Materials for the Study of Variation Treated with Especial Regard to Discontinuity in the Origin of Species

One more case may be given in illustration of the kind of difference which normal forms may present. 74, Pipa (the Surinam Toad). In the majority of the Batrachia, the most anterior pair of spinal nerves leaves the vertebral column between the first and second vertebra?, no sub-occipital being present. The • ml j.air leaves between the second and third vertebrae, and the third pair leaves bet \\i-.-n the third and fourtli vertebra?. The braehial M- i> formed by the whole of the second pair together with parts nf

tintirst and third pairs. (The details of the arrangement are compli- • 1 and \ary L'l-eatly in different forms.) In PIJMI a different arrangeni'-nt exists. The most anterior pair of nerves leaves the spinal HUM by [» ft'orntinij the tirst vertebra, and the pair which leaxes between tinlirst and second vertebra1 is therefore ordinally the se<-<>inl pair- "f -pinal nerves in this form; the pair which leaves between the • nd and third \.-it. -br:e is the third, and so on. The brachial plexus is mad. • up ' -r the wlmle of the second nerve, nearly the whole of the third nerve and nf a branch of the first.

It tht-ii it wen- to be supposed that the pair of nerves which leaves tincc.liiiun between tinfirst and second vertebrae in 1'ipa is homowit h tinpair of nerves which leaves in the same place in /.'"/"/. A'-, it is rlcar that lirtween the skull and the 2nd vertebra of J'ijm, thenis an extra pair nf nerves not found in Rana. The number of v.-rtebi;.- in /'i/,<i is however less than in liana. For in the former thenanonly M-M-M nf these, making with the united sacral vertebra and urn-t\le eight pieces in all; but in Rana there are eight prae- .-aeraU. nne sacral, and counting the urostyle, ten pieces in all. In ' only nne spinal nerve, the 10th, leaves the urostyle, while in l'iri two pair>, the '.'th and loth. pass out through the terminal piece "f the \ertebral enliiiiin. suggesting that the diminution in the niiinlier nf \eitebr-e i - due to the absence of separation between the (.»th vertebi-a and the ui-n>tyle. The whole number of spinal nerves i* theret'nr.- the -aine in bnth Rana and Pipa, but in the latter the Ut pair pei-fnrate tlie Ut vertebra in addition to the 1'nd pair whi.-h ; ,t between the 1st and L'ncl vertebra. l'"i I;I;HIM;I:]; ',

It waa suggested by STASTNII - (A.7//-6. d. vergl. Anat., p. 130, Xote) that pcrhaj^ the l>t \ertebia nf /'//»' n-pi-esents two coalesced vertebr:e, but in an anatniniral e\aminat imi nf two specimens of I'ijiit, I i. ringer (/..-. IsTl.p. IMIJ, fmind ao confirmation of this suggestion. and developmental evidence also went to sheu that no such fusion A Fttrbringer saj - there is no satisfactory \\ay of bringing this case nf /'//„/ into accord \\ith the .-ondition seen in lui,,,i. In the Urodela tliere is of coarse a >ul.occiital be

is of coarse a >ul.occipital nerve between the skull and the 1st vertebra \\hi.-h i> ao1 present in /,'»///,/. and some resemblance to Pipa '- 'I'"- bed; ll1" in ih.- Orodela the 1st spinal does not actually Lmphibia. rel tiiat thi^ papn did nut aj-p.-ar in time to enable nie to anastomose with the plexus, though it gives off the superior thoracic, which in both liana and Pipa comes off at a point peripheral to the formation of the plexus (Fiirbringer).

If the two spinal nerves which come out of the urostyle in Pi]>«, may be taken to shew that this bone contains n + 2 vertebra; while tinsingle pair in Rana shews the urostyle to consist of n + 1, there is in Pipa (as compared with liana), a diminution of one in the total number of vertebrae, together with a backward Homwosis, which is seen in the fact that the 8th vertebra bears the pelvic girdle. Turning now to the nervous system, the fact that the last spinal nerves to join the brachial plexus in Pipa are the Illrd, while in liana they are the IVth, is again an evidence of backward Homoeosis. But if this process were completely carried out, the pair of nerves which in Pipa pass out through the 1st vertebra should pass out between this vertebra and the skull, i.e. in the position of the suboccipital of the Urodela. Beyond this analysis cannot be carried, and this case is a good illustration of the fact that the hypothesis of an individual homology between the segments does not satisfy all the conditions of the problem.

Relation between the ordinal position of spinal nerves and their distribution to the limbs. This subject is introduced partly because it further illustrates the nature of the relations which the spinal nerves maintain towards each other, and thus bears indirectly on the phenomena of their Variation ; but chiefly because it presents a view of some of the complexities which arise in the apportionment of organs centrally disposed in Meristic Series, to the parts of peripheral appendages having no clear or coincident relation to the primary or fundamental segmentation of the body. The facts have thus a value as furnishing a kind of commentary on the nature of Meristic Repetitions in vertebrates. In any attempt to interpret or comprehend Meristic Repetition as a whole, they must be taken into account.

The principles of the distribution of the spinal nerves to the muscles of the fore-limb have been thus eiiuntiated by HERRINGHAM. 1. "Of two muscles, or of two parts of a muscle, that which is nearer the head-end of the body tends to be supplied by the higher, that which is nearer the tail-end by the lower nerve. 2. "Of two muscles, that which is nearer the long axis of the body tends to be supplied by the higher, that which is nearer the periphery by the lower nerve.

3. "Of two muscles, that which is nearer the surface tends to be supplied by the higher, that which is further from it by the lower nerve." HERRINGHAM, W. P., Proc. Roy. Soc., XLI. 1886, p. 437. Details are given shewing the manner in which the innervation of the muscles in. Man bears out these principles. FORGUE and LAXNEGRACE ', who worked with clogs and monkeys by physiological methods, arrived at conclusions identical with those which Herringham came to by human dissection.

1. -i.i two -p..t- on the skin, that which is nearer the prc-axial ii..i-diT ii-niU t" \«- -upplied I iy tinhigher nerve. i'. --(if two -pots in tinpre-axial area the lower tends to be -upplie.1 liv tinlower nerve, and of two spots in the post-axial area the lower t.-ii-l- t., 1,.- supplied by tinhigher nerve." "Tim-, if the liml' !>«• seen from tinfront, the two highest nerves on the outer ami iniifi- >idere-peetively are the IVth and Xth. I. .-i than the-,- tli«- Vth and Vlth take the outer, the IXtli and Xtli the inner .-id.-. llelow tinelbow the Vlth alone takes the outer, and tli.- [Xth alone the inner. In the hand, while tin- Vlthand IXth continue their po-itions, the V I Ith and VTIIth for the first time join in the supply." 1'a i titulars from which this general statement is made are given. II UMIIM.H \M, I.e. p. 43!>.

A.-eeinliiiL,' to -uli-ei|iient investigations of SHERRINGTON's on the l,'iiiil-liii<l>. the iiinei-\atioii of the muscles of the posterior aspect of the thL'h and \<-^ do riot follow the third of Herringham's principles, for in their case tli.- deep lay.-i- of muscles is innervated by roots anterior to tin.-,- w 1 1 i.-li innervate the -uperiieial muscles. The same experiments also, though eleai-ly -hewing th:it the nerve-supply of the skin of the hallux i- anterior bo that of th.- ."nh digit, gave only equivocal evidence that the -am.- \\as true of tinmii-eul.-it ures of these two digits; and in the thiudi the gra.-ilis is not supplied before the vastus externus, whose relation i- iMther that ..f \.-ntral to dorsal than of anterior to posterior-. SHEBBINGTON, C. >.. Proc. /.'"//.>'"'•., 1^'Ji', LI. p. 77.

S.iin.- r.-a t ures in the Meristie X'ai'iat ion of the spinal nerves, as illu.-trateil by tinforegoing .-vuleuee. may be briefly summarized In tintii-st place, as inio-ht be anticipated from the compound natui-.- of a »pinal Derve, when lloiiiieotie Variation takes place, it <l»e» not commonly occur by the fcransformatioD of I'ntire nerves, but rather by change in the distribution and functions of parts of ner\.-. In tinrespect, thereloiv, th.-iv is a dit't'ercnce between lloimrosis iii spinal oerves and that in \ertcbrie, for in the latter,

• In-! as in making up the chapters of a book into volumes, \\hohchapter.-, ma\ be put into one \olmne or into the next, and the following chapters renumbered, so it IIKIV be with the Variati f vertebrffij for these may belong \\liolly to one region of the -pun- or i" another. Hut tinoerves anlike chapters made up of sections : |>ariicnlar sections or groups <>f sections mav come in an earlier chapter ,,r in a Mibse([iieiit one, and the places of those that

hypothesis that both forms are descended from a common ancestor, that such changes and renumbering of whole nerves must have happened, though there is evidence to shew that this may happen piecemeal, as in cases given. Of course in speaking of such changes among the vertebrae it will not be forgotten that partial changes occur too, but there is still greater Discontinuity in their case than in that of the nerves. But that there is Discontinuity in the case of nerves also is clear ; for a given fibre, supplying a given muscle, must leave the spinal cord either by one foramen and one spinal nerve, or by another. Conversely the ?ith motor nerve must supply either one muscle or another, and the transition between the two, however finely it may be subdivided, must ultimately be discontinuous in the case of individual fibres. It would be interesting to know to what extent fibres vary in bundles, but this can hardly be determined.

There is, however, some evidence that the group of fibres supplying a limb does to some extent vary up and down the series as a group, though much rearrangement may occur also within the limits of the group itself. Lastly, there is important evidence that Variation in other parts may be correlated with change in the ordinal positions at which nerves with given distributions emerge from the spinal cord. With Variation in the ordinal positions at which the nerves come out, change in other parts, notably in the ribs, may happen too ; so that we may say that in a sense there may be, at least within the limits of single species (see cases Nos. 24, 65 and 71), a correlation between the apportionment of their functions among the nerves and the contour of the body, both changing together, the ribs rising and falling with the rise and fall of the brachial plexus. The nerves do not merely come out through the foramina like stitches through the welt of a shoe, the shape of the shoe remaining the same wherever the threads pass out. The arrangement is, rather, like that of the strings of such an instrument as a harp or piano, in which there is a correlation between the curves of the frame and the positions of the several notes : so long as the frame is the same, the strings cannot be moved up or down, the instrument still retaining the same compass and the same number of notes.

Tin: occurrence of Hoimeosis among the appendages of Ar- thropoda is illustrated by a small but compact body of evidence. TO tinevidence special value may be attached, not because it i- likelv that in the evolution of the Arthropods variations have reallv taken plan-, in magnitude comparable with those now to In- described, but rather because these cases give a forcible illustration of possibilities that underlie the common and familiar phenomena of Mrristic Repetition. Of these possibilities they are indeed " Instances Prerogative," salient and memorable examples, eiiuntiating conditions of the problem of Variation in a form that cannot be forgotten. Facts of this kind, so common in flowering plants, but in their higher manifestations so rare in aiiimaU. hold a place in the study of Variation comparable pel ha ] i>. uith that \vlnch the phenomena of the prism held in the studv of the nature of Light1. They furnish a test, an elenchus, which any hypi'the-i- profiling to deal with the nature of organic I!. | ii titioM and Meristic Division must needs endure.

•;:,. Cimbex axillaris (a Saw-tly), having the peripheral parts of the left antenna developed as a foot. The right antenna is normal, ending in a club-shaped terminal joint. In the left antenna the terminal joint is entirely replaced by a well-formed f..ot, ha\ing a pair of normal claws and the jtltintula between them ( Fig. III). This foot is rather smaller than a normal foot, but is perfectly formed. The iv>t of the antenna, so far as the point at which the club should begin is normal in form, but is a little Mnaller and thinner than the same parts in the right antenna. KUAATX. (',.. /V///. ,„/. /f/.xv///-., LsTU, XX., p. 377, PL

This specimen was most kindly lent to me for examination by Dr Kraatz, but to this description I am unable to add anything1. FIG. 16. Cimbex axillaris : right antenna normal ; left antenna bearing a foot. II. the left antenna seen from in front. III. the same from above. After KRAATZ. It should be noted that the plantar surface of the foot was turned rather forwards as shewn in the figure, and not downwards like the normal feet. *76. Bombus variabilis £ (a Humble-bee). A specimen taken beside the hedge of a park in Munich, having the left antenna partially developed as a foot. The first two joints were normal. They were followed by two joints which were rather compressed and increased in thickness and breadth. Of these the first was oblong and somewhat narrowed towards its apex by two shallow constrictions, giving it an appearance as of three joints united into one ; below it presented a projecting and tooth-like point. This joint was only slightly shiny. The next joint to it was almost triangular, and was reddish-brown, shiny, and having hairs on its lower surface. Posteriorly it was prolonged inwards, covering the previous joint so that both seemed to form one joint : the posterior edge was somewhat thickly covered with hairs. The upper part of the first of these two joints and the prolongation of the second were together covered by a hairy, scale-like third joint, which seemed to be only attached at its base. From the apex of the second joint arose a shortened claw-joint, like the clawjoint of a normal foot. This joint was reddish-brown and shiny, bearing a pair of regularly formed claws, like the claws of the foot. KRIECHBAUMER, Entom. Nachr., 1889, xv. No. 18, p. 281.

1 Some to whom I have spoken of this specimen, being unfamiliar with entomological literature, and thus unaware of the high reputation of Dr Kraatz among entomologists, have expressed doubt as to its genuineness. I may add therefore that the specimen, when in Cambridge, was illuminated as an opaque object and submitted to most careful microscopical examination both by Dr D. Sharp, F.R.S., and myself, and not the slightest reason was found for supposing that it was other than perfectly natural and genuine. The specimen was also carefully relaxed and washed •with warm water, but no part of it was detached by this treatment.

The tw>, following cases must be given here, inasmuch as tli.-v relat,- t.. H..ni.i"-i- of the appendages in Insects; but in tin-" case "t tli-- first the evidence is unsatisfactory, and in the case ,,t' ili,- second there is considerable doubt whether the variation is reallv "I' the nature of Homoeosis. 77. Prionus coriarius £ : having elytra represented by leg-. 'I'll.- following i- a translation of an announcement in the Hi-ftu.i-r L'i't. Ztg., 1^40, vol. I. p. 48, which is copied from the original communication t'. the Preussische Provinzial-Blatter, Bd. \\. (Th.- latter journal not seen, \Y. B.] :- " One of my pupils 1. n.u-lit me to-day a male 7V/o////.s- coriarius, Fbr., the thorax of \\hich is remarkably constructed. The homy covering of the mesothorax is ali-'-nt. and in place of the elytra is a pair of t'ullv developed legx which are directed upwards and backwards. Tli,-,- legs anin.-< rt«-d at the points of articulation of the elytra. Th'- metathorax >upports the wings as usual and the abdomen is not hardened more than it usually is. In trying to fly, the creature mov.-d these upwardly directed legs simultaneously with its win^s. Th«- -eutdliim is absent and the prothorax has i.nlv two .-pines; other parts iionnally developed." Dr SAAGE, Urami.-berg, 1.S39 : — Hagen, in quoting this case, mentions that th«- >p.-eiint-n was att.-rwards seen by von Siebold, but gives no reference to any writing of von Sii-hold on the subject.

[If this >p,-<-im«-M >till exists, it is to be hoped that a description of it may be published. In the absence of further information tin -re .-i-enis to be no good reason tor accepting the case a- genuine.] •;s Zygaena filipendulae </. Specimen possessing a supernumerary wing arising in such a jiosition as to suggest that it replaced a leg. This -].,-, -mien was originally described by UiCHARDSOX, N. M.r /•'/</,/ Cl,il. I.s!»l. and was exhibited at a meeting of the Ehitomological Society of London, 1MM. J'roc. p. x. The extra

wing was in ^,-nei-al form and appearance like a somewhat folded hind wing but its colour was rather yellower, though it was more red than yellow. I have to thank Mr Richardson for allowing me to examine this specimen in company with Dr Sharp. In compliance with Mr Richardson's wish we did not strip the wing or remove the thick hairs which surrounded its base, and it is therefore not possible to speak with certainty a& to its precise point of origin. The following description of it was drawn up for me by Dr Sharp : " The supernumerary wing projects on the under side of the body, and at its base there intervenes a space between it and the dorsal region of the body about equal to the length of the metathoracic side-piece. The exact attachment of the base of the supernumerary wing cannot be seen owing to the hairiness of the body, but so far as can be seen it is to be inferred that the wing is attached along the length of the posterior coxa, the outer edge of the point of attachment may be inferred to extend as far as the suture between the coxa and thoracic side-piece ; if this view be correct the abnormality may be described as the absence of the hind femur and parts attached to it, and the addition of a reduced wing to the hind-margin of the coxa. It is, however, just possible that if the parts could be clearly distinguished it might be found that the real point of attachment of the abnormal wing is the suture between the metathoracic side-piece and the hind coxa."

It should be distinctly stated that there is no'empty socket or other suggestion that the rest of the leg had been lost, and it was in fact practically certain that it had never been present. There is thus a strong prima facie case for the view that the leg has been developed as a wing, however strong may be the theoretical objections to this conclusion. On the other hand, as will be shewn in a later chapter, supernumerary wings are known in specimens having a full complement of legs, and it is conceivable that one of these supernumerary wings may have arisen in such a way as to prevent the proper development of the leg from the imaginal disc. If the specimen were carefully stripped of hairs some light might perhaps be thrown on this question. The figure (Fig. 17) is from a drawing kindly lent me by Mr Richardson.

*79. Cancer pagurus. Specimen having the right third maxillipede developed as a chela. This animal was brought by a fisherman to the Laboratory of the Marine Biological Association at Plymouth. It is a male, measuring five inches from one side of the carapace to the other. All the parts appear to be normal with the exception of the third maxillipede of the right side. This structure, however, has the form shewn in Fig. 18, A, differing entirely from the ordinary condition of the appendage. Fig. 18, B, is taken from the third maxillipede of the left side and shews the ordinary structure of the same parts. On comparing the two figures it will be seen that the protopodite does not differ in the

limbs of tin.- two side^ ; that the exopodite of the right side is Fn.. is Cancer i»i,inms • • tinright and left third maxillipedes, that of the ; -i.l.- having t!,. endopodite in tinlikmt-ss of the endopodite of a chela. part- n |>n -. iitinr isrhi..]iuiiite and meropodite, </'. groove representing the suture lit which a normal cli.-la is thmwn olT it injured. From P. Z. S., 1890. niially like that of the left, but that it lacks the inner process and the tii-'-lluni \\hich anborne by the normal part. There was some indication that this branch of the limb had been injured, and perhaps tintlagellum may have been torn away, but the appearances \\vnnot Midi as to warrant a (-(inclusion on this point. The braiirhial .-].ipo.liti-s (not shewn in the figures) were normal in both cases. Tinrnd<>p<nlite of the right side was entirely peculiar, and was in tact, literally transmuted into the likeness of one of the great chela?. It consists of a single joint (ntl), articulating with the basipodite centi-ally and bearing the carpopodite. This >in-le joint repi-esents, as it were, the ischiopodite and meropodite of an ordinary chela, but these two parts are ankylosed together and the articulation between them is only represented by a groove i //). Another groove (f/') represents the groove upon the ischiopodite of the chela, at which the limb is commonly thrown off by tinanimal if it is injured. The carpopodite, propodite and dactylopodite are freely mi >\ cable on each other and hardly differ, save in absolute si/e, from those of the normal chehe. The shape, proportions and texture are all tho>e of the chela. B.VTESOX, W., Proc. /•»,/. Nor., 1MM). p. :,.s(i, fig. 1.

SO. A similar case1 of ('<n,,-i r j»i<iitrnt< $. 4 inches across carapace, mature, right pcdipalp [i.e. 3rd maxillipede] normal, left pedipalp modified into a chela having all the joints clearly defined, CORNISH, T., Zoologist, S. :}, vni. p. 349. *«S1. Palinurus penicillatus. The left eye bearing an antennalike tla^ellnm, growing up from the surface of the eye as shewn in the figure (Fig. lit). The eye-stalk and cornea, as represented, appear to h.u,' hren of the normal shape but reduced in size.

FIG. 19. Palinurus peniciilatus, the left eye bearing an antenna-like flagellum. After HOWES. 82. Hippolyte fabricii differs from other species of the genus in being usually without epipodites at the bases of all the cephalothoracic legs except the first pair, while in the other species these appendages are usually present upon the bases of the first and second, or upon the first, second and third pairs, and on this character it was placed by KROYER in a separate section of the genus.

Of 52 individuals (18 males varying in length from 27 mm. to 39 mm. and 34 females varying from 16'5 mm. to 50 mm.), from various localities on the New England coast, 47 had the normal number of epipodites, while 5 had epipodites on one or both of the second pairs of legs. Of the latter 3 were from the Bay of Fundy ; one </, 35 mm. long, has well-developed epipodites on each side of the 2nd pair of legs ; another </, 36 mm. long, has a short epipodite on the left side and none on the right ; the other, $ , 47 mm. long, has a well-developed epipodite on the left side and none on the right. The two others were from Casco Bay ; a $ , 36 mm. long, with a short epipodite on the left side, and a </, 28 mm. long, with a rudimentary one on the right side. As the measurements shew, the presence of these epipodites is not characteristic of the young. SMITH, S. J., Trans. Connecticut Acad., v. 1879, p. 64.

\'iirintt»n in (lie n timber of generative openings in Crayfishes. Astacus fluviatilis. A female having the normal pair of o\ iducal openings on tinbases of the antepenultimate pair of walking legs. ami in addition to them another pair of similar openingplaced njmii tincorresponding joints of the penultimate pair of walking legs. <>n dissection it was found that the ovary uanormal, and that from each side of it a normal oviduct was given off; Inn each of tip-.- oviducts divided a little lower down to (MIIII two Hnaller oviducts, one of which went to each of the four oviducal openings. 1 )i>\iA iirsT1, E., Ann. Soc. Ent. France, 1848, Ser. •_'. vi. p. 47!». /'/.

Astacus fluviatilis $, having a supernumerary pair of oviducal opening placed mi tin' luxt pair of thoracic legs. The normal oviducal openings wenin the usual position and of the usual shape and si/e. but in addition to them there was an extra pair placed on the last thoracic legs. It should be remarked that though the-e are the appendages upon which the openings of the male organs are placed, the oviducal openings were not in this case -ituated at the posterior surface of the joint as the male op, nin-- are, but were placed relatively to the leg in the same situation ax the female openings on the antepenultimate legs. The penultimate legs and the abdominal appendages were normal. On dissection it was found that each oviduct after passing for the greater part of its course as a single tube, divided into two parts, one of which went to each oviducal opening. The ovary itself was normal. HiiMi \M. \V. B., Ann. M<I:/. .V. //.. ]s!)l, Ser. 6, VII. p. '2't>\. /'/. ill. ! I am ui'catly obliged to Mr Beiiham for an opportunity of examining this specimen. Attention is called to the tact that in this specimen Homoaosis occurs in an unusual way, having a gap in the series: for the openings are on the ante- j» unlfiimitf and fnxt thoracic legs respectively.]