, arising not only from suppression but partly also from an\nessentially new differentiation (suctorial head of the _Muscid\u00e6_), and\nif these structural changes show a close adaptation to the existing\nconditions of life, from these considerations alone it is difficult\nto conceive how such transformations can depend upon the action of a\nphyletic force. The latter must have foreseen that at precisely this\nor that fixed period of time the ancestors of these larv\u00e6 would have\nbeen placed under conditions of life which would make it desirable\nfor them to be modified into the maggot-type. But if at the same\ntime the imagines are removed in a less degree from those of the\ncaterpillar-like larv\u00e6, this divergence being in exact relation with\nthe deviations in the conditions of life, I at least fail to see how we\ncan escape the consequence that it is the external conditions of life\nwhich produce the transformations and induce the organism to change.\nIt is to me incomprehensible how one and the same vital force can in\nthe same individual induce one stage to become transformed feebly and\nthe other stage strongly, these transformations corresponding in extent\nwith the stronger or weaker deviations in the conditions of life to\nwhich the organism is exposed in the two stages; to say nothing of the\nfact that by such unequal divergences the idea of a perfect system\n(creative thought) is completely upset.\n\nNor can the objection be raised that we are here only concerned with\ninsignificant changes--with nothing more than the arrested development\nof single organs and so forth, in brief, only with those changes which\ncan be ascribed to the action of the environment.\n\nWe are here as little concerned with a mere suppression of organs\nthrough arrested development as in the case of the Cirripedia;\nthe transformation and reconstruction of the whole body goes even\nmuch further than in these Crustacea, although not so conspicuous\nexternally. Where do we elsewhere find insects having the head inside\na cavity of the body (sectorial head of the _Muscid\u00e6_), and of\nwhich the foremost segment--the physiological representative of the\nhead--consists entirely of the coalesced antenn\u00e6 and pairs of maxill\u00e6?\n\nThe incongruences in the form-relationships are, however, exceedingly\nnumerous in the case of the Diptera, and a special treatise would\nbe necessary to discuss them thoroughly. I may here mention only one\ncase, because the inequality shows itself in this instance in a quite\nopposite sense.\n\nGerst\u00e4cker, who is certainly a competent entomologist, divides the\nDiptera into three tribes, viz. the _Diptera genuina_, the _Pupipara_,\nand the _Aphaniptera_. The latter, the fleas, possess in their divided\nthoracic segments and in their jointed labial appendages characters so\nwidely divergent from those of the true Diptera and of the _Pupipara_\nthat Latreille and the English zoologists have separated them entirely\nfrom the Diptera and have raised them into a separate order.[210] Those\nwho do not agree in this arrangement, but with Gerst\u00e4cker include the\nfleas under the Diptera, will nevertheless admit that the morphological\ndivergence between the _Aphaniptera_ and the two other tribes is far\ngreater than that which exists between the latter. Now the larv\u00e6 of the\nfleas are completely similar in structure to those of the gnat-type,\nsince they possess a corneous head with the typical mouth parts and\nantenn\u00e6 and a 13-segmented body devoid of legs. Were we only acquainted\nwith the larv\u00e6 of the fleas we should rank them with the true Diptera\nunder the sub-order _Nemocera_. On first finding such a larva we should\nexpect to see emerge from the pupa a small gnat.\n\nWhile the imagines of the _Nemocera_ and _Aphaniptera_ thus show but a\nvery remote form-relationship their larv\u00e6 are very closely allied. Can\nany one doubt that in this case it is not the larva but the imago which\nhas diverged to the greatest extent? Have not the fleas moreover become\nadapted to conditions of life widely different from those of all other\nDiptera, whilst their larv\u00e6 do not differ in this respect from many\nother Dipterous larv\u00e6?\n\nWe have here, therefore, another case of unequal phyletic development,\nwhich manifests itself in the entirely different form-relationship\nof the larv\u00e6 and the imagines. Thus in this case, as in that of the\nLepidoptera, it is sometimes the larval and at other times the imaginal\nstage which has experienced the greatest transformation, and, as in the\norder mentioned, the objection that a phyletic vital force produces\ngreater and more important differentiations in the higher imaginal\nstage than in the lower or less developed larval stage, is equally\nineffectual.\n\nIf, however, it be asked whether the unequal phyletic development\ndepends in this case upon an unequal number of transforming impulses\nwhich the two stages may have experienced during an equal period of\ntime, this must be decidedly answered in the negative. The unequal\ndevelopment obviously depends in this case, as in the higher systematic\ngroups of the Lepidoptera, upon the unequal value of the parts affected\nby the changes. These parts are on the one side of small importance,\nand on the other side of great importance, to the whole structure of\nthe insect. This is shown in the last-mentioned case of the fleas,\nwhere, of the typical parts of the body, only the wings have become\nrudimentary, whilst the antenn\u00e6, mouth-parts, and legs, and even the\nform and mode of segmentation (free thoracic segments), must have\nsuffered most important modifications; their larv\u00e6, on the other hand,\ncan have experienced only unimportant changes, since they still agree\nin all typical parts with those of the gnat-type.\n\nAlthough therefore in this and in similar cases a greater number of\ntransforming impulses may well have occurred on the one side than on\nthe other--and it is indeed highly probable that this number has not\nbeen absolutely the same--nevertheless the chief cause of the striking\nincongruence is not to be found therein, but rather in the _strength_\nof the transforming impulses, if I may be permitted to employ this\nfigure, or, more precisely expressed, in the importance of the parts\nwhich become changed and at the same time in the amount of change.\n\nIn this conclusion there is implied as it appears to me an important\ntheoretical result which tells further against the efficacy of a\nphyletic force.\n\nIf the so-called \u201ctypical parts\u201d of an animal disappear completely\nthrough the action of the environment only, and still further, if these\nparts can become so entirely modified as to give rise to quite new and\nagain typical structures (suctorial head of the _Muscid\u00e6_) without\nthe typical parts of the other stage of the same individual being\nthereby modified and transformed into a new type of structure, how can\nwe maintain a distinction between typical and non-typical parts with\nrespect to their origin? But if a difference exists with respect only\nto the physiological importance of such parts, _i.e._ their importance\nfor the equilibrium of the whole organization, while, with reference to\ntransformation and suppression, exactly the same influences appear to\nbe effective as those which bring about a change in or a disappearance\nof the so-called adventitious parts, where is there left any scope\nfor the operation of the supposed phyletic force? What right have we\nto assume that the typical structures arise by the action of a vital\nforce? Nevertheless this is the final refuge of those who are bound\nto admit that a great number of parts or characters of an animal can\nbecome changed, suppressed, or even produced by the action of the\nenvironment.\n\n\n\n\nIV.\n\nSUMMARY AND CONCLUSION.\n\n\nThe question heading the second section of this essay must at the\nconclusion of the investigation be answered in the negative. The\nform-relationship of the larv\u00e6 does not always coincide with that\nof the imagines, or, in other words, a system based entirely on the\nmorphology of the larv\u00e6 does not always coincide with that founded\nentirely on the morphology of the imagines.\n\nTwo kinds of incongruence here present themselves. The first arises\nfrom the different amount of divergence between two systematic groups\nin the larv\u00e6 and in the imagines, these groups being of equal extent.\nThe second form of incongruence consists essentially in that the\ntwo stages form systematic groups of different extents, either the\none stage constituting a group of a higher order than the other and\ntherefore forming a group of unequal value, or else the two stages form\ngroups of equal systematic value, these groups, however, not coinciding\nin extent, but the one overlapping the other.\n\nThis second form of incongruence is very frequently connected with the\nfirst kind, and is mostly the direct consequence of the latter.\n\nThe cause of the incongruences is to be found in unequal phyletic\ndevelopment, either the one stage within the same period of time having\nbeen influenced by a greater number of transforming impulses than the\nother, or else these impulses have been different in strength, _i.e._\nhave affected parts of greater or less physiological value, or have\ninfluenced parts of equal value with unequal strength.\n\nIn all these cases in which there are deep-rooted form-differences,\nit can be shown that these correspond exactly with inequalities in\nthe conditions of life, this correspondence being in two directions,\nviz. in strength and in extent: the former determines the _degree_ of\nform-difference, the latter its _extent_ throughout a larger or smaller\ngroup of species.\n\nThe different forms of incongruence are manifested in the following\nmanner:--\n\n(1.) Different amount of form-divergence between the larv\u00e6 on the one\nside and the imagines on the other. Among the Lepidoptera this is found\nmost frequently in varieties and species, and there is evidence to\nshow that in this case the one stage has been affected by transforming\ninfluences, either alone (varieties), or at any rate to a greater\nextent (species). In the last case it can be shown in many ways that\none stage (the larva) has actually remained at an older phyletic grade\n(_Deilephila_ species). Incongruences of this kind depending entirely\nupon the more frequent action of transforming impulses can only become\nobservable in the smaller systematic groups, in the larger they elude\ncomparative examination. In the higher groups unequal form-divergence\nmay be produced by the transforming impulses affecting parts of unequal\nphysiological and morphological value, or by their influencing parts\nof equal value in different degrees. All effects of this kind can,\nhowever, only become manifest after a long-continued accumulation of\nsingle changes, _i.e._ only in those systematic groups which require\na long period of time for their formation. By this means we can\ncompletely explain why the incongruences of form-divergence continually\ndiminish from varieties to genera, and then increase again from genera\nupwards through families, tribes, and sub-orders: the first diminishing\nincongruence depends upon an _unequal number_ of transforming impulses,\nthe latter increasing incongruence depends upon the _unequal power_ of\nthese impulses.\n\nCases of the second kind are found among the Lepidopterous families,\nand especially in the higher groups (_Rhopalocera_ and _Heterocera_),\nand appear still more striking in the higher groups of the Hymenoptera\nand Diptera. Thus the caterpillar shaped and maggot-formed larv\u00e6 of\nthe Hymenoptera differ from one another to a much greater extent\nthan their imagines, since the latter have experienced a complete\ntransformation of typical parts; whilst in the caterpillar-formed larv\u00e6\nthese parts vary only within moderate limits. Similarly in the case\nof the Diptera, of which the gnat-like larv\u00e6 diverge more widely from\nthose of the grub type than do the gnats from the true flies. On the\nother hand the divergence between the imagines of the fleas and gnats\nis considerably greater than that between their larv\u00e6--indeed the larv\u00e6\nof the fleas would have to be ranked as a family of the sub-order of\nthe gnat-like larv\u00e6 if we wished to carry out a larval classification.\nBy this it is also made evident that these unequal divergences, when\nthey occur in the higher systematic groups, always induce at the same\ntime the second form of incongruence--that of the formation of unequal\nsystematic groups.\n\nIn general whenever such unequal divergences occur in the higher groups\nthey run parallel with a strong deviation in the conditions of life.\nIf these differ more strongly on the side of the larv\u00e6, we find that\nthe structure of the latter likewise diverges the more widely, and that\ntheir form-relationship is in consequence made more remote (saw-flies\nand ichneumons, gnats and flies); if, on the other hand, the difference\nin the conditions of life is greater on the side of the imagines,\nwe find among the latter the greater morphological divergence\n(butterflies and moths, gnats and fleas).\n\n(2.) The second chief form of incongruence consists in the formation\nof different systematic groups by the larv\u00e6 and the imagines, if the\nlatter are grouped simply according to their form-relationship without\nreference to their genetic affinities. This incongruence again shows\nitself in two forms--in the formation of groups of unequal value, and\nthe formation of groups equal in value but unequal in extent, _i.e._ of\noverlapping instead of coinciding groups.\n\nOf these two forms the first arises as the direct result of a different\namount of divergence. Thus the larv\u00e6 of the fleas, on account of their\nsmall divergence from those of the gnats, could only lay claim to the\nrank of a _family_, whilst their imagines are separated from the gnats\nby such a wide form-divergence that they are correctly ranked as a\ndistinct _tribe_ or _sub-order_.\n\nThe inequalities in the lowest groups, varieties, can be regarded in a\nprecisely similar manner. If the larva of a species has become split up\ninto two local forms, but not the imago, each of the two larval forms\npossesses only the rank of a _variety_, whilst the imaginal form has\nthe value of a _species_.\n\nLess simple are the causes of the phenomenon that in the one stage the\nlower groups can be combined into one of higher rank, whilst the other\nstage does not attain to this high rank. Such a condition appears\nespecially complicated when the two stages can again be formed into\ngroups of a still higher rank.\n\nThis is the case in the tribe _Rhopalocera_, which is founded on the\nimagines alone, the larv\u00e6 forming only families of butterflies. Both\nstages can however be again combined into the highest systematic group\nof the Lepidoptera.\n\nIn this case also the difference in the value of the systematic groups\nformed by the two stages corresponds precisely with the difference in\nthe conditions of life. This appears very distinctly when there are\nseveral sub-groups on each side, and not when, as in the fleas, only\none family is present as a tribe on the one side and on the other as\na family. Thus in the butterflies, on the one side there are numerous\nfamilies combined into the higher rank of a sub-order (imagines),\nwhilst on the other side (larv\u00e6) a group of the same extent cannot\nbe formed. In this instance it can be distinctly shown that the\ncombination of the families into a group of a higher order, as is\npossible on the side of the imagines, corresponds exactly with the\nlimits in which the conditions of life deviate from those of other\nLepidopterous families. The group of butterflies corresponds with an\nequally large circle of uniform conditions of life, whilst a similar\nuniformity is wanting on the side of the larv\u00e6.\n\nThe second kind of unequal group formation arises from the circumstance\nthat groups of equal value can be formed from the two stages, but these\ngroups do not possess the same limits--they overlap, and only coincide\nin part.\n\nThis is most clearly seen in the order Hymenoptera, in which both larv\u00e6\nand imagines form two well-defined morphological sub-orders, but in\nsuch a manner that the one larval form not only prevails throughout the\nwhole of the one sub-order of the imagines, but also extends beyond and\nspreads over a great portion of the other imaginal sub-order.\n\nHere again the dependence of this phenomenon upon the influence\nof the environment is very distinct, since it can be demonstrated\n(by the embryology of bees) that the one form of larva--the\nmaggot-type--although the structure now diverges so widely, has\nbeen developed from the other form, and that it must have arisen by\nadaptation to certain widely divergent conditions of life.\n\nThis form of incongruence is always connected with unequal divergence\nbetween the two stages of the one systematic group--in this\ncase the _Terebrantia_. The larv\u00e6 of this imaginal group partly\npossess caterpillar-like (_Phytospheces_) and partly maggot-formed\n(_Entomospheces_) larv\u00e6, and differ from one another to a considerably\ngreater extent than the saw-flies from the ichneumons.[211] The final\ncause of the incongruence lies therefore in this case also in the fact\nthat one stage has suffered stronger changes than the other, so that\na deeper division of the group has occurred in the former than in the\nlatter.\n\nThe analogous incongruences in single families of the Lepidoptera may\nhave arisen in a similar manner, as has already been more clearly shown\nabove; only in these cases we are as yet unable to prove in detail that\nthe larval structure has become more strongly changed through special\nexternal conditions of life than that of the imagines.\n\nIn the smallest systematic group--varieties, it has been possible to\nfurnish some proof of this. The one-sided change here depends in part\nupon the _direct action_ of external influences (seasonal dimorphism,\nclimatic variation), and it can be shown that these influences\n(temperature) acted only on the one stage, and accordingly induced\nchange in this alone whilst the other stage remained unaltered.\n\nIt has now been shown--not indeed in every individual case, but for\neach of the different kinds of incongruence of form-relationship--that\nthere is an exact parallelism corresponding throughout with the\nincongruence in the conditions of life. Wherever the forms diverge\nmore widely in one stage than in the other we also find more widely\ndivergent conditions of life; wherever the morphological systemy of\none stage fails to coincide with that of the other--whether in the\nextent or in the value of the groups--the conditions of life in that\nstage also diverge, either more widely or at the same time within\nother limits; whenever a morphological group can be constructed from\none stage but not from the other, we find that this stage alone is\nsubmitted to certain common conditions of life which fail in the other\nstage.\n\nThe law that the divergence in form always corresponds exactly with\nthe divergence in the conditions of life[212] has accordingly received\nconfirmation in all cases where we have been able to pronounce\njudgment. Unequal form-divergences correspond precisely with unequal\ndivergence in the conditions of life, and community of form appears\nwithin exactly the same limits as community in the conditions of life.\n\nThese investigations may thus be concluded with the following law:--In\ntypes of similar origin, _i.e._ having the same blood-relationship,\nthe degree of morphological relationship corresponds exactly with the\ndegree of difference in the conditions of life in the two stages.\n\nWith respect to the question as to the final cause of transformation\nthis result is certainly of the greatest importance.\n\nThe interdependence of structure and function has often been insisted\nupon, but so long as this has reference only to the agreement of each\nparticular form with some special mode of life, this harmony could\nstill be regarded as the result of a directive power; but when in\nmetamorphic forms we not only see a double agreement between structure\nand function, but also that the transformation of the form occurs in\nthe two chief developmental stages in successive steps at unequal\nrates and with unequal strength and rhythm, we must--at least so it\nappears to me--abandon the idea of an inherent transforming force; and\nthis becomes the more necessary when, by means of the opposite and\nextremely simple assumption that transformations result entirely from\nthe response of the organism to the actions of the environment, all the\nphenomena--so far as our knowledge of facts at present extends--can be\nsatisfactorily explained. A power compelling transformation, _i.e._\na phyletic vital force, must be abandoned, on the double ground that\nit is incapable of explaining the phenomena (incongruence and unequal\nphyletic development), and further because it is superfluous.\n\nAgainst the latter half of this argument there can at most be raised\nbut the one objection that the phenomena of transformation are not\ncompletely represented by the cases here analysed. In so far as this\nsignifies that the whole organic world, animal and vegetable, has not\nbeen comprised within the investigation this objection is quite valid.\nThe question may be raised as to the limit to which we may venture to\nextend the results obtained from one small group of forms. I shall\nreturn to this question in the last essay.\n\nBut if by this objection it is meant that the restricted field of\nthe investigation enables us to actually analyse only a portion of\nthe occurring transformations,[213] and indeed only those cases,\nthe dependence of which upon the external conditions of life would\nbe generally admitted, I will not let pass the opportunity of once\nmore pointing out at the conclusion of the present essay that the\nincongruences shown to exist by no means depend only upon those more\nsuperficial characters the remodelling of which in accordance with\nthe external conditions of life may be most easily discerned and\nis most difficult to deny, but that in certain cases (maggot-like\nDipterous larv\u00e6) it is precisely the \u201ctypical\u201d parts which become\npartly suppressed and partly converted into an entirely new structure.\nFrom the ancient typical appendages there have here arisen new\nstructures, which again have every right to be considered as typical.\nThis transformation is not to be compared with that experienced\nby the swimming appendages of the _Nauplius_-like ancestor of an\n_Apus_ or _Branchipus_ which have become mandibulate, nor with the\ntransformation which the anterior limbs must have gone through in\nthe reptilian ancestors of birds. The changes in question (Dipterous\nlarv\u00e6) go still further and are more profound. I lay great emphasis\nupon this because we have here one of the few cases which show that\ntypical parts are quite as dependent upon the environment as untypical\nstructures, and that the former are not only able to become adapted to\nexternal conditions by small modifications--as shown in a most striking\nmanner by the transformations of the appendages in the Crustacea and\nVertebrata--but that these parts can become modelled on an entirely\nnew type which, when perfected, gives no means of divining its mode\nof origin. I may here repeat a former statement:--With reference to\nthe causes of their origination we have no grounds for drawing a\ndistinction between typical and untypical structures.\n\nIt may be mentioned in concluding that quite analogous although less\nsharply defined results are arrived at if, instead of fixing our\nattention upon the different stages of a systematic group in their\nphyletic development, we only compare the different functional parts\n(organs in the wide sense) of the organisms.\n\nA complete parallel can be drawn between the two classes of\ndevelopmental phenomena. From the very different systematic values\nattached by taxonomists to this or that organ in a group of animals,\nit may be concluded that the individual parts of an organism are to\na certain extent independent, and that each can vary independently,\nwhen affected either entirely alone or in a preponderating degree by\ntransforming impulses, without all the other parts of the organism\nlikewise suffering transformation, or at least without their becoming\nmodified in an equal degree. Did all the parts and organs in two groups\nof animals diverge from each other to the same extent, the systematic\nvalue of such parts would be perfectly equal; we should, for example,\nbe able to distinguish and characterize two genera of the family of\nmice by their kidneys, their liver, their salivary glands, or by the\nhistological structure of their hair or muscles, or even by differences\nin their myology, &c. equally as well as by their teeth, length of\ntoes, &c. It is true that such a diagnosis has yet to be attempted; but\nit may safely be predicted that it would not succeed. Judging from all\nthe facts at present before us, the individual parts--and especially\nthose connected in their physiological action, _i.e._ the system of\norgans--do not keep pace with reference to the modifications which the\nspecies undergoes in the course of time; at one period one system and\nat another period some other system of organs advances while the others\nremain behind.\n\nThis corresponds exactly with the result already deduced from the\nunparallel development of the independent ontogenetic stages. If the\ninequality in the phyletic development is more sharply pronounced in\nthis than in the last class of cases, this can be explained by the\ngreater degree of correlation which exists between the individual\nsystems of organs in any single organism as compared with that existing\nbetween the ontogenetic stages, which, although developed from one\nanother, are nevertheless almost completely independent. We should have\nexpected _\u00e0 priori_ that a strong correlation would have here existed,\nbut as a matter of fact this is not the case, or is so only in a very\nsmall degree.\n\nJust as in the stages of metamorphosis the inequality of phyletic\ndevelopment becomes the more obliterated the more distant and\ncomprehensive, or, in other words, the greater the period of existence\nof the groups which we compare, so does the unequal divergence of the\nsystems of organs become obliterated as we bring into comparison larger\nand larger systematic groups.\n\nIt is not inconceivable--although a clear proof of this is certainly\nas yet wanting--that a variety of the ancestral species would differ\nonly in one single character, such as hairiness, colour, or marking,\nand such instances would thus agree precisely with the foregoing cases\nin which only the caterpillar or the butterfly formed a variety.\nAll the more profound modifications however--such for instance as\nthose which determine the difference between two species--are never\nlimited to one character, but always affect several, this being\nexplicable by correlation, which, as Darwin has shown in the case of\ndogs, may cause modifications in the skull of those breeds having\nhanging ears in consequence of this last character alone. It must be\nadmitted however that one organ only would be originally affected by\na modifying influence. Thus, I am acquainted with two species of a\ngenus of Daphniacea which are so closely allied that they can only be\ndistinguished from one another by a close comparison of individual\ndetails. But whilst most of the external and internal organs are almost\nidentical in the two species the sperm-cells of the males differ in a\nmost striking manner, in one species resembling an Australian boomerang\nin form and in the other being spherical! An analogous instance is\nfurnished by _Daphnia Pulex_ and _D. Magna_, two species which were\nfor a long time confounded. Nearly all the parts of the body are here\nexactly alike, but the antenn\u00e6 of the males differ to a remarkable\nextent, as was first correctly shown by Leydig.\n\nSimilarly in the case of genera there may be observed an incongruence\nof such a kind that individual parts of the body may deviate to\na greater or to a less extent than the corresponding parts in an\nallied genus. If, for instance, we compare a species of the genus\nof Daphniacea, _Sida_, with a species of the nearly allied genus\n_Daphnella_, we find that all the external and internal organs are in\nsome measure dissimilar--nevertheless certain of these parts deviate to\nan especially large extent, and have without question become far more\ntransformed than the others. This is the case, for example, with the\nantenn\u00e6 and the male sexual organs. The latter, in _Daphnella_, open\nout at the sides of the posterior part of the body as long, boot-shaped\ngenerative organs, and in _Sida_ as small papill\u00e6 on the ventral side\nof this region of the body. If again we compare _Daphnella_ with the\nnearly allied genus _Latona_, it will be found that no part in the one\nis exactly similar to the corresponding part in the other genus, whilst\ncertain organs differ more widely than others. This is the case for\ninstance with the oar-like appendages which in _Latona_ are triramous,\nbut in _Daphnella_, as in almost all the other Daphniacea, only\nbiramous.\n\nIn families the estimation of the form-divergence of the systems of\norgans and parts of the body becomes difficult and uncertain: still it\nmay safely be asserted that the two Cladocerous families _Polyphemid\u00e6_\nand _Daphniid\u00e6_ differ much less from one another in the structure of\ntheir oar-like appendages than in that of their other parts, such as\nthe head, shell, legs, or abdominal segments. In systematic groups of\na still higher order, _i.e._ in orders, and still more in classes, we\nmight be inclined to consider that all the organs had become modified\nto an equally great extent. Nevertheless it cannot be conclusively\nsaid that the kidneys of a bird differ from those of a mammal to the\nsame extent as do the feathers from mammalian hair, since we cannot\nestimate the differences between quite heterogeneous things--it can\nonly be stated that both differ greatly. Here also the facts are not\nsuch as would have been expected if transformation was the result of\nan internal developmental force; no uniform modification of _all_\nparts takes place, but first one part varies (variety) and then others\n(species), and, on the whole, as the systematic divergence increases\nall parts become more and more affected by the transformation and\nall tend continually to appear changed to an equal extent. This is\nprecisely what would be expected if the transforming impulses came\nfrom the environment. An equalization of the differences caused by\ntransformation must be produced in two ways; first by correlation,\nsince nearly every primary transformation must entail one or more\nsecondary changes, and secondly because, as the period of time\nincreases, more numerous parts of the body must become influenced by\nprimary transforming factors.\n\nA tempting theme is here also offered by attempting to trace the\ninequality of phyletic development to dissimilar external influences,\nand by demonstrating that individual organs have as a rule become\nmodified in proportion to the divergence in the conditions of life by\nwhich they have been influenced, this action, during a given period of\ntime, having been more frequent in the case of one organ than in that\nof the others, or, in brief, by showing the connection between the\ncauses and effects of transformation.\n\nIt would be quite premature, however, to undertake such a labour at\npresent, since it will be long before physiology is able to account\nfor the fine distinctions shown by morphology, and further because we\nhave as yet no insight into those internal adjustments of the organism\nwhich would enable us _\u00e0 priori_ to deduce definite secondary changes\nfrom a given primary transformation. But so long as this is impossible\nwe have no means of distinguishing correlative changes from the primary\nmodifications producing them, unless they happen to arise under our\nobservation.\n\n\n\n\nAPPENDIX I.[214]\n\nADDITIONAL NOTES ON THE ONTOGENY, PHYLOGENY, &C., OF CATERPILLARS.\n\n\n_Ontogeny of the Noctua larv\u00e6._--References have already been given\nin a previous note (67, p. 166) to observations on the number of legs\nand geometer-like habits of certain _Noctua_-larv\u00e6 when newly hatched.\nThis interesting fact in the development of these insects furnishes\na most instructive application of the principle of ontogeny to the\ndetermination of the true affinities, _i.e._ the blood-relationship of\ncertain groups of Lepidoptera. While the foregoing portions of this\nwork have been in course of preparation for the press, some additional\nobservations on this subject have been published, and I may take the\npresent opportunity of pointing out their systematic bearing--not,\nindeed, with a view to settling definitively the positions of the\ngroups in question, as our knowledge is still somewhat scanty--but with\nthe object of stimulating further investigation.\n\nMr. H. T. Stainton has lately recorded the fact that the young larva\nof _Triph\u00e6na Pronuba_ is a semi-looper (Ent. Mo. Mag. vol. xvii. p.\n135); and in a recently published life-history of _Euclidia Glyphica_\n(_Ibid._ p. 210) Mr. G. T. Porritt states that this caterpillar is\na true looper when young, but becomes a semi-looper when adult. To\nthese facts Mr. R. F. Logan adds (_Ibid._ p. 237) that \u201cnearly all\nthe larv\u00e6 of the _Trifid\u00e6_ are semi-loopers when first hatched.\u201d The\n_Cymatophor\u00e6_ appear to be an exception, but Mr. Logan points out\nthat this genus is altogether aberrant, and seems to be allied to the\n_Tortricid\u00e6_. Summing up the results of these and the observations\npreviously referred to, it will be seen that this developmental\ncharacter has now been established in the case of species belonging\nto the following families of the section _Genuin\u00e6_:--_Leucaniid\u00e6_,\n_Apameid\u00e6_, _Caradrinid\u00e6_, _Noctuid\u00e6_, _Orthosiid\u00e6_, _Hadenid\u00e6_,\nand _Xylinid\u00e6_, as well as the other _Trifid\u00e6_ (excepting\n_Cymatophora_).[215] The larv\u00e6 of the _Minores_ and _Quadrifid\u00e6_ are\nas a rule semi-loopers when adult and may be true loopers when young,\nalthough further observations on this point are wanted. These facts\npoint to the conclusion that the _Noctu\u00e6_ as a whole are phyletically\nyounger than the _Geometr\u00e6_, whilst the _Genuin\u00e6_ and _Bombyciformes_\nhave further advanced in phyletic development than the _Minores_ and\n_Quadrifid\u00e6_. The last two sections are therefore the most closely\nrelated to the _Geometr\u00e6_, as correctly shown by the arrangement given\nin Stainton\u2019s \u201cManual;\u201d whilst that adopted in Doubleday\u2019s \u201cSynonymic\nList,\u201d where the _Geometr\u00e6_ precede the _Noctu\u00e6_, is most probably\nerroneous.\n\n_Additional descriptions of Sphinx-larv\u00e6._--In the foregoing essay on\n\u201cThe Origin of the Markings of Caterpillars,\u201d Dr. Weismann has paid\nspecial attention to the larv\u00e6 of the _Sphingid\u00e6_ and has utilized\nfor this purpose, in addition to his own studies of the ontogeny of\nmany European species, the figures in the chief works dealing with\nthis family published down to the time of appearance of his essay\n(1876).[216] In order to amplify this part of the subject I have added\nreferences to more recent descriptions and figures of Sphinx-larv\u00e6\npublished by Burmeister and A. G. Butler, and I have endeavoured in\nthese cases to refer the caterpillars as far as possible to their\ncorrect position in the respective groups founded on the ontogeny\nand phylogeny of their allies. It is, however, obvious that for the\npurposes of this work figures or descriptions of adult larv\u00e6 are of but\nlittle value, except for the comparative morphology of the markings;\nand even this branch of the subject only becomes of true biological\nimportance when viewed in the light of ontogeny. As our knowledge of\nthe latter still remains most incomplete in the case of exotic species,\nit would be at present premature to attempt to draw up any genealogy\nof the whole family, and I will here only extend the subject by adding\nsome few descriptions of species which are interesting as having been\nmade from the observations of field-naturalists, and which contain\nremarks on the natural history of the insects.\n\nMr. C. V. Riley in his \u201cSecond Annual Report on the Noxious,\nBeneficial, and other Insects of the State of Missouri, 1870,\u201d gives\nfigures and describes the early stages and adult forms of certain\ngrape-vine feeding larv\u00e6 of the sub-family _Ch\u00e6rocampin\u00e6_. The\nfull-grown larva of _Philampelus Achemon_, Drury, \u201cmeasures about 3\u00bd\ninches when crawling, which operation is effected by a series of sudden\njerks. The third segment is the largest, the second but half its size,\nand the first still smaller, and when at rest the two last-mentioned\nsegments are partly withdrawn into the third.... The young larva is\ngreen, with a long slender reddish horn rising from the eleventh\nsegment and curving over the back.\u201d Mr. Riley then states that full\ngrown specimens are sometimes found as green as the younger ones, but\n\u201cthey more generally assume a pale straw or reddish-brown colour, and\nthe long recurved horn is invariably replaced by a highly polished\nlenticular tubercle.\u201d The specimen figured was the pale straw variety,\nthis colour deepening at the sides, and finally merging into a rich\nbrown. The markings appear to consist of an interrupted brown dorsal\nline, a continuous subdorsal line of the same colour, and six oblique\nscalloped white bars along the side. Whether the colour and marking\nis adapted to the vine, as is the case with the two varieties of the\ndimorphic _Ch\u00e6rocampa Capensis_ (_q.v._), is not stated. The larva of\n_Philampelus Satellitia_, Linn., when newly hatched, and for some time\nafterwards is \u201cgreen with a tinge of pink along the sides, and with an\nimmensely long straight pink horn at the tail. This horn soon begins\nto shorten, and finally curls round like a dog\u2019s tail.\u201d The colour\nof the insect changes to a reddish-brown as it grows older, and the\ncaudal horn is entirely lost at the third moult. The chief markings\nappear to be five oblique cream-yellow patches with a black annulation\non segments 6-10, and a pale subdorsal line. The caterpillar crawls\nby a series of sudden jerks, and often flings its \u201chead savagely from\nside to side when alarmed.\u201d \u201cWhen at rest, it draws back the fore\npart of the body and retracts the head and first two joints into the\nthird.\u201d Two points in connection with these species are of interest\nwith respect to the present investigations. The green colour and the\npossession of a long caudal horn when young shows that these larv\u00e6,\nlike those of _Ch\u00e6rocampa Elpenor_ (p. 178), _C. Porcellus_ (p. 184),\nand _Philampelus Labrusc\u00e6_ (p. 195, note), are descended from ancestors\nwhich possessed these characters in the adult state.[217] The next\npoint of interest is the attitude of alarm assumed by these larv\u00e6,\nand effected by withdrawing the head and two front segments into the\nthird.[218] The importance of this in connection with the similar\nhabit of ocellated species will be seen on reading the remarks on page\n367 bearing upon the initial stages of eye-spots. The other species\nfigured by Mr. Riley are _Ch\u00e6rocampa Pampinatrix_, Smith and Abbot, and\n_Thyreus Abboti_, Swains. The latter has already been referred to (p.\n256).\n\nIn a paper \u201cOn a Collection of Lepidoptera from Candahar\u201d (Proc. Zoo.\nSoc., May 4th, 1880), Mr. A. G. Butler has described and figured, from\nmaterials furnished to him by Major Howland Roberts, the larv\u00e6 of\nthree species of _Sphingid\u00e6_. _Ch\u00e6rocampa Cretica_, Boisd., feeds on\nvine; out of 100 specimens examined, there was not one black variety,\nwhile in another closely allied species, found at Jutogh and Kashmir,\nthe larva is stated to be as often black as green. The general colour\nof the caterpillar harmonizes with that of the underside of the vine\nleaves; it possesses a thread-like dorsal, and a pale yellow subdorsal\nline; also \u201ca subdorsal row of eye-spots, each consisting of a green\npatch in a yellow oval, the first spot on the fifth segment being the\nlargest and most distinct, those on each following segment becoming\nsmaller, more flattened, and less distinct, till lost on the twelfth\nsegment, sometimes becoming indistinct after the seventh or eighth\nsegment; these spots are only distinct as eye-spots on the fifth and\nsixth segments, that on the sixth being flatter than that on the fifth,\nthose on the remaining segments appearing like dashes while the larv\u00e6\nis green, but more like eyes on its changing colour when full fed.\u201d\nThe change here alluded to is the dark-brown coloration so generally\nassumed by green Sphinx-larv\u00e6 previous to pupation, and which, as I\nhave stated elsewhere (Proc. Zoo. Soc., 1873, p. 155), is probably an\nadaptation advantageous to such larv\u00e6 when crawling over the ground\nin search of a suitable place of concealment. Making the necessary\ncorrection for the different mode of counting the segments, it will be\nseen that the primary ocelli of this species are in the same position\nas those of the other species of this genus as described in a previous\npart of this essay, and that it belongs to the second phyletic group\ntreated of at p. 193. The interesting fact that this species does not\ndisplay dimorphism, whilst the closely allied form from Kashmir is\ndimorphic, shows that in the present species the process of double\nadaptation has not taken place; and this will probably be found to be\nconnected with the habits of life, _i.e._ the insect being well adapted\nto the colour of its food-plant may not conceal itself on the ground\nby day. The caterpillar of _Deilephila Robertsi_, Butl., is found at\nCandahar on a species of _Euphorbia_ growing on the rocky hills, and\nis so abundant that at the end of May every plant with any leaves left\non it had several larv\u00e6 feeding upon it. \u201cThe larv\u00e6 are very beautiful\nand conspicuous, and are very different in colouring according to their\ndifferent stages of growth.\u201d The general colour is black with white\ndots and spots; a subdorsal row of large roundish spots, one on each\nsegment, either white, yellow, orange or red; dorsal stripe variable\nin colour, and sometimes only partially present or altogether absent.\n\u201cAt the end of May most of the larv\u00e6 found presented a different\nappearance; the black disappears more or less, and with it many of\nthe small white spots. In some cases the black only remains as a ring\nround the larger white spots; the ground-colour therefore becomes\nyellowish-green or yellow, varying very considerably.\u201d The larva does\nnot change colour previous to pupation. This species, according to\nthe outline figure given (_loc. cit._, Pl. XXXIX., Fig. 9), appears\nto belong to the first of Dr. Weismann\u2019s groups, comprising _D.\nEuphorbi\u00e6_, _D. Dahlii_ and _D. Nic\u00e6a_ (see p. 199), and is therefore\nin the seventh phyletic stage of development (p. 224). From the\nrecorded habits it seems most probable that the colours and markings\nof this caterpillar are signals of distastefulness. It is much to\nbe regretted that Major Roberts has not increased the value of his\ndescription of this species by adding some observations or experiments\nbearing on this point. _Eusmerinthus Kindermanni_, Lederer, feeds on\nwillow. \u201cGeneral colour green, covered with minute white dots and seven\nlong pale yellow oblique lateral bands. (The ground-colour is the same\nas the willow-leaves on which the larva feeds, the yellow stripes\nthe same as the leaf-stalks, and the head and true legs like the\nyounger branches).\u201d As no subdorsal line is mentioned or figured, this\nspecies must be regarded as belonging to the third stage of phyletic\ndevelopment (see p. 242).\n\nI have recently had an opportunity of inspecting a large number of\ndrawings of Sphinx-larv\u00e6 in the possession of Mr. F. Moore, and of\nthose species not mentioned in the previous portions of this work the\nfollowing may be noticed:--_Ch\u00e6rocampa Theylia_, Linn., like _Ch.\nLewisii_ (note 82, p. 194), appears to be another form connecting the\nsecond and third phyletic groups of this genus. _Ch. Clotho_, Drury,\nbelongs to the third group (figured by Semper; see note 3 to this\nAppendix). The larva of _Ch. Lucasii_, Walk., offers another instance\nof the retention of the subdorsal line by an ocellated species. The\nlarva of _Ch. Lycetus_, Cram., of which Mr. Moore was so good as to\nshow me descriptions made at the various stages of growth, presents\nmany points of interest. It belongs to the third phyletic group, and\nall the ocelli appear at a very early stage. The dimorphism appears\nalso in the young larv\u00e6, some being green, and others black, a fact\nwhich may be explained by the law of \u201cbackward transference\u201d (see\np. 274). A most suggestive feature is presented by the caudal horn,\nwhich in the young caterpillar is stated to be _freely movable_.\nIt is possible that this horn, which was formerly possessed by the\nancestors of the _Sphingid\u00e6_, and which is now retained in many genera,\nis a remnant of a flagellate organ having a similar function to the\nhead-tentacles of the _Papilio-larv\u00e6_, or to the caudal appendages of\n_Dicranura_ (see p. 289).\n\n_Lophostethus Dumolinii_, Angas.--The larva of this species differs so\nremarkably from those of all other _Sphingid\u00e6_, that I have thought it\nof sufficient interest to publish the following description, kindly\nfurnished by Mr. Roland Trimen, who in answer to my application sent\nthe following notes:--\u201cMy knowledge of the very remarkable larva\nof this large and curious Smerinthine Hawk-moth is derived from a\nphotograph by the late Dr. J. E. Seaman, and from drawings and notes\nrecently furnished by Mr. W. D. Gooch. The colour is greenish-white,\ninclining to grey, and in the male there is a yellow, but in the female\na bluish, tinge in this. All the segments but the second and the head\nbear strong black spines, having a lustre of steel blue, and springing\nfrom a pale yellow tubercular base. The longest of these spines are in\ntwo dorsal rows from the fourth to the eleventh segment, the pairs on\nthe fourth and fifth segments being longer than the rest, very erect,\nand armed with short simple prickles for three-fourths of their upper\nextremity. The anal horn, which is shorter than the spines, is of the\nsame character as the latter, being covered with prickles, and much\ninclined backwards. Two lateral rows of similar shorter spines extend\nfrom the fourth to the 12th segment, and on each of the segments\n6-11 the space between the upper and lower spines is marked with a\nconspicuous pale yellow spot. Two rows of smaller similar spines extend\non each side (below the two rows of larger ones) from the second to the\nthirteenth segment, one spine of the lowermost row being on the fleshy\nbase of each pro-leg. All the pro-legs are white close to the base, and\nrusset-brown beyond. Head smooth, unarmed in adult, greenish-white with\ntwo longitudinal russet-brown stripes on face.\n\n\u201cThe young larv\u00e6 have proportionally much longer and more erect spines\nwith distinct long prickles on them. There is a short pair besides,\neither on the back of the head or on the second segment. Moreover,\nthe dorsal spines of the third and fourth segments, and the anal horn\n(which is quite erect, and the longest of all), are longer than the\nrest, and distinctly _forked_ at their extremity.\n\n\u201cMr. Gooch notes that these young larv\u00e6 might readily be mistaken for\nthose of the _Acr\u00e6\u00e6_, and suggests that this may protect them. He also\nstates that the yellow lateral spots are only noticed after the last\nmoult before pupation, and that the general resemblance of the larva as\nregards colour is to the faded leaves of its food-plant, a species of\n_Dombeia_.\u201d\n\nThe forked caudal horn in the young larva of this species is of\ninterest in connection with the similar character of this appendage in\nthe young caterpillar of _Hyloicus Pinastri_, p. 265.\n\n_Retention of the Subdorsal Line by Ocellated Larv\u00e6._--It has already\nbeen shown with reference to the eye-spots of the _Ch\u00e6rocampa_-larv\u00e6,\nthat these markings have been developed from the subdorsal line, and\nthat, in accordance with their function as a means of causing terror,\nthis line has in most species been eliminated in the course of the\nphylogeny from those segments bearing the eye-spots in order to give\nfull effect to the latter (see p. 379). In accordance with the law\nthat a character when it has become useless gradually disappears, the\nsubdorsal is more or less absent in all those species in which the\nocelli are most perfectly developed; and it can be readily imagined\nthat in cases where adaptation to the foliage exists the suppression\nof this line would under certain conditions be accelerated by natural\nselection. On the other hand, it is conceivable that the subdorsal\nline may under other conditions be of use to a protectively coloured\nocellated species by imitating some special part of the food-plant,\nunder which circumstances its retention would be secured by natural\nselection.\n\nSuch an instance is offered by _Ch\u00e6rocampa Capensis_, Linn.; and as\nthis case is particularly instructive as likewise throwing light upon\nthe retention of the subdorsal by certain species having oblique\nstripes (see p. 377, and note 166, p. 378), I will here give some details\nconcerning this species which have been communicated to me by Mr.\nRoland Trimen, the well-known curator of the South African Museum, Cape\nTown. The caterpillar of _C. Capensis_, like so many other species\nof the genus, is dimorphic, one form being a bright (rather pale)\ngreen, and the other, which is much the rarer of the two, being dull\npinkish-red. Both these forms are adapted in colour to the vine on\nwhich they feed, the red variety according to some extent with the\nfaded leaves of the cultivated vines, but to a greater extent with the\nyoung shoots and underside of the leaves of the South African native\nvine (_Cissus Capensis_), on which it also feeds. There are two\neye-spots in this species in the usual positions; they are described as\nbeing blue-grey in a white ring, and raised so as to project a little.\nThe subdorsal is white, and is bordered beneath by a wide shade of\nbluish-green irrorated with white dots, and crossed by an indistinct\nwhite oblique ray on each segment. These last markings are probably\nremnants of an oblique striping formerly possessed by the progenitor of\nthis and other species of the genus (see, for instance, Fig. 25, Pl.\nIV., one of the young stages of _C. Porcellus_). It is possible that\nthese rudimentary oblique stripes are now of service in assisting the\nadaptation of the larva to its food-plant, but this cannot be decided\nwithout seeing the insect _in situ_.\n\nThe subdorsal line extends from immediately behind the second eye-spot\nto the base of the very short and much curved violet anal horn.\nWith reference to the protective colouring Mr. Trimen writes:--\u201cThe\ndifficulty of seeing these large and beautifully-coloured larv\u00e6 on\nthe vines is quite surprising; six or more may be well within sight,\nand yet quite unnoticed. The subdorsal stripe greatly aids in their\nconcealment, as it well represents in its artificial light and shade\nthe leaf-stalks of the vine.\u201d When this larva withdraws its front\nsegments the eye-spots stand out very menacingly; but in spite of this\nit is greedily eaten by fowls and shrikes (_Fiscus Collaris_), and\nMr. Trimen also found that a tame suricate (_Rhyz\u00e6na Suricata_) and a\nlarge monitor lizard (_Regenia Albogularis_) did not refuse them. The\nfailure of the eye-spots in causing terror in these particular cases\ncannot be regarded as disproving their utility in all instances. It\nmust always be borne in mind that no protective character can possibly\nbe of service against _all_ foes; natural selection only requires that\nsuch characters should be advantageous with respect to the majority\nof the enemies of any species, and further experiments with this\ncaterpillar may show that in the case of smaller foes the eye-spots\nare effective as a means of causing alarm. The dimorphism of the larva\nof _C. Capensis_ is of special interest, although we are not yet\nsufficiently acquainted with the habits of this species to offer a\ncomplete explanation. According to Dr. Weismann\u2019s conclusions (p. 297),\nthe dimorphism of the _Ch\u00e6rocampa_-larv\u00e6 is due to a double adaptation,\nthe insects first having acquired the habit of concealing themselves\nby day, and the dark form having then been produced by the action of\nnatural selection, in order to adapt such varieties to the colour of\nthe soil, whilst others retained the green colour which adapts them\nto the foliage of their food-plants. In accordance with this, _C.\nCapensis_ may have a similar habit of concealment, or (should this be\nfound not to be the case) it is possible that this insect at a former\nperiod possessed this habit and fed upon some other plant, when it\nwould have become dimorphic in the manner explained, and the _existing_\ndimorphism may be a survival of the more ancient dimorphism, the red\nform (corresponding to the older dark form) having been subsequently\nmodified so as to become also adapted to the new food-plant. Much light\nwould be thrown upon this by studying the ontogeny of the species.\n\n_Phytophagic Variability._--A number of observations bearing on the\nphytophagic variability of the Sphinx-larv\u00e6 and other caterpillars\nhave been recorded in a previous note (p. 305), and reference has also\nbeen made to the food-plants of _Acherontia Atropos_ in South Africa\n(note 121, p. 263). I am now enabled to add some further observations\non this species, from notes furnished to me by Mr. Roland Trimen, who\nstates that for many years he has noticed that at the Cape this larva\nvaries greatly in the depth and shade of the green ground-colour, the\nvariability being in strict accordance with the colour of the leaves\nof the particular plant on which the individual feeds. The phenomenon\nwas particularly noticeable in larv\u00e6 feeding on _Buxia Grandiflora_, a\nshrub in common cultivation in gardens, and of which the foliage is of\na very dull pale greyish-green. Another striking instance was noticed\nin some very fine caterpillars feeding on a large shrubby _Solanum_,\nwhich, excepting the bright yellow bands bordering the dorsal violet\nbars, were generally dull ochreous-yellow, like the leaves and stalks\nof the _Solanum_. On plants with bright green or deep green leaves,\nthe colour of the larv\u00e6 is almost in exact agreement. Mr. Trimen\nadds:--\u201cThese remarks apply principally to the underside and pro-legs\nand lower lateral regions, the dorsal colours of violet and yellow\nvarying but little. The protection afforded is very considerable, as\nthe larv\u00e6 almost always cling to the lower side of the twigs of their\nfood-plants, so that their uniformly-coloured under-surface is upwards,\nand turned towards the light, and their variegated upper surface turned\ndownwards.\u201d\n\nThese observations are of the highest importance, not only as adding\nanother instance to the recorded cases of phytophagic variation, but\nlikewise as showing that with this variability a protective habit\nhas been acquired. It is to be hoped that such a promising field\nfor experimental investigation as is offered by this and analogous\ncases will not long remain unexplored. In attacking the problem two\nchief questions have in the first place to be settled: (1) Is the\nvariability truly phytophagic, _i.e._ are the colour variations\nactually brought about by the chemico-physiological action of the\nfood-plant? and (2) Are the larv\u00e6 at any period of growth susceptible\nto the action of phytophagic influences? The first question could be\ndecided by feeding larv\u00e6 from the same batch of eggs on different\nfood-plants from the period of their hatching. The second question\ncould be settled by changing the food-plants of a series of selected\nspecimens at various stages of growth, and observing whether any\nchange of colour was produced. In accordance with the principles\nadvocated in a previous note (p. 305), it is conceivable _\u00e0 priori_\nthat phytophagic variability may occur by direct chemico-physiological\naction, quite irrespective of any of the changes of colour being of\nprotective use. In the case of brightly-coloured distasteful species\nphytophagic variability might thus have full play, but in the case of\nprotectively-coloured edible species, phytophagic variability would be\nunder the control of natural selection. These considerations raise a\nquestion of the greatest theoretical interest in connection with this\nphenomenon. If phytophagic variability can have full play uncontrolled\nby natural selection in brightly-coloured caterpillars, ought not this\nphenomenon to be of more common occurrence in such species than in\nthose protectively coloured? Although our knowledge of this subject is\nstill very imperfect, as a matter of fact brightly coloured larv\u00e6, so\nfar as I have been able to ascertain, do not appear to be susceptible\nof phytophagic influences. But this apparent contradiction, instead\nof opposing actually confirms the foregoing views, as will appear on\nfurther consideration. The colours of protected species are as a whole\nmuch inferior in brilliancy to those of inedible species, so that any\nphytophagic effect would be more perceptible in the former than in the\nlatter, in which the highest possible standard of brilliancy appears\nin most cases to have been attained. Now phytophagic variations of\ncolour appear to be of but small amount, or, in other words, such\nvariations fluctuate within comparatively restricted limits, and as\nthe cases at present known are mostly _adaptive_ it is legitimate\nto conclude that they have been produced and brought to their\npresent standard by natural selection, _i.e._ that they have arisen\nfrom phytophagic influences as a cause of variability. The initial\nstages of phytophagic variations must therefore have been still less\nperceptible than the now perfected final results; and this leads to the\nconclusion that minute variations of this character were of sufficient\nimportance to protectively-coloured species to be taken advantage of\nby natural selection. But minute variations in a dull-coloured larva\nwould, as previously pointed out, produce a comparatively much greater\neffect than such variations in a brilliantly-coloured species; and as\nprotection is required by the former, the initial phytophagic effects\nwould be accumulated, and the power of adaptability conferred by the\ncontinued action of natural selection, whilst in vividly-coloured\nspecies where no power of adaptability is required this cause of\nvariation would not only produce a result which, as compared with its\neffects upon dull species, may be regarded as a \u201cvanishing quantity,\u201d\nbut this result would be too insignificant to be taken advantage of\nby natural selection, which is in these cases dealing only with large\n\u201cquantities,\u201d and striving to make the caterpillars as brilliant as\npossible. The fact that vividly-coloured distasteful larv\u00e6 do not show\nphytophagic variation is to my mind explained proximately by these\nconsiderations; the ultimate cause of phytophagic variability regarded\nas a chemico-physiological action requires further investigation.\n\n_Sexual Variation in Larv\u00e6._--Since most of the markings of\ncaterpillars can be explained by the two factors of adaptation and\ninheritance, or, in other words, by their present and past relations\nto the environment, and since sexual selection can have played no\ndirect part in producing these colours and markings, I feel bound to\nrecord here some few observations on the sexual differences in larv\u00e6 in\naddition to the cases of _Anap\u00e6a_ and _Orgyia_ already recorded (note\ni., p. 308) and of _Lophostethus Dumolinii_ (p. 527).\n\nMr. C. V. Riley states[219] with reference to the larva of _Thyreus\nAbboti_ that the ground-colour appears to depend upon the sex, Dr.\nMorris having described the insect as \u201creddish-brown with numerous\npatches of light green,\u201d and having expressly stated that \u201cthe _female_\nis of a uniform reddish-brown with an interrupted dark-brown dorsal\nline and transverse stri\u00e6.\u201d Mr. W. D. Gooch, who has reared the South\nAfrican butterflies _Nymphalis Cith\u00e6ron_ and _N. Brutus_ from their\nlarv\u00e6, states[220] that these \u201cdiffered sexually in both instances.\u201d\nOf _Brutus_ only a few were bred, but of _Cith\u00e6ron_ many. \u201cThe sexual\ndifference of the latter was that the females had a large dorsal\nsub-cordate cream mark, which was wanting, or only shown by a dot, in\nthe males, and the colour was more vivid in the edgings to the frontal\nhorns.\u201d\n\nAlthough such cases appear to be at present inexplicable, they are of\ninterest as examples of those \u201cresidual phenomena\u201d which, as is well\nknown, have in many branches of science so often served as important\nstarting-points for new discoveries and generalizations.[221]\n\n\n\n\nAPPENDIX II.\n\n\nThe following paper by Dr. Fritz M\u00fcller[222] forms the third of\na series of communications on Brazilian butterflies published in\n\u201cKosmos,\u201d and as it bears upon the investigations made known in the\nthird essay of the present work, I will here give a translation, by\npermission of the publisher, Herr Karl Alberts.\n\n\n\u201cACR\u00c6A AND THE MARACUJ\u00c1 BUTTERFLIES AS LARV\u00c6, PUP\u00c6, AND IMAGINES.\n\n\u201cIn a thoughtful essay on \u2018Phyletic Parallelism in Metamorphic\nSpecies,\u2019 Weismann has shown that in the case of Lepidoptera the\ndevelopmental stages of larva, pupa, and imago vary independently,\nand that a change occurring in one stage is without influence upon\nthe preceding and succeeding stages, so that the course which has\nbeen followed by the individual stages in their developmental history\nhas not been in all cases identical. This want of agreement may\nmanifest itself both by unequal divergence of form-relationship, and\nby unequal group formation. With respect to unequal form-divergence\nthe caterpillars are sometimes more closely related in form than their\nimagines, and at other times the reverse is the case. With respect\nto unequal group formation again, two cases are possible; the larv\u00e6\nand imagines may form groups of unequal value, the one stage forming\nhigher or lower groups than the other, or they may form groups of\nunequal size, _i.e._, groups which do not coincide but which overlap.\nForm-relationship and blood-relationship do not therefore always agree;\nthe resemblances among the caterpillars would lead to a quite different\narrangement to that resulting from the resemblances among the imagines,\nand it is probable that neither of these arrangements would correspond\nwith the actual relationships.\n\n\u201cStarting from this fact, which he establishes by numerous examples,\nWeismann proceeds to show most convincingly that an innate power of\ndevelopment or of transformation, such as has been assumed under\nvarious names by many adherents of the development theory, has no\nexistence, but that every modification and advancement in species has\nbeen called forth by external influences.\n\n\u201cA most beautiful illustration of the want of \u2018phyletic parallelism,\u2019\nas Weismann designates the different form-relationships of the\nlarv\u00e6, pup\u00e6, and imagines, is furnished by the five genera _Acr\u00e6a_,\n_Heliconius_, _Eueides_, _Col\u00e6nis_, and _Dione_ (= _Agraulis_). This\ninstance seems to me to be of especial value, because it offers the\nrare case of pup\u00e6 showing greater differences than the larv\u00e6 and\nimagines.\n\n\u201cThe species of which I observed the larv\u00e6 and pup\u00e6 are _Acr\u00e6a Thalia_\nand _Alalia_, _Heliconius Eucrate_, _Eueides Isabella_, _Col\u00e6nis Dido_\nand _Julia_, _Dione Vanill\u00e6_ and _Juno_; besides these I noticed the\npupa of _Eueides Aliphera_.\n\n\u201cThe following remarks apply only to these species, although we may\nsuppose with great probability that the whole of the congeneric\nforms--excepting perhaps the widely ranging species of _Acr\u00e6a_--would\ndisplay similar characters to their Brazilian representatives.\n\n\u201cThe imagines of the five genera mentioned form two sharply defined\nfamilies, the _Acr\u00e6id\u00e6_ and the butterflies of the Maracuj\u00e1 group.[223]\nThe latter comprises the three genera _Heliconius_, _Eueides_, and\n_Col\u00e6nis_, which differ only in very unimportant characters; _Eueides_\nis distinguished from _Heliconius_ by its shorter antenn\u00e6, and\n_Col\u00e6nis_ differs from _Eueides_ in having the discoidal cell of the\nhind-wings open. The genus _Dione_ is further removed by the different\nstructure of the legs, and the silvery spots on the underside of\nthe wings. Certain species resemble those of other genera in a most\nstriking manner, and much more closely both in colour and marking, and\neven in the form of their wings, than they do their own congeners. This\nis the case with _Acr\u00e6a Thalia_ and _Eueides Pavana_, with _Heliconius\nEucrate_ and _Eueides Isabella_, and with _Eueides Aliphera_ and\n_Col\u00e6nis Julia_, which are deceptively alike, and the last two are\nconnected with _Dione Juno_, at least by the upper side of the wings.\nThe difficulty of judging of the relationships of the single species is\nthus much aggravated; it cannot be said how much of this resemblance\nis to be attributed to blood-relationship, and how much to deceptive\nimitation.\n\n\u201cAs larv\u00e6 all the Brazilian species must be placed _in one genus_, as\nthey agree exactly in the number and arrangement of their spines (4\nspines, not in a transverse row, on segments 2 and 3; 6 spines, in a\ntransverse row, on segments 4-11; 4 spines, not in a transverse row,\non the last (12th) segment). They differ from one another much less\nin this respect than do the German species of _Vanessa_, such, for\ninstance, as _V. Io_ or _Antiopa_ from _V. Polychloros_, _Urtic\u00e6_,\nand _Atalanta_.[224] The larv\u00e6 of _Acr\u00e6a Thalia_ are certainly\nwithout the two spines on the head which the others possess, and, on\nthe other hand, they have a well-developed pair of spines on the\nfirst segment, which, in most of the other species, are completely\nabsent; but this does not justify their separation, since the head\nspines of _Heliconius_, _Eueides_, and _Col\u00e6nis Dido_, which are of\na considerable length, are shorter than those of the next segment in\n_Col\u00e6nis Julia_, and _Dione Vanill\u00e6_, and in _Dione Juno_ they dwindle\ndown to two minute points, this last species also bearing a short pair\non the first segment. The larva of _Dione Juno_ is thus as closely\nrelated to that of _Acr\u00e6a Thalia_ as it is to that of its congener\n_Dione Vanill\u00e6_.\n\n\u201cIf it were desired to form two distinct larval groups this could not\nbe effected on the basis of their differences in form, but could only\nbe based on their food-plants. The larv\u00e6 of _Heliconius_, _Eueides_,\n_Col\u00e6nis_, and _Dione_ live on species of Maracuj\u00e1 (Passiflora);\nthose of _Acr\u00e6a Thalia_ and _Alalia_ on Composit\u00e6 (_Mikania_ and\n_Veronia_). These larval groups would agree with those founded on the\nform-relationships of the imagines, but unlike the imaginal groups,\nwhich can be formed into families, they would scarcely possess a\ngeneric value.\n\n\u201cIf we arrange the single species of caterpillars according to their\nresemblances, this arrangement does not agree with that based on the\nresemblances of the imagines, even if we disregard the different values\nof the groups. The result is somewhat as follows:--\n\n[Illustration:\n\n                                         IMAGINES.\n                   (Nymphalideous butterflies with tufts on wing-veins.)\n                   /---------------------------------------------------\\\n (Families.)                MARACUJ\u00c1-GROUP.                     ACR\u00c6ID\u00c6.\n                    /---------------------------\\                  |\n            /-------------------\\    /-------------------\\         |\n (Genera.)  Heliconius.  Eueides.   Col\u00e6nis.         Dione.      Acr\u00e6a.\n              |                    /-------\\       /-------\\       |\n (Species.) Eucrate.  Isabella.  Dido.  Julia.  Vanill\u00e6.  Juno.  Thalia.\n              |          |         |      |        |        |      |\n              \\----------+---------/      \\--------/        \\------/\n                         \\--------------------------------------/\n                                         LARV\u00c6.\u201d\n]\n\n       *       *       *       *       *\n\n[Here follow the remarks on the habits of the larv\u00e6 in connection with\ntheir colours, &c., which have already been quoted in illustration of\nthe use of the spiny protection (note 133, p. 293). From these facts\nthe author draws the conclusion that the form-relationships of the\ncaterpillars depend rather upon their mode of life than upon their\nblood-relationships, assuming the latter to be correctly expressed by\nthe arrangement of the imagines at present adopted.]\n\n[Illustration: Figs. 1-4. Pup\u00e6 of _Acr\u00e6a Thalia_; _Heliconius Eucrate_;\n_Eueides Isabella_, and _Col\u00e6nis Dido_; life size.]\n\n\u201cA glance at the above figures of the pup\u00e6 of _Heliconius Eucrate_\n(Fig. 2), _Eueides Isabella_ (Fig. 3), and _Col\u00e6nis Dido_ (Fig. 4),\nwill show how great are the differences between these pup\u00e6 as compared\nwith the close form-relationship of all the Maracuj\u00e1 butterflies, and\nwith the no less close resemblance of their larv\u00e6. A family which\ncomprised three such dissimilar pup\u00e6 would also be capable of including\nthat of _Acr\u00e6a Thalia_ (Fig. 1).\n\n\u201cThe pupa of this last species has nothing peculiar in its general\nappearance, but possesses the ordinary pupal form; it is tolerably\nrounded, without any great elevations or depressions; a minute pointed\nprojection is situated on the head over each eye-cover, and a similar\nprocess projects from the roots of the wings. Its distinguishing\ncharacters are five pairs of spines on the back of the abdominal\nsegments. These spines are found also in _Acr\u00e6a Alalia_, but appear\nto be absent in other species, _e.g._ in the Indian _A. Viol\u00e6_.\nLast summer, among some batches of _Thalia_ larv\u00e6--each batch being\nthe progeny from one lot of eggs--I found certain individuals which\ndiffered from the others in having much shorter spines, and these\nchanged into pup\u00e6 in which the five pairs of spines were proportionally\nshorter than usual, thus being an exception to the rule that changes in\none stage of development are without influence on the other stages. I\nmay remark, by the way, that this law, enunciated by Weismann, can only\nbe applied to imagines and pup\u00e6 with certain restrictions. The skin of\nthe pupa forms a sheath or cover for the eyes, antenn\u00e6, trunk, legs,\nand wings of the imago, and if these parts undergo any considerable\nmodification in the latter, corresponding changes must appear in the\npupa. This is shown, for instance, by many \u2018Skippers\u2019 (_Hesperid\u00e6_),\nin which the extraordinarily long trunk necessitates a sheath of a\ncorresponding length. The colour of the pupa of _Acr\u00e6a Thalia_ is\nwhitish, the wing-veins with some other markings and the spines are\nblack; metallic spots are absent.\n\n\u201cIn the pupa of _Heliconius Eucrate_ the laterally compressed region of\nthe wings is raised into a large projection, the antennal sheaths lying\non the edges of the wings are serrated and beset with short pointed\nspines; instead of the minute projections of _Acr\u00e6a Thalia_, the head\nbears two large humped processes; the body is raised on each side into\na foliaceous border carrying five spines of different lengths, the\nforemost pair, directed towards the head, being the longest. The pupa\nis brown, and ornamented with four pairs of brilliant metallic spots,\none pair close behind the antenn\u00e6, and three pairs, almost coalescent,\non the back before the longest pair of spines. A short spine projects\nfrom the middle of each of the latter somewhat arched metallic patches.\n\n\u201cIn the pupa of _Col\u00e6nis Dido_ (which resembles that of _Col\u00e6nis\nJulia_, and to which may be added those of _Dione Vanill\u00e6_ and _Juno_)\nthe spines are absent, the wing region is but moderately arched, and\nthe antenn\u00e6 marked only by small elevations; instead of the leaf-like\nborder, there are on each side of the back five knotty or humped\nprocesses. The metallic spots are similar in number and position to\nthose of _Heliconius Eucrate_; those on the back have a wart-like\nprocess in the middle, instead of a spine.\n\n\u201cThe pup\u00e6 of _Heliconius_ and _Col\u00e6nis_ when moving their posterior\nsegments rapidly, as they do whenever they are disturbed, produce a\nvery perceptible hissing noise by the friction of these segments,\nthis sound, which is especially noticeable in the case of _Heliconius\nEucrate_, perhaps serving to terrify small foes. (So loud is the sound\nproduced in this manner by the pup\u00e6 of _Epicalia Numilia_, that my\nchildren have named them \u2018_Schreipuppen_.\u2019)\n\n\u201cThe pup\u00e6 of _Heliconius_ and _Col\u00e6nis_ thus differ to a much greater\nextent than the imagines or larv\u00e6, and the same holds good for\n_Eueides_ in a much higher degree as compared with its above-mentioned\nallies. The larv\u00e6 of _Eueides_ have no distinctive characters, and\neven the generic rank of the imagines is doubtful; as pup\u00e6, on the\nother hand, they are far removed (even by their mode of suspension)\nnot only from the remainder of the Maracuj\u00e1 group and from the whole\nof the great Nymphalideous group (_Danain\u00e6_, _Satyrin\u00e6_, _Elymniin\u00e6_,\n_Brassolin\u00e6_, _Morphin\u00e6_, _Acr\u00e6in\u00e6_ and _Nymphalin\u00e6_), but from almost\nall other butterflies. The larva pupates on the underside of a leaf;\nthe pupa is fastened by the tail, but does not hang down like the pup\u00e6\nof the other _Nymphalid\u00e6_,--its last segments are so curved that the\nbreast of the chrysalis is in contact with the underside of the leaf.\nI am not acquainted with any other pupa among those not suspended by\na girdle which assumes such a position. Something similar occurs,\nhowever, in the pupa of _Stalachtis_, which is without a girdle, and\naccording to Bates, is \u2018kept in an inclined position by the fastening\nof the tail.\u2019 By this peculiarity Bates distinguishes the _Stalachtin\u00e6_\nfrom the _Libythe\u00e6_ with pup\u00e6 \u2018freely suspended by the tail.\u2019\n\n\u201cBesides through this peculiar position of the body, the pupa of\n_Eueides Isabella_ is distinguished by short hooked and long narrow\nsabre-like pairs of processes on the back and head. Its colour is\nwhitish, yellowish, or sordid yellowish-grey; in the last variety both\nthe four long dorsal processes and the surrounding portions, as well as\nthe points of the other processes, remain white or yellowish. The pupa\n_Eueides Aliphera_ is very similar, only all the processes are somewhat\nshorter, the four longest (dorsal) and some other markings being black.\n\n\u201cNow if, as Weismann has attempted to show for larv\u00e6 and imagines,\nthe form-divergence always \u2018corresponds exactly with the divergence\nin the mode of life,\u2019 the question arises as to what difference\nin the conditions of life has brought about such a considerable\nform-divergence between the pup\u00e6 of such closely allied species as the\nMaracuj\u00e1 butterflies. In pup\u00e6 which do not eat or drink, and which\nhave neither to seek in courtship nor to care for progeny, it is only\nprotection from foes that can concern us. But in the pup\u00e6 of nearly\nallied species of which the larv\u00e6 feed on kindred plants in the same\ndistricts at the same periods of the year, can the enemies be so\ndifferent as to produce such a considerable divergence in form? One\nmight answer this question in the negative with some confidence, and\naffirm that in this case the difference in the pup\u00e6 does not result\nfrom the \u2018divergence in the mode of life,\u2019 or from the difference\nin the external conditions, but is accidental, _i.e._ a consequence\nof some fortunate variation induced by some external cause, which\nvariation afforded protection against common foes--to one species in\none way, and to the other species in some other way; this course,\nonce entered upon, having been urged on by natural selection, until\nat length the wide divergence now shown is attained. How in the case\nof any of the species the peculiarity in colour or form can actually\nserve as a protection, I must confess myself at fault in answering.\nOnly in the case of the pupa of _Eueides Isabella_ will I venture to\noffer a supposition. That it is not green like other pup\u00e6 which suspend\nthemselves among foliage (_Siderone_, _Epicalia,_ _Callidryas_, &c.),\nbut contrasts more or less brightly with the dark green of the leaves,\nprecludes the idea of concealment; on the other hand its colour is\ntoo dull to serve as a conspicuous sign of distastefulness. In either\ncase the meaning of the wonderful processes of the pupa would remain\nunexplained.\n\n\u201cWe are thus compelled to seek another possibility in mimicry, by which\nfoes would be deceived by deceptive resemblance. But what is the object\nimitated? Dead insects overgrown by fungi are often found on leaves,\nthe whitish or yellowish fungi growing from their bodies in various\nfantastic forms. Such insects of course no longer serve as tempting\nmorsels. The processes of the pupa of _Eueides_ suggest such fungoid\ngrowths, although I certainly cannot assert that to our eyes in broad\ndaylight the resemblance is very striking. But the pup\u00e6 hang among the\nshadows of the leaves, and a less perfect imitation may deceive foes\nthat are not so sharp-sighted; protective resemblance must commence\nmoreover with an imperfect degree of imitation.\u201d\n\n\n\n\nEXPLANATION OF THE PLATES.\n\n\nPLATE III.\n\nFigs. 1-12 represent larv\u00e6 of _Macroglossa Stellatarum_, all bred from\none batch of eggs. Most of the figures are enlarged, but sometimes to a\nvery small extent only; the lines show the natural length.\n\nFig. 1. Stage I.; a caterpillar immediately after hatching. Natural\nlength, 0.2 centim.\n\nFig. 2. Stage II.; shortly after the first moult. Natural length, 0.7\ncentim.\n\nFigs. 3-12. Stage V.; the chief colour-varieties.\n\nFig. 3. The only lilac-coloured specimen in the whole brood. Natural\nlength, 3.8 centim.\n\nFig. 4. Light-green form (rare) with subdorsal shading off beneath.\n\nFig. 5. Green form (rare) with strongly-pronounced dark markings\n(dorsal and subdorsal lines). Natural length, 4.9 centim.\n\nFig. 6. Dark-brown form (common). Natural length, 4 centim. In this\nfigure the fine shagreening of the skin is indicated by white dots;\nin the other figures these are partially or entirely omitted, being\nrepresented only in Figs. 8 and 10.\n\nFig. 7. Light-green form (common). Natural length, 4 centim.\n\nFig. 8. Light-brown form (common). Natural length, 3.5 centim.\n\n[Illustration:\n\n  Plate III.\n\n  Aug. Weismann pinx.      Lith. J.A. Hofmann, W\u00fcrzburg.\n]\n\n[Illustration:\n\n  Plate IV.\n\n  Aug. Weismann pinx.      Lith. J.A. Hofmann, W\u00fcrzburg.\n]\n\n[Illustration:\n\n  Plate V.\n\n  Aug. Weismann pinx.      Lith. J.A. Hofmann, W\u00fcrzburg.\n]\n\n[Illustration:\n\n  Plate VI.\n\n  Aug. Weismann pinx.      Lith. J.A. Hofmann, W\u00fcrzburg.\n]\n\nFig. 9. Parti-coloured specimen, the only one out of the whole brood.\nNatural length, 5.5 centim.\n\nFig. 10. Grey-brown form (rare).\n\nFig. 11. One of the forms intermediate between the dark-brown and green\nvarieties, dorsal aspect.\n\nFig. 12. Light-green form with very feeble dorsal line (shown too\nstrongly in the figure), dorsal aspect.\n\nFigs. 13-15. _Deilephila Vespertilio._\n\nFig. 13. Stage III.(?); the subdorsal bearing yellow spots. Natural\nlength, 1.5 centim.\n\nFig. 14. Stage IV.; the subdorsal interrupted throughout by complete\nring-spots, the white \u201cmirrors\u201d of which are bordered with black, and\ncontain in their centres a reddish nucleus. Natural length, 3 centim.\n\nFig. 15. Stage V.; shortly after the fourth moult. Subdorsal line\ncompletely vanished; ring-spots somewhat irregular, with broad black\nborders; natural length, 3.5 centim.\n\nFig. 16. _Sphinx Convolvuli_, Stage V., brown form. Subdorsal line\nretained on segments 1-3, on the other segments present only in small\nremnants; at the points where the (imaginary) subdorsal crosses the\noblique stripes there are large bright spots; natural length, 7.8\ncentim.\n\n\nPLATE IV.\n\nFigs. 17-22. Development of the markings in _Ch\u00e6rocampa Elpenor_.\n\nFig. 17. Stage I.; larva one day after hatching. Natural length, 7.5\nmillim.\n\nFig. 18. Stage II.; larva after first moult. Length, 9 millim.\n\nFig. 19. Stage II.; immediately before the second moult (Fig. 30\nbelongs here). Length, 13 millim.\n\nFig. 20. Stage III.; after second moult. Length, 20 millim.\n\nFig. 21. Stage IV.; after third moult (Figs. 32 and 33 belong here).\nLength, 4 centim.\n\nFig. 22. Stage V.; after fourth moult. A feeble indication of an\neye-spot can be seen on the third segment besides those on the fourth\nand fifth. Ocelli absent on segments 6-10.\n\nFig. 23. Stage VI.; after fifth moult. The subdorsal line is feebly\npresent on segments 6-10, and very distinctly on segments 11 and 1-3.\nOcelli repeated as irregular black spots above and below the subdorsal\nline on segments 6-11; a small light spot near the posterior border of\nsegments 5-10 (dorsal spots) and higher than the subdorsal line. Larva\nadult.\n\nFigs. 24-28. Development of the markings of _Ch\u00e6rocampa Porcellus_.\n\nFig. 24. Stage I.; immediately after emergence from the egg. Length,\n3.5 millim.\n\nFig. 25. Stage II.; after first moult. Length, 10 millim.\n\nFig. 26. Stage III.; after second moult. Length, 2.6 centim.\n\nFig. 27. Eye-spots at this last stage; subdorsal much faded, especially\non segment 4. Position the same as in last Fig.; magnified.\n\nFig. 28. Stage IV.; after third moult; corresponds exactly with Stage\nVI. of _C. Elpenor_. Dorsal view, with front segments partly retracted\n(attitude of alarm). Ocelli on segment 5 less developed than in\n_Elpenor_; repetitions of ocelli as diffused black spots on all the\nfollowing segments to the 11th; two light spots on each segment from\nthe 5th to the 11th, exactly as in _Elpenor_; subdorsal line visible\nonly on segments 1-3. Length, 4.3 centim.\n\nFig. 29. _Ch\u00e6rocampa Syriaca_. From a blown specimen in Lederer\u2019s\ncollection, now in the possession of Dr. Staudinger. Length, 5.3\ncentim.\n\nFig. 30. First rudiments of the eye-spots of _Ch\u00e6rocampa Elpenor_,\nStage II. (corresponding also with Fig. 19 in position, the head of\nthe caterpillar being to the left). Subdorsal line slightly curved on\nsegments 4 and 5.\n\nFig. 31. Eye-spots at Stage III. of the larva Fig. 20 somewhat further\ndeveloped (larva immediately before third moult). Position as in Fig.\n20.\n\nFig. 32. Eye-spots at Stage IV. corresponding to Fig. 21, A being the\neye-spot of the fourth and B that of the fifth segment.\n\nFig. 33. Eye-spot at Stage V. of the larva of _C. Elpenor_; fourth\nsegment.\n\nFigs. 30-33 are free-hand drawings from magnified specimens.\n\nFig. 34. _Darapsa Ch\u00e6rilus_ from N. America. Adult larva with front\nsegments retracted. Copied from Abbot and Smith.\n\nFig. 35. _Ch\u00e6rocampa Tersa_, from N. America. Adult larva copied from\nAbbot and Smith.\n\n\nPLATE V.\n\nFig. 36. Sixth segment of adult _Papilio_-larv\u00e6; A, _P. Hospiton_,\nCorsica; B, _P. Alexanor_, South France; C, _P. Machaon_, Germany; D,\n_P. Zolicaon_, California.\n\nFigs. 37-44. Development of the markings of _Deilephila Euphorbi\u00e6_.\n\nFig. 37. Stage I.; young caterpillar shortly after emergence. Natural\nlength, 5 millim.\n\nFig. 38. Similar to the last, more strongly magnified. Natural length,\n4 millim.\n\nFig. 39. Stage II.; larva immediately after first moult. The row of\nspots distinctly connected by a light stripe (residue of the subdorsal\nline). Natural length, 17 millim.\n\nFig. 40. Stage III.; after second moult; magnified drawing of the last\nfive segments. Only one row of large white spots on a black ground\n(ring-spots); subdorsal completely vanished; the shagreen-dots formerly\nabsent now appear in vertical rows interrupted only by the ring-spots.\nBelow the latter are some enlarged shagreen-dots which subsequently\nbecome the second ring-spots. Natural length of the entire caterpillar,\n21 millim.\n\nFig. 41. Stage IV.; the same larva after the third moult.\nTransformation of the ground-colour from green to black, owing to the\nspread of the black patches proceeding from the ring-spots in Fig. 40\nin such a manner as to leave between them only a narrow green triangle.\nThe shagreen dots below the ring-spots have increased in size, but have\nnot yet coalesced.\n\nFig. 42. Stage III.; larva, same age as Fig. 40, but with _two rows_ of\nring-spots. Natural length of the whole caterpillar, 32 millim.\n\nFig. 43. Stage V.; larva from Kaiserstuhl. Variety with only one row of\nring-spots, and with red nuclei in the mirror-spots. Natural length, 5\ncentim.\n\nFig. 44. Stage V.; larva from Kaiserstuhl (like the three preceding).\nThe green triangles on the posterior edges of the segments in Fig. 42\nhave become changed into red. Natural length, 7.5 centim.\n\nFig. 45. _Deilephila Galii_; Stage IV. Subdorsal with open ring-spots.\nNatural length, 3.4 centim.\n\nFig. 46. _D. Galii_; adult larva; Stage V. Brown variety with feeble\nshagreening; subdorsal completely vanished. Natural length, 6 centim.\n\n\nPLATE VI.\n\nFig. 47. The same species at the same stage. Black variety strongly\nshagreened; similar to _Deil. Euphorbi\u00e6_.\n\nFig. 48. Similar to the last. Yellow var. without any trace of\nshagreening.\n\nFig. 49. _Deilephila Vespertilio._ Three stages in the life of the\nspecies, representing three phyletic stages of the genus. A, life-stage\nIII.=phyletic stage 3 (subdorsal with open ring-spots); B, life-stage\nIV.=phyletic stage 4 (subdorsal with closed ring-spots); C, life-stage\nV.=phyletic stage 5 (subdorsal vanished, only _one_ row of ring-spots).\n\nFig. 50. _Deilephila Zygophylli_, from S. Russia; stage V. From a\nblown specimen in Staudinger\u2019s collection. In this specimen the\nring-spots are difficult to distinguish on account of the extremely\ndark ground-colour; they are nevertheless present, and would probably\nbe more distinct in the living insect. A, open ring-spot from another\nspecimen of this species in the same collection.\n\nFig. 51. _Deilephila Nic\u00e6a_, from South France; Stage V. Copied from\nDuponchel.\n\nFig. 52. _Sphinx Convolvuli_; Stage V., segments 10-8. Brown variety,\nwith distinct white spots at the points of intercrossing of the\nvanished subdorsal with the oblique stripes.\n\nFig. 53. _Anceyrx Pinastri_; A and B, larv\u00e6 immediately after hatching.\nNatural length, 6 millim.\n\nFig. 54. Same species; Stage II. Subdorsal, supra- and infra-spiracular\nlines developed. Natural length, 15 millim.\n\nFig. 55. _Smerinthus Populi_; Stage I. Immediately after hatching; free\nfrom all marking. Length, 6 millim.\n\nFig. 56. Same species at the end of first stage; lateral aspect.\nLength, 1.3 centim.\n\nFig. 57. Same species; Stage II. Subdorsal indistinct; the first and\nlast oblique stripes more pronounced than the others. Length, 1.4\ncentim.\n\nFig. 58. _Deilephila Hippopha\u00ebs_; Stage III. Subdorsal with open\nring-spot on the 11th segment. A, segment 11 somewhat enlarged. Length,\n3 centim.\n\n\nPLATE VII.\n\nFig. 59. _Deilephila Hippopha\u00ebs_; Stage V. Secondary ring-spots on six\nsegments (10-5).\n\nFig. 60. Same species; Stage V. One or two red shagreen dots on\nsegments 10-4 in the position of the ring-spots of Fig. 59. Length, 6.5\ncentim.\n\nFig. 61. Same species; Stage V. Segments 9-6 of another specimen, more\nstrongly magnified. A ring-spot on segments 9 and 8 showing its origin\nfrom two shagreen-dots; two red shagreen-dots on segment 7, on segment\n6 only one.\n\nFig. 62. _Deilephila Livornica_ (Europe) in the last stage. Green form.\nCopied from Boisduval.\n\nFig. 63. _Pterogon \u0152nother\u00e6_; Stage IV. Length, 3.7 centim.\n\nFig. 64. The same species at the same stage; dorsal view of the last\nsegment.\n\nFig. 65. The same segment in Stage V. Eye-spot completely developed.\n\nFig. 66. _Saturnia Carpini_, larva from Freiburg; Stage III. Natural\nlength, 15 millim.\n\nFig. 67. Same species; larva from Genoa; Stage IV. Length, 20 millim.\n\nFig. 68. Same species; larva from Freiburg; Stage III. Segments 8 and 9\nin dorsal aspect. Length, 15 millim.\n\nFig. 69. The same caterpillar; lateral view of segment 8.\n\nFig. 70. _Smerinthus Ocellatus_; adult larva with distinct subdorsal\non the six foremost segments. The shagreening is only shown in the\ncontour, elsewhere omitted. Length, 7 centim.\n\n[Illustration:\n\n  Plate VII.\n\n  Aug. Weismann pinx.      Lith. J. A. Hofmann, W\u00fcrzburg.\n]\n\n[Illustration:\n\n  Plate VIII.\n\n  Aug. Weismann pinx.      Lith. J. A. Hofmann, W\u00fcrzburg.\n]\n\n\nPLATE VIII.\n\nFigs. 71-75 represent segments 8 and 9 of the larva of _Saturnia\nCarpini_ (German form) in dorsal aspect, all at the fourth stage.\nThe head of the caterpillar is supposed to be above, so that the top\nsegment is the eighth.\n\nFig. 71. _Saturnia Carpini._ Darkest variety.\n\nFig. 72. Lighter variety.\n\nFig. 73. Still lighter variety.\n\nFig. 74. One of the lightest varieties; the black extends further on\nsegments 9 and 10 than on the 8th.\n\nFig. 75. Lightest variety.\n\nFigs. 76-80 are only represented on a smaller scale than the remaining\nFigs. in order to save space; were they enlarged to the same scale they\nwould be larger than the other figures.\n\nFig. 76. _Saturnia Carpini_ (Ligurian form); Segment 8; Stage V.\n\nFig. 77. Same form; same segment in stage VI.\n\nFigs. 78, 79, and 80. _Saturnia Carpini_ (German form); dorsal aspect\nof 8th segment in Stage V. (the last of this form).\n\nFig. 78. Darkest variety.\n\nFig. 79. Lighter variety.\n\nFig. 80. Lightest variety.\n\nFigs. 81-86. _Saturnia Carpini_ (German form); Stage IV. Side view of\nthe 8th segment in six different varieties. Fig. 81 shows only two\nsmall green spots at the bases of the upper warts besides the green\nspiracular stripes. Fig. 82 shows the spots enlarged and increased by a\nthird behind the warts; the pro-legs have also become green.\n\nFig. 83. Two of the three green spots, which have become still more\nenlarged, are coalescent.\n\nFig. 84. All three spots coalescent; but here, as also in\n\nFig. 85, various residues of the original black colour are left as\nboundary-marks.\n\nFig. 86. Lightest variety.\n\n\nEND OF PART II.\n\n\n\n\nSTUDIES IN THE THEORY OF DESCENT.\n\n=Part III.=\n\nON THE FINAL CAUSES OF TRANSFORMATION.\n\n\n\n\n=III.=\n\nTHE TRANSFORMATION OF THE MEXICAN AXOLOTL INTO AMBLYSTOMA.\n\n\nINTRODUCTION.\n\nSince the time when Dum\u00e9ril made known the transformation of a number\nof Axolotls into the so-called Amblystoma form, this Mexican Amphibian\nhas been bred in many European aquaria, chiefly with the view to\nestablish the conditions under which this transformation occurred, so\nas to be enabled to draw further conclusions as to the true causes of\nthis exceptional and enigmatical metamorphosis.\n\nAlthough the Amphibians propagated freely, the cases in which\ntransformation occurred remained extremely rare, and it was not once\npossible to reply to the main question, viz. whether this metamorphosis\nwas determined by external conditions or by purely internal causes;\nto say nothing of the possibility of there perhaps being discoverable\ncertain definite external influences by means of which the\nmetamorphosis could have been induced with certainty. But while these\npoints are undecided all attempted theoretical interpretations of the\nphenomenon must be devoid of a solid basis.\n\nIt appeared to me from the first that the history of this\ntransformation of the Axolotl was of special theoretical value; indeed\nI believed that it might possibly furnish a special case for deciding\nthe truth of those ground-principles, according to which the origin of\nthis species is represented by the two conflicting schools as a case of\ntransformation or as one of heterogenesis. I therefore determined to\nmake some experiments with the Axolotl myself, in the hopes of being\nfortunate enough to be able to throw some light upon the subject.\n\nIn the year 1872 Prof. v. K\u00f6lliker was so good as to leave with me\nfive specimens of his Axolotls, bred in W\u00fcrzburg, and these furnished\na numerous progeny in the following year. With these I carried out the\nidea, the theoretical bearing of which will be shown subsequently,\nwhether it would not be possible to force all the larv\u00e6, or at any\nrate, the greater majority, to undergo transformation by exposing them\nto conditions of life which made the use of gills difficult, and that\nof lungs more easy; in other words, by compelling them to live partly\non land at a certain stage of life.\n\nDuring that year indeed I obtained no results, most of the larv\u00e6\nperishing before the time for such an experiment had arrived, and the\nfew survivors did not undergo transformation, but lived on to the\nfollowing spring and then also died one after the other. Through long\nabsence from Freiburg, necessitated by other labours, I had evidently\nleft them without sufficient care and attention. I was thus led to\nthe conviction, which was more fully confirmed subsequently, that no\nresults can be obtained without the greatest care and attention in\nrearing, towards which single object all one\u2019s interest should be\nconcentrated, and it must not be considered irksome to have to devote\ndaily for many months a large amount of time to this experiment. As it\nwas evident that I could not afford this time without calling in other\naid, I hailed with pleasure an opportunity of witnessing the experiment\nperformed by other hands.\n\nA lady living here (Freiburg), Fr\u00e4ulein v. Chauvin, undertook to rear\na number of my larv\u00e6 of the following year which had just hatched,\nand in accordance with my idea to make the experiment of forcibly\ncompelling them to adopt the Amblystoma form. How completely this was\naccomplished will be seen from the following notes by the lady herself,\nand it will no less appear that these results were only obtained by\nthat care in treatment and delicacy of observation which she devoted to\nthe experiments.\n\n\nEXPERIMENTS.\n\n\u201cI began the experiments on June 12th, 1874, with five larv\u00e6 about\neight days old, these being the only survivors out of twelve. Owing\nto the extraordinary delicacy of these creatures, the quality and\ntemperature of the water, and the nature and quantity of their food\nexerts the greatest influence, especially in early life, and one cannot\nbe too cautious in their treatment.\n\n\u201cThe specimens were kept in a glass globe of about thirty centimeters\nin diameter, the temperature of the water being regulated; as food at\nfirst Daphnids, and afterwards larger aquatic animals were introduced\nin large numbers. By this means all the five larv\u00e6 throve excellently.\nAt the end of June the rudiments of the front legs appeared in the most\nvigorous specimens, and on the 9th of July the hind legs also became\nvisible. At the end of November I noticed that one Axolotl remained\nconstantly at the surface of the water, and this led me to suppose that\nthe right period had now arrived for effecting the transformation\ninto Amblystoma. For brevity I shall designate this as No. I., and the\nsucceeding specimens by corresponding Roman numerals.\n\n\u201cIn order to bring about this metamorphosis, on December 1st, 1874, No.\nI. was placed in a large-sized glass vessel containing earth arranged\nin such a manner that, when the vessel was filled with water, only one\nportion of the surface of the earth was entirely covered by the liquid,\nand the creature in the course of its frequent perigrinations was thus\nmore or less exposed to the air. The water was gradually diminished on\nthe following days, during which period the first changes made their\nappearance in the Amphibian--_the gills commenced to shrivel up_, and\nat the same time the creature showed a tendency to seek the shallowest\nspots. On December 4th, it took entirely to the land, and concealed\nitself among some damp moss which I had placed on a heap of sand on the\nhighest portion of the earth in the glass vessel. At this period the\nfirst ecdysis occurred. Within the four days from the 1st to the 4th of\nDecember, a striking change took place in the external appearance of\nNo. I., the gill-tufts shrivelled up almost entirely, the dorsal crest\ncompletely disappeared, and the tail, which had hitherto been broad,\nbecame rounded and similarly formed to that of a land salamander. The\ngrey-brown colour of the body changed gradually into a blackish hue;\nisolated spots, at first of a dull white, made their appearance and\nthese in time increased in intensity.\n\n\u201cWhen the Axolotl left the water on December 4th the gill-clefts were\nstill open, but these closed gradually, and after about eight days were\novergrown with skin and no longer to be seen.\n\n\u201cOf the other larv\u00e6 three appeared at the end of November (_i.e._ at\nthe same time when No. I. came to the surface of the water) to have\nkept pace in development with No. I., an indication that for these\nalso the right period had arrived for accelerating the developmental\nprocesses. They were therefore submitted to the same treatment as No.\nI. No. II. became transformed at the same time and exactly in the same\nmanner as the latter; its gill-tufts were complete when it was first\nplaced in the shallow water, but after four days these had almost\nentirely disappeared; in the course of about ten days after it took to\nthe land, the overgrowth of skin on the gill-clefts and the complete\nassumption of the salamander form occurred. During this last period the\ncreature took food, but only when urged to do so.\n\n\u201cIn Nos. III. and IV. the development proceeded more slowly. Neither of\nthese so frequently sought the shallow spots, nor did they as a rule\nremain so long exposed to the air, so that the greater part of January\nhad expired before they took entirely to the land. Nevertheless the\ndessication of the gill-tufts did not take a longer time than in Nos.\nI. and II. as the first ecdysis occurred as soon as they took to the\nland.\n\n\u201cNo. V. showed still more striking deviations in its transformation\nthan Nos. III. and IV., but as this specimen appeared much weaker\nthan the others from the beginning and was retarded in growth to a\nmost notable extent, this is by no means surprising. It took fourteen\ninstead of four days before the transformation had advanced far enough\nto enable it to leave the water. It was especially interesting to\nobserve the behaviour of this specimen during this period. Its weak\nand delicate constitution evidently made it much more susceptible to\nall external influences than the others. If exposed to the air for too\nlong a time it acquired a light colour, and when annoyed or alarmed it\nemitted a peculiar odour, similar to that of a salamander. As soon as\nthese phenomena were observed it was at once placed in deeper water,\ninto which it immediately plunged and gradually recovered itself,\nthe gills always becoming again expanded. The same experiment was\nrepeated several times and always led to the same result, from which\nwe may venture to conclude that by accelerating the transformation\ntoo energetically, the process may come to a standstill, and even by\ncontinued compulsion may end in death.\n\n\u201cIt yet remains to be mentioned with respect to Axolotl No. V. that\nthis specimen, unlike all the others, did not emerge from the water at\nthe first ecdysis, but at the time of the fourth.\n\n\u201cAll the Axolotls are now (July, 1875) living, and are healthy and\nvigorous, so that with respect to their state of nourishment there is\nnothing to prevent their propagating. Of the first four the largest is\nfifteen centim. long; Axolotl No. V. measures twelve centim.\n\n\u201cThe preceding statements appear to demonstrate the correctness of the\nviews advanced in the Introduction:--Axolotl larv\u00e6 generally but not\nalways complete their metamorphosis if, in the first place, they emerge\nsound from the egg and are properly fed; and if, in the next place,\nthey are submitted to the necessary treatment for changing aquatic\ninto a\u00ebrial respiration. It is obvious that this treatment must only\nbe applied very gradually, and in such a manner as not to overtax the\nvital energy of the Amphibian.\u201d\n\n       *       *       *       *       *\n\nTo the foregoing remarks of Fr\u00e4ulein v. Chauvin I may add that in all\nfive cases the transformation was complete, and not to be confounded\nwith that change which occurs more or less in all Axolotls in the\ncourse of time when confined in small glass vessels. In this last case\nthere frequently appear changes in the direction of the Amblystoma form\nwithout the latter being actually reached. In the five adult Axolotls\nwhich I possessed for a short time, and of which two were at least four\nyears old, the gills were much shrivelled, but the aquatic tail and\ndorsal crest were unchanged. The crest may, however, also disappear,\nand the tail become shortened without these changes being due to a\ntransformation into Amblystoma, as will be shown further on.\n\nWith respect to the duration of the transformation, this amounted\nin Axolotls Nos. I. to IV. altogether to twelve or fourteen days.\nOf these, four days were taken by the first changes which occurred\nwhile the creature was still in the water; the remaining time, to the\ncompletion of the metamorphosis, was passed on land. Dum\u00e9ril gives the\nduration of the metamorphosis as sixteen days.\n\nThe following results of the experiments just described appear to me\nto be especially noteworthy:--The five Axolotl larv\u00e6 which can alone\nbe taken into consideration, the others having soon perished, all\nexperienced metamorphosis, and without an exception became Amblystomas.\nOnly one of them, No. I., by persistently swimming at the surface, as\nwas observed at the end of six months, showed a decided tendency to\nundergo metamorphosis and to adopt a\u00ebrial respiration. With respect to\nthis specimen it may therefore be confidently assumed that it would\nhave taken to the land, and that metamorphosis would have occurred\nwithout artificial aid, just as was the case in the thirty specimens\nwhich Dum\u00e9ril altogether observed.\n\nRespecting Nos. II., III., and IV., on the other hand, such a\nsupposition is but little probable. These three larv\u00e6 endeavoured\nto keep in deep water and avoided as long as possible the shallow\nplaces which would have enforced them to take entirely to lung\nbreathing. Metamorphosis thus occurred more than a month later in these\nindividuals.\n\nFinally, there can scarcely be any doubt that No. V. would not have\nbecome transformed without forcible adaptation to an a\u00ebrial life.\n\nFrom these results we may venture to conclude that most Axolotl larv\u00e6\nchange into the Amblystoma form when, at the age of six to nine\nmonths, they are placed in such shallow water that they are compelled\nto respire chiefly by their lungs. The experiments before us are\ncertainly at present but very few in number, but such a conclusion\ncannot be termed premature if we consider that out of several hundred\nAxolotls (the exact number is not given) Dum\u00e9ril obtained only about\nthirty Amblystomas, while v. K\u00f6lliker bred only one Amblystoma out of a\nhundred Axolotls.\n\nIt now only remains questionable whether _each larva_ could have been\nforced to undergo metamorphosis, but this could only be decided by\nnew experiments. It was originally my intention to have delayed the\npublication of the experiments till Fr\u00e4ulein v. Chauvin had repeated\nthem in larger numbers, but as my Axolotls have not bred this year\n(1875) I must abandon my scheme, and this can be done the more readily\nbecause, for the theoretical consideration of the facts, it is\nimmaterial whether _all_ or only _nearly all_ the Axolotls could have\nbeen compelled to undergo transformation. I must not, however, omit to\nmention that Herr Gehrig, the curator of our Zoological Museum, bred\na considerable number of larv\u00e6 from the same brood as that with which\nFr\u00e4ulein v. Chauvin experimented, and that of these larv\u00e6 six lived\nover the winter _without undergoing metamorphosis_. They were always\nkept in deep water and thus furnished the converse experiment to those\nrecorded above; they further prove that this whole brood did not have a\nprevious tendency to undergo metamorphosis.\n\nIf these new facts are to be made use of to explain the nature of this\nextraordinary process of transformation in accordance with our present\nconception, the data already known must in the first place be called to\nour aid.\n\nIt has first to be established that _Siredon Mexicanus_ never, as\nfar as we know, undergoes metamorphosis in its native country. This\nAmphibian is there only known in the _Siredon_ form, a statement which\nI have taken from De Saussure,[225] who has himself observed the\nAxolotl in the Mexican lakes. This naturalist never found a single\nAmblystoma in the neighbourhood of the lakes, \u201cnevertheless the larva\n(Axolotl) is so common there that it is brought into the market by\nthousands.\u201d De Saussure believes that in Mexico the Axolotl does not\nundergo transformation.[226] The same statement is distinctly made by\nCope,[227] whose specimens of _Siredon Mexicanus_ bred in America,\neven in captivity showed \u201cno tendency to become metamorphosed.\u201d On the\nother hand Tegetmeier observed[228] that one out of five specimens\nobtained from the Lake of Mexico underwent metamorphosis, and this\naccordingly establishes the second fact, viz. that the true Axolotl\nbecomes transformed under certain conditions into an Amblystoma when in\ncaptivity.\n\nThis last remark would be superfluous if, as was for a long time\nbelieved, the Paris Axolotls, of which the metamorphosis was first\nobserved and which at the time made such a sensation, were actually\n_Siredon Mexicanus_, _i.e._ the _Siredon_ which alone in its native\ncountry bears the name of Axolotl. In his first communication Dum\u00e9ril\nwas himself of this opinion; he then termed the animal \u201c_Siredon\nMexicanus vel Humboldtii_,\u201d[229] but subsequently, in his amplified\nwork[230] on the transformation of the Axolotl observed in the Jardin\ndes Plantes, he retracted this view, and after a critical comparison of\nthe five described species of _Siredon_, he came to the conclusion that\nthe species in the possession of the Paris Museum was probably _Siredon\nLichenoides_ (Baird). All the transformations of Axolotls observed\nin Europe must consequently be referred to this species, since they\nwere--at least as far as I know--all derived from the Paris colony. My\nown experimental specimens were also indirectly descended from these.\n\nNow it must be admitted that this does not coincide with the fact that\nthe Amblystoma form which Dum\u00e9ril first obtained from his Axolotls\nagreed with Cope\u2019s species, _A. Tigrinum_, while on the other hand we\nlearn from Marsh[231] that _Siredon Lichenoides_ (Baird), when it does\nundergo metamorphosis, becomes transformed into _Amblystoma Mavortium_\n(Baird).\n\nMarsh found _Siredon Lichenoides_ in mountain lakes (7000 feet above\nthe sea) in the southwest of the United States (Wyoming Territory),\nand obtained from them, by breeding in aquaria, _Amblystoma Mavortium_\n(Baird). He considers it indeed doubtful whether the Amphibian\nundergoes this transformation in its native habitat, although he\ncertainly states this opinion without rigorous proof on purely\ntheoretical considerations, because, according to his view, \u201cthe low\ntemperature is there less favourable.\u201d[232]\n\nIf I throw doubt upon this last statement it is simply because\n_Amblystoma Mavortium_ is found native in many parts of the United\nStates, viz:--in California, New Mexico, Texas, Kansas, Nebraska, and\nMinnesota. It is indeed by no means inconceivable that in the mountain\nlakes where Marsh obtained this species, it may behave differently\nwith respect to metamorphosis than in other habitats, and this appears\nprobable from certain observations upon _Triton_ which will be\nsubsequently referred to.\n\nMeanwhile, in the absence of further observations, we must admit that\nthe Paris Axolotls were not _Siredon Lichenoides_, but some nearly\nallied and probably new species. But little information is furnished\nby observing the course of the transformation, although it is at least\nestablished that this Axolotl in its native habitat does not undergo\nmetamorphosis or does so as exceptionally as in Europe. Unfortunately\nin his papers Dum\u00e9ril gives no precise statement respecting the\nlocality of this species imported from \u201cMexico\u201d--it is probable that\nhe was himself unacquainted with it, so that I can only state on the\nauthority of Cope that Amblystoma has never been brought from south of\nthe provinces of Tamaulipas and Chilhuahua, _i.e._ south of the Tropic\nof Cancer.[233]\n\nThis last statement, however, gives no certainty to the matter. Of much\nmore importance is the above-mentioned fact, that the true Axolotl\nof the lakes about the city of Mexico does not, as a rule, become\ntransformed into an Amblystoma in that locality, although this species\nin certain cases undergoes metamorphosis when in confinement. From\nthis circumstance and from the fact that the Paris Axolotl likewise\nexperienced but a very small percentage of metamorphosis in captivity,\nwe may venture to conclude that this species also, in its native\nhabitat, either does not become transformed at all or does so only\nexceptionally.\n\nBut there is yet another body of facts which come prominently into\nview on considering the history of the transformations. I refer to the\nexistence of numerous species of _Amblystoma_ in a natural state. In\nthe \u201cRevision of the genera of _Salamandrid\u00e6_,\u201d published some years\nago by Strauch,[234] this author, following Cope,[235] gives twenty\nspecies of _Amblystoma_ as inhabiting North America. Although some\nof these species are based on single examples, and consequently, as\nStrauch justly remarks, \u201cmay well have to be reduced in the course\nof time,\u201d there must nevertheless always remain a large number of\nspecies which live and propagate as true Amblystomas, and of which the\nhabitat extends from the latitude of New York to that of New Mexico.\nThere are therefore true species of _Siredon_ which regularly assume\nthe Amblystoma form under their natural conditions of life, and which\npropagate in this form, while, on the other hand, there are at least\ntwo species which, under their existing natural conditions of life,\nalways propagate as _Siredon_. It is but another mode of expression\nfor the same facts if we say that the Mexican Axolotl and the Paris\n_Siredon_--whether this is _Lichenoides_ or some other species--stand\nat a lower grade of phyletic development than those species of\n_Amblystoma_ which propagate in the salamander form. No one can raise\nany objection to this statement, while the alternative view maintained\nby all authors contains a theory either expressed or implied which is,\nas I believe, incorrect, viz. that the Mexican Axolotl _has remained_\nat an inferior stage of phyletic development.\n\nAll zoologists[236] who have expressed an opinion upon the\ntransformation of the Axolotl, and who are not, like the first observer\nof this fact, embarrassed by Cuvier\u2019s views as to the immutability of\nspecies, regard the phenomenon as though a species, which owing to some\nspecial conditions had hitherto remained at a low stage of development,\nhad now through some other influences been compelled to advance to a\nhigher stage.\n\nI believed for a long time that the phenomenon could not otherwise be\ncomprehended, so little was I then in a position to bring all the facts\ninto harmony with this view. Thus in the year 1872 I expressed myself\nas follows[237]:--\u201cWhy should not a sudden change in all the conditions\nof life (transference from Mexico to Paris) have a direct action on\nthe organization of the Axolotl, causing it suddenly to reach a higher\nstage of development, such as many of its allies have already attained,\nand which obviously lies in the nature of its organization--a stage\nwhich it would perhaps itself have reached, although later, in its\nnative country? Or is it inconceivable that the sudden removal from\n8000 feet above the sea (Mexican plateau) to the altitude of Paris, may\nhave given the respiratory organs an impetus in the direction of the\ntransformation imminent? In all probability we have here to do with the\ndirect action of changed conditions of life.\u201d\n\nThat the substance of this last statement must still hold good is\nobvious from the experiments previously described, which show that\nby the application of definite external influences, we have it to\na certain extent in our power to produce the transformation. It is\nprecisely in this last point that there lies the new feature furnished\nby these experiments.\n\nBut are we also compelled to interpret the phenomenon in the above\nmanner? _i.e._ as a sudden advance in the phyletic development of the\nspecies occurring, so to speak, at one stroke? I believe not.\n\nWhat first made this view appear to me erroneous, was the appearance\nof the living Amblystomas bred from my Axolotl larv\u00e6. These creatures\nby no means differed from the Axolotls merely in single characters,\nbut they were distinct from the latter in their entire aspect; they\ndiffered in some measure in all their parts, in some but slightly and\nin other parts strongly--in brief, they had become quite different\nanimals. In accordance with this, their mode of life had become\ncompletely modified; they no longer lived in the water, but remained\nconcealed by day among the damp moss of the vivarium, coming forth only\nby night in search of food in dry places.\n\nI had been able to perceive the great difference between the two\nstages of development from the anatomical data with which I had long\nbeen familiar, and which Dum\u00e9ril had made known with respect to the\nstructure of his Amblystomas. But the collecting of numerous details\ngives no very vivid picture of the metamorphosis; it was the appearance\nof the living animal that first made me conscious how deep-seated was\nthe transformation which we have here before us, and that this change\nnot merely affected those parts which would be directly influenced\nby the change in the conditions of life, such as the gills, but that\nmost if not all the parts of the animal underwent a transformation,\nwhich could in part be well explained as morphological adaptation to\nnew conditions of life, and partly as a consequence of this adaptation\n(correlative changes), but could not possibly be regarded as the sudden\naction of these changed conditions.\n\nSuch at least is my view of the case, according to which a _per saltum_\ndevelopment of the species of such a kind as must here have taken\nplace, is quite inconceivable.\n\nI may venture to assume that most observers of the metamorphosis of\nAxolotl have, like myself, not been hitherto aware of the extent of\nthe transformation, and it may thus be explained why the theoretical\nbearings of the case have on all sides been incorrectly conceived. We\nhave here obviously a quite extraordinary case of the first order of\nimportance. I believe that it can easily be shown that the explanation\nof the history of the metamorphosis of the Paris Axolotl which has\nhitherto been pretty generally accepted, necessarily comprises a very\nfar-reaching principle. If this interpretation is correct, then in my\nopinion must also hold good the ideas of those who, like K\u00f6lliker,\nAskenasy, N\u00e4geli, and, among the philosophers, Hartmann and H\u00fcbner,\nwould refer the transformation of species in the first instance to a\npower innate in the organism, to an active, _i.e._ a self-urging \u201claw\nof development\u201d--a phyletic vital force.\n\nThus, if the Axolotls transformed into Amblystomas are regarded as\nindividuals which, impelled by external influences, have anticipated\nthe phyletic development of the others, then this advance can only\nbe ascribed to a phyletic vital force, since the transformation is\nsudden, and leaves no time for gradual adaptation in the course of\ngenerations. The _indirect_ influence of the external conditions of\nlife, _i.e._ natural selection, is thus excluded from the beginning.\nBut the _direct_ action of the changed conditions of life by no means\nfurnishes an explanation of the complete transformation of the whole\nstructure, such as I have already alluded to, and which I will now\nenter into more closely.\n\nThe differences between the Paris Axolotl and its Amblystoma according\nto Dum\u00e9ril, K\u00f6lliker, and my own observations are as follow:--\n\n1. The gills disappear; the gill-clefts close up, and of the branchial\narches only the foremost remain, the posterior ones disappearing. At\nthe same time the os _hyoideum_ becomes changed (Dum\u00e9ril).\n\n2. The dorsal crest completely disappears (Dum\u00e9ril).\n\n3. The aquatic tail becomes transformed into one like that of the\nsalamanders (Dum\u00e9ril), which, however, is not as in the salamander\ncylindrical, but somewhat compressed laterally (Weismann).\n\n4. The skin becomes yellowish white, irregularly spotted on the sides\nand back (Dum\u00e9ril), whilst at the same time its former grey-black\nground-colour changes into a shining greenish black (Weismann); it\nloses, moreover, the slimy secretion of the skin, and the cutaneous\nglands become insignificant (K\u00f6lliker).\n\n5. The eyes become prominent and the pupils narrow (K\u00f6lliker), and\neyelids capable of completely covering the eyes are formed; in Axolotl\nonly a narrow annular fold surrounds the eyes, so that these cannot be\nclosed (Weismann).\n\n6. The toes become narrowed and lose their skin-like appendages\n(K\u00f6lliker), or more precisely, the half webs which connect the proximal\nextremities of the toes on all the feet (Weismann).\n\n7. The teeth are disposed in this species, as in all other\n_Amblystom\u00e6_, in transverse series; whilst in Axolotl, as in _Triton_\nlarv\u00e6, they are arranged at the sides of the jaw in the form of a\ncurved arch-like band beset with several rows of teeth.[238] (Dum\u00e9ril.\nSee his fig., _loc. cit._ p. 279).\n\n8. In Axolotl the lower jaw, in addition to the teeth on the upper edge\nof the bone, also bears \u201cde tr\u00e8s petites dents dispos\u00e9es sur plusieurs\nrangs;\u201d these last disappear after metamorphosis (Dum\u00e9ril). I will add\nthat the persistent teeth belong to the _os dentale_ of the lower jaw,\nand those that are shed to the _os operculare_.[239]\n\n9. The surface of the posterior moveable part of the body is slightly\nconcave both before and after transformation; the anterior part is,\nhowever, less concave in _Amblystoma_ than in _Siredon_ (Dum\u00e9ril).\n\nI have not yet been able to verify Dum\u00e9ril\u2019s 7th and 9th statements,\nas I did not want to kill any of my living Amblystomas,[240] simply\nin order to confirm the observations of a naturalist in whom one may\ncertainly place complete confidence. Neither have I as yet observed the\ntransformation of the branchial arches, but all the other statements of\nK\u00f6lliker and Dum\u00e9ril I can entirely corroborate.\n\nThe structural differences between Axolotl and Amblystoma are\nconsiderably greater and of more importance than those between allied\ngenera, or indeed than between the families of the Urodela. The genus\n_Siredon_ undoubtedly belongs to a different sub-order to the genus\n_Amblystoma_ into which it occasionally becomes transformed. Strauch,\nthe most recent systematic worker at this group, distinguishes\nthe sub-order _Salamandrida_ from that of the _Ichthyodea_ by the\npossession of eyelids, and by the situation of the palatine teeth in\nsingle rows on the posterior edge of the palatal bone: in _Ichthyodea_\nthe eyelids are wanting and the palatine teeth are either \u201csituated on\nthe anterior edge of the palatal bone,\u201d or \u201ccover the whole surface of\nthe palatal plates in brush-like tufts.\u201d\n\nHow is it possible to regard such widely divergent anatomical\ncharacters as changes suddenly produced by the action (but once\nexerted) of deviating conditions of life? Hand in hand with the\nshedding of the old and the appearance of new palatine teeth, there\noccurs a change in the anatomical structure of the vertebral column,\nand also--as we may fairly conclude from K\u00f6lliker\u2019s correct observation\nof the cessation of the slimy secretion--in the histological structure\nof the skin. Who would undertake to explain all these profound\nmodifications as the direct and sudden action of certain external\ninfluences? And if any one were inclined to explain such changes as a\nconsequence of the disappearance of the gills, _i.e._ as correlative\nchanges, what else is such a correlation than the phyletic vital force\nunder another name?\n\nIf from one change arising from the direct action of external\nagencies, the whole body can in two days become transformed in all\nits parts, in the precise manner which appears best adapted for the\nnew conditions of life under which it is henceforward to exist, then\nthe word \u201ccorrelation\u201d is only a phrase which explains nothing, but\nwhich prevents any attempt at a better explanation, and it would be\npreferable to profess simply the belief in a phyletic vital force.\n\nMoreover, it is hardly permissible to seek such an explanation, since\nUrodela are known which have no gills in the adult state, and which\nnevertheless possess all the other characters of the _Ichthyodea_,\nviz. want of eyelids, characteristic palatine teeth, and the tongue\nbone. This is the case with the genera _Amphiuma_ (Linn.), _Menopoma_\n(Harl.), and _Cryptobranchus_ (v. d. Hoev.). The two first genera, as\nis known, still possess gill-clefts, but _Cryptobranchus_ has even\nlost these clefts, which, as in _Amblystoma_, are overgrown by skin;\nnevertheless _Cryptobranchus_ is, according to the concurrent testimony\nof all systematists, a true salamander in habits, tongue bone, palatine\nteeth,[241] &c. It must further be added that the Axolotl itself can\nlose the gills without thereby becoming transformed into an Amblystoma.\nI have previously mentioned that in Axolotls which were kept in shallow\nwater the gills frequently became diminutive, and it also sometimes\nhappens that they completely shrivel up. I possess an Axolotl preserved\nin alcohol in which the gills have shrivelled up into small irregular\nbunches, and the dorsal crest is also so completely absent that its\nplace is occupied by a long furrow, and even on the tail the crest\nhas entirely disappeared from the lower edge and about half from the\nupper edge. Notwithstanding this, the creature is widely removed from\nAmblystoma in structure; it possesses the arched branchial apparatus,\nthe palatine teeth, the skin, &c., of the Axolotl.\n\nThese facts prove, therefore, that the shedding of the gills by no\nmeans always entails all the other modifications which we observe in\nthe metamorphosis of Axolotl, so that these modifications are thus\nnot by any means the necessary and immediate consequence of such gill\nshedding.\n\nWhether these modifications will occur after a long series of\ngenerations--whether the successors of _Cryptobranchus_ will also one\nday acquire the salamandriform structure is another question, and one\nwhich I could not exactly answer in the negative. But this question\ndoes not here come into consideration, as we are now only concerned\nwith the _immediate result_ of the shedding of the gills.\n\nThe problem appears therefore to be as follows:--Either the hitherto\nreceived interpretation of the transformational history of the Axolotl\nas a further development of the species is incorrect, or else the case\nof Axolotl incontestably proves the existence of a phyletic vital force.\n\nWe have now to ask whether the facts of this transformational history\nare not capable of another explanation.\n\nI believe that this is certainly possible, and that another\ninterpretation can be shown to be correct with some degree of\nprobability.\n\nI am of opinion that those Amblystomas which have been developed\nin captivity in certain instances from _Siredon Mexicanus_ (_S.\nPisciformis_), as well as from the Paris Axolotls, are not progressive,\nbut reversion forms; I believe that the Axolotls which now inhabit\nthe Mexican lakes were Amblystomas at a former geological (or better,\nzoological) epoch, but that owing to changes in their conditions of\nlife, they have reverted to the earlier perennibranchiate stage.\n\nI was undoubtedly first led to this conception by the results which\narose from my studies on the seasonal dimorphism of butterflies.[242]\nIn this case we were also concerned with the two different forms under\nwhich one and the same species appears, and of which it was shown to\nbe probable that the one is phyletically older than the other. The\nyounger summer form, according to my view, has arisen, through the\ngradual amelioration of the climate, from the winter form, which at an\nearlier zoological epoch was the only one in existence; but the latter,\nthe primary form, has not for this reason ceased to exist, but now\nalternates in each year as a winter form with the secondary summer form.\n\nNow with seasonally dimorphic butterflies, it was easily possible to\ninduce the summer brood to assume the winter form by exposing their\npup\u00e6 for a long time to a low temperature; and it was shown to be\nhighly probable that this abrupt and often very extensive change or\ntransformation, only apparently takes place suddenly, and is but the\napparent result of the action of cold upon this generation, whilst in\nfact it depends upon reversion to the primary form of the species, so\nthat the low temperature, which is only once applied, gives but the\nimpetus to reversion, and is not the true cause of the transformation.\nThis cause must rather be sought in the long continued action of the\ncold to which the ancestors of our existing butterflies were subjected\nfor thousands of generations, and of which the final result is the\nwinter form.\n\nIf we assume for an instant that my interpretation of the\ntransformation of Axolotl as just offered is correct, we should have\nconditions in many respects analagous to those of seasonal dimorphism.\nIt is true that in this case the two forms no longer alternate\nregularly with each other, but the primary form may occasionally\nappear instead of the secondary form, owing to the action of external\nconditions.\n\nJust as in the case of seasonal dimorphism it is possible to compel\nthe summer generation to abandon the summer form, and to assume the\nwinter guise by the action of cold; so in the present case we are able\nto induce the Axolotl to adopt the Amblystoma form by making a\u00ebrial\nrespiration compulsory at a certain stage of life; and further, just as\nin seasonal dimorphism it can be shown that this artificially produced\nchange is only apparently an abrupt transformation, and is actually a\nreversion to the much older winter form; so here we have not an actual,\nbut only an apparent remodelling of the species--a reversion to the\nphyletically older form.\n\nThis certainly appears a paradox, inasmuch as a form here arises by\nreversion which must yet undoubtedly rank as the more highly developed.\nI believe, however, that much which seems paradoxical in this statement\nwill disappear on further examination.\n\nIt must in the first place be taken into consideration that the\nphyletic development of species need not by any means always take place\nby advancement. We have indeed many cases of retrogressive development,\nalthough in a somewhat different sense, as with parasites and those\nforms which have degenerated from free locomotion to a sedentary mode\nof life.[243] I do not confuse this kind of retrogressive development,\narising from the arrest of certain organs and systems of organs, with\ntrue reversion. The latter is a return to a form which has already\nbeen once in existence; but in the former case, in spite of all\nsimplification of the organization, some entirely new feature always\ncomes into existence. But I am not able to see any absurdity in the\nassumption that even true reversion, whether of a whole species or of\nthe individuals of a certain district, may be regarded as possible,\nand I require no further concession. Why, for example, should it be\ninconceivable that at a very remote period the Axolotl was adapted to\na life on land; that through the direct and indirect action of changed\nconditions of life it gradually acquired the salamander form, but that\nsubsequently, through new and unfavourable changes in the conditions of\nlife, it again relapsed to the older form, or at least to one nearly\nrelated thereto?\n\nAt any rate such an assumption contains nothing opposed to known facts,\nbut can be supported in many ways, and finally it commends itself, at\nleast in my opinion, as offering the only admissible explanation of the\nfacts before us.\n\nThe existence of a whole series of species of _Amblystoma_, as already\nmentioned, at once shows that species of _Siredon_ can become elevated\ninto the salamander form, and can propagate regularly in this state,\nand further, that this phyletic advance has already actually taken\nplace in many species.\n\nThat degeneration may also occur from this high stage to a lower stage\nof development, is shown by many observations on our water-salamanders.\nIt is known that under certain circumstances Tritons, as it is\ngenerally expressed, become \u201csexually mature in the larval condition.\u201d\n\nIn the year 1864 De Filippi[244] found fifty Tritons in a pool at\nAndermatten, in the neighbourhood of Puneigen, and of these only two\nshowed the structure of the adult water-salamander; all the others\nstill possessed gills, but notwithstanding this, they agreed in both\nsexes, in size and in the development of the sexual organs, with mature\nanimals. De Filippi established that these \u201csexually mature larv\u00e6\u201d not\nonly resembled larv\u00e6 externally through the possession of gills, but\nthat they also possessed all the other anatomical characters of the\nlarv\u00e6, _i.e._ the characteristic bunches of palatine teeth situated\non both sides in the position of the subsequent single rows, and a\nvertebral column represented throughout its whole length by the _chorda\ndorsalis_.\n\nAccording to my view this would be a case of the reversion of the\nTriton to the immediately anterior phyletic stage, _i.e._ to the\nperennibranchiate stage, and in the present instance the majority\nof zoologists who take their stand by the theory of descent, would\ncertainly concur in this view. I should at least consider it to be a\nuseless play upon words did we here speak of larval reproduction, and\nthereby believe that we had explained something. The animal certainly\nbecomes sexually mature in the same condition as that in which it\nfirst appears as a larva, but we first get an insight into the nature\nof this process by considering that this so-called \u201csexually mature\nlarva\u201d has the precise structure which must have been possessed by\nthe preceding phyletic stage of the species, and that an individual\nreversion to the older phyletic stage of the species is consequently\nbefore us. I maintain that Dum\u00e9ril is in error in regarding this case\nof the Triton as parallel with the true larval reproduction of Wagner\u2019s\n_Cecidomyia_ larva. In this last case it is certainly not reversion\nto an older phyletic stage that confers the power of reproduction\nupon the larv\u00e6, since the latter do not represent an older phyletic\nstage of the species, but must have arisen contemporaneously with this\nlast stage. The enormous structural difference between the larv\u00e6 and\nthe imagines is not explained by the latter having arisen from the\nformer supplementarily as a finished production, but by both having\nbeen contemporaneously adapted to continually diverging conditions\nof life.[245] Considered phyletically, these larv\u00e6 are by no means\nnecessarily transitional to the origination of the flies. They could\nhave been quite different without the form of the imagines having\nbeen thereby modified, since the stages of insect metamorphosis vary\nindependently of each other in accordance with the conditions of life\nto which they are subjected, and exert scarcely any, or only a very\nsmall form-determining influence upon each other, as has been amply\nproved in the preceding essay. In any case the power of these larv\u00e6\n(the _Cecidomyi\u00e6_) to propagate themselves asexually was first acquired\nas a secondary character, as appears from the fact that there exist\nnumerous species of the same genus which do not \u201cnurse.\u201d In the form\nwhich they now possess they could never have played the part of the\nfinal stage of the ontogeny, nor could they formerly have possessed\nthe power of sexual reproduction.[246] In brief, we are here concerned\nwith true larval reproduction, whilst in Triton we have reversion to an\nolder phyletic stage.[247]\n\nI cannot agree with my friend Professor Haeckel when he occasionally\ndesignates the reversion of the Tritons as an \u201cadaptation\u201d to a purely\naqueous existence.[248] We could here only speak of \u201cadaptation\u201d if\nwe took the word in a quite different sense to that in which it was\nfirst introduced into science by Darwin and Wallace. These naturalists\nthereby designate a gradual bodily transformation appearing in the\ncourse of generations in correspondence with the new requirements\nof altered conditions of life or, in other words, the action of\nnatural selection, and not the result of a suddenly and direct acting\ntransforming cause exerted but once on a generation.\n\nJust because the word \u201cadaptation\u201d can be used in ordinary language in\nmany senses, it is desirable that it should have only _one_ precise\nsignification, and above all that we should not speak of adaptation\nwhere scarcely any morphological change occurs, but only a kind of\nfunctional change in the sense used by Dohrn.[249] This is the case\nfor example, when Forel[250] shows that fresh water Pulmonifera,\nthe organization of which is attributed to the direct respiration\nof air, can nevertheless become settled in the greatest depths of\nmountain lakes through their lungs being again employed as gills.\nThat not the least change in the lungs hereby takes place is shown\nby the observations of Von Siebold,[251] who saw the shallow water\nPulmonifera using their lungs alternately for direct a\u00ebrial and\naquatic respiration, according to the amount of air contained in the\nwater. If with Von Siebold we merely apply the word \u201cadaptation\u201d to\nsuch cases, this expression would lose the special sense which it\noriginally conveyed, and the word would have to be abandoned as a\n_terminus technicus_; still, such cases may perhaps be spoken of as\n_physiological_ adaptation.\n\nIn any case the reproductive \u201clarv\u00e6\u201d of the Tritons as little present\na case of true adaptation as the Axolotl, which occasionally becomes\ntransformed into an Amblystoma. In both cases the transformation\nreferred to is by no means indispensable to the life of the individual.\nMature Tritons (devoid of gills) can exist, as I have myself seen,\nfor many months, and probably also for a year in deep water, although\nadapted for purely pulmonary respiration; whilst Axolotls, as I have\nalready mentioned, can live well for a year in shallow water poor in\nair. If their gills by this means become shrivelled up or completely\ndisappear, even this is not adaptation in the Darwinian sense, but\nthe effect of directly acting external influences, and chiefly of\ndiminished use.\n\nA case entirely analagous to that of Filippi\u2019s was observed by\nJullien in 1869. Four female larv\u00e6 of _Lissotriton Punctatus_\n(Bell)--(synonymous with _Triton T\u00e6niatus_, Schnd.), taken from a\npool, proved to be sexually mature. They contained mature eggs in\ntheir ovaria ready for laying, and two of them actually deposited\neggs. Four male larv\u00e6 found in the same pool, appeared to be equally\ndeveloped with respect to size, but their testicles contained no free\nspermatozoa, but only sperm-cells.[252]\n\nI have met with a third case of a similar kind mentioned by Leydig in\nhis memoir, rich in interesting details, \u201con the tailed Amphibians\nof the Wurtemburg fauna.\u201d[253] Schreibers, the former director of\nthe Vienna Museum, also found \u201clarv\u00e6\u201d of Tritons with well-developed\ngills, but of the size of the \u201cadult male individuals,\u201d and, as shown\nby anatomical investigation, with well \u201cdeveloped sexual organs,\u201d the\novaria especially being distended with eggs.\n\nIt is thus established that species which long ago reached the\nsalamander stage in phyletic development, may occasionally degenerate\nto the perennibranchiate stage. This fact obviously makes my\nconception of the Axolotl as a reversion form appear much less\nparadoxical--indeed, the cases of reversion in Triton are precisely\nanalagous to the process which I suppose to have taken place in the\nAxolotl. We have only to substitute Amblystomas for Tritons, to\nimagine the pool in which De Filippi found his \u201csexually mature Triton\nlarv\u00e6\u201d enlarged to the size of the Lake of Mexico, and to conceive the\nunknown, and perhaps here transitory, causes of the reversion to be\npermanent, and we have all that is necessary, so far as we at present\nknow, for the restoration of the Axolotl; we obtain a perennibranchiate\npopulation of the lake.\n\nIt has not yet been determined whether the perennibranchiate form of\nthe Triton actually prevailed permanently in De Filippi\u2019s pool, since,\nso far as I know, this has not since been examined.\n\nLet us, however, assume for an instant that this is really the\ncase, and that there exists at that spot a colony of sexually\nreproductive perennibranchiate Tritons: should we wonder if a true\nTriton occasionally appeared among their progeny, or if we were able\nto induce the majority of the individuals of this brood to become\nmetamorphosed into Tritons by keeping them in shallow water? According\nto my view this is precisely the case of the Mexican Axolotl.\n\nI need not, however, restrict myself to this in order to support my\nhypothesis, but must also directly combat the view hitherto received,\nsince the latter is in contradiction with facts.\n\nDid there really exist in the Axolotl a tendency to sudden phyletic\nadvancement, then one fact would remain quite incomprehensible, viz.\nthe sterility of the Amblystomas.\n\nOut of about thirty Amblystomas obtained by Dum\u00e9ril down to the\nyear 1870, there was not one in a state of sexual maturity; neither\ncopulation nor deposition of eggs took place, and the anatomical\ninvestigation of single specimens showed that the eggs were immature,\nand that the spermatozoa, although present, were without the undulating\nmembrane characteristic of the salamanders, but were not devoid of\nall power of movement, only, as established by Quatrefages, were\n\u201cincompletely motile.\u201d[254]\n\nSo also the five Amblystomas about which I have been writing, show up\nto the present time no appearance of reproduction.\n\nThe objection raised by Sacc,[255] that the sterility of the\nAmblystomas bred from Axolotls is attributable to \u201cbad nourishment,\u201d is\nobviously of but little avail. How is it that the Axolotls, which are\nfed in a precisely similar manner, propagate so readily? Moreover, I am\nable to expressly assert that my Amblystomas were very well fed. It is\ntrue that they have as yet scarcely reached the age of two years, but\nthe Axolotl propagates freely in the second year, and some of Dum\u00e9ril\u2019s\nAmblystomas were five years old in 1870.\n\nThis fact of the sterility is strongly opposed to the idea that these\nAmblystomas are the regular precursors of the phyletically advancing\ngenus _Siredon_.[256] I will by no means assert that my theory of\nreversion actually explains the sterility, but it is at least not\ndirectly opposed to it. Mere reversion forms may die off without\npropagating themselves; but a _new form_ called forth by the action\nof a phyletic vital force should not be sterile, because this is the\nprecise \u201caim\u201d which the vital force had in view. The conception of a\nvital force comprises that of teleology.\n\nThe sterility of Amblystoma moreover, although not completely\nexplicable from our standpoint, can be shown to be a phenomenon\nnot entirely isolated. In the above mentioned case of _Lissotriton\nPunctatus_, the female \u201clarv\u00e6\u201d were certainly sexually mature and laid\neggs, but the males of the same period contained in their testicles no\nfully developed spermatozoa.\n\nOther cases of this kind are unknown to me; at the time when I made the\nexperiments with butterflies already recorded (see the first essay),\nthis point of view was remote, and I therefore neglected to examine\nthe artificially bred reversion forms with respect to their organs of\nreproduction. But general considerations lead to the supposition that\natavistic forms may easily remain sterile.\n\nDarwin[257] finds the proximate causes of sterility in the first place\nin the action of widely diverging conditions of life, and in the next\nplace in the crossing of individuals widely different in constitution.\nNow it is certainly deviating conditions of life which lead to the\nmetamorphosis of the Axolotl, and from this point of view it cannot be\nsurprising if we find those individuals sterile which show themselves\nso especially affected by these changed conditions as to revert to the\nsalamander form.\n\nBy this it is not in any way meant to be asserted that reversion\nis invariably accompanied by sterility, and one cannot raise as an\nobjection to my interpretation of the metamorphosis of the Axolotl,\nthat a reproductive colony of Axolotls could never have arisen by\nreversion. On the contrary, Jullien\u2019s egg-depositing female Triton\nlarv\u00e6 show that also with reversion the power of reproduction may be\ncompletely preserved.[258] From the above-mentioned general causes\nof sterility, it may even be inferred that fertility can be lost in\ndifferent degrees, and it can be further understood to a certain\nextent why this fertility is more completely lost by reversion\nto the Amblystoma, than by the reversion of the Triton to the\nperennibranchiate form.\n\nIf in these cases the reversion is brought about by a change in the\nconditions of life, we may perhaps suppose that the magnitude of this\nchange would determine the degree of fertility, and the preservation\nof the reversion form. Still more, however, would the fertility be\ninfluenced by the extent of the morphological difference resulting\nfrom the reversion. We know that the blending of very different\nconstitutions (_e.g._ the crossing of different species) produces\nsterility. Something similar results from the sudden reversion to a\nstage of development widely different in its whole structure. Here\nalso we have in a certain sense the union of two very different\nconstitutions in one individual--a kind of crossing.\n\nFrom this point of view it can in some measure be comprehended why\nsterility may be a result of reversion; on the other hand, we thereby\nobtain no explanation why, with the same amount of morphological\ndifference, in one case complete sterility, and in another relative\nfertility occurs. The morphological difference between Axolotl and\nAmblystoma is exactly the same as between Triton and its \u201csexually\nmature larva;\u201d the difference between the two cases of reversion\ndepends entirely upon the direction of the leap, that taken in the\nformer case being precisely opposite in direction to that taken in the\nlatter.\n\nHerein might be sought the explanation of the different strength with\nwhich the reproductive power is affected; not indeed in the direction\nof the leap itself, but in the differences in the ontogeny which\nare determined by the differences in the direction of the leap. The\nreversion of the Triton to an older phyletic stage coincides with\nthe arrest at a younger ontogenetic stage; or, in other words, the\nolder stage of the phylogeny to which reversion takes place is still\nentirely comprised in the ontogeny of each individual. Each Triton is\nperennibranchiate throughout a long period of its life; the reverting\nindividual simply reverts to the older phyletic stage by remaining at\nthe larval stage of its individual development.\n\nBut it is quite different with the reversion of the Axolotl to the\nformerly acquired, but long since abandoned Amblystoma form. This is\nnot retained in the ontogeny of Axolotl, but has been completely lost;\nfor a long series of generations--so must we suppose--the ontogeny\nhas always only attained to the perennibranchiate form. Now if at the\npresent time certain individuals were compelled to revert to the\nAmblystoma form, certainly no greater leap would have been made from\na morphological point of view, than in the reversion of Triton to the\nperennibranchiate form, but at the same time the leap would be in\nanother direction, viz. over a long series of generations back to a\nform which the species had not produced for a long period, and which\nhad to a certain extent become foreign to it. We should thus have here\nalso the grafting of a widely different constitution upon that of the\nAxolotl, or, if one prefers it, the commingling of two widely different\nconstitutions.\n\nOf course I am far from wishing to pretend that this \u201cexplanation\u201d is\nexact; it is nothing more than an attempt to point out the direction\nin which the causes affecting the reproductive powers in different\ndegrees are to be looked for. A deeper penetration into and special\ndemonstration of the manner in which these causes bring about such\nresults, must be reserved for a future period. For the present it\nmust suffice to have indicated that there is an essential distinction\nbetween the two kinds of reversion, and to have made it to some\nextent comprehensible that this distinction may be the determining\nimpulse with respect to the question of sterility. Perhaps the law\nhere concealed from us may one day be thus formulated:--Atavistic\nindividuals lose the power of reproduction the more completely, the\ngreater the number of generations of their ancestors whose ontogeny no\nlonger comprises the phyletically older stage to which the reversion\ntakes place.\n\nThe hypothesis which interprets the transformation of the Axolotl as\na case of reversion, thus holds out the possibility of our being able\nto comprehend the sterility of the Amblystomas arising in this manner,\nwhilst, on the other hand, for the adherents of a phyletic vital\nforce, not only is this observed sterility as Dum\u00e9ril expresses it \u201cun\nv\u00e9ritable \u00e9nigme scientifique,\u201d but an absolute paradox. We should\nexpect such a directive and inciting principle to call into existence\nnew forms having vitality and not destined to perish, the more so when\nit is concerned with a combination of structural characters which, when\noriginating in another manner (viz. from other species of _Siredon_),\nhave long since shown themselves to have vitality and reproductive\npower. We are indeed acquainted with species of _Amblystoma_ which\npropagate as such, and each of which arises from an Axolotl-like larva.\nThus we cannot regard the sterile Amblystomas produced by the Paris\nAxolotls as abortive attempts of a vital force--an interpretation which\nis certainly in itself already sufficiently rash.\n\nNow if it be asked what change in the conditions of life could have led\nto the reversion in the Lake of Mexico[259] of the Amblystoma to the\nSiredon form, I must admit that I can only offer a conjectural reply,\nhaving but a conditional value so long as it is not supported by a\nprecise knowledge of the conditions there obtaining, and of the habits\nboth of the Axolotl and of the Amblystoma.\n\nIt may be supposed generally that reversion is brought about by\nthe same external conditions as those which formerly produced the\nperennibranchiate stage. This supposition is in the first place\nsupported by the experiments here recorded, since it is evidently the\ninducement to a\u00ebrial respiration which causes the young Axolotl to\nrevert to the Amblystoma form, _i.e._ the inciting cause under whose\ndomineering influence the Amblystoma form must have arisen.\n\nHere again the case is quite similar to that of seasonally dimorphic\nbutterflies. Reversion of the summer brood to the winter form is there\nmost easily caused by the action of cold, _i.e._ by the same influence\nas that under whose sway the winter form was developed.\n\nWe know indeed that reversion may also arise by the crossing of\nraces and species, and I have attempted to show that reversion in\nbutterflies may also be brought about by other influences than cold;\nbut still the most probable supposition obviously is, that reversion\nwould be caused by the persistent action of the same influences as\nthose which in a certain sense created the perennibranchiate form.\nThat the latter was produced under the influence of an aquatic life\nthere can be no doubt, and thus, in accordance with my supposition, the\nhypothetical _Amblystoma Mexicanum_, the supposed ancestral form of\nthe Axolotl of the Mexican Lake, might have been caused to revert to\nthe perennibranchiate form by a reduction in the possibilities of its\nliving upon land, and by its being compelled to frequent the water.\n\nI will not here return to the consideration of every other opinion\n_ab initio_. It is very advisable to distinguish between the mere\nimpulses which are able to produce sudden reversion, and between\nactual transforming causes which result directly or indirectly in the\nremodelling of a species. Thus, it is conceivable _\u00e0 priori_ that\nreversion may occur by the action of an inciting cause having nothing\nto do with the origin of the phyletically older form. Temperature\ncan certainly have played no part, or only a very small part, in the\nformation of the perennibranchiate form; nevertheless cold may well\nhave been one of the inciting causes which induced the Amblystoma at\none time to revert to the _Siredon_ form, and we cannot at present\nconsider De Saussure to be incorrect when he maintains that the low\ntemperature of the Mexican winter might prevent that transformation\n(of the Axolotl into the Amblystoma) which would occur \u201cin the warm\nreptile-house\u201d of the Jardin des Plantes. He supports this view by\nstating that \u201cTschudi has found the Amblystoma\u201d (of course another\nspecies) \u201cin the hottest parts of the United States.\u201d \u201cOn the Mexican\nplateau, however, it snows every winter, and if the lake does not\nactually freeze, its temperature must fall very considerably in the\nshallowest parts.\u201d\n\nBut although this view is not opposed by any theoretical\nconsiderations, I still hold it to be incorrect. I doubt whether it\nis temperature that has brought about the reverse transformation of\nthe Amblystoma into the Axolotl, or which, according to De Saussure\u2019s\nconception, at the present time prevents the transformation of the\nAxolotl in the Lake of Mexico. I doubt this because Amblystomas are now\nknown from all parts of the United States as far north as New York, a\nproof that a winter cold considerably greater than that of the Mexican\nplateau is no hindrance to the metamorphosis of the Axolotl, and that\nthe genus does not show itself to be in this respect more sensitive\nthan our native genera of _Salamandrid\u00e6_.\n\nThe following observations of De Saussure, in which he calls attention\nto the nature of the Mexican Lake, appear to me to be more worthy of\nconsideration:--\u201cThe bottom of this lake is shallow, and one passes\nimperceptibly from the lake into extensive marshy regions before\nreaching solid ground; perhaps this circumstance makes the Axolotl\nincapable of reaching dry land, and prevents the transformation.\u201d\n\nIn any case the Lake of Mexico offers very peculiar conditions for\nAmphibian life. My esteemed friend Dr. v. Frantzius has called my\nattention to the fact that this lake--as well as many other Mexican\nlakes--is slightly saline. At the time of the conquest of Mexico by\nFerdinand Cortez, this circumstance led to the final surrender of the\ncity, as the Spaniards cut off the supply of water to the besieged, and\nthe water of the lake is undrinkable. The ancient Mexicans had laid\ndown water-conduits from the distant mountains, and the city is still\nsupplied with water brought through conduits.\n\nNow this saltness cannot in itself be the cause of the degeneration\nto the perennibranchiate form, but it may well be so in combination\nwith other pecularities of the lake. The narrowest part of the lake\nis the eastern, and it is only in this part that the Axolotl lives.\nNow in winter, violent easterly gales rush down from the mountains and\nblow continuously, driving the water before them to such an extent\nthat it becomes heaped up in the western portion of the lake, where it\nfrequently causes floods, whilst 2000 feet of the shallow eastern shore\nare often laid completely dry.[260]\n\nNow if we consider these two peculiarities, viz. salineness and\nperiodical drying up of a part of the bottom of the lake through\ncontinuous gales, we certainly have for the Axolotl, conditions of\nlife which are only to be found in few species. One might certainly\nattempt to apply these facts in a quite opposite sense, and to regard\nthem as unfavourable to my theory, since the retreat of the water\nfrom a great portion of the bottom of the lake would--so one might\nthink--rather facilitate transition to a life upon land, and indeed\ncompel the adoption of such a mode of existence. But we should thus\nforget that the exposed bottom of the lake is a sterile surface without\nfood or place of concealment, and, above all, without vegetation;\nand further, that owing to the considerable salineness of the water\n(specific gravity = 1.0215),[261] the whole of the exposed surface must\nbe incrusted with salt, a circumstance which would render it quite\nimpossible for the creatures to feed upon land. Sodic chloride and\ncarbonate are dissolved in the water in such considerable quantities,\nthat they are regularly deposited upon the shores of the lake as a\ncrust, which is collected during the dry season of the year and sent\ninto the market under the name of \u201ctequisquite\u201d (M\u00fchlenpfordt).[262]\n\nThus the supposition is not wanting in support, that peculiar\nconditions make it more difficult for the creature to obtain its food\nupon land than in the water, and this alone may have been sufficient\nto have induced it to acquire the habits of a purely aquatic existence,\nand thus to revert to the perennibranchiate or Ichthyodeous form.\n\nBut enough of supposition. We must not complain that we are unable\nfrom afar to discover with precision the causes which compelled the\nAxolotl to abandon the Amblystoma stage, as long as we are not able to\nexplain the much nearer cases of reversion in Filippi\u2019s and Jullien\u2019s\nTritons; nevertheless, in these cases also, the causes affecting the\nwhole colony of Tritons must be general, since--at least in the case\nnoticed by Filippi--the greater majority of the individuals remained\nin the larval condition. Experiments with Triton larv\u00e6 could throw\ngreater light upon this subject; it would have in the first place to be\nestablished whether reversion could be artificially induced, and if so,\nby what influences.\n\nFrom the previously mentioned experiments with butterflies, as well\nas from the results obtained with Axolotls, we should expect that\nin Tritons, reversion to the Ichthyodeous form would take place if\nwe allowed the inciting cause, viz. the bathing of the gills and of\nthe whole body with water, to act persistently, and at the same time\nwithheld that influence under whose action the salamander form became\ndeveloped, viz. the bathing of the gills, the skin, and the surfaces of\nthe lungs with air.\n\nOld experiments of this kind are to be met with, but they were never\ncarried on for a sufficient time to entirely allay the suspicion, that\nthe specimens concerned would perhaps have undergone the ordinary\nmetamorphosis if their existence had been prolonged.\n\nThus, Schreibers[263] relates that \u201cby confining tadpoles of the\nsalamander found at large in their last stage of growth, under water by\nmeans of an arrangement (net?), and feeding them with finely chopped\nearthworms, he was able to keep them for several months--and indeed\nthroughout the winter--in this condition, and in this way to forcibly\ndefer their final change, and their transition from the tadpole stage\nto that of the perfected creature during this period.\u201d It is not stated\nwhether the animals finally underwent transformation, so that it cannot\nbe decided whether we have here a case of reversion or simply one of\nretarded development. That metamorphosis may occur after a long period\nof time, is shown by experiments upon the tadpole of _Pelobates_\nconducted by Professor Langer in Vienna.[264] The creatures were kept\nin deep water in such a manner that they were not able to land, and\nby this means three out of a large number of individuals had their\nmetamorphosis delayed till the second summer; notwithstanding this,\ntransformation then occurred.\n\nIt cannot be objected to my reversion hypothesis, that it opposes on\nthe one side what on the other it postulates, viz. a _per saltum_\nchange of structure. Reversion is characterized by the sudden\nacquisition of an older, _i.e._ a formerly existing phyletic stage.\nThat reversion occurs is a fact, whilst nobody has hitherto been able\nto prove, or even to make probable, that a stage of the future (_sit\nvenia verbo_) has been attained at once (_per saltum_).\n\nNow if it is possible to find influences in the present conditions\nof life of the Axolotl which make it difficult or quite impossible\nfor it to live upon land, and which therefore appear as incentives\nto the reversion to the Ichthyodeous form, the other portion of my\nhypothesis--the assumption that the Axolotl had become an Amblystoma at\na former period--can also be supported by facts.\n\nWe know from Humboldt[265] that the level of the Lake of Mexico at a\ncomparatively recent period was considerably higher than at present. We\nknow further that the Mexican plateau was covered with forest, which\nhas now been destroyed wherever there are human, and especially Spanish\nsettlements. Now if we suppose that at some post-glacial period the\nmountain forests extended to the borders of the lake, at that time\ndeep, with precipitous sides and much less saline, not only should we\nthus have presented different conditions of life to those at present\nexisting, but also such as would be most favourable for the development\nof a species of salamander.\n\nOn the whole, I believe that my attempt to explain the exceptional\nmetamorphosis of the Axolotl of the Mexican lake cannot be objected\nto as being a too airy phantasy. In any case it is the only possible\nexplanation which can be opposed to that which supposes that the\noccasional transformation of the Axolotl is not reversion, but an\nattempt at advancement. This last assumption must, in my judgment, be\nrejected on purely theoretical grounds by those who hold that a sudden\ntransformation of a species, when connected with adaptation to new\nconditions of life, is inconceivable--by those who regard adaptation,\nnot as the sudden work of a magic power, but as the end result of a\nlong series of natural, although minute and imperceptible causes.\n\nIf my interpretation of the facts be correct, there arises certain\nconsequences which I may here briefly mention in conclusion.\n\nFirst, with regard to more obvious results. If _Siredon Mexicanus_,\nShaw, only by occasional reversion assumes the Amblystoma form,\nand never, or only exceptionally, propagates as such, but only\nas _Siredon_, the more recent systematists are not justified in\nstriking out the genus _Siredon_ and in placing _S. Mexicanus_ as an\nundeveloped form in the genus _Amblystoma_. So long as there exists\nnot one only, but several species of _Siredon_ which as such regularly\npropagate themselves, the genus exists; and although we would not\ndeprive systematists of all hope of these species of _Siredon_ being\none day re-elevated to _Amblystom\u00e6_, it nevertheless better accords\nwith the actually existing state of affairs if we allow the genus\n_Siredon_ to remain as before among the genera of _Salamandrina_, and\nto include therein all those species which, like the Paris Axolotl,\n_S. Mexicanus_, Shaw, and probably also _S. Lichenoides_, Baird, only\nexceptionally, or through artificial influences, assume the Amblystoma\nform, but without propagating regularly in this condition. On the other\nhand, we should correctly comprise under the genus _Amblystoma_ all\nthose species which propagate in this state regularly, and in which the\nperennibranchiate stage occurs only as a larval condition.\n\nTo arrive at a decision in single cases would chiefly concern the\nAmerican naturalists, whose ever increasing activity may lead us\nto hope soon for a closer investigation of the reproduction of the\nnumerous species of _Amblystoma_ of their native country. I should\nrejoice if the facts and arguments which I have here offered should\ngive an impetus to such researches.\n\nThe second consequence to which I may refer, is of a purely\ntheoretical nature, and concerns a corollary to the \u201cfundamental\nbiogenetic law\u201d first enunciated by Fritz M\u00fcller and Haeckel. This, as\nis well known, consists of the following law:--The ontogeny comprises\nthe phylogeny, more or less compressed and more or less modified.\n\nNow according to this law, each step in phyletic development when\nreplaced by a later one, must remain preserved in the ontogeny, and\nmust therefore appear at the present time as an ontogenetic stage\nin the development of each individual. But my interpretation of the\ntransformation of the Axolotl appears to stand in contradiction to\nthis, since the Axolotl, which at a former period was an Amblystoma,\nretains nothing of the latter in its ontogeny. The contradiction is,\nhowever, only apparent. As long as we are concerned with an actual\nadvance in development, and therefore with the attainment of a new step\nnever formerly reached, the older stages will be found in the ontogeny.\nBut this is not the case when the new stage is not an actual novelty,\nbut formerly represented the final stage of the individual development;\nor, in other words, when we are concerned with the reversion, not\nof single individuals, but of the species as such, to the preceding\nphyletic stage, _i.e._ with a phyletic degeneration of the species.\nIn this case the former end-stage of the ontogeny would be simply\neliminated, and we should then only be able to recognize its former\nexistence by its occasional appearance in a reversion form. Thus, under\ncertain conditions the Triton sinks back to the perennibranchiate\nstage; not in such a manner that the individual first becomes a Triton\nand then undergoes perennibranchiate re-modification, but simply, as I\nhave already shown above, by its remaining at the Ichthyodeous stage\nand no longer attaining to the Salamander form. So also, according\nto my hypothesis, the salamandrine _Amblystoma Mexicanum_, formerly\ninhabiting the shores of the Lake of Mexico, has degenerated to the\nperennibranchiate stage, and the only trace that remains to us of its\nformer developmental status is the tendency, more or less retained\nin each individual, to again ascend to the salamander stage under\nfavourable conditions.\n\nThe third and last consequence which my interpretation of the facts\nentails, is the change in the part played by reversion in organic\nnature. Whilst atavistic forms have hitherto been known only as\nisolated and exceptional cases, interesting indeed in the highest\ndegree, but devoid of significance in the course of the development\nof organic nature, a real importance in this last respect must now be\nattached to them.\n\nI may assume that reversion can in two ways be effectual for the\npreservation or re-establishment of a living form. In the first\nplace, where, as in Axolotl, the new and organically higher form\nbecomes untenable through external influences, instead of simply\nperishing--since advancement in another direction does not appear to\nbe possible--a reversion of the species to the older and more lowly\norganized stage occurs. In the second place, the older phyletic form\nmay not be abandoned while a newer form is being developed therefrom,\nbut the former may alternate with the latter, as we see in the case\nof seasonally dimorphic butterflies. It can hardly be objected if I\nregard the alternation of the summer and winter form in this case as a\nperiodic reversion to the phyletically older (winter) form.\n\nAlthough the reversion of an entire species, such as I suppose to have\nbeen the case with the Axolotl, may be of rare occurrence, this is\ncertainly not the case with periodic or cyclical reversion; the latter\nplays a very important part in the development of the various forms of\nalternating or cyclical propagation.[266]\n\n\nPOSTSCRIPT.\n\nIn the previous portion of this essay it was pointed out that the\ncauses to which I attributed the reversion of the hypothetical\n_Amblystoma Mexicanum_ to the existing Axolotl, did not appear to me\nto amount to a complete explanation of the phenomenon. In the first\nplace these seemed to me too local, since they could only be applied\nwith any certainty to the Axolotl of the lake of the Mexican capital,\nwhilst the Paris Axolotls obtained from other parts of Mexico still\nrequired an explanation. On the other hand, these causes did not appear\nto me sufficiently cogent. Should we even learn subsequently that the\nParis Axolotl is also derived from a salt lake which is exposed to\nsimilar winds to the Lake of Mexico, we still have in this peculiarity\nof the lakes only a cause tending to make it difficult for the larva\nto undergo metamorphosis, and to reach a suitable new habitat on the\nland. The impossibility of doing this, or the complete absence of such\nhabitat, does not however follow as a necessary consequence.\n\nIt would obviously be a much more solid support for my hypothesis if it\nwere possible to point to some physical conditions of the land which\nthere precluded the possibility of the existence of Amblystomas.\n\nFor a long time I was indeed unable to discover such causes, and I\ntherefore concluded the previous portion of this essay and went to\npress. Afterwards, when residing in one of the highest valleys of our\nAlps in the Upper Engadine, an idea accidentally occurred to me, which\nI do not now hesitate to regard as correct after having tested it by\nknown facts.\n\nIt happens that in the Upper Engadine there live only such Amphibia\nas persistently, or at least frequently resort to the water. I found\nfrogs up to nearly 7000 feet above the sea, and Tritons at 6000 feet\n(Pontresina and Upper Samaden). On the other hand, the land-living\nmountain salamander, _S. Atra_,[267] was absent, although suitable\nstations for this species were everywhere present, and it would have\nwanted for food as little as do its allies the water-newts. Neither\nwould the great elevation above the sea offer any obstacle to its\noccurrence, since it occasionally ascends to a height of 3000 metres\n(Fatiot).[268]\n\nNow it is well known that the atmosphere of the Upper Engadine,[269]\nlike that of other elevated Alpine valleys enclosed by extensive\nglaciers, is often extraordinarily dry for a long period, a condition\nwhich appears to me to explain why the black land-salamander is there\nabsent,[270] whilst its near water-living ally occurs in large numbers.\nThe skin of the naked Amphibia generally requires moisture, or else\nit dries up, and the creature is deprived of a necessary breathing\napparatus, and often dies as rapidly as though some important internal\norgan had been removed. Decapitated frogs hop about for a long time,\nbut a frog which escapes from a conservatory and wanders about for one\nnight in the dry air of a room, is found the following day with dry and\ndusty skin half dead in some nook, and perhaps perishes in the course\nof another day if left without moisture.\n\nAll that we know of the biology of the Amphibia is in accordance with\nthis. Thus, all the land-salamanders of southern Italy avoid the hot\nand dry air of summer by burying in the ground, where they undergo\na summer sleep. This is the case with the interesting _Salamandrina\nPerspicillata_,[271] and with the land-living Sardinian Triton, the\nremarkable _Euproctus Rusconii_, Gen\u00e9,[272] (_Triton Platycephalus_,\nSchreiber). With respect to _Geotriton Fuscus_ I learn from Dr.\nWiedersheim, who has studied the life conditions of this, the lowest\nEuropean Urodelan, in its own habitat, that in Sardinia it sleeps\nuninterruptedly from June till the winter; whilst on the coast of\nSpezia and at Carrara, where it also occurs, it avoids the summer sleep\nin a very peculiar manner. It makes use of the numerous holes in the\ncalcareous formation of that region, and for some months in the year\nbecomes a cave-dweller. As soon as the great heat occurs, often in May,\nit withdraws into the holes, and again emerges in November during the\nwet weather. In these lurking holes it does not fall into a sleep, but\nis found quite active, and its stomach, filled chiefly with scorpions,\nshows that it goes successfully in search of food; the moist air of the\nholes makes it unnecessary for it to bury in the earth.\n\nIn the same sense it appears to me must be conceived the fact\nthat the solitary species of frog of the Upper Engadine, _Rana\nTemporaria_,[273] the brown grass frog, is there much more a frequenter\nof the water than in the plains. It is true that I can find no remark\nto this effect in the excellent work of Fatiot, already referred to\nabove, and I am therefore obliged to resort to my own observations,\nwhich, although often repeated, have always been carried on for only a\nshort time. I was much struck with the circumstance that the Engadine\nfrogs were to be found in numbers in the water long after the pairing\nseason, which, according to Fatiot, lasts at most to the end of June.\nIn the numerous pools around Samaden I found them in July and August,\nwhilst in the plains they only take to the water at the time of\nreproduction, and seek winter quarters in the mud on the first arrival\nof this season. (Fatiot, p. 321.) In the Engadine they have therefore\nin some measure adopted the mode of life of the aquatic frogs, but this\nof course does not prevent them from returning in damp weather to their\nold habits and roving through meadows and woods.\n\nAfter these considerations had made it appear to me very probable that\nthe dry air of the Upper Engadine accounted for the absence of the\nblack land-salamander, the question at once arose whether the absence\nof Amblystomas from the Mexican plateau might not perhaps be due to\nthe same cause, _i.e._ whether such a dryness of the atmosphere might\nnot perhaps prevail also in that region, so that Amphibia, or at least\nsalamander-like Amphibia, could not long exist on the land. The height\nabove the sea is still greater (7000 to 8000 feet), and the tropical\nsun would more rapidly dessicate everything in a country poor in water.\n\nAs I was at the time without any books that might have enlightened\nme on the meteorological conditions of Mexico, I wrote to Dr. v.\nFrantzius, who, by many years residence in Central America was familiar\nwith the climate of this region, and solicited his opinion. I received\nthe reply that on the high plains of Mexico an extraordinary dryness\nof the atmosphere certainly prevails. \u201cThe main cause of the dryness\nof the high plains is to be found in the geographical position,\nthe configuration of the land, and the physical structure. The\nnorth-eastern trade-wind drives the clouds against the mountains, on\nthe summits of which they deposit their moisture, so that no vapour is\ncarried over; as long as the north-east trade-wind blows, the streams\nfeeding the rivers flowing into the Atlantic Ocean are abundantly fed\nwith water, whilst on the western slopes, and especially on the high\nplains, the clouds give no precipitation. In the second half of the\nyear also, during our summer, the so-called rainy season brings but\nlittle rain[274]--little in comparison with the more southern regions,\nwhere the heavy tropical thunderstorms daily deluge the earth with\nwater. Mexico lies much too northerly, and does not reach the zone of\ncalms, within which region these tropical rains are met with.\u201d\n\nThus, in the high degree of dryness of the air lasting throughout the\nyear, I do not doubt that we have the chief cause why no Amblystomas\noccur on these elevated plains; they simply cannot exist, and would\nbecome dried up if taken there, supposing them not to be able to change\ntheir mode of life and to take to the water. If therefore in former\ntimes Amblystomas inhabited Mexico, the coming on of the existing\nclimatic conditions left them only the alternative of becoming extinct,\nor of again taking to the aquatic life of their Ichthyodeous ancestors.\nThat this was not _directly_ possible--that the Amblystoma form was not\nable to become aquatic without a change of structure, is shown by the\nfact that even in the Lake of Mexico no Amblystoma occurs. A retreat to\nan aqueous existence could, as it appears, only be effected by complete\nreversion to the Ichthyodeous form, which then also took place.\n\nBut my hypothesis of the transformation of the Axolotl not only\nrequires the proof that Amblystomas cannot exist under present\nconditions in Mexico, but also the further demonstration that at a\nformer period other conditions prevailed there, and these of such a\nnature as to make the existence of land-salamanders possible.\n\nWith respect to my question, whether we might not perhaps assume that\nat some post-glacial period the conditions of atmospheric moisture on\nthe high plains of Mexico were essentially different from those at\npresent prevailing, I recollected Dr. v. Frantzius and the above-quoted\nobservation of Humboldt\u2019s,[275] who discovered in the neighbourhood\nof the Lake of Tezenco (Mexico) distinct evidence of a much higher\nformer level of the water. \u201cAll such elevated plains were certainly at\na former period so many extensive water-basins, which gradually became\nfilled, and are still filling up with detritus. The evaporation from\nsuch large surfaces of water must at that time have caused a very moist\natmosphere, favourable to vegetation and adapted for the life of naked\nAmphibia.\u201d\n\nFrom this side also my hypothesis thus receives support, and we may\nassume with some certainty that at the beginning of the diluvial\nperiod[276] the woods surrounding the Mexican lakes were inhabited\nby Amblystomas, which, as the lakes subsequently became more and more\ndried up and the air continually lost moisture, found it more difficult\nto exist on the land. They would at length have completely died out,\nhad they not again become aquatic by reversion to the Ichthyodeous\nform. It may perhaps be supposed that the above-mentioned physical\nconditions--desolate, salt-incrusted shores--co-operated in the\nproduction of the reversion, by making it difficult for the larv\u00e6 to\nquit the water; but we can only judge with certainty upon this point\nwhen, by means of experiment, we have discovered the causes which\nproduce reversion in the Amphibia.\n\n\nADDENDUM.\n\nI have lately met with another interesting notice on the reproduction\nof the native North American Amblystomas. Professor Spence F. Baird, of\nWashington, has often observed the development from the egg of various\nspecies, and especially of _Amblystoma Punctatum_ and _A. Fasciatum_.\nHis observations do not appear to be as yet published, so that I was\nunable to discover any account of the development of _Amblystoma_ in\nexisting literature.[277] I am authorized to extract the following\nbrief data from a letter addressed to Dr. v. Frantzius.\n\nIn order to deposit their eggs the Amblystomas go into the water,\nwhere the eggs are laid enclosed in a jelly-like mass, but never more\nthan fifteen to twenty together. The spherical eggs are very large,\nperhaps a quarter of an inch in diameter. They soon develop into a\n_Siredon_-like larva, which remains several months in this condition.\nThe gills then shrivel up, the creature begins to crawl, and gradually\npasses through the different transformations to the complete Amblystoma\nform.\n\nIt appears from this communication that the Amblystomas lay much larger\nand much fewer eggs than the Axolotl, and that their development\nthroughout resembles that of our salamanders.\n\nIn concluding I may mention an anatomical fact which most strongly\nsupports my view that the Mexican Axolotl is a reverted Amblystoma.\nI learn from Dr. Wiedersheim that the Axolotl possesses the\n\u201cintermaxillary gland\u201d which occurs in all the land Amphibia. This\norgan, lying in the intermaxillary cavity, appears, whenever it occurs,\nto produce a kind of birdlime, _i.e._ a very glutinous secretion,\nwhich serves to attach the prey to the rapidly protrusible tongue.\nAlthough this secretion may perhaps also have another function, from\nthe absence of the intermaxillary gland in all exclusively aquatic\nAmphibia, it follows that it must be devoid of importance for, and\ninapplicable to feeding in the water. The intermaxillary gland is\nabsent in all _Perennibranchiata_ and _Derotremata_ which Wiedersheim\nhas hitherto investigated, viz. in _Menobranchus_, _Proteus_,\n_Siren_, _Cryptobranchus_, _Amphiuma_, and _Menopoma_, all of which\nare indeed without the cavity in which the gland is situated in the\n_Salamandrina_, _i.e._ the _cavum intermaxillare_.\n\nNow in the _Salamandrina_ the gland appears at an early stage. It is\npossessed in a well-developed state by the larv\u00e6 both of species of\n_Triton_ and of _Amblystoma_, where indeed the glandular structure\ncompletely fills the _cavum intermaxillare_.\n\nWere the Axolotl a species retarded in phyletic development,\nthe presence of a gland which does not occur in any other\n_Perennibranchiata_, and which is only of use for life upon land, would\nbe quite inexplicable.\n\nThe matter becomes still more enigmatical through the fact that\nthe gland, although present, is quite rudimentary. Whilst in the\n_Salamandrina_ the capacious intermaxillary cavity is entirely filled\nby the tubes of the gland in question, in Axolotl this cavity is almost\ncompletely filled with a closely woven connective tissue, in which\nthere can only be found a small number of gland-tubes--in the extreme\nfront, and at the base immediately over the intermaxillary teeth--these\ntubes agreeing in the details of their histological structure with the\nelements of the same gland in the _Salamandrid\u00e6_.\n\nI give these anatomical details from Dr. Wiedersheim\u2019s verbal\ncommunication. An amplified account will subsequently appear in another\nplace.[278]\n\nAn explanation of this rudimentary intermaxillary gland in the Axolotl\nonly appears to me possible on the supposition that the latter is an\natavistic form. From this point of view it is evident that the gland\nalready present in all _Amblystoma_-larv\u00e6 must have been taken over\nby the perennibranchiate form of the existing Axolotl, through the\nreversion of the hypothetical _Amblystoma Mexicanum_ of the \u201cdiluvial\nperiod.\u201d[279] It can also be easily understood that this organ would\nbecome more and more rudimentary in the course of time, since it has\nno further use in the water, and the gap thus arising in the formerly\npresent _cavum intermaxillare_ would become filled with connective\ntissue.\n\nWhile the German edition of this work was going through the press\nI obtained, through the kindness of my friend Dr. Emil Bessels of\nWashington, the Mexican memoir upon the new Axolotl,[280] which even in\nMexico regularly, or at least in many cases, becomes developed into the\nAmblystoma form.\n\nThe facts are briefly as follows:--The small Lake of Santa Isabel\nis some hours\u2019 journey from the Mexican capital. In this lake there\nlives a species of Axolotl which had hitherto remained unknown, and\nwas described by Se\u00f1or Velasco as _Siredon Tigrinus_. This species\npropagates itself indeed in the Axolotl state, but in many cases it\nbecomes transformed into Amblystoma and takes to the land. Although\npropagation in the Amblystoma condition was not observed, it can hardly\nbe doubted that it also propagates in this form.\n\nAt first sight these facts appear to refute my hypothesis, that\nthe extreme dryness of the air of the Mexican plateau precludes\nthe existence of land Amphibia. Nevertheless I do not abandon this\nhypothesis for the former one, since a closer study of the data\nfurnished by Velasco confirms rather than refutes my supposition.\n\nVelasco expressly corroborates the statement that the Axolotl hitherto\nknown from the great Mexican lake which never dries up (Lake of\nXochimilco and Chalco), is only met with in its native habitat in the\n_Siredon_ form, _i.e._ as _Siredon Humboldtii_. According to Velasco\nthe cause of the frequent assumption of the Amblystoma form by the new\n_Siredon Tigrinus_, is to be found in the local conditions of life of\nthis species. The Lake Santa Isabel is shallow, its greatest depth\namounting to three meters, and it is liable to a periodical drying up,\nwhich is so complete that one can pass dry-shod through it in several\nplaces. The species must therefore have long since died out had it\nnot been able to adapt itself periodically to a land life. Now it\ncould have become transformed into a land Amphibian--as Se\u00f1or Velasco\nobserved--at various stages of growth; and indeed this author believes\nthat \u201cthe Creator has implanted an instinct in this creature,\u201d which\nenables it to always undergo metamorphosis at the right time.\n\nThis last assumption may or may not be taken as correct, but this much\nis established, viz. that numerous individuals of this species take to\nthe land, and remain there during a period of many months.\n\nBut does this contain the proof that salamander-like animals are\nactually able to lead a land life in Mexico--that the dry air is\nadvantageous, or at least supportable to them? It does not appear so\nto me, but rather that all which has been reported of this Amblystoma\nby Se\u00f1or Velasco goes to show that the animal does not, properly\nspeaking, live upon land like the North American Amblystomas, or like\nour land-salamanders, but that _it only experiences a summer sleep\nlasting over the period of drought_. These Amblystomas were observed\nas they left the dried-up lake at night in order to seek some moist\nlurking-place in the neighbourhood, where they might remain concealed.\nThey are only known in the villages situated near the lake, and were\nonly seen there at large just when they were wandering from the lake\nto their place of concealment. At other times they were mostly found\nin the earth, buried under walls, the pavement of the market-place,\n&c. When laying down a line of railway, a workman found in the earth\na whole nest of twelve Amblystomas lying close together. All these\nare not mere lurking-holes which could be abandoned at any moment;\nit would rather appear that we have here places of refuge for the\nentire duration of the period of drought, and that these would only\nbe forsaken when the water of the rainy season penetrated the soil.\nI am not myself in a favourable position for investigating these\nsuppositions more closely, but this could be done by Se\u00f1or Velasco,\nwho lives in Mexico, and science would be much indebted to him if he\nwould examine as precisely as possible into the habits and conditions\nof life of this, and of the other species of Mexican Axolotls.\nUnfortunately this gentleman can, it would appear, have seen only the\nFrench publications upon the transformation of the Axolotl, and could\nnot therefore have asked himself questions arising from my conception\nof the facts; otherwise many of his observations would have led to more\ndefinite results. The above conclusion can however be still further\nsupported by Se\u00f1or Velasco\u2019s data.\n\nOne might indeed insist that with us also the land-salamanders conceal\nthemselves in moist places during dry weather, and often lie hidden,\nas in Mexico, in a hole, in a cluster of as many as ten together;\nbut with us they leave their lurking-place from time to time and go\nin search of food. Se\u00f1or Velasco mentions nothing with respect to\nthis. What especially struck me was the statement that the Mexican\nAmblystomas _were also to be found in the water_.[281] When Lake Santa\nIsabel is drained, the fishermen stretch large nets across the exit\nchannels, and in these they not only find ordinary Axolotls, but also\nsome \u201c_sin aretes_,\u201d which they also designate \u201c_mochos_,\u201d _i.e._\nhornless Axolotls, because they have no gills, but have already reached\nthe Amblystoma stage. Our land-salamanders live in the water only\nas larv\u00e6, but they also love and require moisture. Only the female\nenters the water when she wants to deposit her young (eggs with mature\nlarv\u00e6), and then only at the margin of shallow pools or small brooks.\nThe Mexican Amblystoma thus much more resembles in its habits our\nwater-salamanders (Tritons), which remain in the water at least during\nthe whole period of reproduction. These also leave the water later,\nand, like the land-salamander, seek concealment in the earth. They have\nthis habit also in those districts which possess a very dry atmosphere;\nand especially in the Engadine, where I first conceived the idea of\ntaking into account the dryness of the air, I found in the pools at the\nend of August and the beginning of September only larv\u00e6 of Tritons. The\nolder Amphibians must therefore have been on the land, presumably in\ntheir places of winter concealment.\n\nFrom what we have hitherto learnt from Se\u00f1or Velasco, the mode of\nlife of _Amblystoma Tigrinum_ must resemble that of our Tritons,\nalthough its structure is that of a land-salamander. I would thus\noffer the following explanation of the facts at present known:--Owing\nto the periodic drying up of the lake of Santa Isabel, the _Siredon\nTigrinus_ would be again compelled to undergo metamorphosis. Whether\nthis was formerly entirely abandoned, or whether it always occurred\nin solitary individuals, is almost immaterial; in any case the habit\nof metamorphosis must have been very rapidly acquired through natural\nselection, and must have again become general, if the faculty was only\npresent in the species, although latent. Through the dryness of the\nair, the Amblystomas that had taken to the land would be compelled to\nbury themselves at once, and to remain asleep till the recurrence of\nthe rainy season, when they would hasten back into the water and would\nthere live as a species of Triton.\n\nNow one might feel inclined to ask why the species of the great Mexican\nlake has not also taken to this mode of life. To this it may be simply\nreplied that the water of this lake never dries up, and that the\nAxolotls have thus never been reduced to the alternative of undergoing\nmetamorphosis or of perishing. If therefore the conditions of existence\nin water were more favourable than on land, the tendency to abandon\nmetamorphosis would increase from generation to generation, and the\ndeportment at present observed would finally result, _i.e._ propagation\nwould take place exclusively in the Axolotl state. As has already been\nmentioned above, the latest observations of Velasco furnish further\nconfirmation that the Axolotl of the great lake is never met with in\nthe Amblystoma condition, \u201calthough it (the Axolotl) is brought daily\nfrom Mexico into the market throughout the whole year.\u201d I should not\nhowever regard it as a refutation of my view if prolonged investigation\nshould show that this species also (_Siredon Humboldtii_) occasionally\ndeveloped into an Amblystoma; on the contrary, it would not at all\nsurprise me if such cases of reversion occurred in Mexico as well as in\nEurope. The fact that an immense majority of the Amphibians propagate\nin the Axolotl state would not be thereby affected, and would still\nrequire an explanation: this I am still inclined to see in the dryness\nof the air of the high plains, which is so unfavourably adapted for a\nlife passed entirely on land.\n\n\n\n\n=IV.=\n\nON THE MECHANICAL CONCEPTION OF NATURE.\n\n\n\n\nINTRODUCTION.\n\n\nIn the first of the three preceding essays it was attempted to solve\nthe question whether the transformations of a given complex of\ncharacters in a certain systematic group could be completely explained\nby the sole aid of Darwinian principles. It was attempted to trace\nthe origin of the marking and colouring of the Sphinx-caterpillars to\nindividual variability, to the influences of the environment, and to\nthe laws of correlation acting within the organism. These principles\nas applied to the origin of a certain well-defined, although narrowly\nrestricted range of forms, were tested in order to see whether they\nwere alone sufficient to explain the transformation of the forms.\n\nIt appeared that this was certainly the case. In all instances, or\nat least where the facts necessary to obtain a complete insight were\navailable, the transformations could be traced to these known factors;\nthere remained no inexplicable residual phenomena, and we therefore had\nno reason for inferring the existence of some still unknown modifying\ncause lying concealed in the organism. In this region of the marking\nand colouring of caterpillars, the assumption of a phyletic vital force\nhad to be abandoned, as being superfluous for the explanation of the\nfacts.\n\nIn the second essay the attempt was next made with reference to double\nform-relationship, as presented for observation in metamorphic insects,\nto draw conclusions as to the causes of the transformations. It\nappeared here that form- and blood-relationship do not always coincide,\nsince the larv\u00e6 of a species, genus, or family, &c., may show quite\ndifferent form-relationships to their imagines. These facts alone told\nvery decisively against the existence of an internal developmental\npower, so that the latter had likewise to be set aside by the method\nof elimination, since the observed incongruences as well as the\ncongruences of form-relationship, found sufficient explanation in the\naction of the environment on the organism.\n\nThis investigation thus also led to the denial of a phyletic vital\nforce.\n\nIn the third essay I finally sought to prove that the only case\nof transformation of one species into another at present actually\nobserved[282], could not without further evidence be interpreted as\nthe result of the action of a phyletic vital force, but that more\nprobably we had here only an apparent case of new formation, which was\nin reality but a reversion to a stage formerly in existence.\n\nIf this last investigation removes the only certain observation which\ncould have been adduced in favour of the hypothesis of a phyletic vital\nforce, so also do the two former essays show that this hypothesis, at\nleast in the case of insects, must be abandoned as inadequate.\n\nThe question now arises whether this conclusion, based on such a\nlimited range of inquiry, can also be applied to the other groups of\nthe organic world without further evidence.\n\nThe supporters of a principle of organic development will deny this in\neach individual case, and will demand special proof for each group of\norganisms; I believe this position, however, to be incorrect. Here,\nif anywhere, it appears to me justifiable to apply the conclusions\ninductively from special cases to general ones, since I cannot at all\nsee why a power of such pre-eminent and fundamental importance as a\nphyletic vital force should have its activity limited to solitary\ngroups in the organic world. If such a power exists it must be the\ninciting cause of organic development in general, and must be equally\nnecessary in every part of creation, as no advancement could take place\nwithout it. In this case, however, the force would be recognizable and\ndemonstrable at every point; the phenomena should nowhere stand in\nopposition to its admission, and should in no case be explicable or\ncomprehensible without it. The same laws and forces which caused the\ndevelopment of one group of forms must underlie the development of the\nwhole organic world.\n\nI therefore believe that we are correct in applying to the whole living\nworld the results furnished by the investigation of insects, and in\nthus denying the existence of an innate metaphysical developmental\nforce.\n\nThere is, however, a quite distinct method which leads to the same\nresults, and to the preliminary, if not to the complete and definitive\nrejection of such a principle; _the admission of this power is directly\nopposed to the laws of natural science_, which forbid the assumption of\n_unknown_ forces as long as it is not demonstrated that _known_ forces\nare insufficient for the explanation of the phenomena. Now nobody will\nassert that this has in any case been proved; the test of applying the\nknown factors of transformation has only just commenced, and wherever\nit has been made they have proved sufficient as causal forces. Thus,\neven without the foregoing special investigations we should deny a\nphyletic vital force; the more so as its admission is fraught with\nthe greatest consequences, since it involves a renunciation of the\npossibility of comprehending the organic world. We should, on this\nassumption, at once cut ourselves off from all possible mechanical\nexplanation of organic nature, _i.e._ from all explanation conformable\nto law. But this signifies no less than the renunciation of all further\ninquiry; for what is investigation in natural science but the attempt\nto indicate the mechanism through which the phenomena of the world\nare brought about? Where this mechanism ceases science is no longer\npossible, and transcendental philosophy alone has a voice.\n\nThis conception represents very precisely the well-known decision of\nKant:--\u201cSince we cannot in any case know _\u00e0 priori_ to what extent the\nmechanism of Nature serves as a means to every final purpose in the\nlatter, or how far the mechanical explanation possible to us reaches,\u201d\nnatural science must everywhere press the attempt at mechanical\nexplanation as far as possible. This obligation of natural science will\nbe conceded even by those who lay great stress upon the necessity for\nassuming a designing principle. Thus, Karl Ernst von Baer states that\nwe have no right \u201cto assert of the individual processes of Nature, even\nwhen these evidently lead to a definite result, that some Mind has\noriginated them designedly. The naturalist must always commence with\ndetails, and may then afterwards ask whether the totality of details\nleads him to a general and final basis of intentional design.\u201d[283]\n\nBut even if we are precluded on these grounds only from assuming\nthe existence of a directive power, _i.e._ a phyletic vital\nforce, for explaining detailed phenomena, and are at the same\ntime debarred from the possibility of arriving at a physical or\nmechanical explanation--which amounts to no less than the abandoning\nof the scientific position--it certainly cannot be asserted that\nthe development of the organic world is already conceived of as\na mechanical process. We rather acquiesce in the belief that the\nprocesses both of organic and of inorganic nature depend most probably\nupon purely causal powers, and that the attempt to refer these to\nmechanical principles should not therefore be abandoned. There is no\nground for renouncing the possibility of a mechanical explanation, and\nthe naturalist _must not_ therefore resign this possibility; for this\nreason he cannot be permitted to assume a phyletic power so long as it\nis not demonstrated that the phenomena can never be understood without\nsuch an assumption.\n\nIt cannot be raised as an objection that even for the explanation of\nindividual life a vital power was long ago admitted, as there was not\nthen sufficient material at hand to enable the phenomena of life to be\ntraced to physical forces. It is now no longer questionable that this\nassumption was a useless error--a false method--at the time when made\ncertainly very excusable, since the aspect of the question was then,\nowing to the imperfect basis of facts, very different to the present\nanalogous question as to the causes of derivative development. Thus,\nalthough it is now easy to prove this assumption to be erroneous, it\nwas in the former sense correct, as it was in accordance with the\nexisting state of knowledge. At that time there was hardly one of the\nnumerous bridges which now connect inorganic with organic nature,\nso that the supposition that life depended upon forces which had no\nexistence outside living beings was sufficiently near.\n\nIn any case the philosophers of that period cannot be blamed for\nfilling up the gaps in the existing knowledge by unknown powers,\nand in this manner seeking to establish a finished system. The task\nof philosophy is different to that of natural science; the former\nstrives at every period to set up a completely finished representation\nof the universe in accordance with the existing state of knowledge.\nNatural science on the other hand is only concerned in collecting\nthis knowledge; she need not therefore always finish off, and indeed\ncan never close her account, since she will never be in a position to\nsolve all problems.[284] But science must not for this reason pronounce\nany question to be insoluble simply because it has not yet been\ncompletely solved; this she does, however, as soon as she renounces\nthe possibility of a mechanical explanation by invoking the aid of a\nmetaphysical principle.\n\nThat this is the correct mode of scientific investigation is seen\nby the abandoning of the (ontogenetic) vital force. The latter is\nno longer admitted by anybody, now that we have turned from mere\nspeculation to the investigation of Nature\u2019s processes; nevertheless\nits non-existence has not been demonstrated, nor are we yet in a\nposition to prove that all the phenomena of life must be traced to\npurely physico-chemical processes, to say nothing of our being actually\nable to thus trace them. Von Baer also states \u201cthat the abolishment of\nthe vital force is an important advance; it is the reduction of the\nphenomena of life to physico-chemical processes, although these indeed\nstill contain many gaps.\u201d He points out how very far we are still\nremoved from being able to reduce to physical causes, the processes\nthrough which the fertilized yelk of an egg becomes developed into a\nchicken.\n\nHow comes it therefore that we all have a conviction that such a\ncomplete reduction will in time become possible, or if not this, that\nthe development of the individual depends entirely upon the same forces\nwhich are in operation without the organism? For what reason have we\nrejected the \u201cvital force\u201d?\n\nSimply because we see no reason for assuming that known forces are\ninsufficient for explaining the phenomena, and because we are not\njustified in admitting directive forces as long as we have any hope of\none day furnishing a mechanical explanation.\n\nBut if it is not only permissible, but even necessary, to explain the\n_ontogenetic_ vital power by known forces, and to commence to indicate\nthe mechanism which produces the individual life, why should it not\nbe equally necessary to abandon that assumption of a _phyletic_ vital\nforce which stifles any deeper inquiry, and to attempt to point out\nthat here also the co-operation of mechanical forces has brought about\nthe multitudinous and wonderful phenomena of the organic world?\n\nThe renunciation of the old vital force was certainly an immediate\nconsequence of the acquisition of new facts--of the knowledge that the\nsame compounds which compose organic bodies can be produced without the\nlatter. This discovery, due to W\u00f6hler and his followers, showed that\norganic products could be prepared artificially.[285] In brief, the\ndecline of the vital force followed from the knowledge that at least\none portion of the processes of life was governed by known forces.\n\nBut in the domain of the development of the organic world have we not\nquite analogous proofs of the efficacy of known forces? Is not the\n_variability_ of all types of forms a fact? and must not this under the\naction of natural selection and heredity lead to permanent changes? Has\nnot the problem of explaining the subserviency of all organic form to\nlaw as a _result_ without invoking its aid as a _principle_ been thus\nsuccessfully solved? It is true that we have not directly observed the\nprocess of natural selection from beginning to end; neither has anybody\ndirectly observed the mode in which the heat of the animal body is\ngenerated by the processes of combustion going on in the blood and\nin the tissues; nevertheless, this is believed as a certainty, and a\n\u201cvital force\u201d is not invoked.\n\nNow the above-mentioned Darwinian principles of transmutation are\ncertainly not simple forces of nature like those underlying the\ndevelopment of the individual, _i.e._ chemico-physical forces, and it\ncannot be said _\u00e0 priori_ whether in one of these principles--perhaps\nin variability or in correlation--there may not lie concealed a\nmetaphysical principle in addition to the physical forces. In fact it\nhas lately been asserted by Edward von Hartmann[286] that the theory\nof selection is not a mechanical explanation, since it combines forces\nwhich are only partly mechanical and in part directive.\n\nIt must therefore be next investigated whether this assertion is\ntenable.\n\n\n\n\nI.\n\nARE THE PRINCIPLES OF THE SELECTION THEORY MECHANICAL?\n\n\nEdward von Hartmann may justly claim that his views should be\nconsidered and tested by naturalists.[287] He would be correctly\nclassed with those philosophers who have approached this question with\na many-sided scientific preparation. It can nevertheless be perceived\nin his case how difficult, and indeed how impossible, it is to estimate\nthe true value of the facts furnished by the investigation of nature,\nwhen we attempt to take up only the results themselves, without being\npractised in the methods by which these are reached, _i.e._ without\nbeing completely at home in one of the scientific subjects concerned\nthrough one\u2019s own investigations. It appears to me that the denial of\nthe purely mechanical value of the Darwinian factors of transformation\narises in most part from an erroneous classification of the scientific\nfacts with which we have to deal. There can certainly be no mistake\nthat the entire philosophical conception of the universe, as laid down\nby Von Hartmann in his \u201cPhilosophy of the Unconscious,\u201d is unfavourable\nto an unprejudiced estimate of scientific facts and to their mechanical\nvaluation.\n\nVariability, heredity, and above all correlation, would not be regarded\nby Von Hartmann as purely mechanical principles, but he would therein\nassume a metaphysical directive principle.\n\nIn the first place, as regards variability, Von Hartmann endeavours\nto show that it is only a quite unlimited variability which suffices\nfor the explanation of necessary and useful adaptations by means\nof selection and the struggle for existence. But this does not\nexist--variation rather takes place in a fixed direction only (in\nAskenasy\u2019s sense), and this can be nothing else than the expression of\nan innate law of development, _i.e._ a phyletic vital force.\n\nThis deduction appears to me in two ways erroneous. In the first place\nit is incorrect that a quite unlimited variability is a postulate\nof the theory of selection, and in the next place the admission of\nvariability, which is in a certain sense \u201cfixed in direction,\u201d does not\nnecessitate the assumption of a phyletic vital force.\n\nA mere unsettled variability, uniform in all possible directions, is,\naccording to Von Hartmann, necessary for the theory of selection,\nbecause only then does the variability offer a certain guarantee \u201cthat\nunder given conditions of life the variations necessary for complete\nadaptation will not be wanting.\u201d But it is hereby overlooked that the\nnew life conditions to which the adaptation must take place are as\nlittle fixed and unchangeable as the organism itself. In such a case of\ntransformation we have not to deal with a type of organization which\nwas before fixed and immutable, and which has to be squeezed into new\nlife-conditions as into a mould. The adaptation is not one-sided, but\nmutual; a species in some measure selects its new conditions of life,\ncorresponding with those possible to its organization, _i.e._ with the\nvariations actually occurring. I will choose an instance which will\neven be conceded by Von Hartmann as being only explicable by natural\nselection, viz., a case of mimicry.\n\nSupposing that among the South American _Heliconiid\u00e6_ there occurred\na species of _Pieris_ which had no resemblance to these protected\nbutterflies, either in form, marking, or colouring; who can deny\nthat it would be most useful to this species to acquire the form and\ncolouring of a Heliconide, and thus, by taking to new conditions of\nlife, to avoid the persecutions of its foes? But if the physical\nnature of the Pieride concerned precluded the occurrence of Heliconoid\nvariations, would this incapability of insinuating itself into these\nnew conditions necessitate the decline of the species? Could not its\nexistence be secured in some other manner? could not the destruction of\nnumerous individuals by foes be compensated for by increased fertility?\nto say nothing of the numerous other means through which the number\nof surviving individuals might become increased, and the existence\nof the species secured. This case is not arbitrarily chosen; in the\ndistricts where the _Heliconiid\u00e6_ occur there are actually a large\nnumber of Whites which do not possess the protective colours of the\nformer nauseous family. In the adoption of these new life conditions\nwe have not to deal therefore with survival or extermination, but only\nwith amelioration. It is not every species of \u201cWhite\u201d that can become\nadapted to these conditions, because every species does not give rise\nto the necessary colour variations; those that do, become in this way\nmodified, because they are thus better protected than before. And so\nit is throughout; wherever we find protected insects enjoying immunity\nfrom foes we see also mimickers, sometimes only single, sometimes\nseveral, and generally from very diverse groups of insects, according\nto the general resemblance which existed before the commencement of\nthe process of adaptation, and to the variations made possible by the\nphysical nature of the species concerned.\n\nIn the first essay of the second part of this work it was shown that in\ncertain Lepidopterous larv\u00e6 a process of adaptation is at the present\ntime still in progress, this depending upon the fact that while the\nyoung caterpillar is very well protected by the leaf-green colour of\nits body, this colour becomes insufficient to conceal the insect as\nsoon as it exceeds the leaf in size. All such caterpillars--and there\nis a whole series of species--as they increase in size acquire the\nhabit of concealing themselves on the earth by day, and of feeding\nonly at night. New conditions of life are thus imposed, and these\nare even compulsory, _i.e._ they could not be abandoned without\nrisking the existence of the species. Now in accordance with these\nnew conditions, some individuals in these species have lost the green\ncolouring of the young stages, and have acquired the brown coloration\nof the dark surroundings of the insects which conceal themselves\nby day. In one species this change has now occurred in almost all\nindividuals, in others in only a larger or smaller proportion of\nthem. Now supposing that among these species there occurred one, the\nphysical nature of which did not admit of the production of brown\nshades of colour, would the species for this reason succumb? Is it not\nconceivable that the want of colour adaptation might be compensated\nfor by better concealment, _i.e._ by burrowing into the earth, or by\na greater fertility of the species, or by the development of warning\nsignals--supposing the species to be unpalatable--or finally, by the\nacquisition of a terrifying marking? In other words, could not the\ncaterpillar itself modify the new condition of life--that of being\nconcealed by day--in accordance with variations made possible by its\nphysical nature?\n\nAs a matter of fact in one of these species the green colour remains\nunchanged in spite of the altered mode of life, and this species,\nwherever it occurs, notwithstanding the persecution of entomologists,\nis always common (_Deilephila Hippopha\u00ebs_); it conceals itself better\nand deeper however than those other species which, like _Sphinx\nConvolvuli_, are difficult to detect on account of their brown colour.\nIn another species the striking yellowish green colouring is likewise\nretained in the majority of individuals, but this species buries\nitself by day in the loose soil (_Acherontia Atropos_).\n\nTo this it may be objected that there are also compulsory changes in\nthe conditions of life from which the species cannot withdraw itself,\nbut in which adaptation must necessarily follow, or extermination would\ntake place.\n\nSuch compulsory conditions of life do most assuredly occur, and there\nis indeed no doubt that many living forms have perished through not\nbecoming transformed. I believe, however, that such conditions occur\nmuch more rarely than one is inclined to admit at first sight. As\na rule the alternative of immediate change or of extermination is\noffered only by such changes in the conditions of life as occur very\nrapidly. The sudden appearance of a new and dominant enemy, such as\nman, has already caused the extinction of the Dodo (_Didus ineptus_),\nand of Steller\u2019s Sea Cow (_Rhytina Stelleri_), and of other vertebrate\nanimals, and constantly leads to the extermination of many other\nspecies of different classes. When in America hundreds of thousands of\nacres of primeval forest are annually destroyed, the conditions of life\nof a numerous fauna and flora must be thereby suddenly changed, leaving\nno choice but extermination.\n\nSuch abrupt changes in the conditions of life occur, however, but\nseldom in nature unless caused by man, and must therefore have very\nrarely happened in former epochs of the earth\u2019s history. Even climatic\nchanges, which we might at first regard as of this character, and\nwhich produce a modification in one fixed direction, occur always so\ngradually that the species has time either to adapt itself to the\nconditions in this or that direction, according to the variations\npossible to its physical nature, or else to emigrate.\n\nIt thus appears to me erroneous to suppose that variability must be\n\u201cmerely undetermined\u201d in order to complete its part in Darwin\u2019s theory\nof selection, and its \u201cillimitedness\u201d seems to me also as little\nnecessary for this purpose. Von Hartmann imagines that it is only\nunlimited variability that furnishes a guarantee that any type, to\nwhatever extent diverging from its point of departure, will be reached\nby the Darwinian method of gradual transmutation by means of selection\nand the struggle for existence.\n\nBut who has ever asserted that _any_ type can be reached from any\npoint? Or if anybody has said such nonsense, who can prove that its\nadmission is necessary for the theory of selection? Nowhere in systemy\ndo we see any point of support for such an assumption. But when Von\nHartmann imagines that the \u201cunlimited\u201d variability which he postulates\nfor Darwin \u201cis in itself unlimited, the limits of its divergence in\na given direction being found, not in itself, but only in external\nobstacles,\u201d he conceives variability to be something independent of,\nand in some way added to, the animal body, and not a mere expression\nfor the fluctuations in the type of the organism. If, however, we\nconceive variability in this latter, the true scientific sense, it\nis in no way \u201cquantitatively unlimited,\u201d nor are its limits even\ndetermined by _external_ influences, but essentially by _internal_\ninfluences, _i.e._ by the underlying physical nature of the organism.\nDarwin has indeed already shown this in a most beautiful manner in his\ninvestigations upon the correlations of organs and systems of organs\nof the body. To make use of a metaphor, the forces acting within the\nbody are in equilibrium; if one organ becomes changed this causes a\ndisturbance in the forces, and the equilibrium must be restored by\nchanges in other parts, and these again entail other modifications,\nand so forth. Herein lies the reason why the primary change cannot\nexceed a certain amount if the restoration of the equilibrium is\nnot to be quite impossible. This is but a metaphor, and I do not\nwish to assert that we are at present in a position to formulate and\ndemonstrate mathematically for any particular case, how much an organ\ncan become changed in any one species before an interruption of the\ninternal harmony of the body takes place. But such impossibility of\ndemonstration does not appear to me to furnish a sufficient reason\nfor regarding variability as the expression of a directive power--as\nan \u201cinnate tendency to variation conformable to law.\u201d[288] On the\ncontrary, it is to me easily conceivable that we only learn to analyse\nthe processes of nature in detail very slowly, because of their\nnecessary complexity. It thus appears to me quite useless when in this\nsense Wigand makes use of the objection, that \u201cthe gooseberry has not\nundergone any enlargement since 1852, although it is inconceivable\nwhy it should not attain the size of a pumpkin if variability was\nnot internally limited.\u201d It may well be that this is for the present\n\u201cinconceivable;\u201d nevertheless, this does not justify us in setting\nup a hypothetical \u201cforce of variation\u201d which will not admit of the\ngooseberry surpassing the pumpkin in size. We are bound to maintain\nthat it is the action and reaction of known forces which sets a limit\nto the enlargement of this fruit.\n\nIn more simple instances the causes of such limitations to growth can\nbe well perceived. Several decades have passed since Leuckart proved\nin how exact a relation the proportion of volume and surface stood to\nthe degree of organization of an animal. In animals of a spherical\nform the surface is quite sufficient for respiration, so long as they\nare of microscopic size. But such an organism cannot become enlarged\nat pleasure, because the ratio of the surface to the volume would\nbecome quite different. The surface increases as the square, whilst\nthe volume increases as the cube, so that very soon the surface of\nthe more rapidly increasing bodily mass can no longer suffice for\nrespiration.[289] This sort of limitation is in no way equivalent to\nthat purely external kind which, for instance, manifests itself in such\na manner as to prevent the indefinite lengthening of the tail feathers\nof the Bird of Paradise. In this case feathers that were too long would\nhinder flight, and such individuals would accordingly be eliminated by\nnatural selection. The cause is in the former case purely internal,\ndepending upon the equilibrium of the forces governing the organism.\n\nVon Hartmann is entirely in the right when he asserts that variability\nis neither qualitatively nor quantitatively unlimited. In both\nsenses it is limited (in direction as well as in amount) by the\nphysico-chemical forces acting in some contrary way in each specific\norganism--by the physical nature of each living form. He errs,\nhowever, both in making absolute illimitability a necessary postulate\nof the theory of selection, as also in inferring the existence of\na directive principle from that limitation of variability which is\ncertainly present. \u201cTendencies to variation\u201d do however exist, not in\nthe sense of a directive power, but as expressions of the different\nphysical constitutions of species, which necessarily cause unequal\nreactions to the same external actions, as will be more clearly proved\nbelow.[290]\n\nThis is, of course, a modification of Darwin\u2019s original assumption of\nan unbounded variability not limited in direction; but Darwin himself\nhas later coincided in the view that the quality of the variations is\nessentially determined by the nature of the organism.[291]\n\nI now turn to the consideration of the second factor of the theory of\nselection--heredity. This also, according to Von Hartmann is not a\nmechanical principle. Darwin himself has now become convinced how great\nis the probability against the hereditary retention of modifications\nwhich, whether feebly or strongly pronounced, appear _only in single\nindividuals_, _i.e._ of those so-called \u201cfortuitous\u201d variations which\nare not the expression of a directive developmental principle. \u201cBut as\namong the numberless possible directions of an indefinite variability,\nuseful modifications can only occur in single cases, Darwin has by this\nsupplementary admission himself retracted an inadmissible assumption of\nhis theory of selection,\u201d and so forth. A \u201cregular, designed tendency\nto variation, acting from within and contemporaneously affecting a\nlarge number of individuals,\u201d must therefore be assumed \u201cin order to\ninsure the by itself improbable inheritance.\u201d\n\nBut even from the unbounded variability laid down by the author, it\nby no means follows that useful variations can only occur in single\nindividuals. In the whole category of _quantitative_ variations the\nreverse is always the case. Is it the lengthening of some part that\nis concerned; so would a large number of individuals always possess\nthe useful variation, since we are not dealing with an _absolute_\nenlargement, but only with the fact that the part concerned is longer\nthan in other individuals.[292]\n\nBut if _qualitative_ variations come into consideration, it may be\nasked whether Darwin\u2019s \u201csupplementary admission\u201d does not go too\nfar. Such calculations as those quoted by Darwin from the article in\nthe _North British Review_ of March 1867 are extremely deceptive,\nsince we have no means of measuring the amount of protection afforded\nby a useful variation, and we can therefore hardly compute with\nany certainty, in how great a percentage of individuals a change\nmust contemporaneously occur in order to have a chance of becoming\ntransferred to the following generation. If our blue rock-pigeon could\nexist in a polar climate, and if we had the power of introducing it\ngradually, but not suddenly, into these regions in a wild state, who\ncan doubt that it would assume the white colour of all polar animals?\nNevertheless, among wild rock-pigeons white varieties do not occur\nmore frequently than among swallows, crows, or magpies. Or must the\nwhite colour of polar animals, the yellow colour of desert species, and\nthe green colour of leaf-frequenting forms, be always referred to a\n\u201cregular, designed, fixed tendency to variation acting from within,\u201d\nand causing a \u201clarge number of individuals\u201d to vary in a similar manner?\n\nThere is, however, a grain of truth in the foregoing; variations which\noccur singly have but little chance of becoming predominant characters,\nand this is obviously what Darwin concedes. But this is by no means\nequivalent to the assumption that only those variations which from the\nfirst occur in numerous individuals have a chance of being perpetuated.\nLet us keep to the facts. We have not the slightest reason either\nfor regarding the white colour of polar animals as the direct action\nof cold, or for considering that the green colour of foliage-living\ncaterpillars depends upon direct action arising from the habit of\nresting upon the leaves;[293] both these characters are explicable only\nby natural selection, and there is nothing to favour the assumption\n(which Von Hartmann postulates as necessary for success) that many\nindividuals varied into white at the same time. We know no single\nextra-polar species of a dark colour which frequently, _i.e._ in many\nindividuals of every generation, varies into white, but we know many\nspecies which from time to time produce single white individuals. Now\nwhen, on the other hand, we find that all polar animals to which the\nwhite coloration is advantageous, and indeed none but species of which\nthe nearest allies vary only individually into white, possess this\ncolour, must we not conclude from this alone that _single_ variations\ncan, under favourable conditions, become predominant characters?\n\nIt appears to me that in this question one weighty factor has been too\nlittle regarded, even by the supporters of the selection theory, viz.,\nthe slowness of most, and especially of climatic changes, which I have\nalready insisted upon. If the transformation of a temperate into an\narctic climate occurred so rapidly that the species exposed to it had\nthe alternative either of becoming white in ten or twenty generations\nor of being unable to exist, then the hasty intervention of a directive\npower could alone save them from extermination by causing hundreds of\nthousands of individuals to become similarly coloured with all speed.\nBut it is quite different if the change of climate takes place only in\nthe course of several thousand generations; and this, according to the\ngeological evidence, must have been the true state of the case.\n\nLet us take a definite example--the well-known one of the hare. With\nus this animal remains brown in the winter and but seldom produces\nwhite varieties, whilst its ally the Alpine hare is white during\nseven months of the year, the Norwegian hare during nine months, and\nthe Greenland hare throughout the whole year. If our climate became\ntransformed into an arctic one, after a given time there would arrive\na period when the older coloration no longer possessed any advantage\nover the occasional and singly-appearing white variations; the winter\ndays during which the ground was covered with snow would have become so\nnumerous, that the protection afforded to the white animals would be\nequal to the protection enjoyed by the brown individuals on the equally\nnumerous days free from snow. From this time forth the hares that\nwere white in winter would not be subjected to a greater decimation\nby foxes, &c., than the brown individuals. This period must however\nbe represented as consisting of one or more centuries, and it would\nbe strange if from the individual white hares, which now had an equal\nchance of existing, some white families did not become established.\nBut the state of affairs would gradually become reversed--the brown\nhares would experience greater decimation, and wherever there were\nwhite families these would possess an advantage in the struggle for\nexistence. It does not follow that the dark individuals would be\nforthwith extirpated; on the contrary, the advantage in favour of the\nwhite would be but small throughout a long period of time, and these\nindividuals would only gradually increase to a higher percentage of\nthe total population; nevertheless their numbers would constantly but\nvery slowly augment. In the course of time this increase would become\nmore rapid for two reasons--first, because even a very small advantage\nin favour of the increasing number of individuals would always leave a\ngreater number of these victorious; and secondly, because on the whole\nas the climate became more arctic, the advantage of being white would\ncontinually become more decisive in determining which should live and\nwhich should succumb.\n\nThus I see no reason why individual variations which do not appear only\nonce, but which frequently recur in the course of generations, should\nnot acquire predominance under favourable conditions. All facts are\nin accord with this. Even the common hare shows us that it would be\nquite capable of becoming coloured in a similar manner. In the museum\nof Stuttgart there are three specimens of _Lepus timidus_, killed in\nWurtemburg, which are completely white, and several others which are\nsilver-grey or spotted with white. In eastern Russia the common hare\npossesses a light grey, almost white, winter coat, and Seidlitz[294]\nmakes known the interesting observation that such light specimens occur\n_singly_ in Livonia, where \u201cthe common hare has become naturalized\nsince the commencement of the century.\u201d\n\nAs I have already insisted upon above, from the point of view\nof the conditions of life there is no reason for assuming rapid\ntransformations; the change of conditions is almost always extremely\nslow; and indeed in numerous instances no objective change occurs,\nbut simply a subjective one, if we may thus designate those cases in\nwhich the alteration in the conditions of life depends upon a change\nin the animal form which is undergoing transformation, and not in that\nof the environment. This is the case in the above-mentioned instances\nof mimicry, where the whole change in the conditions of life arises\nfrom one species becoming similar to another. The process of natural\nselection has here as long a period of time as it requires to perfect\nits results. It is quite similar in all cases of special protective\nadaptations of form and colour. In all these it is always _improvement_\nthat is concerned, and not the question \u201cto be or not to be\u201d with which\nwe have to deal.\n\nIt is just cases of this last kind, however, which are best fitted for\nexposing the improbability and insufficiency of the assumption of a\nvariational tendency as a distinct directive power. We have only to\nfix our attention upon some particular case of sympathetic colouring,\nor, still better, of mimicry. A \u201ctendency to variation\u201d implies that a\nlarge number of individuals produce varieties resembling the model to\nbe imitated, and this--at least according to Von Hartmann--must take\nplace in each of the successive generations, so that by this means,\ncombined with heredity, the useful variation becomes increased. But how\ncomes it that this \u201ctendency to variation\u201d coincides with the existence\nof the model both in time and place? Can this be due to accident if\nthe two have not a common cause? The upholders of a directive power\nwill certainly not admit this; so that there remains only Leibnitz\u2019s\nassumption of a pre-established harmony contained in the first organic\ngerm, which, after innumerable transformations of the organic form\nand after millions of years, gave rise in the midst of the Amazonian\nregion to an inedible Heliconide with certain yellow, black, and white\nmarkings on the wings, and at precisely the same time developed the\ntendency in a Pieride at the same spot on the globe to imitate this\nHeliconide as a model!\n\nIn addition to this assumption, which is certainly but little worthy\nof consideration, there is perhaps one other remaining, viz., that all\nor many Pierides and other species of butterflies possessed the same\ntendency to a Heliconoid variation and were always everywhere striving\nto develop this type, but succeeded only where they accidentally\ncoincided in time and place with the model, the \u201ctendency\u201d being thus\nfurthered by natural selection. But the facts negative this assumption,\nsince such imitative variations have never been observed to a\nperceptible extent in other species.[295]\n\nAll variations which are demonstrably useful can be similarly dealt\nwith if their origin is explained by variational tendencies.\n\nWe perceive that the objection which Von Hartmann brings against\nheredity is only valid on the ground that this process affords no\nsecurity for the preservation of variations which occur singly. That\nheredity itself is a mechanical process is not directly disputed; it is\nsimply assumed that new characters can be transferred by inheritance\nonly when they are produced by the metaphysical \u201cdevelopmental\nprinciple,\u201d and not when they arise \u201caccidentally.\u201d This critic does\nnot therefore direct his attack against heredity, but rather against\nthe mechanical origin of variability.\n\nVon Hartmann might have said here that a reference of the phenomenon\nof heredity to purely mechanical causes, _i.e._ a mechanical theory\nof heredity, is up to the present time wanting. That he has not done\nso proves on the one hand that he despised the dialectical art, but,\non the other hand, that he himself has not overlooked the subserviency\nof the total phenomenon to law, and that he grants the possibility of\nfinding a mechanical explanation therefor. If, in fact, the power of\ninheritance does not depend upon mechanical principles, I know not\nwhat organic processes we are entitled to regard as mechanical, since\nthey are all dependent in essence upon heredity, with which process\nthey are at one, and from which they cannot be thought of as isolated.\nHaeckel correctly designates reproduction as surplus individual\ngrowth, and accordingly refers the phenomena of heredity to those of\ngrowth. Conversely, growth may also be designated reproduction, since\nit depends upon a continuous process of multiplication of the cells\ncomposing the organism, from the germ-cell to the innumerable congeries\nof variously differentiated cells of the highly developed animal body.\nWho can fail to see that these two processes, the reproduction of the\ngerm-cell and its offspring in the economy of the individual, and the\nreproduction of individuals and species in the economy of the organic\nworld, show an exact and by no means simply superficial analogy?[296]\nBut whoso grants this must also conceive both processes to depend upon\nthe same cause--he cannot assume for the one a causal power and for\nthe other a directive principle. If nutrition and cell-multiplication\nare purely mechanical processes, so also is heredity. Although it\nhas not yet been possible to demonstrate the mechanism of this\nphenomenon, it can nevertheless be seen broadly that by means of a\nminimum of living organic matter (_e.g._ the protoplasm of the sperm\nand germ-cell) certain motions are transferred, and these can be\nregarded as directions of development, as I have already briefly laid\ndown in a former work.[297] The power of organisms to transmit their\nproperties to their offspring appears to me to be only conceivable in\nsuch a manner \u201cthat the germ of the organism by its chemico-physical\ncomposition together with its molecular structure, has communicated to\nit a fixed direction of development--the same direction of development\nas that originally possessed by the parental organism....\u201d (_loc.\ncit._ p. 24). This is confessedly nothing more than a hint, and we do\nnot learn therefrom the means by which developmental direction can be\npossibly transferred to another organism.\n\nRecently Haeckel, that indefatigable pioneer to whom we are indebted\nfor such a rich store of new ideas, has attempted to bridge over this\ngap in his essay on \u201cThe Perigenesis of the Plastidule,\u201d Berlin, 1876.\nThe basic idea, that heredity depends upon the transference of motion,\nand variability upon a change of this motion, completely corresponds\nwith the conviction gained in the province of physical science, that\n\u201call laws must finally be merged in laws of motion\u201d (Helmholtz[298]).\nI hold this view to be the more completely justifiable--although\ncertainly not in the remotest degree as proved--because I formerly\ndesignated the acquired individual variations as the \u201cdiversion of the\ninherited direction of development.\u201d Haeckel\u2019s hypothesis in so far\naccomplishes more than Darwin\u2019s pangenesis, in which a transference\nof matter, and not of a species of motion peculiar to this matter, is\nassumed. But although the germ of a mechanical theory of heredity may\nbe contained in Haeckel\u2019s hypothesis, this nevertheless appears to me\nto be somewhat remote from completely solving the problem. It brings\nwell into prominence one portion of the process of inheritance; under\nthe image of a molecular motion of the plastidule, which motion is\nmodifiable by external influences, we can well understand the fact of\na change gradually taking place in the course of generations. On the\nother hand, the assumption of consciousness in the plastidule,--however\nadmissible philosophically--although only as a formula, scarcely\nfurnishes any deeper knowledge. In the light of a theory, detailed\ninstances which were formerly obscure should become comprehensible.\nI fail to see, however, how the various forms of atavism, _e.g._ the\nreversions which so commonly occur by crossing different races, become\nmore comprehensible by assuming consciousness in the plastidule. If\nin both parents the plastidule long ago acquired different molecular\nmotions, why, in its rencounters in the germ, does it recollect past\ntimes and reassume the older and long abandoned motion? That it does\nacquire the latter is indeed a fact if we once refer the directional\ndevelopment of the individual to molecular motion of the plastidule;\nthe wherefore does not appear to me, however, to become clearer by\nassuming consciousness in the plastidule. A mechanical theory of\nheredity must rather be able to show that the plastidule movements of\nthe male and female germ-cells, in their rencounter in the case of the\ncrossing of widely divergent forms, become mutually modified in such\na manner that the motion of the common ancestral form must occur as\nthe resultant. To such demonstration there is however as yet a long\nstep. Haeckel himself moreover points out that his hypothesis is by\nno means a \u201cmechanical theory of heredity,\u201d but only an introduction\nto this theory, which he hopes \u201cwill be capable of being elevated to\nthe rank of a genetic molecular theory\u201d (_loc. cit._ p. 17). But\nalthough we must also confess with the critic of the \u201cPhilosophy of the\nUnconscious,\u201d that \u201cthe facts of heredity have hitherto defied every\nscientific explanation,\u201d[299] this furnishes us with no excuse for\nflying to a metaphysical explanation, \u201cwhich is here certainly least\nable to satisfy the inability to understand the connection arising from\nnatural laws.\u201d\n\nIt is not to be wondered at that Von Hartmann, on the ground of the\n\u201cUnconscious\u201d on which he takes his stand, speaks of the law of\ncorrelation as an unconscious acknowledgment of a \u201cnon-mechanical\nuniversal principle on the side of Darwinism.\u201d By \u201ccorrelation\u201d he\nunderstands something quite different to the idea which we attach to\nthis expression. He supposes that \u201cDarwinism sees itself compelled\nto acknowledge through empirical facts the uniform correlation of\ncharacters pertaining to the specific type; but it thereby contradicts\nits mechanical principles of explanation, all of which amount to the\nsame thing as conceiving the type as a mosaic, chequered, superficial,\nand accidental aggregate of characters, which have been singly\nacquired, contemporaneously or successively, by selection or habit.\u201d\nI do not believe, however, that any such conception has ever been\nadmitted either by Darwin or any one else. The admission that not all,\nbut only every deep-seated _physiological_ detailed modification, is\nor may be bound up with a system of correlated changes, indeed implies\nthat we on our side also acknowledge an internal harmony of parts--an\nequilibrium, as I have above expressed it.\n\nBut does this include the admission of a teleological principle, or\nexclude a mechanical explanation? Do we thereby acknowledge a \u201cspecific\ntype\u201d in the sense of an inseparably connected complex of characters,\nnone of which can be taken away without all the others becoming\nmodified? Does such a view agree generally with the empirical facts?\n\nNeither of these views appears to me to represent the case.\n\nI will first answer the second question. On all possible sides the\nearlier view of the absolute nature of species is contradicted; there\nis no boundary between species and varieties. But when Von Hartmann\nassumes that by the transformation of one species \u201cinto another\u201d the\n\u201cwhole uniformly connected complex must become changed,\u201d he falls\nback into the old doctrine of the absolute nature of species, which\nis sharply contradicted by multitudes of facts. We not unfrequently\nobserve varieties which differ from the parent-form by only a single\ncharacter, whilst others show numerous differences, and again others\nmay be seen in which the differences predominate. This last deviation\nwould then be designated by many systematists as a new species, but not\nso by others.\n\nThe \u201cspecific type\u201d is thus indeed a kind of mosaic-work, but it is\na structure to which all the single characters--the stones of the\nmosaic--belong and build up one harmonious whole, and not a meaningless\nconfusion. Some of the stones or groups of stones can be taken away\nand replaced by others differently coloured without the structure\nbeing thereby necessarily distorted, _i.e._ destroyed as a structure;\nbut the larger the stones which are exchanged the more necessary will\ncorrections in the other parts of the structure become, in order that\nthe harmony of the whole may be preserved.\n\nStill more weighty than those insensible transitions which in various\ngroups of animals so frequently connect species with species, appear\nto me, however, the facts made known in the second essay of the\nsecond part of this volume, which prove that the two forms in which\none species appears can change entirely independently of one another.\nThe caterpillar changes and becomes a new variety or even species\n(according to the form-value of the change), whilst the butterfly\nremains unaltered. How could this occur if some other law than that\nof physiological equilibrium linked together the parts or characters\nand permitted them to become severed? Must not the two stages become\nchanged with and through one another, like the parts of one body, since\nthey first together constitute the specific type? Is not the fact of\nthis not happening a proof that the whole \u201cuniformly connected complex\u201d\nof the specific type is not bound and held together by a metaphysical\nprinciple, but simply by natural laws?\n\nNow when Von Hartmann comprises the relations of different species\nto one another under the idea of correlation, such for instance as\nthe relation of dependence in which orchidaceous flowers stand with\nrespect to the insects which visit them, he completely abandons the\nscientific conception which should be associated with this expression,\nand compares together two heterogeneous things which have nothing in\ncommon excepting that they are both considered by him as a result\nof the \u201cUnconscious.\u201d The consequence which is then deduced from\nthis correlation of his own construction, viz., that an organic law\nof correlation is only another expression for a \u201claw of organic\ndevelopment\u201d in the sense of a metaphysical power, obviously cannot be\nadmitted.\n\nBy correlation we understand nothing more than the dependence of one\npart of the organism upon the others and the mutual inter-relations\nof these parts, which depend entirely upon a \u201cphysiological relation\nof dependence,\u201d as Von Hartmann himself has correctly designated it.\nHerein is evidently comprised the total morphology of the organism--the\nstructure as a whole, the length, thickness and weight of the single\nparts, as well as the histological structure of the tissues, since upon\nall these depends the performance of the single parts. But when, under\ncorrelation, Von Hartmann comprises \u201calso a morphological, systematic,\ninter-action of all the elements of the organism with reference both\nto the typical ground-plan of the organization as well as to the\nmicroscopic anatomical structure of the tissues,\u201d he drags into the\nidea something foreign to it, not on the ground of facts, but actually\nin opposition to them, and supported only by a supposed \u201cinnate\ndevelopmental principle\u201d which \u201cis not of a mechanical nature.\u201d\n\nThe living organism has already been often compared with a crystal, and\nthe comparison is, _mutatis mutandis_, justifiable. As in the growing\ncrystal the single molecules cannot become joined together at pleasure,\nbut only in a fixed manner, so are the parts of an organism governed\nin their respective distribution. In the crystal where nothing but\nhomogeneous parts become grouped together their resulting combination\nis likewise homogeneous, and it is obvious that they offer but very\nlittle possibility of modification, so that the governing laws thus\nappear restricted and immutable. In the organism, whether regarded\nmicroscopically or macroscopically, various parts become combined, and\nthese therefore offer numerous possibilities of modification, so that\nthe governing laws are more complex, and appear less restricted and\nunchangeable. In neither instance do we know the final causes which\nalways lead to a given state of equilibrium; in the case of a crystal\nit has not occurred to anybody to ascribe the harmonious disposition\nof the parts to a teleological power; why then should we assume such\na force in the organism, and thus discontinue the attempt, which has\nalready been commenced, to refer to its natural causes that harmony of\nparts which is here certainly present and equally conformable to law?\n\nOn these grounds the assertion that the theory of selection is not an\nattempt at a \u201cmechanical\u201d explanation of organic development appears to\nme to be incorrect. Variability and heredity, as well as correlation,\nadmit of being conceived as purely mechanical, and must be thus\nregarded so long as no more cogent reasons can be adduced for believing\nthat some force other than physico-chemical lies concealed therein.\n\nBut we certainly cannot remain at the purely empirical conception as\nlaid down by Darwin in his admirable work on the \u201cOrigin of Species.\u201d\nIf the theory of selection is to furnish a method of mechanical\nexplanation, it is essential that its factors should be formulated in a\nprecise mechanical sense. But as soon as we attempt to do this it is\nseen that, in the first enthusiasm over the newly discovered principle\nof selection, the one factor of transformation contained in this\nprinciple itself has been unduly pushed into the background, to make\nway for the other more apparent and better known factors.\n\nI have for many years insisted that the first, and perhaps most\nimportant, or in any case the most indispensable, factor in every\ntransformation, is _the physical nature of the organism itself_.[300]\n\nIt would be an error to believe that it is entirely the external\nconditions which determine what changes shall appear in a given\nspecies; the nature of these changes depends essentially upon the\nphysical constitution of the species itself, and a modification\nactually arising can obviously be only regarded as the resultant of\nthis constitution and of the external influences acting thereon.\n\nBut if an essential or perhaps even a preponderating share in\ndetermining new characters is to be undoubtedly ascribed to the\norganism itself, for a mechanical representation of organic\ndevelopmental processes everything depends upon our being able\nto conceive this most important factor in a definite theoretical\nmanner, and to comprise under one common point of view its apparently\ncontradictory manifestations of constancy and variability.\n\nNow every change of considerable extent is certainly considered by\nDarwin to be the direct or indirect consequence of external actions;\nbut indirect action always presupposes a certain small variability\n(individual variability), without which larger modifications cannot\nbe brought about. Empirically this small amount of variability is\ndoubtless present, but the question arises, upon what does it depend?\nCan it be conceived as arising mechanically, or is it perhaps just at\nthis point that the metaphysical principle steps in and offers those\nminute variations which make possible that course of development which,\naccording to this view, is immutably pre-determined? It is certainly\nthe absence of a theoretical definition of variability which always\nleaves open a door for smuggling in a teleological power. A mechanical\nexplanation of variability must form the basis of this side of the\ntheory of selection.\n\nThis explanation is not difficult to find. All dissimilarities of\norganisms must depend upon the individuals having been affected by\ndissimilar external influences during the course of the development of\norganic nature. If we ascribe to the organism the power of giving rise\nby multiplication only to exact copies of itself, or, more correctly,\nthe power of transmitting unaltered to its successors the motion of its\nown course of development, each \u201cindividual variation\u201d must depend upon\nthe power of the organism to react upon external influences, _i.e._ to\nrespond by changes of form and of function, and consequently to modify\nits original (inherited) developmental direction.\n\nIt has sometimes been insisted upon, that the \u201cindividuals of the same\nspecies\u201d or the offspring of one mother cannot be absolutely equal,\nbecause, from the commencement of their existence, they have been\nsubjected to dissimilar actions of the environment. But this implies\nthat by perfectly equal influences they would become equal, _i.e._\nit supposes that variability is not inseparably bound up with the\nessence of the organism, but is only the consequence of developmental\ntendencies which are in themselves equal being unequally influenced. As\na matter of fact the first germs of an individual certainly cannot be\nsupposed to be perfectly equal, because the individual differences of\nthe ancestors must be contained therein in different degrees according\nto their constitution, and we should have to go back to the primordial\norganism of the earth in order to find a perfectly homogeneous root,\na _tabula rasa_ from which the descendants would commence their\ndevelopment. Whether such a homogeneous root ever existed is however\ndoubtful; it is much more probable that _numerous_ organisms first\narose spontaneously,[301] and these cannot be presumed to have been\nabsolutely equal, since the conditions under which they came into\nlife cannot have been perfectly identical. Let us, however, for the\nsake of simplicity assume a single primordial organism; the first\ngeneration which took its rise from this by reproduction could only\nhave possessed such individual differences as were produced by the\naction of dissimilar external influences. But the third generation,\ntogether with self-acquired, would also have shown _inherited_,\ndissimilarities, and in each succeeding generation the number of\ntendencies to individual difference imparted to the germ by heredity\nmust have increased to a certain degree, so that it may be said that\nall germs, from their first origination, bear in themselves a tendency\nto show individual peculiarities, and would develop these even if\nthey should not be again affected by dissimilar influences. This is\nobviously the case, since the youngest egg-cells in the ovary of an\nanimal are, as can be demonstrated, always exposed to unequal external\nconditions with respect to nutrition and pressure.[302] Hence, if it\nwere possible that two germs were exactly equal with respect to the\ndirection of development imparted to them by heredity, they would\nnevertheless furnish two incongruent individuals; and if, conversely,\nit were possible that two individuals could be exposed to absolutely\nthe same external influences from the formation of the embryo, these\nalso could not be identical, because the individual differences of the\nancestors would entail small differences, even in asexual reproduction,\nin the direction of development transmitted to the egg. The differences\nbetween individuals of similar origin thus finally depend entirely upon\nthe dissimilarity of external influences--on the one side upon those\nwhich divert the development of the progenitors, and on the other side\nupon those which divert the individual itself from its course, _i.e._\nfrom the developmental direction transmitted hereditarily. Although\nI thus essentially agree with Darwin and Haeckel in so far as these\nauthors refer the \u201cuniversal individual dissimilarity\u201d to dissimilar\nexternal actions, I differ from Darwin in this, that I do not see an\nessential distinction between the direct and indirect production of\nindividual differences, if by the latter is meant only the unequal\ninfluencing of the germ in the parental organism. Haeckel is certainly\ncorrect in referring the \u201cprimitive differences of the germs produced\nby the parents\u201d to the inequalities of nutrition to which the single\ngerms must inevitably have been exposed in the parent organism;\nbut another dissimilarity of the germs must evidently be added--a\ndissimilarity which has nothing to do with unequal nutrition, but\nwhich depends upon unequal inheritance of the individual differences\nof the ancestors, a source of dissimilarity which must arise to a\ngreater extent in sexual than in asexual reproduction. Just as in\nsexual propagation there occurs a blending of the characters (or\nmore precisely, developmental directions) of two _contemporaneous\nindividuals_ in _one_ germ, so in every mode of reproduction there meet\ntogether in the same germ the characters of a whole _succession of\nindividuals_ (the ancestral series), of which the most remote certainly\nmake themselves but seldom felt in a marked degree.\n\nThe fact of individual variability can in this way be well understood;\nthe living organism contains in itself no principle of variability--it\nis the _statical element_ in the developmental processes of the\norganic world, and would always reproduce exact copies of itself if the\ninequality of the external influences did not affect the developmental\ncourse of each new individual; these influences are therefore the\n_dynamical elements_ of the process.\n\nFrom this conception of variability two important empirically\nestablished facts can be theoretically deduced, viz. the limitability\nof variation with respect to quality, which has already been previously\nmentioned, and the origination of transformations by the direct action\nof external conditions of life.\n\nIf the differences in individuals of the same origin depend upon\nthe action of unequal influences, variation itself is nothing else\nthan the reaction of the organism to a definite external inciting\ncause, the quality of the variation being determined by the quality\nof the inciting cause and by that of the organism. In the cases of\nindividual variation hitherto considered, the quality of the organism\nis equal but that of the inciting cause is unequal, and in this way\nthere arise minute differences in organisms of an equal physical\nconstitution--variations of a different quality.\n\nThe same result, viz., different qualities of variation, may also arise\nin a reverse manner by organisms of a different physical nature being\naffected by equal external influences. The response of the organism to\nthe cause inciting change would be different according to its nature,\nor, in other words, organisms of different natures react differently\nwhen affected by equal modifying influences. The physical nature of\nthe organism plays the chief part with respect to the quality of the\nvariations; each specific organism can thus give rise to extremely\nnumerous, but not to all conceivable, variations; that is, only to\nsuch variations as are made possible by its physical composition.\nFrom this it follows further that the possibilities of variation in\ntwo species are more widely different, the wider they diverge in\nphysical constitution (including bodily morphology)--that a cycle of\nvariation is peculiar to every species. In this manner we are led to\nthe knowledge that there must certainly exist a \u201cfixed direction of\nvariation,\u201d but not in the sense of Askenasy and Von Hartmann, as the\nresult of an unknown internal principle of development, but as the\nnecessary, _i.e._ mechanical, consequence of the unequal physical\nnature of the species, which must respond even to the same inciting\ncause by unequal variations.\n\nThe facts, as far as we know them, agree very well with this\nconclusion. Allied species vary in a similar manner, whilst species\nwhich are more distantly related vary in a different manner, even\nwhen acted upon by the same external influences. Thus, in the first\npart of these \u201cStudies\u201d I have remarked that many butterflies under\nthe influence of a warm climate acquire an almost black coloration\n(_Polyommatus Phl\u00e6as_), whilst on the other hand others become lighter\n(_Papilio Podalirius_).\n\nWe can thus understand why always certain courses of development are\nfollowed, a fact which cannot be completely explained by the nature of\nthe conditions of life which induce the variations. But as soon as we\nclearly perceive that the quality of the changes essentially depends\nupon the physical nature of the organism itself, we arrive at the\nconclusion that species of widely diverging constitutions must give\nrise to different variations, whilst those of allied constitutions\nwould produce similar variations. But definite courses of development\nare thus traced out, and we perceive that from any point of the organic\ndevelopmental series, it is impossible that any other point can be\nattained at pleasure. Variation in a definite direction thus by no\nmeans necessitates the acknowledgment of a metaphysical developmental\nprinciple, but can be well conceived as the mechanical result of the\nphysical constitution of the organism.\n\nThe manner in which the dissimilar physical constitution of organisms\nmust arise can also be easily shown, although the first commencement of\nthe whole developmental series, _i.e._ the oldest living forms must be\nassumed to have been almost homogeneous in their physical constitution.\nThe quality of the variation is, as said before, not merely the\nproduct of the physical constitution, but the resultant of this and of\nthe quality of the changing external conditions. Thus from the first\n\u201cspecies\u201d there proceeded, through the dissimilar influence of external\nconditions of life, several new \u201cspecies,\u201d and as this took place the\nformer physical nature of the organism at the same time became changed,\nnecessitating also a new mode of reacting upon external influences,\n_i.e._ another direction of variation. The difference from the primary\n\u201cspecies\u201d must certainly be conceived as having been very minute, but\nit must have increased with each new transformation, and must have\nproceeded exactly parallel with the degree of physical change connected\nwith each transformation. Thus, hand in hand with the modifications,\nthe power of modification, or mode of reaction of the organism to\nchanging influences, must have continually become re-modified, and we\nfinally obtain an endless number of differently constituted living\nforms, of which the variational tendencies are different in exact\nproportion to their physical divergence, so that nearly allied forms\nrespond similarly, and widely divergent forms very differently, to the\nsame inciting causes.\n\nIndividual variation arises, as I have attempted to show, by each\nindividual having been continually affected by different, and indeed\nby constantly changing, influences. Let us, however, imagine on the\ncontrary, that a large group of individuals is affected by the same\ninfluences--in fact by such influences as the remaining individuals of\nthe species are not exposed to: this group of individuals would then\nvary in a nearly similar manner, since both factors of variation, viz.\nthe external influence and the physical constitution, are equal or\nnearly so. Such local variations would first become prominent when the\nsame external influence had acted upon a series of generations, and\nthe minima of variation produced in the individual by the once-exerted\naction of the cause inciting change had become augmented by heredity.\nTransformations of some importance (up to the form-value of species)\ncan thus arise simply by the direct action of the environment, in the\nsame way as that in which individual differences are produced--only\nthe latter fluctuate from generation to generation, since the inciting\ninfluences continually change; whilst, in the former, the constant\nexternal cause inciting modification always reproduces the same\nvariation, so that an accumulation of the latter can take place.\nClimatic varieties can be thus explained.\n\nA more efficacious augmentation of the variations arising in the single\nindividual is certainly brought about by the _indirect_ action of the\nenvironment upon the organism. It is not here my intention to explain\nonce more the processes of natural selection; I mention this only in\norder to point out that in these cases transformation depends upon a\n_double action_ of the environment, since the latter first induces\nsmall deviations in the organism by direct action, and then accumulates\nby selection the variations thus produced.\n\nBy regarding variability in this manner--by considering each variation\nas the reaction of the organism to an external action, as a diversion\nof the inherited developmental direction, it follows that without a\nchange in the environment no advance in the development of organic\nforms can take place. If we imagine that from any period in the earth\u2019s\nhistory the conditions of life remain completely unchanged, the species\npresent on the earth at this period would not, according to our view,\nundergo any further modification. Herein is clearly expressed the\ndifference of this view from that other one according to which the\ninciting principle of modification is not in the environment, but lies\nin the organism itself in the form of a phyletic vital force.\n\nI cannot here refrain from once more returning to the old (ontogenetic)\nvital force of the natural philosophers, since the parallel between\nthis and its younger sister, the \u201cphyletic vital force\u201d which\nappears in so many disguises, is indeed striking. Were the inciting\nprinciple of the development of the individual actually an independent\nvital force acting within the organism, the birth and growth of\nthe individual would be able to take place without the continuous\nencroachment of the environment, such as occurs in nutrition and\nrespiration. Now this is known to be impossible, so that those who\nsupport the existence of such a force, if any still exist, would be\ndriven to the obscure idea of a co-operation between the designing\npower and the influences of the environment, just in the same manner\nas such a co-operation is at present postulated by the defenders of\nthe phyletic vital force. I shall further on take the opportunity of\npointing out that this last idea is quite untenable; with respect to\nthe (ontogenetic) vital force any clearer proof cannot well be adduced,\nbut it will be admitted that the confused notion of the co-operation\nand inter-action of teleological and causal powers is, from our\npoint of view, opposed to those very simple and clear ideas which\nare in harmony with the views on phyletic development. As in racial\ndevelopment each change of the organic type is entirely dependent upon\nthe action of the environment upon the organism, so in the development\nof the individual, the totality of the phenomena of the personal life\nmust depend upon similar actions. Physiology, as is known, herein\nentirely supports our view, since this shows that without the continual\nalternating action of the environment and of the organism there can be\nno life, and that vital phenomena are nothing but the reactions of the\norganism to the influences of the environment.\n\nIt will be immediately perceived how exactly the processes of phyletic\nand of ontogenetic development coincide, not merely in their external\nphenomena but in their nature, if we trace the consequences of the\nexisting knowledge of the structure of the animal body. Although we\nmay not entirely agree with Haeckel\u2019s doctrine of individuality in\nits details, its correctness must on the whole be conceded, since it\ncannot be disputed that the notion of individuality is a relative one,\nand that several categories of morphological individuals exist, which\nappear not only _singly_ as _physiological individuals_, _i.e._ as\nindependent living beings of lowest grade, but which can also _combine_\nto form beings of a higher order.\n\nBut if we admit this, we should see with Haeckel nothing but\nreproduction in the origination of a high organism from a single cell,\nthe egg; this reproduction being at the same time combined with various\ndifferentiations of the offspring, _i.e._ with adaptations of the\nlatter to various conditions of life. Not even in the fact that the\ntissues and organs of a single physiological individual stand in great\ndependence upon one another through physical causes,[303] is there any\nstriking difference between this view and the phyletic composition of\nthe animal (and vegetable) kingdom out of physiological individuals\n(Haeckel\u2019s \u201c_Bionten_\u201d), since contemporaneous animals (individuals and\nspecies) are known to influence one another in the most active manner.\n\nNow if we further consider that the same units (cells) which, by\ntheir reproduction and division of labour, at present compose the\nbody of the highest organism, must at one time have constituted as\nindependent beings the beginning of the whole of organic creation, and\nthat consequently the same processes (division of cells) which now\nlead to the formation of a mammal, at that time led only to a long\nseries of different independent beings, it will be admitted that both\ndevelopmental series must depend upon the same inciting powers, and\nthat with reference to the causes of the phenomena it is not possible\nthat any great gap can exist between ontogeny and phylogeny, _i.e._\nbetween the life-phenomena of the individual and those of the type.\nAccording to our view both depend upon that co-operation of the same\nmaterial physical forces which admits of being briefly summarized\nas the reaction of organized living matter to influences of the\nenvironment.\n\nOur opponents either cannot boast of such harmony in their conception\nof nature, or else they must, together with the phyletic vital force,\nre-admit into their theory the old ontogenetic vital force. I know not\nindeed why they should not do so. Whoever inclines to the view that\norganic nature is governed not merely by causal, but at the same time\nby teleological, forces, may admit that the latter are as effective as\ninciting causes of individual, as they are of phyletic, development.\nAccording to my idea they are even bound to admit this, since it cannot\nbe perceived why the adaptations of the ontogeny should not depend upon\nthe same metaphysical principle assumed for each individual, as the\nadaptations of the phylogeny; the latter are indeed only brought about\nby the former. I believe therefore that the vital force (ontogenetic)\nof the ancients stands or falls with the modern (phyletic) vital force.\nWe must admit both or neither, since they both rest on the same basis,\nand are supported or opposed by the same arguments. Whoever feels\njustified in setting up a metaphysical principle where complete proof\nthat _known forces_ are sufficient for the explanation of the phenomena\nhas not yet been adduced, must do the same with respect to individual,\nas he does to phyletic, development, since this proof is in both cases\nvery far from being complete, and still contains large and numerous\ngaps.[304]\n\nThe theoretical conception of variation as the reaction of the organism\nto external influences has also not yet been experimentally shown to\nbe correct. Our experiments are still too coarse as compared with the\nfine distinctions which separate one individual from another; and\nthe difficulty of obtaining clear results is greatly increased by\nthe circumstance that a portion of the individual deviations always\ndepends upon heredity, so that it is frequently not only difficult,\nbut absolutely impossible, to separate those which are inherited from\nthose which are acquired. Still further are we removed from being\nable to refer variation to its final mechanical causes, _i.e._ from a\nmechanical theory of reproduction, which would bring within the range\nof mathematical calculation both the phenomena of stability (heredity)\nand of change (variability).\n\nBut although sufficient proofs of the correctness of the views\nhere advocated cannot at present be adduced, these views are not\ncontradicted by any known facts--they are, on the contrary, supported\nby many facts which they in turn make comprehensible (local forms,\ndifferent cycles of variation in heterogeneous species). These views\nare finally completely justified by their furnishing the only possible\ntheoretical formulation of variability on which a mechanical conception\nof organic development can be based. That such a conception is not only\nadmissible, but is unavoidable, at least to the naturalist, I have\nalready attempted to prove.\n\n\n\n\nII.\n\nMECHANISM AND TELEOLOGY.\n\n\nIn the third volume of his smaller works Karl Ernst von Baer submits\nthe theory of selection to a most searching examination. Without\nactually calling in question its scientific admissibility, he believes\nthat this theory is dependent upon its satisfying one condition, viz.\nthat it should connect the teleological with the mechanical principle.\n\n\u201cThe Darwinian hypothesis, as stated by its supporters, always ends in\ndenying to the processes of nature any relation to a future, _i.e._ any\nrelation of aim or design. Since such relations appear to me _quite\nevident_,\u201d &c. And further:--\u201cIf the scientific correctness of the\nDarwinian hypothesis is to be admitted, it must accommodate itself to\nthis universal striving after a purpose. If it cannot do this we should\nhave to deny its value.\u201d\n\nThese words appear almost equivalent to passing a sentence of doom\nupon the theory of selection and the mechanical conception of nature,\nfor how can one and the same process be effected simultaneously by\nnecessity and by designing powers? The one excludes the other, and we\nmust--so it appears--take our stand either on one side or the other.\n\nNevertheless we cannot set aside Von Baer\u2019s proposition without\nfurther examination simply because it is apparently incapable of being\nfulfilled, since it contains a truth which should not be overlooked,\neven by those who uphold the mechanical theory of nature. It is the\nsame truth which is also made use of by the philosophical opponents of\nthis theory, viz. that the universe as a whole cannot be conceived as\nhaving arisen from blind necessity--that the endless harmony revealed\nin every nook and corner by all the phenomena of organic and of\ninorganic nature cannot possibly be regarded as the work of chance,\nbut rather as the result of a \u201cvast designed process of development.\u201d\nIt is also quite correct when, in reply to the supposed objection\nthat the mechanical theory of nature is not concerned with chances\nbut with necessities, Von Baer answers that the operations of a\nseries of necessities which \u201care not connected together\u201d can only be\ntermed accidents in their opposing relations. He illustrates this by\ninstancing a target. If I hit the latter by a well-aimed shot, nobody\nwould explain this as the result of an accident, but if \u201ca horseman is\nriding along a gravelly road past this target, and one of the pebbles\nthrown up by the hoof of the galloping horse hits the mark, this would\nbe termed an accident of extremely rare occurrence. My target was not\nthe mark for the pebble, therefore the hit was purely accidental,\nalthough the projection of the stone in this precise direction with\nthe velocity which it had acquired, was sufficiently explained by\nthe kick given by the horse. But the hit was accidental because the\nkick of the galloping horse, although it necessarily projected the\npebble, had no relation at all to my target. For the same reason we\nmust regard the universe as an immense accident if the forces which\nmove it are not designedly regulated--the more immense because it is\nnot a single motion of projection that acts here, but a large number\nof heterogeneous powers, _i.e._ a large number of variously acting\nnecessities which are, as a whole, devoid of purpose, but which\nnevertheless accomplish this purpose, not only at any single moment,\nbut constantly. A truly admirable series of desirable accidents!\u201d[305]\n\nThe same idea is expressed, although in a very different manner, by\nVon Hartmann, in the concluding chapter of his work already quoted.\nHe thinks that \u201cdesign is a necessary and certain consequence of the\nmechanical laws of nature.\u201d \u201cWere the mechanism of natural laws not\nteleological there would be no mechanically regulated laws, but a weak\nchaos of obstinate and capricious powers. Not until the causality of\nthe laws of inorganic nature had superseded the expression \u201cdead\u201d\nnature, and had shown itself as the mainspring of life and of a\nconformability to design visible on all sides, did it deserve the\nname of mechanical lawfulness; just as a complication of wheels and\nmachinery made by man, which move in some definite manner with respect\nto one another, only acquires the name of a mechanism or of a machine\nwhen the immanent teleology of the combination and of the various\nmovements of the parts is revealed.\u201d[306]\n\nAgainst the correctness of the idea underlying these statements\nscarcely anything can in my opinion be said. The harmony of the\nuniverse and of that portion of it which we designate organic nature,\ncannot be explained by chance, _i.e._ without a common ground for\nco-operating necessities; by the side of mere mechanism it is\nimpossible not to acknowledge a teleological principle--the only\nquestion is, in what manner can we conceive this as acting without\nabandoning the purely mechanical conception of nature?\n\nThis is obviously effected if, with Von Baer and Von Hartmann, we\npermit the metaphysical principle to interrupt the course of the\nmechanism of nature, and if we consider both the former and the latter\nto work together with equal power. Von Hartmann expressly makes such an\nadmission under the name of an \u201cinternal principle of development,\u201d to\nwhich he attributes such an important share that one cannot understand\nwhy it should have any need for the employment of causal powers, and\nwhy it does not simply do everything itself. Von Baer expresses himself\nmuch less decisively, and even in many places insists upon the purely\nmechanical connection of organic natural phenomena; but that with him\nalso the idea of interruption by a metaphysical principle is present,\nis principally shown by his assuming, at least partly, the _per saltum_\ndevelopment of species. This necessarily involves an actively internal\npower of development.\n\nAlthough I have already brought forward many arguments against the\nexistence of such a power, and although in refuting it every form of\ndevelopment by directive powers is at the same time overthrown, it\nnevertheless appears to me not to be superfluous in such a deeply\nimportant question to show that a _per saltum_ development, and\nespecially the so-called heterogeneous generation, is inconceivable,\nnot only on the ground of the arguments formerly employed against the\nphyletic vital force in general, but quite independently of these.\n\nIn the first place it must be said that the positive basis of this\nhypothesis is insecure. Cases of sudden transformation of the whole\norganism with subsequent inheritance are as yet quite unknown. It has\nbeen shown that the occasional transformation of the Axolotl must most\nprobably be regarded in a different light. Another case, taken for\nheterogeneous generation, viz. the budding of twelve-rayed Medus\u00e6 in\nthe gastric cavity of an eight-rayed species, has lately been shown by\nFranz Eilhard Schulze[307] to be a kind of parasitism or commensalism.\nThe buds of the _Cunin\u00e6_ do not spring, as was supposed, from the\n_Geryonia_, but are developed from a _Cunina_ egg. But even if we\nrecall here the cases of alternation of generation and heterogenesis,\nthis would not be of any value by way of proof; it would only be thus\nindicated how one might picture to oneself a sudden transformation.\nThat in alternation of generation, or generally, in every mode of\ncyclical reproduction, we have not to deal with the abandonment of\none type of organization and the transition to some other, is proved\nby the continual return to the type of departure--by the cyclical\ncharacter of the entire transformation. That two quite heterogeneous\ntypes can belong to one cycle of development is, however, capable\nof a far better and more correct explanation than would be given\nby the supporters of _per saltum_ development. If we trace cyclical\nreproduction to the adaptation of different developmental stages\nor generations to deviating conditions of life, we thus not only\nexplain the exact and often striking agreement between form and mode\nof life--we not only bridge over the gap between metamorphosis and\nalternation of generation, but we can also understand how, within one\nand the same family of Hydrozoa, species can occur with or without\nalternation of generation, and further how other species can exist in\nwhich the alternation of generation (the production of free Medus\u00e6) is\nlimited to the one sex; we can understand in general how one continuous\nseries of forms may lead from the simple sexual organ of the Polypes\nto the independent and free swimming sexual form of the Medus\u00e6, and\nhow hand in hand with this the simple reproduction becomes gradually\ncyclical. It is just these intermediate steps between the two kinds of\nreproduction that make quite untenable the idea that the heterogeneous\nforms in cyclical propagation arise through so-called \u201cheterogeneous\ngeneration,\u201d _i.e._ through sudden _per saltum_ transformation. It\nis excusable if philosophers to whom these facts are strange, or who\nhave to take the trouble of working them up, should adduce alternation\nof generation as an instance of \u201cheterogeneous generation,\u201d but by\nnaturalists this should be once and for ever abandoned.\n\nAll other facts which have hitherto been referred to \u201cheterogeneous\ngeneration\u201d are still less explicable as such, inasmuch as they always\nrelate to changes in single parts of an organism, such as the sudden\nchange of fruit or flower in cultivated plants. The notion of _per\nsaltum_ development, however, demands a total transformation--it\ncomprises (as Von Hartmann quite correctly and logically admits) the\nidea of a _fixed specific type_ which can only be re-modelled _as a\nwhole_, and cannot become modified piecemeal. It must further be added,\nthat the observed variations which have arisen abruptly in single parts\nare not as a rule inherited:[308] fruit-trees are only propagated by\ngrafting, _i.e._ by perpetuating the individual, and not by ordinary\nreproduction by seeds. Now, if we nowhere see sudden variations of\nlarge amount perpetuated by heredity, whilst we everywhere observe\nsmall variations which can all be inherited, must it not be concluded\nthat _per saltum_ modification is not the means which Nature\nemploys in transforming species, but that an accumulation of small\nvariations takes place, these leading in time to large differences?\nIs it logical to reject the latter conclusion because our period of\nobservation is too brief to enable us to directly follow long series\nof accumulations, whilst _per saltum_ variation is admitted, although\nunsupported by a single observation? As long as there remains any\nprospect of tracing large deviations to the continually observed\nphenomenon of small variations, I believe we have no right to resort to\nthe purely hypothetical explanation afforded by _per saltum_ variations.\n\nBut the hypothesis of \u201cheterogeneous generation\u201d is not only without a\nbasis of facts--it can also be directly shown to be untenable. Since\nthe operation of an internal power of transformation does not explain\nadaptation to the conditions of life, the claims of natural selection\nto explain these transformations must be admitted; but the co-operation\nof a phyletic vital force and natural selection is inconceivable if we\nimagine the modifications to occur _per saltum_.\n\nThe supposed \u201cheterogeneous generation\u201d is always illustrated by the\nexample of alternation of generation. The origination of a new animal\nform is thus conceived to take place in the same manner as we now see,\nin the cyclical reproduction of the Medus\u00e6, free swimming, bell-shaped\nMedusoids, produced from fixed polypites, or _Cercari\u00e6_ from Trematode\nworms by internal budding; in brief, it is imagined that one animal\nform suddenly gives rise to another widely deviating form by purely\ninternal causes. Now on this theory it would be an unavoidable\npostulate, that by such a process of _per saltum_ development there\narises not merely a new type of some species, but at the same time\nindividuals capable of living and of persisting under, and fitted to,\ngiven conditions of life. But every naturalist who has attempted to\ncompletely explain the relation between structure and mode of life\nknows that even the small differences which separate one species from\nanother, always comprise a number of minute structural deviations which\nare related to well defined conditions of life--he knows that in every\nspecies of animal the whole structure is adapted in the most exact\nmanner _in every detail_ to special conditions of life. It is not an\nexaggeration when I say in every detail, since the so-called \u201cpurely\nmorphological parts\u201d could not be other than they are without causing\nchanges in other parts which exercise a definite function. I will not\nindeed assert that in the most closely related species all the parts\nof the body must in some manner differ from one another, if only to a\nsmall extent; it seems to me not improbable, however, that an exact\ncomparison would very frequently give this result. That animals which\nare so widely removed in their morphological relations as Medus\u00e6 and\nPolypes, or Trematoda and their \u201cnurses,\u201d are differently constructed\nin each of their parts can, however, be stated with certainty.\n\nNow if this wide deviation in every part were in itself no obstacle to\nthe assumption of a designing and re-modelling power, it would become\nso by the circumstance that all the parts of the organism must stand in\nthe most precise relation to the external conditions of life, if the\norganism is to be capable of existing--all the parts must be exactly\nadapted to certain conditions of life. How can this be brought about\nby a transforming force acting spasmodically? Von Hartmann--who, in\nspite of his clear perception and widely extended scientific knowledge,\ncannot possibly possess a strong conviction of that harmony between\nstructure and life-conditions prevailing throughout the whole system\nof the organism, and which personal research and contemplation are\nalone able to give--simply bridges over the difficulty by permitting\nnatural selection to come to his aid as an \u201cauxiliary principle\u201d of\nthe re-modelling power. It would not be supposed that naturalists\nwould resort to the same device--nevertheless those who support the\nphyletic force and _per saltum_ development generally invoke natural\nselection as the principle which governs adaptation. But when does\nthis agency come into operation? When by germinal metamorphosis a\nnew form has arisen, this, from the first moment of its existence,\nmust be adapted to the new conditions of life or it must perish. No\ntime is allowed for it to continue in an unadapted state throughout\na series of generations until adaptation is luckily reached through\nnatural selection. Let us have either natural selection or a phyletic\nforce--both together are inconceivable. If there exists a phyletic\nforce, then it must itself bring about adaptation.\n\nIt might perhaps be here suggested that the same objection applies to\nthat process of modification which is effected by small steps, but that\nit does so only when the change occurs suddenly. This, however, as I\nhave already attempted to show, but very rarely takes place; in many\ncases (mimicry) the conditions even change in the first place through\nthe change in form and therefore, as is evident, as gradually as the\nlatter. It must be the same in all other cases where transformation of\nthe existing form and not merely extinction of the species concerned\ntakes place. The transmutation must always keep pace with the change\nin the conditions of life, since if the latter change more rapidly the\nspecies could not compete with rival species--it would become extinct.\n\nThe abrupt transformation of species implies sudden change in the\nconditions of life, since a Medusa does not live like a Polype,\nnor a Trematode like its \u201cnurse.\u201d For this reason it is impossible\nthat natural selection can be an aiding principle of \u201cheterogeneous\ngeneration.\u201d If such abrupt transformation takes place it must produce\nthe new form instantly equipped for the struggle for existence,\nand adapted in all its organs and systems of organs to the special\nconditions of its new life. But would not this be \u201cpure magic\u201d? It\nis not thereby even taken into consideration that here--as in the\ncases of mimicry--time and place must agree. The requirements of a\npre-established harmony (\u201c_pr\u00e4stabilirte Harmonie_\u201d) further demand\nthat an animal fitted for special conditions of life should only make\nits appearance at that precise period of the earth\u2019s history when these\nspecial conditions are all fulfilled, and so forth.\n\nBut he who has learnt to perceive the numerous and fine relations\nwhich, in every species of animal, bring the details of structure into\nharmony with function, and who keeps in view the impelling power of\nthese conditions, cannot possibly hold to the idea of a _per saltum_\ndevelopment of animal forms. If development has taken place, it must\nhave occurred gradually and by minute steps--in such a manner indeed\nthat each modification had time to become equilibrated to the other\nparts, and in this way a succession of modifications gradually brought\nabout the total transformation of the organism, and at the same time\nsecured complete adaptation to new conditions of life.\n\nNot only abrupt modification however, but every transformation is\nto be rejected when based upon the interference of a metaphysical\nprinciple of development. Those to whom the arguments already advanced\nagainst such a principle appear insufficient may once more be asked,\nhow and where should this principle properly interfere? I am of\nopinion that one effect can have but one sufficient cause; if this\nsuffices to produce it, no second cause is required. The hand of a\nwatch necessarily turns once round in a circle in a given time as soon\nas the spring which sets the mechanism in movement is wound up; in an\nunwound watch a skilful finger can perhaps give the same movement to\nthe hand, but it is impossible that the latter can receive both from\nthe operator and from the spring at _the same time, the same motion_ as\nthat which it would receive through either of these two powers _alone_.\nIn the same manner it appears to me that the variations which lead to\ntransformation cannot be at the same time determined by physical and by\nmetaphysical causes, but must depend upon either one or the other.\n\nOn no side will it be disputed that at least one portion of the\nprocesses of organic life depends upon the mechanical co-operation of\nphysical forces. How is it conceivable that sudden pauses should occur\nin the course of these causal forces, and that a directive power should\nbe substituted therefor, the latter subsequently making way again for\nthe physical forces? To me this is as inconceivable as the idea that\nlightning is the electric discharge of a thunder-cloud, of which the\nformation and electrical tension depends upon causal forces, and of\nwhich the time and place are purely determined by such forces, but that\nJupiter has it nevertheless in his power to direct the lightning flash\naccording to his will on to the head of the guilty.\n\nNow although I deny the possibility or conceivability of the\ncontemporaneous co-operation of teleological and of causal forces\nin producing any effect, and although I maintain that a purely\nmechanical conception of the processes of nature is alone justifiable,\nI nevertheless believe that there is no occasion for this reason to\nrenounce the existence of, or to disown, a directive power; only we\nmust not imagine this to interfere directly in the mechanism of the\nuniverse, but to be rather behind the latter as the final cause of this\nmechanism.\n\nVon Baer himself points this out to us, although he does not follow\nup the complete consequences of his arguments. He especially insists\nin his book, which abounds in beautiful and grand ideas, that the\nnotions of necessity (causality) and of purpose by no means necessarily\nexclude one another, but rather that they can be connected together in\na certain manner. Thus, the watchmaker attains his end, the watch, by\ncombining the elastic force of a spring with wheel-work, _i.e._ by\nutilizing physical necessities; the farmer accomplishes his purpose,\nthat of obtaining a crop of corn, by sowing the seed in suitable land,\nbut the seed must germinate as an absolute necessity when exposed to\nthe influences of warmth, soil, moisture, &c. Thus, in these instances\na chain of necessities is undoubtedly connected with a teleological\nforce, the human will; and it directly follows from such cases that\nwherever we see an aim or result attained through necessities, the\ndirective force does not interrupt the course of the series of\nnecessities which have already commenced, but is active before the\nfirst commencement of these necessities, since it combines and sets the\nlatter in movement. From the moment when the mechanism of the watch\nis combined harmoniously and the spring wound up, it goes without the\nfurther interference of the watchmaker, just as the corn-seed when once\nplaced in the earth develops into a plant without assistance from the\nfarmer.\n\nIf we apply this argument to the development of the organic world,\nthose who defend mechanical development will not be compelled to deny a\nteleological power, only they would have with Kant[309] to think of the\nlatter in the only way in which it can be conceived, viz. _as a Final\nCause_.\n\nIn the region of inorganic nature nobody any longer doubts the purely\nmechanical connection of the phenomena. Sunshine and rain do not now\nappear to us to be whims of a deity, but divine natural laws. As the\nknowledge of the processes of nature advances, the point where the\ndivine power designedly interrupts these processes must be removed\nfurther back; or, as the author of the criticism of the philosophy\nof the Unconscious[310] expresses it, all advance in the knowledge\nof natural processes depends \u201cupon the continual elimination of the\nidea of the miraculous.\u201d We now believe that organic nature must be\nconceived as mechanical. But does it thereby follow that we must\ntotally deny a final Universal Cause? Certainly not; it would be a\ngreat delusion if any one were to believe that he had arrived at a\ncomprehension of the universe by tracing the phenomena of nature to\nmechanical principles. He would thereby forget that the assumption\nof eternal matter with its eternal laws by no means satisfies our\nintellectual need for causality. We require before everything an\nexplanation of the fact that relationships everywhere exist between the\nparts of the universe--that atoms everywhere act upon one another.[311]\nHe who can content himself with the assumption of matter may do so,\nbut he will not be able to show that the assumption of a Universal\nCause underlying the laws of nature is erroneous.\n\nIt will not be said that there is no advantage in assuming such a\nFinal Cause, because we cannot conceive it, and indeed cannot so much\nas demonstrate it with certainty. It certainly lies beyond our power\nof conception, in the obscure region of metaphysics, and all attempts\nto approach it have never led to anything but an image or a formula.\nNevertheless there is an advance in knowledge in the assumption of this\nCause which well admits of comparison with those advances which have\nbeen led to by certain results of the new physiology of the senses. We\nnow know that the images which give us our sense of the external world\nare not \u201cactual representations having any degree of resemblance,\u201d[312]\nbut are only signs for certain qualities of the outer world, which\ndo not exist as such in the latter, but belong entirely to our\nconsciousness. Thus we know for certain that the world is not as we\nperceive it--that we cannot perceive \u201cthings in their essence\u201d--and\nthat the reality will always remain transcendental to us. But who will\ndeny that in this knowledge there is a considerable advance, in spite\nof its being for the most part of a negative character? But just as we\nmust assume behind the phenomenal world of our senses an actual world\nof the true nature of which we receive only an incomplete knowledge\n(_i.e._ a knowledge corresponding only in reality with the relations of\ntime and space), so behind the co-operating forces of nature which \u201caim\nat a purpose\u201d must we admit a Cause, which is no less inconceivable\nin its nature, and of which we can only say one thing with certainty,\nviz., that it must be teleological. Just as the former first leads\nus to perceive the true value of our sensual impressions, so does\nthe latter knowledge lead us to foresee the true significance of the\nmechanism of the universe.\n\nIt is true that in neither case do we learn more than that there is\nsomething present which we do not perceive, but in both instances this\nknowledge is of the greatest value. The consciousness that behind\nthat mechanism of the universe which is alone comprehensible to us\nthere still lies an incomprehensible teleological Universal Cause,\nnecessitates quite a different conception of the universe--a conception\nabsolutely opposed to that of the materialist. Most correctly and\nbeautifully does Von Baer say that \u201ca purpose cannot be otherwise\nconceived by us than as proceeding from a will and consciousness; in\nthis would the \u2018aiming at a purpose,\u2019 which appears to us as reasonable\nas it is necessary, have its deepest root.\u201d If we conceive in this\nworld a divine Universal Power exercising volition as the ultimate\nbasis of matter and of the natural laws resident therein, we thus\nreconcile the apparent contradiction between the mechanical conception\nand teleology. In the same way that Von Hartmann, somewhere speaks of\nthe immanent teleology of a machine, we might speak of the immanent\nteleology of the universe, because the single forces of matter are\nso exactly adjusted that they must give rise to the projected world,\njust as the wheels and levers of a machine bring forth a required\nmanufactured article. I admit that these are grossly anthropomorphic\nideas. But as mortals can we have any other ideas? Is not the notion\nof purpose in itself an equally anthropomorphic one? and is there\nany certainty that the idea of causality is less so? Do we know that\ncausality is unlimited, or that it is universally valid? In the\nabsence of this knowledge, should it not be permissible to satisfy\nas far as we can the craving of the human mind for a spiritual First\nCause of the universe, by speaking of it in terms conceivable to\nhuman understanding? We can take up such a final position and still\nbe conscious that we thereby form no certain conception, and indeed\ncome no nearer to the reality. The materialist still makes use of the\nnotion of \u201ceternity,\u201d and frequently handles it as though it were a\nperfectly known quantity. We nevertheless do not seriously believe that\nby the expression \u201ceternal matter,\u201d any true idea resulting from human\nexperience is gained.\n\nIf it is asked, however, how that which in ourselves and in the\nremainder of the animal world is _intellectual_ and _perceptive_, which\n_thinks_ and _wills_, is ascribable to a mechanical process of organic\ndevelopment--whether the development of the mind can be conceived as\nresulting from purely mechanical laws? I answer unhesitatingly in the\naffirmative with the pure materialist, although I do not agree with\nhim as to the manner in which he derives these phenomena from matter,\nsince thinking and extension are heterogeneous things, and one cannot\nbe considered as a product of the other. But why should not the ancient\nnotion of \u201cconscious matter\u201d given out by Maupertuis and Robinet, not\nbe again entertained, as pointed out in recent times by Fechner?[313]\nWould there not thus be found a useful formula for explaining phenomena\nhitherto quite incomprehensible?\n\nVon Hartmann in criticizing himself, designates the sensibility of\natoms as an \u201calmost inevitable hypothesis\u201d (p. 62), \u201cinevitable\nbecause if sensibility were not a general and original property of the\nconstituent elements of matter, it would be absolutely incomprehensible\nhow through its potentiality and integration that sensibility known\nto us as being possessed by the organism could have arisen.\u201d \u201cIt is\nimpossible that from purely external elements devoid of all internality\n(_Innerlichkeit_) there should suddenly appear, by a certain mode\nof combination, an internality which becomes more and more richly\ndeveloped. The more certainly science becomes convinced that in\nthe sphere of externality (_\u00c4usserlichkeit_) the higher (organic)\nphenomena are only results of combination, or are the aggregate\nphenomena of the elementary atomic forces, the more surely, when she\nonce seriously concerns herself with this other question, will she not\nfail to be convinced that the sensibility possessed by higher stages\nof consciousness can be only combination-results, or the aggregate\nphenomena of the elementary sensations of atoms, although these\natomic sensations as such always remain below the level of the higher\ncombinations of consciousness.\u201d In confusing this double-sided nature\nof the objective phenomenon \u201clies the main error of all materialism\nand of all subjective idealism. Just as the attempt of the latter\n(subjective idealism) to construct the external phenomena of existence\nin space out of functions of internality and their combinations is\nimpossible, so is the endeavour of the former (materialism) to build up\ninternal sensation out of any combinations of force acting externally\nin space equally impossible.\u201d\n\nI have no intention of going any deeper into these questions. I mention\nthem only in order to point out that even from this side there appears\nto me no obstacle in the way of a purely mechanical conception of the\nprocesses of the universe. The naturalist may be excused if he attempts\nto penetrate into the region of philosophy; it arises from the wish to\nbe able to contribute a little towards the reconciliation of the latest\nknowledge of the naturalist with the religious wants of the human\nmind--towards the aim striven for by both sides, viz. a satisfactory\nand harmonious view of the universe, according with the state of\nknowledge of our time.\n\nI believe that I have shown that the theory of selection by no means\nleads--as is always assumed--to the denial of a teleological Universal\nCause and to materialism, and I thereby hope that I have cleared the\nway for this doctrine, the importance of which it is scarcely possible\nto over-estimate. Many, and not the most ill-informed, do not get so\nfar as to make an unbiassed examination into the facts, because they\nare at the outset alarmed by the to them inevitable consequence of the\nmaterialistic conception of the universe. Mechanism and teleology do\nnot exclude one another, they are rather in mutual agreement. Without\nteleology there would be no mechanism, but only a confusion of crude\nforces; and without mechanism there would be no teleology, for how\ncould the latter otherwise effect its purpose?[314]\n\nVon Hartmann correctly says:--\u201cThe most complete mechanism conceivable\nis likewise the most completely conceivable teleology.\u201d We may thus\nrepresent the phenomenal universe as such a completely conceivable\nmechanism. With this conception vanish all apprehensions that the new\nviews would cause man to lose the best that he possesses--morality and\npurely human spiritual culture. He who, with Von Baer, considers the\nlaws of nature as the \u201cpermanent expressions of the will of a creative\nprinciple,\u201d will clearly perceive that a further advance in the\nknowledge of these laws need not divert man from the path of increasing\nimprovement, but must further him in this course--that the knowledge\nof truth, whatever may be its purport, cannot possibly be considered\na backward step. Let us take our stand boldly on the ground of new\nknowledge, and accept the direct consequences thereof, and we shall\nnot be obliged to give up either morality or the comforting conviction\nof being part of an harmonious world, as a necessary member capable of\ndevelopment and perfection.\n\nAny other mode of interference by a directive teleological power in\nthe processes of the universe than by the appointment of the forces\nproducing them, is however, at least to the naturalist, inadmissible.\nWe are still far removed from completely understanding the mechanism by\nmeans of which the organic world is evoked--we still find ourselves at\nthe very beginning of knowledge. We are, however, already convinced\nthat both the organic and the inorganic worlds are dependent only upon\nmechanical forces, for to this conclusion we are led, not only by the\nresults of investigators who have restricted themselves to limited\nprovinces, but also by the most general considerations. But although\nthe force of these arguments may not be acknowledged, and although\none might maintain that the inductional proofs against the existence\nof a \u201cphyletic vital force\u201d have been directed only against points of\ndetail, or have never been completely demonstrated, _i.e._ for all\npoints, it must nevertheless be conceded, that for the naturalist\nthe mechanical conception of Nature is the only one possible--that\nhe is not at all justified in abandoning this view so long as the\ninterference of teleological forces _in the course_ of the processes\nof organic development has not been demonstrated to him. Thus, it will\nnot be immaterial whether a conception of Nature which to many seems\ninevitable is consistent with the idea of universal design, or a final\ndirective universal principle, since the value which we may attach to\nour own lives and aims, essentially depends thereon. The final and main\nresult of this essay will thus be found in the attempted demonstration\nthat the mechanical conception of Nature very well admits of being\nunited with a teleological conception of the Universe.\n\n\nTHE END.\n\n\n\n\nFOOTNOTES\n\n\n[171] [The slight variability in the colour of this pupa, opens up\nthe interesting question of the photographic sensitiveness of this\nand other species, which is stated to cause them to assimilate in\ncolour to the surface on which the larva undergoes its final ecdysis.\nSome experiments upon this subject have been recorded by Mr. T. W.\nWood, Proc. Ent. Soc. 1867, p. xcix, but the field is still almost\nunexplored. R.M.]\n\n[172] \u201c\u00dcber den Einfluss der Isolirung auf die Artbildung.\u201d Leipzig,\n1872, p. 20.\n\n[173] In some instances _Deilephila Lineata_ has also been seen by day\nhovering over flowers.\n\n[174] It is true that I only reared one brood, but from this fifty\nspecimens were obtained. It would be interesting to know whether this\nvariety of the caterpillar is distributed over the whole of Southern\nEurope.\n\n[175] In this sense Lubbock says:--\u201cIt is evident that creatures which,\nlike the majority of insects, live during the successive periods\nof their existence in very different circumstances, may undergo\nconsiderable changes in their larval organization in consequence of\nforces acting on them while in that condition; not, indeed, without\naffecting, _but certainly without affecting to any corresponding\nextent_, their ultimate form.\u201d--\u201cOrigin and Metamorphoses of Insects,\u201d\nLondon, 1874, p. 39.\n\n[176] \u201cGrundz\u00fcge der Zoologie,\u201d 1875.\n\n[177] [Lepidopterists are of course aware that even these distinctions\nare not absolute, as no single character can be named which does not\nalso appear in certain moths. The definition in this case, as in that\nof most other groups of animals and plants, is only a general one.\nSee, for instance, Westwood\u2019s \u201cIntroduction to the Classification of\nInsects,\u201d vol. ii. pp. 330-332. Also some remarks by C. V. Riley in his\n\u201cEighth Annual Report\u201d on the insects of Missouri, 1876, p. 170. With\nreference to the antenn\u00e6 as a distinguishing character, see Mr. A. G.\nButler\u2019s article in \u201cScience for All,\u201d 1880, part xxvii. p. 65. R.M.]\n\n[178] The genus of _Morphin\u00e6, Discophora_, possesses hairs very similar\nto those of the genus _Cnethocampa_ belonging to the _Bombycid\u00e6_.\n\n[179] [The larv\u00e6 of genera 14, _Phyciodes_, and 35, _Crenis_, are\nlikewise spiny. See Edwards\u2019 \u201cButt. of N. Amer.\u201d vol. ii. for figures\nof the caterpillar of _Phyc. Tharos_: for notes on the larv\u00e6 of\n_Crenis Natalensis_ and _C. Boisduvali_ see a paper by W. D. Gooch,\n\u201cEntomologist,\u201d vol. xiv. p. 36. The larv\u00e6 of genus 55, _Ageronia_, are\nalso spiny. (See Burmeister\u2019s figure of _A. Arethusa_, \u201cL\u00e9p. R\u00e9p. Arg.\u201d\nPl. V. Fig. 4). The larv\u00e6 of genus 98, _Aganisthos_, also appear to be\nsomewhat spiny (see Burmeister\u2019s figure of _A. Orion_, _loc. cit._ Pl.\nV. Fig. 6), and this raises the question as to whether the genus is\ncorrectly located in its present position. The larv\u00e6 of the following\ngenera figured in Moore\u2019s \u201cLepidoptera of Ceylon,\u201d parts i. and ii.,\nare all spiny:--6, _Cirrochroa_ (Pl. XXXII.); 7, _Cynthia_ (Pl. XXVI.);\n27, _Kallima_ (Pl. XIX.); and 74, _Parthenos_ (Pl. XXIV.). Many species\nof caterpillars which are spiny when adult appear to be spineless,\nor only slightly hairy when young. See Edwards\u2019 figures of _Melit\u00e6a\nPhaeton_, _Argynnis Diana_, and _Phyc. Tharos_ (_loc. cit._) and his\ndescription of the larva of _Arg. Cybele_, \u201cCanad. Entom.\u201d vol. xii. p.\n141. The spiny covering thus appears to be a character acquired at a\ncomparatively recent period in the phyletic development. R.M.]\n\n[180] [The larv\u00e6 of the 110th genus, _Paphia_, Fabr. (_An\u00e6a_, H\u00fcbn.)\nare also smoothed-skinned. See Edwards\u2019 figure (_loc. cit._ vol. i. Pl.\nXLVI.) of _P. Glycerium_. Also C. V. Riley\u2019s \u201cSecond Annual Report\u201d on\nthe insects of Missouri, 1870, p. 125. Burmeister figures the larva of\na species of _Prepona_ (genus 99) which is smooth (_P. Demophon_, _loc.\ncit._ Pl. V. Fig. 1). The horns on the head of _Apatura_, &c., may\npossibly be a survival from a former spiny condition. R.M.]\n\n[181] \u201cSynopsis of the described Lepidoptera of North America.\u201d\nWashington, 1862.\n\n[182] \u201cCatalog der Lepidopteren des Europ\u00e4ischen Faunengebietes.\u201d\nDresden, 1871.\n\n[183] This group of moths (\u201cSchw\u00e4rmer\u201d) is regarded as of very\ndifferent extents by systematists; when I here comprise under it only\nthe _Sphingid\u00e6_ proper and the _Sesiid\u00e6_, I by no means ignore the\ngrounds which favour a greater extension of the group; the latter is\nnot rigidly limited. [The affinities of the _Sesiid\u00e6_ (_\u00c6geriid\u00e6_) are\nby no means clearly made out: it appears probable that they are not\nrelated to the _Sphingid\u00e6_. See note 160, p. 370. R.M.]\n\n[184] [For Mr. A. G. Butler\u2019s observations on the genus _Acronycta_,\nsee \u201cTrans. Ent. Soc.\u201d 1879, p. 313; and note 68, p. 169, of the present\nvolume. R.M.]\n\n[185] [The following characters are given in Stainton\u2019s \u201cManual of\nBritish Butterflies and Moths,\u201d vol. i. p. 114:--\u201cLarva of very\nvariable form: at one extreme we find the singular _Cerura_ larv\u00e6,\nwith only fourteen legs, and two long projecting tails from the last\nsegment; at the other extreme we have larv\u00e6 with sixteen legs and no\npeculiarity of form, such as _Chaonia_ and _Bucephala_; most have,\nhowever, the peculiarity of holding the hind segment of the body erect\nwhen in repose; generally quite naked, though downy in _Bucephala_ and\nrather hairy in _Curtulu_; very frequently there are projections on the\nback of the twelfth segment.\u201d R.M.]\n\n[186] Encyl. Meth. ix. p. 310.\n\n[187] [The genus _Vanessa_ (in the wide sense) appears to be in a\nremarkable condition of what may be called phyletic preservation.\nThus, the group of species allied to _V. C.-album_ passes by almost\ninsensible steps into the group of butterflies typified by our\n\u201cTortoiseshells.\u201d The following is a list of some of the intermediate\nspecies in their transitional order:--_I.-album_, _V.-album_, _Faunus_,\n_Comma_, _California_, _Dryas_, _Polychloros_, _Xanthomelas_,\n_Cashmirensis_, _Urtic\u00e6_, _Milberti_, &c. Similarly, our _Atalanta_\nand _Cardui_ are connected by a number of intermediate forms, showing\na complete transition from the one to the other. The following is the\norder of the species so far as I am acquainted with them:--_Atalanta_,\n_Dejeanii_, _Callirho\u00eb_, _Tammeamea_, _Myrinna_, _Huntera_,\n_Terpsichore_, _Carye_, _Kershawii_, and _Cardui_. R.M.]\n\n[188] \u201cProdromus Systematis Lepidopterorum.\u201d Regensburg, 1864.\n\n[189] [The larva of _Acherontia Morta_, figured by Butler (see note 121,\np. 262), possesses the characteristically recurved horn; that of _Ach.\nMedusa_ figured by the same author, does not appear to possess this\ncharacter in any marked degree. R.M.]\n\n[190] [See note 97, p. 233. R.M.]\n\n[191] _Loc. cit._ Pl. XXV. [This species is referred by Butler to the\ngenus _Paonias_, H\u00fcbn. R.M.]\n\n[192] Abbot and Smith, Pl. XXIX. [Placed by Butler in the genus\n_Cressonia_, Grote and Robinson. Abbot and Smith state that this\nlarva is sometimes green. According to Mr. Herman Strecker (Lepidop.\nRhopal. and Hetero, Reading, Pa. 1874, p. 54) it feeds upon black\nwalnut (_Juglans Nigra_), hickory (_Carya Alba_), and ironwood\n(_Ostrya Virginica_). Of the North American species of _Smerinthus_,\nthe following, in addition to _Exc\u00e6catus_, closely resemble our\n_Ocellatus_:--_S. (Calasymbolus) Geminatus_, Say; _(C.) Cerisii_,\nKirby; and _Ophthalmicus_, Boisd. In addition to _S. (Cressonia)\nJuglandis_, _S. (Triptogon) Modesta_ much resembles our _Populi_. The\nlarva of _Geminatus_, according to Strecker, is \u201cpale green, lightest\nabove, with yellow lateral granulated stripes; caudal horn violet;\nstigmata red. It feeds on the willow.\u201d R.M.]\n\n[193] Cat. Brit. Mus.\n\n[194] [This lengthening of the true legs is mimetic according to\nHermann M\u00fcller, and causes the anterior portion of the caterpillar to\nresemble a spider. See note 129, p. 290. R.M.]\n\n[195] [Certain butterflies appear to be crepuscular, if not nocturnal\nin their habits. Thus in his \u201cNotes on the Lepidoptera of Natal,\u201d Mr.\nW. D. Gooch states that he never saw _Melanitis_, _Leda_, or _Gnophodes\nParmeno_ on the wing by day, but generally during the hour after\nsunset. He adds:--\u201cMy sugar always attracted them freely, even up to\n10 or 11 p.m.\u201d Many species of _Hesperid\u00e6_ are also stated to be of\ncrepuscular habits by this same observer. See \u201cEntomologist,\u201d vol xvi.\npp. 38 and 40. R.M.]\n\n[196] I only make this assumption for the sake of simplicity, and not\nbecause I am convinced that the existing _Rhopalocera_ are actually the\noldest Lepidopterous group.\n\n[197] Zeitschrift f\u00fcr wissenschaftl. Zoologie, vol. xx. p. 519.\n\n[198] [See for instance Lubbock\u2019s \u201cOrigin and Metamorphoses of\nInsects,\u201d chap. iii.; and F. M. Balfour\u2019s \u201cComparative Embryology,\u201d\nvol. i., 1880, pp. 327--356. This last work contains an admirable\n_r\u00e9sum\u00e9_ of our knowledge of the embryonic development of insects up to\nthe date of publication. R.M.]\n\n[199] Are not the 4th, 11th, and 12th segments destitute of the\nrudiments of legs as in the larv\u00e6 of all existing saw-flies? I might\nalmost infer this from B\u00fctschli\u2019s figures (see for instance Pl. XXV.,\nFig. 17A).\n\n[200] [The grub-formed Hymenopterous larv\u00e6, like the larv\u00e6 of all\nother holometabolous insects, thus represent an acquired degenerative\nstage in the development, _i.e._ an adaptation to the conditions of\nlife at that stage. Bearing in mind the above-quoted observations of\nB\u00fctschli and the caterpillar-like form of the Terebrantiate group of\nHymenopterous larv\u00e6, the following remarks of Balfour\u2019s (_loc. cit._ p.\n353), appear highly suggestive:--\u201cWhile in a general way it is clear\nthat the larval forms of insects cannot be expected to throw much light\non the nature of insect ancestors, it does nevertheless appear to me\nprobable that such forms as the caterpillars of the Lepidoptera are not\nwithout a meaning in this respect. It is easy to conceive that even a\nsecondary larval form may have been produced by the prolongation of one\nof the embryonic stages; and the general similarity of a caterpillar to\n_Peripatus_, and the retention by it of post-thoracic appendages, are\nfacts which appear to favour this view of the origin of the caterpillar\nform.\u201d See also Sir John Lubbock, _loc. cit._, pp. 93 and 95. R.M.]\n\n[201] [In the most recent works dealing with this order six\ngroups, based on the character of the imagines are recognized,\nviz.:--_Tubulifera_, _Terebrantia_, _Pupivora_, _Heterogyna Fossores_,\nand _Mellifera_. (See, for instance, F. P. Pascoe\u2019s \u201cZoological\nClassification,\u201d 2nd ed. p. 147.) Of these groups the larv\u00e6 of the\n_Terebrantia_ as thus restricted are all of the caterpillar type\n(_Tenthredinid\u00e6_ and _Siricid\u00e6_), whilst those of the other groups are\nmaggot-shaped. For a description of the development of the remarkable\naberrant larva of _Platygaster_, see Ganin in Zeit. f. wissenschaftl.\nZool., vol. xix. 1869. R.M.]\n\n[202] [For recent investigations on the structure of the thorax in\nDiptera, see a paper by Mr. A. Hammond, in Journ. Linn. Soc., Zoology,\nvol xv. p. 9. R.M.]\n\n[203] I am familiar with the fact that the two sub-orders of true\nDiptera, the short-horned (_Brachycera_), and the long-horned\n(_Nemocera_), are not sharply limited; and I am likewise well\nacquainted with the circumstance that there are forms which connect the\ntwo larval types. The connecting forms of the imagines do not, however,\nalways coincide with the intermediate larval forms, so that there\nhere arises a second and very striking incongruence of morphological\nrelationship which depends only upon the circumstance that the one\nstage has diverged in form more widely than the other through a greater\ndivergence in the conditions of life. The difficulty is in these cases\naggravated because an apparent is added to the true form-relationship\nthrough convergence, so that without going into exact details the form\nand genealogical relationships of the Diptera cannot be distinguished.\nIt would be of great interest for other reasons to make this\ninvestigation, and I hope to be able to find leisure for this purpose\nat some future period.\n\n[204] \u201cEntwicklung der Dipteren.\u201d Leipzig, 1864.\n\n[205] Lubbock concludes from the presence of thoracic legs in the\nembryonic larva of bees that these have been derived from a larva of\nthe _Campodea_ type, but he overlooks the fact that the rudiments of\nthe abdominal legs are also present; _loc. cit._, p. 28.\n\n[206] \u201cF\u00fcr Darwin,\u201d Leipzig, 1864, p. 8.\n\n[207] Mem. Peabody Acad. of Science, vol. i. No. 3.\n\n[208] Verhandl. Wien. Zoolog. Botan. Gesellsch. 1869, p. 310.\n\n[209] \u00dcber Ontogenie und Phylogenie der Insekten. Eine akademische\nPreisschrift. Jen. Zeitschrift. Bd. x. Neue Folge, iii. Heft 2. 1876.\n[Some remarks by F. M. Balfour on the origin of certain larval forms\nhave already been quoted in a previous note (p. 485). This author\nfurther states:--\u201cThe fact that in a majority of instances it is\npossible to trace an intimate connection between the surroundings\nof a larva and its organization proves in the clearest way _that\nthe characters of the majority of existing larval forms of insects\nhave owed their origin to secondary adaptations_. A few instances\nwill illustrate this point:--In the simplest types of metamorphosis,\n_e.g._ those of the Orthoptera genuina, the larva has precisely the\nsame habits as the adult. We find that a caterpillar form is assumed\nby phytophagous larv\u00e6 amongst the Lepidoptera, Hymenoptera, and\nColeoptera. Where the larva has not to go in search of its nutriment\nthe grub-like apodous form is assumed. The existence of such an apodous\nlarva is especially striking in the Hymenoptera, in that rudiments\nof thoracic and abdominal appendages are present in the embryo and\ndisappear again in the larva.... It follows from the above that the\ndevelopment of such forms as the Orthoptera genuina is more primitive\nthan that of the holometabolous forms, &c.\u201d Comparative Embryology,\nvol. 1, p. 352. R.M.]\n\n[210] [The _Aphaniptera_ are now recognized in this country as\na sub-order of Diptera. See, for instance, Huxley\u2019s \u201cAnatomy\nof Invertebrated Animals,\u201d p. 425, and Pascoe\u2019s \u201cZoological\nClassification,\u201d 2nd ed. p. 122. R.M.]\n\n[211] [This illustration of course only applies to the old arrangement\nof the Hymenoptera into _Terebrantia_ and _Aculeata_. See also note 201,\np. 488. R.M.]\n\n[212] [Eng. ed. This law is perhaps a little too restricted, inasmuch\nas it is theoretically conceivable that the organism may be able\nto adapt itself to similar conditions of life in different ways;\ndifferences of form could thus depend sometimes upon differences of\nadaptation and not upon differences in the conditions of life, or, as\nI have formerly expressed it, it is not necessary to allow always only\n_one_ best mode of adaptation.]\n\n[213] [It must be understood that the word rendered here and elsewhere\nthroughout this work as \u201ctransformation\u201d is not to be taken in the\nnarrow sense of metamorphosis, but as having the much broader meaning\nof a change of any kind incurred by an organism. Metamorphosis is in\nfact but one phase of transformation. R.M.]\n\n[214] By the Editor.\n\n[215] Mr. C. V. Riley in his excellent \u201cAnnual Reports\u201d already\nquoted in previous notes, states that the larv\u00e6 of _Agrotis Inermis_,\n_Leucania Unipuncta_ (Army-worm), and _L. Albilinea_ are all loopers\nwhen newly hatched. (See First Report, p. 73; Eighth Report, p. 184;\nand Ninth Report, p. 53.)\n\n[216] The following species not referred to in the previous part of\nthis work are figured by Semper (Beit. zur Entwicklungsgeschichte\neiniger ostasiat. Schmet.; Verhandl. d. k.k. zoo. bot. Gesell. in Wien,\n1867):--_Panacra Scapularis_, Walk.; _Ch\u00e6rocampa Clotho_, Drury; and\n_Diludia (Macrosila) Discistriga_, Walk. The following are figured by\nBoisduval and Guen\u00e9e. (Sp\u00e9c. G\u00e9n. 1874):--_Smerinthus Ophthalmicus_,\nBoisd.; _Sphinx Jasminearum_, Boisd.; _S. (Hyloicus) Plebeia_, Fabr.;\n_S. (Hyloicus) Cupressi_, Boisd.; _S. (Pseudosphinx) Catalp\u00e6_, Boisd.;\n_Philampelus Jussiu\u00e6_, H\u00fcbn. (= _Sphinx Vitis_, Linn.?); and _Ceratomia\nAmyntor_, H\u00fcbn. As the works of Abbot and Smith, and Horsfield and\nMoore have been exhausted by Dr. Weismann, it is quite unnecessary\nto extend this note by giving a list of the species figured by these\nauthors.\n\n[217] The same inference has already been drawn with respect to\n_Pterogon (Proserpinus) \u0152nother\u00e6_, see pp. 257, 258.\n\n[218] This would of course be the _fourth_ segment if the head be\nconsidered the first, as on the Continent.\n\n[219] \u201cSecond Annual Report,\u201d 1870, p. 78.\n\n[220] \u201cEntomologist,\u201d vol. xiv. p. 7.\n\n[221] With reference to the habits of _C. Capensis_ (p. 531), I have\nsince been informed by Mr. Trimen that this species does not conceal\nitself by day, so that the dimorphism may be regarded as a character\nretained from an earlier period and adapted to the present life\nconditions.\n\n[222] \u201cKosmos,\u201d Dec. 1877, p. 218. The paper is here introduced chiefly\nwith a view to illustrate an important case of incongruence among\nLepidopterous pup\u00e6.\n\n[223] [Maracuj\u00e1, the local name for the Passiflora. R.M.]\n\n[224] See p. 448.\n\n[225] Verhandl. Schweiz. Naturforsch. Gesellschaft. Einsiedeln, 1868.\n\n[226] [Eng. ed. In 1878 Se\u00f1or Jos\u00e9 M. Velasco published a paper\nentitled \u201cDescription, metamorfosis. y costumbres de una especie nueva\ndel genero _Siredon_.\u201d Memor. Sociedad Mexicana de Historia Natural,\nDecember 26th. See Addendum to this essay.]\n\n[227] Dana and Silliman\u2019s Amer. Journ., 3rd series, i. p. 89. Annals\nNat. Hist. vii. p. 246.\n\n[228] Proc. Zoo. Soc. 1870, p. 160.\n\n[229] Compt. Rend., vol. lx. p. 765 (1865).\n\n[230] Nouvelles Archives du Mus\u00e9um d\u2019Histoire Nat. Paris, 1866, vol.\nii. p. 268.\n\n[231] Proc. Boston Soc., vol. xii. p. 97; Silliman\u2019s Amer. Journ., vol.\nxlvi. p. 364; reference given in \u201cTroschel\u2019s Jahresbericht\u201d for 1868,\np. 37.\n\n[232] Proc. Boston Soc., vol. xii. p. 97; Silliman\u2019s Amer. Journ., vol.\nxlvi. p. 364. I have not been able to get a copy of this paper, and\nquote from a reference in \u201cTroschel\u2019s Jahresbericht.\u201d See preceding\nnote.\n\n[233] Dana and Silliman\u2019s Amer. Journ. See note 3.\n\n[234] Proc. Acad. Philadelph. xix. 1867, pp. 166-209.\n\n[235] M\u00e9m. Acad. Petersb. vol. xvi.\n\n[236] [Eng. ed. Seidlitz is an exception, since in his work on\nParthenogenesis (Leipzig, 1872, p. 13) he states that \u201cIn the Axolotl,\nP\u00e6dogenesis, which is not in this case ... monogamous, but sexual, and\nindeed gyn\u00e6kogenetic, has already become so far constant that it has\nperhaps entirely superseded the orthogenetic reproduction.\u201d]\n\n[237] \u00dcber den Einfluss der Isolirung auf die Artbildung. Leipzig,\n1872, p. 33.\n\n[238] Dum\u00e9ril represents the teeth of the _vomer_ as separated from\nthose of the _os palatinum_ by a gap. This is probably accidental,\nsince Gegenbaur (Friedrich u. Gegenbaur, the skull of Axolotl,\nW\u00fcrzburg, 1849) figures the rows of teeth as passing over from the one\nbone to the other without interruption. This was the case with the\nAxolotls which I have been able to examine on this point; but this\nsmall discrepancy is, however, quite immaterial to the question here\nunder consideration.\n\n[239] See O. Hertwig \u201c\u00dcber das Zahnsystem der Amphibien und seine\nBedeutung f\u00fcr die Genese des Skelets der Mundh\u00f6hle.\u201d Archiv. f\u00fcr\nmicrosc. Anat., vol. xi. Supplement, 1874.\n\n[240] [Eng. ed. These Amblystomas have since died and have been\nminutely described by Dr. Wiedersheim. See his memoir, \u201cZur Anatomie\ndes _Amblystoma Weismanni_,\u201d in Zeit. f\u00fcr wiss. Zool., vol. xxxii. p.\n216.]\n\n[241] See Strauch, _loc. cit._ p. 10.\n\n[242] See Part I. of this volume.\n\n[243] [This is the principle of \u201cDegeneration\u201d recognized by Darwin\n(see \u201cOrigin of Species,\u201d 6th ed. p. 389, and \u201cDescent of Man,\u201d vol.\ni. p. 206), and given fuller expression to by Dr. Anton Dohrn (see\nhis work entitled \u201cDer Ursprung der Wirbelthiere und das Princip des\nFunctionswechsels.\u201d Leipzig, 1875). A large number of cases have been\nbrought together by Prof. E. R. Lankester, in his recent interesting\nwork on \u201cDegeneration, a Chapter in Darwinism.\u201d Nature series, 1880.\nR.M.]\n\n[244] \u201cSulla Larva del _Triton Alpestris_.\u201d Archivio per la Zoologia.\nGenova e Torino, 1861, vol. i. pp. 206-211.\n\n[245] See also Lubbock \u201cOn the Origin and Metamorphoses of Insects,\u201d\nLondon, 1874.\n\n[246] See the first essay \u201cOn the Seasonal Dimorphism of Butterflies,\u201d\np. 82.\n\n[247] [Eng. ed. It has frequently been objected to me that the existing\nAxolotl is not a form resulting from atavism, but a case of \u201carrested\ngrowth.\u201d The expression \u201catavism\u201d is certainly to be here taken in\na somewhat different sense than, for example, in the case of the\nreversion of the existing Axolotl to the Amblystoma form. Further on, I\nhave myself insisted that in the first case the phyletic stage in which\nthe reversion occurred is still completely preserved in the ontogeny\nof each individual, whilst the Amblystoma stage has become lost in the\nontogeny of the Axolotl. If, therefore, we apply the term \u201catavism\u201d\nonly to such characters or stages (_i.e._ complexes of characters) as\nare no longer preserved in the ontogeny, we cannot thus designate the\npresent arrest of the Axolotl at the perennibranchiate stage. Such a\nrestriction of the word, however, appears to me but little desirable,\nsince the process is identical in both cases, _i.e._ it depends upon\nthe same law of heredity, in accordance with which a condition formerly\noccurring as a phyletic stage suddenly reappears through purely\ninternal processes. It is true that the reversion is not _complete_,\n_i.e._ the present sexually mature Axolotl does not correspond in all\ndetails with its perennibranchiate ancestors. Since Wiedersheim has\nshown that the existing Axolotl possesses an intermaxillary gland, this\ncan be safely asserted. This gland occurs only in _land_ Amphibians,\nand therefore originated with the Amblystoma form, afterwards becoming\ntransferred secondarily to the larval stage. Nevertheless, the present\nAxolotl must resemble its perennibranchiate ancestors in most other\ncharacters, and we should be the more entitled to speak of a reversion\nto the perennibranchiate stage as we speak also of the reversion of\nsingle characters. To this must be added that the Axolotl does not\ncorrespond exactly with an Amblystoma larva, since Wiedersheim has\nshown that the space for the intermaxillary gland is present, but that\nthe gland itself is confined to a few tubes which do not by any means\nfill up this space. (\u201cDas Kopfskelet der Urodelen.\u201d Morph. Jahrbuch,\nvol. iii. p. 149). By the expression \u201carrested growth\u201d not much is\nsaid, if at the same time the cause of the arrest is left unstated. But\nwhat can be the cause why the whole organization remains stationary at\nthe perennibranchiate stage, the sexual organs only undergoing further\ndevelopment? Surely only that law or force of heredity known by its\neffects, but obscure with respect to its causes, through which old\nphyletic stages sometimes suddenly reappear, or in other words, that\npower through which reversion takes place. It must not be forgotten\nthat all these cases of \u201clarval reproduction\u201d in Amphibians appear\nsuddenly. The present sexually mature form of the Axolotl has not\narisen by the sexual maturity gradually receding in the ontogeny from\ngeneration to generation, but by the occurrence of single individuals\nwhich were sexually mature in the perennibranchiate stage, these having\nthe advantage over the _Amblystom\u00e6_ in the struggle for existence under\nchanged climatic conditions.\n\nBy admitting a reversion, we perfectly well explain why arrest at the\nperennibranchiate stage can be associated with complete development of\nthe sexual organs; the assumption of an \u201carrested growth\u201d leaves this\ncombination of characters completely unexplained. Moreover, I am of\nopinion that the expressions \u201carrested growth\u201d or \u201creversion\u201d are of\nbut little importance so long as the matter itself is clear.]\n\n[248] See Haeckel\u2019s \u201cAnthropogenie,\u201d p. 449.\n\n[249] \u201cDer Ursprung der Wirbelthiere und das Princip des\nFunctionswechsels,\u201d Leipzig, 1875.\n\n[250] Bull. Soc. Neuch\u00e2tel. vol. viii. p. 192. Reference given in\n\u201cTroschel\u2019s Jahresbericht\u201d for 1869.\n\n[251] Sitzungsberichte d. math. phys. Klasse der Akad. d. Wiss. zu\nM\u00fcnchen, 1875. Heft i.\n\n[252] Compt. Rend. vol. lxviii. pp. 938 and 939.\n\n[253] Archiv f. Naturgeschichte, 1867.\n\n[254] Compt. Rend. vol. v. 1870, p. 70.\n\n[255] Bull. Soc. Neuch\u00e2tel. vol. viii. p. 192. Reference given in\n\u201cTroschel\u2019s Jahresbericht\u201d for 1869.\n\n[256] [Eng. ed. It was mentioned in the German edition of this work\nthat in the spring of 1876 a female Amblystoma of the Jardin des\nPlantes in Paris had laid eggs (see Blanchard in the Compt. Rend. 1876,\nNo. 13, p. 716). Whether these eggs were fertile, or whether they\ndeveloped was not then made known. Thus much was however at the time\nclear, that even if this had been the case, the reproduction of this\nAmblystoma would have been only an _exceptional_ occurrence. At that\ntime there were in the Jardin des Plantes Amblystomas which had been\nkept for more than ten years, and only on one occasion was there a\ndeposition of eggs, and this by only one specimen. That I was correct\nin speaking of the \u201csterility\u201d of these Amblystomas in spite of this\none exception, is proved by the latest communication from the Jardin\ndes Plantes. We learn from this (Compt. Rend. No. 14, July, 1879, p.\n108) that in the years 1877 and 1878 none of the Amblystomas laid any\nmore eggs, although all means were exerted to bring about propagation.\nIn April, 1879, eggs were again laid by one female, and by a second in\nMay. These eggs certainly developed, as did those of 1876, and produced\ntadpoles. These Amblystomas are therefore not absolutely, but indeed\nrelatively sterile. Whilst the Axolotl propagates regularly and freely\nevery year, this occurs with the Amblystoma but rarely and sparsely.\nThe degree of their sterility can only be approximately established\nwhen we know the number of Amblystomas that have since been kept in the\nJardin des Plantes. Unfortunately nothing has been said with respect to\nthis.]\n\n[257] Origin of Species, 6th ed. p. 252.\n\n[258] In plants also reversion forms show sterility in different\ndegrees. Mr. Darwin has called my attention to the fact that the\npeloric (symmetrical) flowers which occasionally appear as atavistic\nforms in _Corydalis solida_ are partly sterile and partly fertile.\nThat in other causes of sterility, and above all by bastardizing, the\nreproductive power is lost in the most varying degrees, has been known\nsince the celebrated observations of K\u00f6lreuter and G\u00e4rtner. [Eng.\ned. An Orchid (_Catasetum tridentatum_) has the sexes separate, and\nthe male flowers (_Myanthus barbatus_) differ considerably from the\nfemale (_Monachanthus viridis_); besides these, there occurs a form\nwith bisexual flowers which must be considered as a reversion (_Cat.\ntridentatum_) and _this is always sterile_. Darwin, \u201cFertilization of\nOrchids,\u201d 2nd ed. p. 199.]\n\n[259] As we do not know the origin of the \u201cParis Axolotl\u201d I must\nrestrict myself in the following remarks to _Siredon Mexicanus_ (Shaw).\n\n[260] M\u00fchlenpfordt, \u201cVersuch einer getreuen Schilderung der Republik\nMejico,\u201d Hanover, 1844, vol. ii. p. 252.\n\n[261] [The specific gravity of sea water (Atlantic), according to the\ndeterminations of Mr. Buchanan on board the \u201cChallenger,\u201d at 15.56\u00b0\nC. varies from 1.0278 to 1.0240. That of the water of the Dead Sea is\n1.17205.--Watts\u2019 \u201cDict. of Chemistry,\u201d vol. v., table, p. 1017. R.M.]\n\n[262] _Loc. cit._ p. 252.\n\n[263] \u201c\u00dcber die specifische Verschiedenheit des gefleckten und des\nschwarzen Erdsalamanders oder Molchs, und der h\u00f6chst merkw\u00fcrdigen, ganz\neigenth\u00fcmlichen Fortpflanzungsweise des Letzteren.\u201d Isis, Jahrg. 1833,\np. 527.\n\n[264] The experiments referred to have not been made known; I am\nindebted for them to a written communication kindly furnished by an\nesteemed colleague.\n\n[265] See M\u00fchlenpfordt\u2019s work already quoted, vol. i.\n\n[266] In the province of botany such a case has already been made\nknown by Fritz M\u00fcller (Botan. Zeitung, 1869, p. 226; 1870, p. 149). I\nmay be here permitted to quote a passage from the letter in which Dr.\nM\u00fcller calls attention to this interesting discovery. \u201cAs a proof of\nthe possibility that a reversion form can again become a persistent\ncharacter in a species or in the allied form of a particular district,\nI may refer you to an _Epidendrum_ of the island of Santa Catharina.\nIn all Orchids (with the exception of _Cypripedium_) only one anther\nis developed; in very rare cases well-formed anthers appear as\nreversions among the aborted lateral anthers of the inner whorl. In the\n_Epidendrum_ mentioned, these are however _always present_.\u201d\n\n[267] [This species is interesting as being ovoviviparous, the young\npassing through the branchiate stage within the body of the mother.\nSome experiments, which were partially successful, were made by\nFr\u00e4ulein v. Chauvin with a view to solve the question whether the\nbranchiate stage could be prolonged by taking the larv\u00e6 directly from\nthe mother before birth and keeping them in water. See \u201cZeit. f\u00fcr\nwissen. Zoo.\u201d vol. xxix., p. 324. R.M.]\n\n[268] See Fatiot, \u201cLes Reptiles et les Batraciens de la haute\nEngadine.\u201d Geneva, 1873.\n\n[269] I can remember at Upper Engadine a peculiar kind of preserved\nbeef, prepared by simply drying in the air; also the mummification of\nentire human bodies by drying in the open air, as is practised at Great\nSt. Bernard.\n\n[270] \u201cFaune des Vert\u00e9br\u00e9s de la Suisse,\u201d vol. iii. \u201cHistoire Naturelle\ndes Reptiles et des Batraciens.\u201d Geneva, 1873.\n\n[271] See Wiedersheim, \u201cVersuch einer gleichenden Anatomie der\nSalamandrinen.\u201d W\u00fcrzburg, 1875.\n\n[272] See Gen\u00e9, \u201cMemorie della Reale Acad. di Torino,\u201d vol. i.\n\n[273] _Rana esculenta_ never reaches Alpine regions, this species not\nhaving been found higher than 1100 meters. (Fatiot, _loc. cit._, p.\n318.)\n\n[274] See also the excellent work upon Mexico by M\u00fchlenpfordt already\nquoted, vol. i., pp. 69-76.\n\n[275] \u201cEssai politique sur le Royaume de la Nouvelle Espagne,\u201d 1805, p.\n291.\n\n[276] [The expression made use of by the author, viz. \u201cDiluvialzeit,\u201d\nwould perhaps be more in harmony with the views of English geologists\nif rendered as the \u201cpluvial period,\u201d thereby indicating the period of\nexcessive rainfall which, according to Mr. Alfred Tylor, succeeded\nto and was a consequence of the thawing of the great glaciers which\naccumulated during the last glacial epoch. There is abundant evidence\nto show that during the latter period glacial action extended in North\nAmerica at least as far south as Nicaragua. See Belt on \u201cThe Glacial\nPeriod in North America,\u201d Trans. Nova Scotian Inst. of Nat. Sci. 1866,\np. 93, and \u201cThe Naturalist in Nicaragua,\u201d pp. 259-265. R.M.]\n\n[277] [Eng. ed. A memoir by Samuel Clarke has since been published upon\nthe embryonic development of _Amblystoma punctatum_, Baird. Baltimore,\n1879.]\n\n[278] [Eng. ed. See this author\u2019s work, \u201cDas Kopfskelet der Urodelen.\u201d\nLeipzig, 1877, p. 149.]\n\n[279] [See preceding note 52. R.M.]\n\n[280] See note 226, p. 566.\n\n[281] [Prof. Semper also remarks (\u201cAnimal Life,\u201d note 47, p. 430) with\nreference to the Axolotl of Lake Como in the Rocky Mountains, which he\nstates always becomes transformed into _Amblystoma Mavortium_, that\nthis metamorphosis \u201ctakes place in the water, and the Amblystomas, so\nlong as they are little, actually live exclusively in the water, as I\nknow by my own experience. A young Amblystoma which I kept alive for a\nlong time, never went out of the water of its own free will, while one\nnearly twice as large lives entirely on land and only takes a bath now\nand then. It always goes into the water when the temperature of the\nair in the cellar, in which my aquaria stand, falls below that of the\nwater--down to about 6\u00b0 or 8\u00b0 C.\u201d This statement appears to suggest\nthat the effect of temperature may be a factor in some way concerned\nin these interesting cases of transformation, and would in any case\nbe well worthy of experimental investigation. Some further details\nconcerning the _Siredon Lichenoides_ of Lake Como have been recently\npublished by Mr. W. E. Carlin (Proc. U.S. National Museum, June, 1881).\nThe lake, which is shallow, is fed by a constant stream of fresh water,\nbut the water of the lake is intensely saline. The Siredon never enter\nthe fresh water stream, but congregate in large numbers in the alkaline\nwaters of the lake. \u201cWhen about one hundred and fifty were placed\nin fresh water they seemed to suffer no inconvenience, but it had a\nremarkable effect in hastening their metamorphosis into the Amblystoma\nform. Of an equal number kept in fresh water and in the lake water,\nquite a change occurred with the former after twenty-four hours, while\nthe latter showed no change after several days of captivity. Those that\nwere kept well fed in jars usually began to show a slight change in\nfrom two to three weeks, and all of them completed the change into the\nAmblystoma inside of six weeks, while in some kept, but not specially\nfed, there were but three changes in three months.\u201d (Nature, Aug. 25th,\n1881, p. 388.) R.M.]\n\n[282] [Some experiments on the transformation of the Crustacean\n_Artemia Salina_ into _A. Milhausenii_ by gradually increasing the\nsaltness of the water, and conversely, the transformation of _A.\nMilhausenii_ into _A. Salina_ by diminishing the saltness of the water,\nhave been made by Schmankewitsch (Zeitschrift f. wiss. Zool. xxv.\nSuppl. 103 and xxix. 429), but the changes which occur here are much\nless considerable than in the case of the Axolotl. R.M.]\n\n[283] \u201cReden und kleinere Aufs\u00e4tze, Th. II.: Studien aus dem Gebiete\nder Naturwissenschaften.\u201d St. Petersburg, 1876, p. 81.\n\n[284] This obviously does not imply that the naturalist should not\ninvestigate Nature\u2019s processes, and not only correlate these, but also\nwork them up into a universal conception; this is indeed both desirable\nand necessary if natural knowledge is to be regarded in its true value.\nThe naturalist by this means becomes a philosopher, and the vitality\nof the so-called \u201cnatural philosopher\u201d has been inspired, not by the\nnecessity for investigation, but by philosophy proper.\n\n[285] [The discovery here referred to is the synthesis of urea by\nW\u00f6hler in 1828 (Pogg. Ann. xii., 253; xv. 619), by the molecular\ntransformation of ammonium cyanate. Since that period large numbers\nof organic syntheses have been effected by chemists, and many of the\ncompounds formerly supposed to be essential products of life have been\nbuilt up in the laboratory from their inorganic elements. The division\nof chemistry into \u201corganic\u201d and \u201cinorganic\u201d is thus purely artificial,\nand is merely retained as a matter of convenience, the former division\nof the science being defined as the chemistry of the carbon compounds.\nR.M.]\n\n[286] \u201cWahreit und Irrthum im Darwinismus.\u201d Berlin, 1875.\n\n[287] [Eng. ed. I have been reproached by competent authorities for\nhaving clothed my ideas upon the theory of selection in the form of\na reply to Von Hartmann. I willingly admit that this author cannot\nbe considered as the leader of existing philosophical views upon the\ntheory of descent in Germany; Frederick Albert Lange has certainly a\nmuch greater claim to this position. Lange does not however combat\nthis theory; he accepts and develops it most beautifully and lucidly\non a sound philosophical basis in such a manner as has never been\ndone before from this point of view (\u201cGeschichte des Materialismus,\u201d\n3rd. ed., 1877, vol. ii. pp. 253-277). On most points I can but agree\nwith Lange. Von Hartmann, however, whose objections appeared to me to\nbe supported by a wide scientific knowledge, afforded me a suitable\nopportunity of developing my own ideas upon some essential points\nin the theory of selection. In this sense only have I attempted to\ninterfere with this author, the refutation of his views, as such,\nhaving been with me a secondary consideration.] [The chief exponent\nof the doctrine of organic evolution in this country is Mr. Herbert\nSpencer, in whose \u201cPrinciples of Biology,\u201d vol. i. chap. xii., will be\nfound a masterly treatment of the theory of descent from a \u201cmechanical\u201d\npoint of view. R.M.]\n\n[288] [The above views on the nature of variability, which were also\nbroadly expressed in the first essay \u201cOn the Seasonal Dimorphism of\nButterflies\u201d (pp. 114, 115), are fully confirmed by Herbert Spencer\n(_loc. cit._ chaps. ix. and x.), and more recently by A. R. Wallace in\nan article on \u201cThe Origin of Species and Genera\u201d (_Nineteenth Century_,\nvol. vii., 1880, p. 93). See also some remarks by Oscar Schmidt in his\n\u201cDoctrine of Descent and Darwinism,\u201d Internat. Scien. Ser. 3rd. ed.\n1876, p. 173. R.M.]\n\n[289] [This law has been beautifully applied by Herbert Spencer in\norder to explain why, with an unlimited supply of food, an organism\ndoes not indefinitely increase in size. \u201cPrinciples of Biology,\u201d vol.\ni. p. 121-126. R.M.]\n\n[290] [Eng. ed. This idea, formerly expressed by me, occurs also\nin Lange (\u201cGeschichte des Materialismus,\u201d ii. 265), and is there\nexemplified in a very beautiful manner by illustrations from modern\nchemistry. Lange compares what I have termed above the \u201cphysical\nconstitution\u201d of the organism to the chemical constitution of one of\nthose organic acids which by substitution of single elements may become\ntransformed into more complicated acids, but which, as it were, always\nundergo \u201cfurther development\u201d in only one determined and narrowly\nrestricted course. Here, as with the organism, the number of possible\nvariations is very great, but is nevertheless limited, since \u201cwhat\ncan or cannot arise is determined beforehand by certain hypothetical\nproperties of the molecule.\u201d]\n\n[291] \u201cOrigin of Species.\u201d 4th German ed., p. 19; 5th English ed., p. 6.\n\n[292] [Mr. A. R. Wallace, in his article last referred to, quotes some\nmost valuable measurements of mammals and birds, showing the amount of\nvariation of the different parts. These observations were published\nby J. A. Allen, in a memoir \u201cOn the Mammals and Winter Birds of East\nFlorida,\u201d &c. (Bulletin of the Museum of Comparative Zoology at Harvard\nCollege, Cambridge, Mass., vol. ii. No. 3.) R.M.]\n\n[293] [See note 142, p. 310. R.M.]\n\n[294] \u201cDie Darwin\u2019sche Theorie,\u201d Dorpat, 1875.\n\n[295] [A certain number of instances of mimicry are known to occur\nbetween species both of which are apparently nauseous. A most able\ndiscussion of this difficult problem is given by Fritz M\u00fcller, in the\ncase of the two butterflies _Ituna Ilione_ and _Thyridia Megisto_, in\na paper published in _Kosmos_, May, 1879 (p. 100). The author shows by\nmathematical reasoning that such resemblances between protected species\ncan be accounted for by natural selection if we suppose that young\nbirds and other insect persecutors have to learn by experience which\nspecies are distasteful and which can be safely devoured. See also\nProc. Ent. Soc. 1879, pp. xx-xxix. R.M.]\n\n[296] See Haeckel\u2019s \u201cGenerelle Morphologie,\u201d ii. 107.\n\n[297] \u201c\u00dcber die Berechtigung der Darwin\u2019schen Theorie,\u201d Leipzig, 1868.\n\n[298] \u201cPopul\u00e4re wissenschaftl. Vortr\u00e4ge,\u201d vol. ii., Brunswick, 1871, p.\n208.\n\n[299] \u201cDas Unbewusste vom Standpunkte der Physiologie u.\nDescendenztheorie,\u201d Berlin, 1872, p. 89. The second edition appeared in\n1877, in Von Hartmann\u2019s own name.\n\n[300] \u201c\u00dcber die Berechtigung,\u201d &c., Leipzig, 1868. In this work will be\nfound briefly laid down the theoretical conception of variability here\npropounded somewhat more broadly. [In the last edition of the \u201cOrigin\nof Species\u201d Darwin states, with respect to the direct action of the\nconditions of life as producing variability, that in every case there\nare two factors, \u201cthe nature of the organism and the nature of the\nconditions.\u201d 6th ed. p. 6. R.M.]\n\n[301] [Although hardly necessary to the evolutionist, it may perhaps\nbe well to remind the general reader, that all experiments upon\nspontaneous generation, or abiogenesis, have hitherto yielded negative\nresults; no life is produced when the proper precautions are taken for\nexcluding atmospheric germs. But although we have so far failed to\nreproduce in our laboratories the peculiar combination of conditions\nnecessary to endow colloidal organic matter with the property of\n\u201cvitality,\u201d the consistent evolutionist is bound to believe, from\nthe analogy of the whole of the processes of nature, that at some\nperiod of the earth\u2019s history the necessary physical and chemical\nconditions obtained, and that some simple form or forms of life arose\n\u201cspontaneously,\u201d _i.e._ by the operation of natural causes. R.M.]\n\n[302] See Haeckel\u2019s \u201cGenerelle Morphologie,\u201d vol. ii. p. 203, and\nSeidlitz, \u201cDie Darwin\u2019sche Theorie,\u201d 1875, p. 92 _et seq._\n\n[303] [In a recently published work by Dr. Wilhelm Roux this author has\nattempted to work out the idea of an analogy between the struggle for\nexistence and survival of the fittest in individuals and species, and\nthe struggle for existence and survival of the parts in the individual\norganism. See \u201cDer Kampf der Theile im Organismus: ein Beitrag zur\nVervollst\u00e4ndigung der mechanischen Zweckm\u00e4ssigkeitslehre,\u201d Leipzig,\n1881. R.M.]\n\n[304] [Eng. ed. Meanwhile it has been shown by Oscar Schmidt that\nVon Hartmann, under the name of \u201cthe Unconscious,\u201d re-invests\nthe old vital force with some portion of its former power. \u201cDie\nnaturwissenschaftlichen Grundlagen der Philosophie des Unbewussten,\u201d\nLeipzig, 1877, p. 41.]\n\n[305] _Loc. cit._ p. 175.\n\n[306] _Loc. cit._ p. 156.\n\n[307] \u201c\u00dcber die Cuninen-Knospen\u00e4hren im Magen von Geryonien.\u201d Reprint\nfrom \u201cMittheil. des naturwiss. Vereines,\u201d Graz, 1875.\n\n[308] [See Darwin\u2019s \u201cOrigin of Species,\u201d 6th ed. pp. 33, 34, and\n201-204. R.M.]\n\n[309] [Eng. ed. See Kant\u2019s \u201cAllgemeine Naturgeschichte und Theorie des\nHimmels.\u201d]\n\n[310] \u201cDas Unbewusste vom Standpunke der Physiologie und\nDescendenz-Theorie,\u201d Berlin, 1872, p. 16.\n\n[311] [Eng. ed. See Lotze\u2019s \u201cMikrokosmos,\u201d 1st ed., vol. iii. pp.\n477-483.]\n\n[312] See Helmholtz\u2019s \u201cPopul\u00e4re wissenschaftl. Vortr\u00e4ge,\u201d vol. ii.,\nBrunswick, 1872.\n\n[313] See also Fr. Vischer\u2019s \u201cStudien \u00fcber den Traum. Beilage zur\nAugsburger Allgem. Zeitung,\u201d April 14th, 1876. Haeckel also includes\nthis idea in his recent essay already quoted, \u201cDie Perigenesis der\nPlastidule,\u201d Berlin, 1876, p. 38 _et seq._\n\n[314] See Von Hartmann, _loc. cit._ p. 158.\n\n\n\n\nINDEX.\n\n\n  A.\n\n  Abbot and Smith, descriptions of larv\u00e6, 192, 216, 261, 263, 268, 452,\n            453.\n\n  _Acherontia Atropos_, South African form of larva, 263, 324, 531;\n    larva in Spain, 324;\n    larva phytophagically variable, 531.\n\n  _Acosmeryx anceus_, larva, 192.\n\n  _Acr\u00e6a_ and Maracuj\u00e0 butterflies as larv\u00e6, pup\u00e6, and imagines, 536.\n\n  _Acronycta_, affinities of genus, 169.\n\n  Adaptation, 281, 589;\n    analogous, 376;\n    physiological, 590;\n    mutual, 647.\n\n  Allen, J. A., variation in birds and mammals, 658.\n\n  Alpine form of _P. Napi_, 40.\n\n  Alpine hare, seasonal change of colour, 7, 660.\n\n  Alternation of generations, 80, 699, 702.\n\n  _Amblystoma._ _A. mavortium_ from _Siredon lichenoides_, 567;\n    _A. tigrinum_ bred by Dum\u00e9ril, 567;\n    distribution of species of genus, 569;\n    habits of species, 573;\n    a reversion form, 531, 592;\n    sterility of, 593;\n    oviposition by, 594, 623;\n    definition of genus, 610;\n    degeneration of, 612;\n    causes of reversion of, 600, 608, 613, 620, 631;\n    _A. punctatum_ and _A. fasciatum_, development of, 623;\n    summer sleep of, 628;\n    _A. mavortium_, metamorphosis of, 629.\n\n  _Ambulyx gannascus_ and _A. liturata_, larv\u00e6, 245.\n\n  Amixia, the principle of, 46, 110.\n\n  Ammonites, 275.\n\n  _Ampelophaga rubiginosa_, larva, 192.\n\n  Amphibia of Upper Engadine, 615.\n\n  _Amphiuma_, characters of genus, 579.\n\n  _Anceryx_, larva, 264;\n    stages 3, 4, and 5, 266.\n\n  _Aphaniptera_, characters of, 498.\n\n  Aphides, organic reproduction of, 97.\n\n  _Araschnia._ _A. levana_ and _A. prorsa_ seasonally dimorphic, 2;\n    dimorphism of larv\u00e6, 6;\n    _A. levana_, first experiments with, 10;\n    _A. var. porima_, artificial production of, 10, 17;\n    _A. levana_, Dorfmeister\u2019s experiments with, 11;\n    explanation of seasonal dimorphism of, 19;\n    northern, extension of, 20;\n    _A. prorsa_ gradual origination of, 22;\n    _A. levana_, reversion of, caused by high temperature, 37;\n    the _levana_ form the most constant, 43, and sexually dimorphic, 43;\n    summer and winter forms compared, 56;\n    experiments with _A. levana_, 117;\n    _A. levana_, variable in all three stages, 405;\n    incongruence in genus, 449.\n\n  Argent, W. J., phytophagic variability of larva of _Sphinx Ligustri_,\n            306.\n\n  _Argynnis paphia_, dimorphism of, 250.\n\n  _Artemia salina_ and _A. Milhausenii_, transformations of, 635.\n\n  _Ascaris nigrovenosa_, propagation of, 90.\n\n  Askenasy, on limits of variability, 114.\n\n  Atavism and arrested growth, 587.\n\n  _Aterica meleagris_, colour variations of, 8.\n\n  Atoms, actions of, 710;\n    sensibility of, 714.\n\n  Axolotl, transformation into Amblystoma, 555;\n    experiments with, 558;\n    duration of transformation, 563;\n    conditions of metamorphosis of, 564;\n    distinguishing characters, 575.\n\n\n  B.\n\n  Baer, Karl Ernst von, on the mechanism of nature, 639;\n    on the vital force, 641;\n    on the theory of selection, 694;\n    on causality and purpose, 708.\n\n  Baird, Prof., on the development of species of _Amblystoma_, 623.\n\n  Balbiani, reproduction of Aphides, 97.\n\n  Balfour, F. M., on insect phylogeny, 485, 494.\n\n  Bates, H. W., local variation in Amazonian Lepidoptera, 60.\n\n  Bees, embryological development of, 483.\n\n  Belt, Thomas, on coloured frogs, 294;\n    on the glaciation of North America, 622.\n\n  Biogenetic law, corollary of, 611.\n\n  Birchall, E., local variation in British Lepidoptera, 60.\n\n  Blanchard, on oviposition of _Amblystoma_, 594.\n\n  Boisduval, figures of larv\u00e6 referred to, 216, 218, 245;\n    on snake-like appearance of the larva of _D. nic\u00e6a_, 342;\n    on the relationship of _Euchloe belia_ and _E. ausonia_, 48.\n\n  _Bombycid\u00e6_, early stage of larv\u00e6, 166.\n\n  _Bombyx._ _B. cynthia_, colour changes in larva, 7;\n    _B. neustria_, caterpillar refused by lizards, 336;\n    _B. Rubi_, caterpillar eaten by lizards, 338.\n\n  Boscher, E., dimorphism of larva of _S. ocellatus_, 241, 306.\n\n  Brauer, insect phylogeny, 493.\n\n  Burmeister, figures of larv\u00e6 referred to, 195, 196, 224, 232, 255,\n            261, 263, 436, 437;\n    early stages of larv\u00e6 of Argentine butterflies, 166.\n\n  Butler, A. G., affinities of genus _Acronycta_, 169;\n    genera of _Ch\u00e6rocampin\u00e6_, 190, 192, 194, 196;\n    genera of _Smerinthin\u00e6_, 233, 243, 244;\n    genera of _Macroglossin\u00e6_, 253, 254;\n    species of _Pterogon_, 256;\n    genera of _Sphingin\u00e6_ and _Acherontiin\u00e6_, 262;\n    on coloured larv\u00e6 being rejected, 336;\n    affinities of _Sesiid\u00e6_, 370;\n    descriptions and figures of new Sphinx-larv\u00e6, 524, 526.\n\n  B\u00fctschli, embryology of bees, 483.\n\n  Butterflies, action of warmth in producing changes in colour and\n            marking, 58;\n    summer and winter larv\u00e6 and pup\u00e6 alike in seasonally dimorphic\n            species, 64;\n    seasonally dimorphic species hibernate as pup\u00e6, 73;\n    odoriferous organs, 102;\n    early stages of larv\u00e6, 165;\n    di- and polymorphism of, 250;\n    crepuscular habits of certain species, 475.\n\n\n  C.\n\n  _Callosune eupompe_, polymorphism of female, 251.\n\n  Cameron, P., on larva of _S. ocellatus_, 241, 282;\n    protective habits of saw-fly larv\u00e6, 290, 294.\n\n  Carlin, W. E., on _Siredon lichenoides_, 630.\n\n  Caterpillars, origin of markings, 161;\n    means of defence, 289;\n    defensive habits, 290;\n    food of brightly coloured species, 294;\n    habit of concealment by day, 291, 296;\n    phytophagic variability, 305, 531;\n    seasonal variability, 305;\n    experiments with ocellated species,\n  330;\n    experiments with annulated species, 336;\n    sexual dimorphism of, 308, 527, 534;\n    independent variability of stages, 416.\n\n  _Cecidomyia_, metagenesis of, 82, 586.\n\n  _Ch\u00e6rocampa._ _C. elpenor_ and _C. porcellus_, relationship of, 171;\n    markings of larv\u00e6 of the genus, 177;\n    _C. elpenor_, 1st and 2nd larval stages, 178;\n      3rd and 4th stages, 180;\n      5th stage, 181;\n      6th stage, 183;\n    _C. porcellus_, 1st larval stage, 184;\n      2nd and 3rd stages, 185;\n      4th stage, 186;\n    _C. elpenor_ and _porcellus_, larval markings compared with other\n            species of _Ch\u00e6rocampa_, 188;\n    _C. elpenor_, experiments with larva, 331, 332;\n    _C. Japonica_ and _C. Lewisii_, larv\u00e6, 194;\n    _C. cretica_, larva, 524;\n    _C. lycetus_, larva, 527;\n    _C. capensis_, larva, 529;\n    eaten by birds, &c., 530.\n\n  Characters, new, backward transference of, 279.\n\n  Chauvin, Fr\u00e4ulein von, experiments with Axolotl, 558;\n    experiments with _Salamandra atra_, 615.\n\n  Chavannes, figure of larva referred to, 255.\n\n  Clarke, development of _Amblystoma punctatum_, 623.\n\n  Classification, objects of, 168.\n\n  Claus, on the origination of heterogenesis, 89.\n\n  Clemens, larva of _A. ello_, 268.\n\n  Climatic variation, 45;\n    in _P. Napi_, 47;\n    in _E. Belia_, 47;\n    nature of causes producing, 52, 105;\n    by sex, 61, 62;\n    climatic varieties very distinct from local, 45;\n    climatic influences operate slowly, 58;\n    climatic changes gradual, 652, 660.\n\n  Cole, B. G., seasonal dimorphism of _E. punctaria_, 4.\n\n  Coleoptera, grub-like larv\u00e6 of, 495.\n\n  _Colias edusa_, dimorphism of female, 250.\n\n  Colour, biological value of, 289.\n\n  Conditions of life, compulsory changes in, 651.\n\n  Congruence in Lepidopterous families, 435;\n    in Lepidopterous genera, 444.\n\n  _Conifer\u00e6_, food-plants of _Anceryx_, 268.\n\n  Convergence of characters in _Ophiuroidea_, 396;\n    in Dipterous larv\u00e6, 494.\n\n  Cope, Prof., habits of _Siredon Mexicanus_, 566;\n    on distribution of _Amblystoma_, 569.\n\n  Correlation of growth, 278, 363, 388, 415;\n    action of, 415, 516;\n    Von Hartmann\u2019s views of, 670;\n    nature of, 673.\n\n  _Cryptobranchus_, characters of, 579.\n\n  Cyclical propagation, 82, 699;\n    and reversion, 613.\n\n\n  D.\n\n  Daphniacea, differences between allied species and genera, 516.\n\n  Darwin, C., influence of external conditions of life, 59;\n    inheritance at corresponding periods, 63, 275, 431;\n    markings of butterflies, 101;\n    definition of eye-spots, 326;\n    adaptation in larv\u00e6, 392;\n    correlation of growth, 516;\n    degeneration, 583;\n    adaptation, 589;\n    causes of sterility, 596;\n    limited variability, 656;\n    factors of variability, 676;\n    abrupt variations, 701.\n\n  Degeneration, 583;\n    phyletic, 611.\n\n  _Deilephila_, division of larv\u00e6 into groups, 199;\n    _D. Euphorbi\u00e6_, 1st larval stage, 202;\n      2nd and 3rd stages, 203;\n      4th and 5th stages, 205;\n    _D. nic\u00e6a_, larva, 207;\n    _D. dahlii_, larva, 208;\n    _D. vespertilio_, larva, 209;\n    _D. Galii_, larva, 211;\n      5th stage, 213;\n    _D. livornica_, larva, 215;\n    _D. Zygophylli_, larva, 217;\n    _D. Hippopha\u00ebs_, larva, 218, 650;\n    division of larv\u00e6 according to phyletic stages, 223;\n    genealogy of, 358;\n    _D. Galii_, larva rejected, 340;\n    _D. Euphorbia_, larva eaten, 340;\n    habits of larv\u00e6 and imagines, 412;\n    _D. Robertsi_, larva, 525.\n\n  Development;\n    of marking of Sphinx-larv\u00e6, laws of, 274;\n    phyletic, of markings of _Sphingid\u00e6_, 370;\n    unequal phyletic, 460.\n\n  Dimorphism and polymorphism;\n    of butterflies, 250;\n    of Sphinx-larv\u00e6, 300;\n    of pup\u00e6, 412.\n\n  Diptera, incongruences among, 488;\n    imaginal characters of, 488;\n    divisions of, 498.\n\n  Dodo, extinction of, 651.\n\n  Dohrn, A.;\n    on degeneration, 583;\n    on functional change, 590.\n\n  Dorfmeister, G.;\n    experiments with _A. levana_, 11.\n\n  Duponchel;\n    figures of larv\u00e6 referred to, 207, 208, 218, 253.\n\n  Dum\u00e9ril;\n    experiments with Axolotl, 555, 564, 566, 593;\n    characters of Axolotl, 375, 576.\n\n\n  E.\n\n  Edwards, W. H., seasonal dimorphism of _P. pseudargiolus_ and\n            _P. violecea_, 4;\n    of _P. Tharos_ and _P. Marcia_, 4;\n    of _P. Ajax_, 30;\n    of _L. Artemis_ and _L. Proserpina_, 32;\n    experiments with _P. Ajax_ and _P. Tharos_, 33, 126, 140;\n    pupal period of winter forms of _P. Ajax_, 36;\n    experiments with _G. interrogationis_, 149;\n    early stages of larv\u00e6 of N. American butterflies, 165;\n    larva of _P. pseudargiolus_ and attendant ants, 290;\n    figures of larv\u00e6 referred to, 165, 436, 437.\n\n  Eimer, colour variation in _Lacerta muralis_, 361.\n\n  _Emmelesia unifasciata_, larva phytophagically variable, 307.\n\n  Environment, direct and indirect action of, 686.\n\n  _Ephyra punctaria_ and _E. omicronaria_, seasonal dimorphism of, 4.\n\n  _Eriogaster lanestris_, larva eaten by lizards, 336.\n\n  _Eriotus Pteridis_, resemblance of larva, 320.\n\n  _Euchelia Jacob\u00e6\u00e6_;\n    larva and imago rejected by lizards, 336;\n    constant in all three stages, 405.\n\n  _Euchloe_:\n    _E. belia_ and _E. ausonia_, seasonal dimorphism of, 3;\n    _E. belia_, climatic variation of, 47;\n    var. _Simplonia_ in Alps, 48;\n    _E. belemia_, seasonal dimorphism and migration of, 78.\n\n  _Euproctus Rusconii_, habits of, 617.\n\n  _Eusmerinthus Kindermanni_, larva, 526.\n\n  Evershed, J., larva of _C. porcellus_, 188.\n\n  Eye-spots;\n    repetition of, 277;\n    development of, 325, 327;\n    terrifying action, 330;\n    resemble flower-buds, 334;\n    useful when rudimentary, 366.\n\n\n  F.\n\n  Fatiot, Amphibia of Upper Engadine, 616.\n\n  Fechner, on conscious matter, 714.\n\n  Filippi, De, _Triton alpestris_, 585.\n\n  Forel, Prof., fresh water _Pulmonifera_, 590.\n\n  Form-relationship and blood-relationship, not always coincident, 395.\n\n  Frantzius, Von, on climate of Mexico, 619.\n\n  Frogs, 615, 616, 618;\n    inedible species, 294.\n\n  Function, change of, 365.\n\n\n  G.\n\n  Gegenbaur, Prof., the skull of Axolotl, 576.\n\n  Gehrig, breeding of Axolotl, 565.\n\n  Gen\u00e9, habits of _Euproctus Rusconii_, 617.\n\n  Genera, Lepidopterous, show greatest congruence, 459.\n\n  Generation, heterogeneous, 679, 698, 702.\n\n  Gentry, T. G., phytophagic variability of caterpillars, 306.\n\n  Gerst\u00e4cker, division of the Diptera, 498.\n\n  _Geotriton fuscus_, habits, 617.\n\n  Glacial epoch; butterflies monogoneutic during, 19, 72;\n    in N. America, 622.\n\n  Gooch, W. D., description of larv\u00e6 referred to, 436;\n    crepuscular butterflies, 476;\n    sexual dimorphism of caterpillars, 535.\n\n  Gooseberry, not increased in size, 654.\n\n  Goss, H., larva of _C. porcellus_, 188.\n\n  Gosse, P. H., embryonic larva of _P. Homerus_, 167.\n\n  _Grapta interrogationis_, experiments with, 149.\n\n  Green colour of caterpillars, 293, 310.\n\n  Growth, law of, 655.\n\n  Gynandromorphism in butterflies, 249.\n\n  _Gynanisa Isis_, young larva, 166.\n\n\n  H.\n\n  Haeckel, Prof., on metagenesis, 80, 83;\n    on ontogeny, 270;\n    on adaptation, 281, 589;\n    on the biogenetic law, 611;\n    on reproduction and growth, 664;\n    the plastidule theory, 667.\n\n  Hare, white variation of, 660.\n\n  Hartmann, E. von; his views examined, 645;\n    on variability, 646, 652, 655, 664;\n    on heredity, 657;\n    his \u201cPhilosophy of the Unconscious,\u201d 670;\n    on correlation, 670, 673;\n    on the nature of species, 671;\n    on design in nature, 696;\n    on the sensibility of atoms, 714.\n\n  Helmholtz, Prof., on natural laws, 668;\n    on the senses, 711.\n\n  _Hemaris hylas_, larva, 255.\n\n  Heredity, alternating and continuous, 24;\n    cyclical, 66;\n    homochronic, 63, 431;\n    not mechanical, 657;\n    nature of, 667.\n\n  Herrich-Sch\u00e4ffer, wing neuration of butterflies, 449.\n\n  Hertwig, on teeth of Amphibia, 376.\n\n  Heterogenesis, definition of, 82, 96;\n    parallelism between and metamorphosis, 86;\n    origination of, 89.\n\n  Heyden, Von, reproduction of _Aphis_, 97.\n\n  Hilgendorf, on fossil shells, 109.\n\n  _Hipparchia semele_, colour variations, 8.\n\n  Hoffmann, Ernst, on immigration of European butterflies, 77.\n\n  Horsfield and Moore, figures of caterpillars referred to, 193, 195,\n            196, 243, 253, 261.\n\n  H\u00fcbner, figures of caterpillars referred to, 193, 208, 211, 218, 231,\n            267.\n\n  Humboldt, Baron, on the Lake of Mexico, 608, 621.\n\n  Huxley, Prof., \u201cAnatomy of Invertebrated Animals\u201d referred to, 498.\n\n  _Hydromedus\u00e6_, alternation of generations in, 83;\n    want of parallelism in different generations of, 395.\n\n  Hydrozoa, asexual reproduction and polymorphosis, 83.\n\n  Hymenoptera, incongruence among, 481;\n    imaginal characters, 481;\n    causes of incongruence, 486.\n\n\n  I.\n\n  Ichthyodea, characters of, 577, 579.\n\n  Incongruence; in Lepidopterous families, 437;\n    genera, 446, 449;\n    species, 451;\n    varieties, 456;\n    two kinds of, 458, 502;\n    in Hymenoptera, 481;\n    in Diptera, 488;\n    different forms of, 503;\n    dependent upon action of environment, 505, 507, 508.\n\n  Inheritance; homochronic, 63, 431;\n    definition of cyclical, 66;\n    limited by sex, 68.\n\n  Insects, ancestry of, 493.\n\n  Intermaxillary gland of Axolotl, 623.\n\n\n  J.\n\n  Jullien, on _Lissotriton punctatus_, 591.\n\n\n  K.\n\n  Kant, referred to, 638, 709.\n\n  Kirby, W. F., \u201cSynonymic Catalogue\u201d referred to, 2, 49, 435, 449.\n\n  Kleemann, larva of _S. Ligustri_, 259.\n\n  Knaggs, Dr. H. G., seasonal dimorphism of _S. illustraria_, 4.\n\n  K\u00f6lliker, Prof., experiments with Axolotl, 564;\n    characters of Axolotl, 575.\n\n\n  L.\n\n  _Lacerta muralis_, colour variations, 361.\n\n  Lake of Mexico, 603, 608, 621.\n\n  Lange, F. A., \u201cHistory of Materialism\u201d referred to, 645, 656.\n\n  Langer, Prof., experiments with _Pelobates_ tadpoles, 607.\n\n  Lankester, Prof. E. R., on degeneration, 583.\n\n  _Lasiocampa Pini_, larva eaten by lizards, 336;\n    variable in two stages, 405.\n\n  Lepidoptera; foes, 8;\n    larva and imago independently variable, 401;\n    species constant in three stages, 405;\n    species variable in three stages, 405;\n    variable in two stages, 405;\n    variable in one stage, 406;\n    causes of variability of pup\u00e6, 411;\n    case of incongruence among pup\u00e6, 540.\n\n  _Leptodora hyalina_, development, 93.\n\n  _Lepus timidus_, white variations of, 662.\n\n  Leuckart, reproduction of _Ascaris nigrovenosa_, 90;\n    on growth, 654.\n\n  Leydig, on sexual larv\u00e6 of _Triton_, 591.\n\n  _Limenitis Artemis_ and _L. Proserpina_, dimorphic in both sexes, 32;\n    _L. Camilla_, constant in three stages, 405.\n\n  Lines, dorsal, subdorsal, supra- and infra-spiracular, 175.\n\n  _Lissotriton punctatus_, sexual larv\u00e6, 591, 595.\n\n  Lizards, colour variation of, 361;\n    rejection of caterpillars by, 336.\n\n  Local forms, distinct from climatic varieties, 45;\n    produced by isolation, 109.\n\n  Lockyer, B., young _Noctua_ larv\u00e6, 166.\n\n  _Lophostethus Dumolinii_, larva, 527.\n\n  _Lophura hyas_, larva, 254.\n\n  Lubbock, Sir John, insect metamorphosis, 65, 431;\n    derivation of metagenesis, 83;\n    colours of caterpillars, 294;\n    ancestry of insects, 493.\n\n  _Lyc\u00e6na pseudargiolus_, larva and ants, 290.\n\n\n  M.\n\n  _Macroglossa_, larv\u00e6 of, 245;\n    _M. stellatarum_, oviposition of, 245;\n    1st, 2nd, and 3rd larval stages, 246;\n    4th and 5th stages, 247;\n    di- and polymorphism of larva, 247;\n    pupa of, 250;\n    _M. belis_, and _M. pyrrhosticta_, larv\u00e6, 255.\n\n  Markings, general biological value of, 285;\n    special biological value of, 308;\n    subordinate, 347;\n    of _Sphingid\u00e6_, evolution of, 380.\n\n  Marsh, Prof., on _Siredon lichenoides_, 567.\n\n  Matter, assumption of, 711;\n    conscious, 714.\n\n  Mayer, Paul, insect phylogeny, 493.\n\n  McLachlan, R., phytophagic variability, 305.\n\n  Mechanism and teleology, 694.\n\n  _Medus\u00e6_, internal parasitism, 699.\n\n  _Melanopsis recurrens_, from _Paludina_ bed, 275.\n\n  _Menopoma_, characters of, 579.\n\n  M\u00e9rian, Madame, figures of caterpillars referred to, 195, 232, 263,\n            268.\n\n  Metagenesis, defined, 81, 96;\n    derivation, 83.\n\n  Metamorphosis, 430.\n\n  Mexico, climate of, 619.\n\n  Milli\u00e8re, figure of larva referred to, 254.\n\n  Mimicry, 648, 663.\n\n  M\u00f6ller, L., phytophagic variability, 305.\n\n  Mono- and digoneutic species, 16.\n\n  Moore, F., figures of caterpillars referred to, 436, 526.\n\n  Morris, affinities of _Apatura_ and _Nymphalis_, 438.\n\n  Moths, seasonal dimorphism in, 4;\n    resembling splinters, 292.\n\n  M\u00fchlenpfordt, on Lake of Mexico, 604.\n\n  M\u00fcller, Fritz, odoriferous organs of butterflies, 102;\n    sexual selection in butterflies, 103;\n    habit and protective resemblance, 290;\n    spiny caterpillars, 293;\n    larva of _Papilio Evander_, 299;\n    reversion in Orchids, 614;\n    maggot-like larv\u00e6 of insects, 493;\n    the biogenetic law, 611;\n    mimicry, &c., 665.\n\n  M\u00fcller, Hermann, larva of _Stauropus Fagi_, 290.\n\n  M\u00fcller, P. E., _Leptodora hyalina_, 94.\n\n  Murray, Andrew, seasonal change of colour, in animals, 7;\n    green caterpillars, 293.\n\n  _Muscid\u00e6_, embryonic development, 492.\n\n\n  N.\n\n  N\u00e4geli, reversion of cultivated plants, 107.\n\n  Natural selection, 112, 361, 704, 705.\n\n  Nature, mechanical conception of, 634;\n    harmony of, 697.\n\n  Neumayr and Paul, on _Melanopsis_, 275.\n\n  _Noctu\u00e6_, change of colour in larv\u00e6, 166;\n    number of legs in young larv\u00e6, 166, 520;\n    ontogeny of larv\u00e6, 520.\n\n  Noll, Dr., larva of _Ach. Atropos_, 324.\n\n  North America, glaciation of, 622.\n\n  Nurse-breeding in Hymenoptera and Diptera, 402.\n\n  _Nymphalid\u00e6_, larval classification, 435.\n\n\n  O.\n\n  Ontogeny, Fritz M\u00fcller and Haeckel on, 270;\n    predominance of new characters in last stage of, 280, 283.\n\n  Ontogeny and Phylogeny of Sphinx-markings, 177.\n\n  Ophiuroidea, want of parallelism in, 396.\n\n  Orchids, reversion in, 614.\n\n  Organic compounds, synthesis of, 643.\n\n  Organism, parts independently variable, 514.\n\n\n  P.\n\n  _Pachylia Ficus_, early stages of larva, 232.\n\n  Packard, insect phylogeny, 493.\n\n  Pangenesis, theory of, 668.\n\n  _Papilio._ _P. Ajax_, seasonal forms, 30;\n    _P. Turnus_, a dimorphic form, 32;\n    _P. Machaon_ and _Podalirius_, seasonally dimorphic in Spain and\n            Italy, 74;\n    _P. Ajax_, experiments with, 126;\n    _P. Merope_, young larva, 166;\n    _P. Homerus_, young larva, 167;\n    _P. Machaon_, larva rejected by lizards, 339.\n\n  Parallelism, phyletic, in metamorphic species, 390;\n    phyletic, law of, 510.\n\n  _Pararga \u00c6geria_, sexually dimorphic, 68;\n    climatic variation of, 68.\n\n  Pascoe, F. P., \u201cZoological Classification\u201d referred to, 488, 498.\n\n  _Pelobates_, experiments with tadpoles, 607.\n\n  _Pergesa Mongoliana_, larva, 194.\n\n  _Philampelus achemon_ and _P. satellitia_, larv\u00e6, 523;\n    _P. labrusc\u00e6_, larva, 195;\n    _P. vitis_, and _P. anchemolus_, larv\u00e6, 232.\n\n  Philosophy, function of, 640.\n\n  _Phyciodes Tharos_ and _P. Marcia_, seasonal dimorphism of, 4;\n    experiments with, 140.\n\n  Phylogeny, conclusions from, 270;\n    and ontogeny of Sphinx-markings, 177;\n    of insects, 493.\n\n  _Pierin\u00e6_, seasonal dimorphism of, 13;\n    analogous seasonal dimorphism in, 60, 108;\n    experiments with, 122.\n\n  _Pieris Napi_, experiments with, 13;\n    summer form the younger, 29;\n    reversion of, caused by mechanical vibration, 38;\n    var. _Bryoni\u00e6_, the potential winter form, 39;\n    experiments with this var., 40;\n    this var. the parent form of _Napi_, 41;\n    a climatic var., 41;\n    very variable, 43;\n    variability due to crossing, 43;\n    var. _\u00e6stiva_ more variable than var. _vernalis_, 43;\n    _P. Napi_, summer and winter forms compared, 56;\n    incongruence in this species, 457;\n    _P. Krueperi_, seasonal dimorphism of, 78;\n    _P. Brassic\u00e6_, larva rejected by lizards, 337;\n    constant in three stages, 405;\n    photographic sensitiveness of pupa, 405;\n    _P. Napi_, variable in two stages, 405.\n\n  Plants, cultivated, 107;\n    sterility of reverted, 596.\n\n  Plastidule theory, 667.\n\n  _Plebeius amyntas_ and _P. polysperchon_, seasonal dimorphism, 2;\n    _P. Alexis_, do., 3;\n    _P. pseudargiolus_ and _P. violacea_, do., 4;\n    _P. agestis_, do., 50;\n    polymorphism of females in this genus, 251.\n\n  Pluvial period, 621.\n\n  Polar animals, colour of, 658, 660.\n\n  _Polyommatus phl\u00e6as_, distribution and climatic variation, 49;\n    seasonal dimorphism, 50;\n    Italian summer and winter forms compared, 55.\n\n  _Polyptychus dentatus_, larva, 244.\n\n  Ptarmigan, seasonal colours of, 6.\n\n  _Pterogon_, larv\u00e6 of, 255;\n    _P. \u00e6nother\u00e6_, larva, 256.\n\n  _Pulmonifera_, functional change in lungs, 590.\n\n  _Pyg\u00e6ra bucephala_, larva rejected by lizards, 337.\n\n\n  Q.\n\n  Quatrefages, De, spermatozoa of _Amblystoma_, 593.\n\n\n  R.\n\n  _Rana temporaria_ of Upper Engadine, 618.\n\n  Ratzeburg, larva of _Anc. Pinastri_, 265.\n\n  Reproduction and growth, 664.\n\n  Reversion, of species, 611;\n    two modes of, 612;\n    periodic, 613.\n\n  _Rhopalocera_, characters of, 433, 471.\n\n  _Rhytina Stelleri_, extinction of, 651.\n\n  Riley, C. V., descriptions of caterpillars referred to, 437, 521, 522;\n    on sexual dimorphism in caterpillar, 535.\n\n  Ring-spots, 325;\n    development, 327;\n    signs of distastefulness, 341;\n    resembling berries, 344.\n\n  R\u00f6sel, breeding of _Aras. Levana_ by, 6;\n    on larva of _Deil. Euphorbi\u00e6_, 206;\n    of _D. Galii_, 213;\n    of _Smer. Tili\u00e6_, 236;\n    of _S. ocellatus_, 240;\n    of _Anc. Pinastri_, 265, 268.\n\n  Roux, Dr. W., struggle of parts in the organism, 689.\n\n  Rutherford, D. G., colour variations of _Ater. meleagris_, 8.\n\n\n  S.\n\n  Sacc, on sterility of _Amblystoma_, 593.\n\n  Salamanders, habits of Italian land species, 617;\n    oviposition of, 630.\n\n  _Salamandra atra_, experiments with, 615.\n\n  _Salamandrina_, characters of, 577;\n    _S. perspicillata_, habits of, 617.\n\n  Sars, development of _Leptodora hyalina_, 94.\n\n  _Saturnia._\n    _S. Yamamai_, variable in first stage only, 406;\n    _S. carpini_, development of larva of local forms, 419.\n\n  _Satyrin\u00e6_, hibernation of larv\u00e6, 73.\n\n  Saussure, De, on _Siredon Mexicanus_, 565, 602;\n    physical character of Mexican Lake, 603.\n\n  Saw-flies, protective habits of larv\u00e6, 290.\n\n  Schmankewitsch, transformation of _Artemia_, 635.\n\n  Schmidt, Oscar, on convergence of character, 396, 459;\n    on variability, 654.\n\n  Schreibers, sexual _Triton_ larv\u00e6, 591;\n    arrested tadpoles, 607.\n\n  Schulze, F. E., parasitism in _Medus\u00e6_, 699.\n\n  Science, function of, 640.\n\n  Scudder, S. H., embryonic larv\u00e6 of butterflies, 166.\n\n  Seasonal dimorphism, origin and significance, 1;\n    seasonal change of colour in animals, 7;\n    and climatic variation, 45;\n    seasonal adaptation in caterpillars, 305.\n\n  Seidlitz, reproduction of Axolotl, 571;\n    white var. of hare, 662.\n\n  _Selenia tetralunaria_, _S. illunaria_, _S. lunaria_, and\n            _S. illustraria_, seasonal dimorphism, 4.\n\n  Semper, C., origin of alternation of generations, 84;\n    transformation of _Amblystoma mavortium_, 629.\n\n  Semper, G., young Sphinx-larv\u00e6, 166;\n    figures of caterpillars referred to, 522.\n\n  Senses, physiology of, 711.\n\n  Sepp, figure of caterpillar referred to, 262.\n\n  _Sesiid\u00e6_, affinities of, 370.\n\n  Sexual dimorphism, of butterflies, 32, 250;\n    of caterpillars, 308, 527, 534;\n    secondary sexual characters, 62;\n    sexual selection, 69, 102.\n\n  Siebold, Von, on _Pulmonifera_, 590.\n\n  _Siredon._ _S. Mexicanus_, habits, 565;\n    _S. lichenoides_, transformation, 567, 630;\n    phyletic advance of genus, 584;\n    retention of genus, 610;\n    _S. tigrinus_ from L. St. Isabel, 626.\n\n  Slater, J. W., food of gaily coloured caterpillars, 294.\n\n  _Smerinthus_, larv\u00e6 of, 232;\n    _S. Tili\u00e6_, 233;\n    4th stage, 235;\n    _S. Populi_, 236;\n    and stage, 237;\n    3rd and 4th stages, 238;\n    5th stage, 239;\n    _S. ocellatus_, 240;\n    dimorphism of larva, 241;\n    _S. Tili\u00e6_, variable in two stages, 405;\n    North American species of, 453;\n    _S. tatarinovii_, and _S. planus_, larv\u00e6, 244.\n\n  Species, causes of transformation of, 116;\n    nature of, 671.\n\n  Specific characters, 91;\n    may originate through direct action, 100.\n    Specific constitution, 112.\n\n  Spencer, Herbert, theory of descent, 645;\n    variability, 654;\n    law of growth, 655.\n\n  Spermatozoa, of _Amblystoma_, 593;\n    of _Lissotriton punctatus_, 595.\n\n  Speyer, Dr., seasonal dimorphism of moths, 4;\n    of _P. phl\u00e6as_ in Germany, 50.\n\n  _Sphingid\u00e6_, oviposition, 164;\n    congruence and incongruence in genera, 450.\n\n  _Sphinx_-markings, ontogeny and morphology, 177.\n    Larv\u00e6 of genus, 259;\n    _S. Convolvuli_, 650.\n    Origination of markings, 273;\n    law of development, 274;\n    forms of markings, 309;\n    _S. Ligustri_, constant in three stages, 405.\n\n  Stainton, H. T., larval characters of _Notodontid\u00e6_, 443;\n    young larv\u00e6 of _Triph. pronuba_, 520.\n\n  Staudinger, Dr., seasonal dimorphism of _Euch. Belia_ and\n            _E. Ausonia_, 3, 48;\n    butterflies of Finnmark, 20;\n    _P. Napi_ var. _Bryoni\u00e6_ from Lapland, 44, 49;\n    larv\u00e6 of _Ch. alecto_, 193;\n    affinities of _Nymphalis_ and _Apatura_, 438.\n\n  _Stauropus Fagi_, protective habits of larva, 290.\n\n  Sterility, causes of, 596;\n    a result of reversion, 597;\n    law of, in reversion forms, 599.\n\n  Stoat, seasonal change of colour, 7.\n\n  Strauch, \u201cRevision of _Salamandrid\u00e6_\u201d referred to, 570, 577.\n\n  Strecker, H., N. American species of _Smerinthus_, 453.\n\n  Stripes, longitudinal, 312, 372;\n    oblique, 317, 373.\n\n\n  T.\n\n  Tadpoles, arrested development, 607.\n\n  Tegetmeier, W. B., metamorphosis of _Siredon_, 566.\n\n  Teleology and Mechanism, 694.\n\n  _Terebrantia_, larv\u00e6, 508.\n\n  Transformation, causes of, 460, 496;\n    factors of, 676.\n\n  _Trematoda_, alternation of generations in, 83.\n\n  Trimen, Roland;\n    larva of _Lopho. Dumolinii_, 527;\n    of _Ch. Capensis_, 529;\n    phytophagic variability of larva of _Ach. Atropos_, 531.\n\n  _Triptogon roseipennis_, larva, 244.\n\n  _Triton_;\n    _T. alpestris_, sexual larva, 585;\n    reproductive larv\u00e6, 590.\n    Degeneration in genus, 612.\n    Species of Upper Engadine, 631.\n\n  Typical parts, modified by environment, 501, 513.\n\n\n  U.\n\n  Upper Engadine;\n    Amphibia, 615;\n    climate, 616;\n    frog from, 618;\n    Tritons of, 631.\n\n  Urea, synthesis of, 643.\n\n\n  V.\n\n  _Vanessa._\n    _V. Urtic\u00e6_, climatic variation, 60;\n    _V. Atalanta_, _Urtic\u00e6_, and _polychloros_, variable in two stages,\n            405;\n    _V. Io_, variable in one stage, 407.\n    Congruence and incongruence in genus, 446;\n    phyletic completeness of, 447;\n    arrangement of spines on larv\u00e6, 448.\n\n  Variability, origin, 107;\n    limited, 114, 362, 653, 655;\n    independent in different stages, 402;\n    definition of, 404;\n    a relative term, 408;\n    causes of, in pup\u00e6, 411;\n    primary and secondary, 416;\n    independent in different larval stages, 416;\n    phytophagic, 305, 531;\n    Von Hartmann\u2019s views, 646;\n    not unlimited, 647, 652, 655, 683;\n    factors of, 676, 684;\n    mechanical theory of, 677;\n    individual, 681, 685.\n\n  Variation; climatic, 45, 686;\n    causes of, 105;\n    analogous, 683;\n    abrupt not inherited, 701.\n\n  Varieties, climatic, distinct from local, 45.\n\n  Velasco, Se\u00f1or, Axolotl of L. St. Isabel, 626.\n\n  Vital force, phyletic, elimination of, 287;\n    objections to, 352, 461, 511, 518;\n    ontogenetic, abandonment of, 641, 687.\n\n\n  W.\n\n  Wallace, A. R.;\n    introduction of the expression \u201cseasonal dimorphism,\u201d 1;\n    on the dull colours of female butterflies, 8;\n    sexual dimorphism of butterflies, 32;\n    local variation in colour, 60;\n    comparison of seasonal dimorphism with alternation of generations,\n            80;\n    the bright colours of caterpillars, 293;\n    on adaptation, 589;\n    on variability, 654, 658.\n\n  Wallace, Dr., colour changes in larva of _B. Cynthia_, 7.\n\n  Walsh, B. D., phytophagic variation, 305.\n\n  Weale, J. P. M., on the young larv\u00e6 of _Pap. merope_ and _Gyn. Isis_,\n            166;\n    the larva of _Ach. Atropos_ in S. Africa, 263;\n    habits of S. African Sphinx-caterpillars, 290.\n\n  Weir, J. Jenner, the colour variations of _Hip. semele_, 8;\n    experiments with brightly-coloured, hairy, and spiny caterpillars,\n            336.\n\n  Westwood, Prof. J. O., seasonal dimorphism of _Eph. punctaria_, 4;\n    figure of caterpillar referred to, 243;\n    gynandromorphic butterfly, 249.\n\n  White colour, protection afforded by, 6, 659, 661.\n\n  White, Dr. F. B., local variation in British Lepidoptera, 60;\n    young _Notua_ larv\u00e6, 166.\n\n  Wiedersheim, on the anatomy of _Amblystoma_, 577;\n    habits of _Geotriton fuscus_, 617;\n    anatomy of Axolotl, 623.\n\n  Wigand, on the non-increase in size of the gooseberry, 654.\n\n  Wilde, O., on larva of _Dei. Galii_, 213;\n    of _Dei. Hippopha\u00ebs_, 218;\n    larval characters of the _Notodontid\u00e6_, 443.\n\n  Wilson, Owen, figure of larva of _Ch. porcellus_ referred to, 188.\n\n  W\u00f6hler, Prof., the synthesis of urea, 643.\n\n  Wood, T. W., photographic sensitiveness of Lepidopterous pup\u00e6, 405.\n\n  W\u00fcrtemberger, on Ammonites, 275.\n\n\n  Y.\n\n  Yenisei, Swedish Expedition referred to, 20.\n\n\n  Z.\n\n  Zeller, P. C., seasonal dimorphism of _Pl. Polysperchon_ and\n            _Pl. Amyntas_, 2;\n    of Italian butterflies, 75.\n\n\n\n\nERRATA.\n\n\n  Page 81, line 8 from top, and throughout essay, for \u201c_Daphnid\u00e6_\u201d read\n      \u201c_Daphniid\u00e6_.\u201d\n\n  Page 95, line 3 from bottom, for \u201cDaphnoidea\u201d read \u201cDaphniacea.\u201d\n\n  Page 166, line 7 from bottom (note), for \u201cp. 438\u201d read \u201cp. 433.\u201d\n\n  Page 245, line 17 from bottom (note), for \u201cSt\u00e4ll\u201d read \u201cStoll.\u201d\n\n  Page 263, after the word \u201cinsects\u201d (bottom line of note), add, \u201cbut\n      the whole marking is suggestive of distastefulness.\u201d\n\n  Page 296, line 3 from bottom, for \u201cSt\u00e4helina--collector\u201d read\n      \u201cSt\u00e4helin--a collector.\u201d\n\n  Page 305, line 5 from bottom (note), for \u201cIn 1869\u201d read \u201cIn 1865.\u201d\n\n  Page 434, bottom line of note, for \u201cGeometr\u00e6\u201d read \u201cBombycid\u00e6.\u201d\n\n  Page 494, line 2 from top, for \u201cfrom which a larval form\u201d read \u201cfrom\n      a larval form which.\u201d\n\n  Page 542, line 12 from top, for \u201c_Dione Vanilla_\u201d read \u201c_Dione\n      Vanill\u00e6_.\u201d\n\n  Page 544, line 15 from bottom, for \u201c_Siderome_\u201d read \u201c_Siderone_.\u201d\n\n\n\n\nTranscriber\u2019s Notes\n\n\nPunctuation, hyphenation, and spelling were made consistent when a\npredominant preference was found in this book; otherwise they were not\nchanged.\n\nSimple typographical errors were corrected; occasional unbalanced\nquotation marks were corrected.\n\nAmbiguous hyphens at the ends of lines were retained.\n\n\u201cErrata\u201d at the end of this Volume have been applied to the relevant\ntext of this eBook.\n\nThis is Volume II of a two-volume set. The Table of Contents for both\nvolumes is in Volume I.\n\nThe Index for both volumes is in this one. References to pages 1-400\nwill be found in Volume I.\n\nFootnotes and references to them have been renumbered into one\ncontinuous sequence and moved to the end of the text of this eBook,\nimmediately preceding the Index. The sequence begins at \u201c171\u201d because\nthere are 170 footnotes in Volume I. Some \u201cSee note\u201d\" references may be\nincorrect.\n\nReading devices that cannot display some of the characters in this\neBook may substitute question marks or other placeholders.\n\nPage 414: \u201cHippophae\u201d was printed that way, rather than as \u201cHippoph\u00e6\u201d.\n\nPage 423: \u201cStage VII. Pupation.\u201d was printed in the second column.\n\nPage 436: \u201cPhaeton\u201d was printed that way, rather than as \u201cPh\u00e6ton\u201d.\n\nPage 642: \u201cyelk\u201d was printed that way.\n\nPage 539: The names in the \u201cSpecies\u201d line were italicized, but the\nunderscores used here to represent italics have been omitted from that\nline to make the diagram narrower.\n\nVolume II is available at no charge from Project Gutenberg,\nwww.gutenberg.org, as eBook number 47847. An eBook combining both\nvolumes is available as eBook number 47849.","title":"Studies in the Theory of Descent, Volume II","tradition":"philosophy-ethics","translated_text":null,"word_count":79718}