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

Possibly Continuous numerical Variation in Digits: miscellaneous Under this heading are placed in connexion a few cases of great interest. Whatever may be held as to the relation to the problem of Species of the phenomena hitherto described, it can scarcely be doubted that the following are instances of Variation which at least may be of the kind by which new forms are evolved. Great interest would attach to a determination whether the reduction of the digits in these cases is a continuous or a discontinuous process, but unfortunately these phenomena have been statistically studied by no one, and it is not possible to do more than make bare mention of the fact that such Variation is known to occur. There is no statistical evidence as to whether the individuals in any one locality may not fall into groups, dimorphic or polymorphic in respect of the degree to which the digits are developed (compare the case of the Earwig, Introduction, p. 40). As an inquiry into the Continuity of Variation such an investigation would be exceptionally valuable. In the case, for instance, of Cistudo mentioned below, such a statistical inquiry should surely not be hard to make.

'606. Chalcides. This is a genus of Lizards belonging to the family Scincidas. In several genera of this family the limbs are reduced or absent, differences in this respect being frequent among species of the same genera. (See BOULENGER, Catalogue of Lizards in Brit. Mus., 1887, in. pp. 398, &c.) Mr BOULENGER kindly shewed me a number of Lizards of the genus Chalcides from the shores of the Mediterranean basin which strongly resemble each other in colour and general appearance, but which contained almost a complete series of conditions in respect

of the development of the limbs and digits, ranging from C. ocellatu* and • '. lin/i-iiii/ii' with pentadactyle limbs fairly developed, through C. lineatus (tridartyl.-) and C. tridactylus to C. gueniheri in which the limbs are minutf conical rudiments. Amongst the species of this seriegreat individual variations occur. 007. Chalcides mionecton : normally four digits on each foot. A specimen in Brit. Mus. kindly shewn to me by Mr BOULENGER 008. C. sepoides : Mr BOULEXGER tells me that the normal number of digits on each toot is five, but that specimens occur having four digitmi each t<>ot.

009. Cistudo. This genus includes the North American Box- t ui-tlcs as defined by AGASSIZ (X. Amer. Testudinata, (.'<>ntrib. to X. H. of U. S., I. p. 444). These animals are widely distributed to the E. of Rocky Mountains. On the hind feet of some of them there are three digits, while others have four. GRAY (P. Z. S., IM-'.t. p. 10) described two Mexican specimens which agreed in having three large claws on the hind foot with no appearance of a fourth claw, and even scarcely any rudiment of the fourth toe, which was then believed to be present in the other members of the genus. To this three-toed form he gave the generic name Onychotria, but in Hrlt. Mns. Cat., 1855, he gave up this name as a generic distinction, describing the Mexican form as Cistudo mexicana, giving three toes on the hind foot as a definite character.

AGASSI/ in is"; 7 (I.e.) divided Cistudo into four species, giving to the Mexican form the name C. triniif/td.^, and he states that the western and south-western type is remarkable for having almost universally only three toes on the hind feet. The toe which is missing is the outer toe and " it fades away so gradually that the genus Onychotria cannot stand." The form found from New Kngland to the Carolinas is called by Agassiz C.virginea = C. Carolina, and he states that he received a three-toed specimen from N. Carolina which agreed in all other respects with those from New England.

PUTNAM (Proc. Boston, N. H. £,x. p. 65) stated that the threetoed form found in the South is only a variety of C. virainea, and that he had seen two specimens which had three toes on one hind foot and four on the other. 010. Rissa1. The commoD Kittiwake (R. tridactyla) as found in 1 In illustration of the possible bearing of these facts on the problem of Species reference in:iy In- made to the fact that among birds there are several examples of >]'<•<•!< s differing from their near allies by reason of the absence of the hallux. Speaking of this feature in Jocamaralcyon triilni-ti/hi, SCLATER observes: "In the pn -rnt bird we meet with :iin>tlnT example of the same character [viz. a monotypie form], and with one, perhaps more isolated in its structure than any of those above mentioned, Judimtir<iti->/<»i being notably different from all other members of the (ialbnlidne in the absence of the hallux. At the same time we must be careful not to put too high a value upon this at first sight seemingly important

this country and in N. Atlantic has no hallux, but only a small knob without a nail in its place. No variation in respect of this digit is recorded1. Birds not distinguishable from the Atlantic Kittiwake occur in the North Pacific, but amongst these Pacific specimens birds are found occasionally as rarities having a hallux "as large as it is in any species of Larus" (CoUES, p. 64-6). This feature also exhibits gradations. Specimens are described by COUES and also by SAUNDEHS having the hallux including the nail "2 in. long, with a perfect claw. These are given as extreme examples. SAUNDERS remarks that this hallux is small for the size of the bird, stating that another species of similar size, L. canus, had a hallux "5 in. long. Of these specimens of R. tridactyla from Alaska one had the nail of the hallux developed, though less so than in the extreme case. Saunders states further that the variation is not always equal in extent on both feet of the same individual : he considers that the extreme form is probably rare and local. COUES, E., Birds of North- West (U. S. Geol. Surv. Terr.), 1874, p. 646; and SAUNDERS, HOWARD, P. Z. &, 1878, pp.

611. Rissa brevirostris : a species from the N. Pacific distinct from R. tridactyla shews a similar variation in the development of the hallux, though in a smaller degree. A specimen has no claw on right hind toe and only minute speck on left ; another has no hind nail whatever ; another has small black nails of unequal size on the two hind toes. SAUNDERS, H., /. c., p. 165. (J12. Erinaceus. E. europceus has a large hallux, while in E. diadematus it is only 4 mm. in length, and in E. albiventris it is normally absent in adults. An adult female E. albiventris had a minute hallux in the left hind foot, represented by a claw and ligauientous structures, the phalanges being absent 2. In a female a few months old a minute hallux with usual number of phalanges was present on both sides. The presence or absence of a hallux has often been considered a sufficient ground for the formation of a new s;enus. DOBSON, G. E., P. Z. H., 1884, p. 402.

613. Elephas. In both the Indian and African elephant the number of digits represented by bones is five, both in the fore and the hind foot. The number of hoofs differs in the two species. The African elephant has normally four on the fore foot and three character, as the same feature occurs as is well known, not only in certain genera of other allied families (such as Alcedinidag and Picidas), but even in a genus of Oscines (Cholornis), in which group the foot-structure is generally of a very uniform character." SCLATEB, P. L., Monograph of the Jacamars and Puff-Birds, 1879 — 82, p. 50.

1 Mr A. H. EVANS has called my attention to a recent paper by CLARKE (Ibis, 1892, p. 442) giving an account of a minute rudiment of the hallux in embryos of E. tridactyla from Scotland. on the hind foot, and I am not aware that variations from this number have been seen. In the Indian elephant there is variation, and though I cannot give any complete account of the matter the following particulars may be <>f interest. According to Bun n\ the 'Elephant' has generally five hoofs on both t<>re and hind feet, but sometimes there are four, or even three1. He gives a particular case of an Indian elephant with four hoofs on each foot, both fore and hind feet.

TACMAUD-, to whom Buffon refers, was desired by the French Academy to notice on his journey in Siam, whether elephants had hoofs, and he states that all that he saw had five on each foot. I'o^ibly tlutour-toed variety does not occur in Siam. I am indebted t<» Mr \V. T. BLANFORD for the information that the natives of India attach importance to the number of hoofs, and also for the following references. HODGSON3 gives a sketch of elephants with four hoofs on each foot, marked " Elephas Indie UK, var. isodactylus nob., Hab. the Saul forest," together with the following note : " The natives of Nepal distinguish between the breeds with four toes [sic] on all the feet and those with five to four toes." SANDERSON* speaking of this says that some elephants have but sixteen hoofs, the usual number being five on each fore foot and four on each hind foot ; and that in the native opinion 'a less number than eighteen hoofs in all disqualifies the best animals.' FousYTH5 also alludes to the same fact.

Taken together these accounts seem to shew that five on the fore foot and four on the hind foot is the most usual number, but that both the number on the fore foot may diminish to four and that on the hind foot may increase to five. Several text-books mention the subject but I know no statistics regarding it. In view of the different number characteristic of the African elephant this variation has some interest. In particular it would be of'use to know whether the variation exhibits Discontinuity, and also to what extent it is symmetrical.

614. ItecuiTciico of digital Variation in strains or families is frequent, but though many observations on the subject have been made no guiding principle has been recognized. To the general statement that digital Variation, whether taking the form of polyclactylisni or other- 1 BUKKHN, Hist. Nat., xxviii. p. 201. The mention of three hoofs must I think refer to the African species, which Buffoii does not distinguish from the Indian. In the Camlni.li/i' Museum (Catal. (){M) is an old preparation of the skin of an elephant's foot having three hoofs. This is declared by the Catalogue to be the fore foot of an Indian elephant. Perhaps this is a mistake.

3 HODGSON, B. H., Mammals of India, MS. in Zool. Soc. Library. wise, does very commonly appear in the offspring or kindred of the varying individuals I can add nothing. It should be mentioned that though in families exhibiting digital Variation the forms that the change takes may differ (in some cases widely even among individuals nearly related) yet on the whole the variation, if recurring at all, more often recurs in a like form. This holds good apart from the rarity of the particular form of variation. The facts described by FAROE (I.e., infra) are exceptionally interesting in this connexion. In the family described by him duplicity of the thumbs occurred in the paternal grandmother, while the father and three children had their thumbs of the three-phot anged form as in No. 483. This case strikingly illustrates the well-known principle that Meristic variability may appear in the same strain or family under forms morphologically very dissimilar.

Attention is also called to the circumstance that in the case of the three toes in the ox (No. 558) the descent was wholly through females, and the same was almost certainly true in the polydactyle cats (No. 480). In the case of the syndactyle pigs the evidence of maintenance of the variation in the strain is very clear (No. 584). See also No. 564. As regards digital Variation in Man the following are the best genealogical accounts : Association of digital Variation with other forms of Abnormality.

615. In the great majority of cases of polydactylism the rest of the body is normal, the limb or limbs varying alone. There are however a certain number of examples of polydactylism in association with other abnormalities ; as for instance with phocomely, cyclopia, double uterus, hare-lip, defective dentition, defect of tibia, tic., but there is nothing as yet to indicate any special connexion between these several variations. Diminution in number of digits and syndactylism is on the contrary very often associated with general deformity and with many forms of arrested development. To this no doubt is largely due the fact that cases of ectrodactylism are commonly irregular, whereas polydactylism is generally fairly regular in its manifestations, for numerous cases of diminution in number of digits occur in bodies or in limbs otherwise amorphous.

1 The notorious case of a village in Isere where the majority of the inhabitants are said to have been polydactyle. Most modern writers on the subject quote this statement but I have never found original authority for the fact. By some it is referred to DEVAY, Du danger des mariages consanguins, 1862, p. 95, but I can find no mention of the facts in that work. Ix the remarks preliminary to the evidence of digital Variation it was stated that this group of facts is interesting rather as bearing on morphological conceptions than from any more direct relation to the problem of Species. The indications to be gained from the evidence will be treated under the following heads:

animals. C2) Particular forms of digital Variation proper to particular i h The maims and pes as .systems of Minor Symmetry. ( ."> ) Duplicity of limbs. (7) The absence of a strict distinction between duplicity of (!)) Relation of the facts of digital Variation to the problem of Species. (1) Comparative frequency of digital Variation in different animals. In reviewing much of the evidence of Variation and especially in the evidence concerning the variations of teeth it has been seen that the frequency of these variations is immensely greater in some classes or species than in others. This is remarkably clear in the case of the variations of digits. Compare for instance the great frequency of polydactylism in the Horse with the complete absence of recorded cases in the Ass. It is true that the latter is the rarer animal, but it might still be expected that some record would have liecn ioimd if the variation were as frequent in the Ass as in the Horse. Again polydactyle Cats are certainly not very rare and specimens are in several collections having been acquired at many

dates. On the other hand digital Variation in the Dog seems to be confined to the formation of a hallux in the hind foot, and to duplicity of hallux and pollex J. Similarly though digital Variation is so common in the Pig it is very rare in the Sheep, only one or two clear cases being so far known to me. Note again that polydactylism is common in the Fowl and has been often seen in the Pheasant, while in other birds it is very rare. Some one will of course remark that the Fowl is a domesticated bird and the Pheasant is partially so ; but pigeons, ducks - and geese3 are as much domesticated and in them digital Variation does not seem to be known. The cases in Apes deserve mention in connexion with this matter. One case of syndactylism was quoted in Pithecia No. 525, a case of polydactylism in Macacus No. 50-i, in Orang No. 511, and in Hylobates No. 508, and a case of ectrodactylism in Macacus No. 526. These five cases surely suggest that Meristic Variation is something more than a mere result of high feeding or of " unnatural " conditions. It is not a little strange that among Apes Meristic Variation should be frequently met with in so many systems of organs.

(2) Particular forms of digital Variation proper to particular Of more significance than the frequency with which digital Variation recurs in certain animals is the frequency with which in particular animals it approaches to particular forms, or to particular conditions in a series or progression of forms. This has been seen in the Cat, Man, Horse, Pig, Ox, &c. In each of these the mode of occurrence of Variation has in it something distinctive, something that marks the phenomenon as in some way different from the similar phenomena in other forms. Taking for instance the curious series of cases found in the human manus, ranging

1 Both these variations are of course very common and may be seen in any walk in the streets. The hallux is very frequently present in the Dachshund and is common in Collies, Mastiffs and other large breeds. In the Mastiff dew-claws (hallux) are not a disqualification (SHAW, Book of the Dog}. In the St Bernard the hallux is very often double, perhaps more often than not. This is largely due to the fact that the monks of the Hospice considered the presence of the dew-claw of the utmost importance and preferred it double if possible (SHAW, I.e.). The same writer states that ' the more fully the dew-claws are developed the more the feet are out-turned.' This fact suggests that there may be a change of Symmetry like that in the Cat, but I have no observations on the point. I have several times seen simultaneous duplicity of hallux and of pollex in the same individual (Dachshund, &c.). Other digital variations must be rare in dogs as there are hardly any recorded cases. A problematical case of ectrodactylism is given by BACM, Deut. Ztschr. f. Thicrm., xv. 1889, p. 709, Jig. [q. r.]. I once saw a mongrel Fox-terrier with no pollex on either inarms, but I was not satisfied that they had not been cut off, though there was no suggestion of this.

- For an interesting account of a Duck with the webs of the toes almost wholly absent see MOBIUS, Zool. Gart., xvm. 1877, p. 223. Another case of the same kind MORRIS, F. 0., Zool., iv. p. 1214. from tinaddition of a phalanx t" the polh-x up to the condition ol' X,,,,. 4s.s ,,r 4!»o, ;uid comparing them with the essentially similar series of cuses in tinhind foot of the Cat, there is this remarkable difference: that though both progressions lead up to a similar kind of Symmetry in the series of digits, in the human maims an approach is m;ide to a system of Symmetry whose axis lies internal to the index, while in the Cat's feet the axis lies external to the index (see Section (4)). The series of forms in the manus of the Cat is still more peculiar and is not like any case of polydactylism in other animals.

From the evidence it will have been seen that digital Variation in most of its manifestations may be similar and simultaneous in the limbs of the two sides of the body, though not rarely it affects the limb of one side only ; and still more frequently the form which it assumes on one side differs in degree from that found on the other side. Considerable difference in kind between Variation on the right side and on the left is much rarer. Almost the same statement may be made respecting simultaneity of Variation between the manus and the pes, though in the pes the manifestation of Variation is rarely identical with that in the manus of the same individual. Some variations, as for instance duplicity of pollex and hallux, or extra digit external to minimus, are not rarely found simultaneously in both pes and manus, but there are many cases in which no such agreement is found. The frequency of this simultaneous variation in the case of syndactylism in the Pig may be specially noticed.

Certain variations in certain animals seem to be almost or quite restricted either to hind limb or to fore limb. The form taken on by the pes of the Cat upon increase in number of digits is distinct from that assumed by the manus. The development of the digit II in the Horse is much more common in the manus. The extra digit (or pair of digits) in the Pig is so far as I know Beep only in the rnanus. On the contrary the three-toed state in the Ox is found in the manus and also in the pes. Generally speaking, Meristic Variation is much commoner in fore limbs than in hind limbs.

One fact here calls for special notice. Though general statements are hazardous, we are perhaps justified in affirming the principle that large Meristic Variation, involving great departure from the normal, very rarely affects exclusively one side of a bilaterally symmetrical body. In cases of variation in vertebrae, in spinal nerves, in teeth, in the oviducts of Astacus, and many more, it is seen that on the occurrence of great variation the change is seldom restricted wholly to one side of the body, though the condition reached by the two sides is frequently of differing degree. Now in

the extreme forms of double-hand as seen in Man there is a curious exception to this principle. For in nearly all the extreme cases the abnormality was on one side only, the' other being normal. This was seen in Nos. 492 — 500 and 501 — 503, and also in Macacus No. 504. The case No. 500 is probably an exception to this general statement. As to the significance of this absence of correspondence between the right and left sides in extreme cases of digital Variation I can make no conjecture. It has seemed that perhaps in such cases the absence of symmetry between the two sides of the body may be connected with the fact that in these extreme forms of double-hand an approach is made to a bilateral symmetry completed within the series of digits. But against this suggestion must be noticed first the fact that a similar bilateral symmetry is established in the six-toed pes of the Cat (Condition IV of the pes, p. 316), but the variation is nevertheless found on both sides of the body ; and secondly the case of double-foot in the lamb (No. 566), though for reasons stated this latter case may perhaps be open to question.

This is a subject to which it is most difficult to give adequate treatment. Several of the phenomena have as yet been studied in far too small a range of cases to justify sound generalization, and with further knowledge the suggestions arising from the facts now before us may not improbably be found to have been misleading wholly or in part. Besides this there is a serious difficulty in finding modes of expressing with clearness even those principles of form which seem to underlie the phenomena. This difficulty proceeds first from the vague and contradictory character of the indications, and next from the total absence of a terminology by which diversities of symmetry and the form-relations of parts may be expressed. Nevertheless it has seemed best to abstain from the introduction of new terms until the ideas to be expressed shall have been more clearly apprehended. It need scarcely be said that the remarks which follow merely represent an attempt to state some of the lines of inquiry along which the facts point.

On p. 88 mention was made of the fact that in a Bilateral Symmetry the organs which occur as a pair, one on the right and the other on the left, in so far as they are symmetrical are optical images of each other, this relation of images being what is implied by the statement that these organs are bilaterally symmetrical. The hands and feet of vertebrates are organs of this kind, the right hand and the right foot being approximately images of the left hand and foot respectively. But beyond their symmetrical relations to each other in the Major Symmetry of the whole body each manus and each pes may exhibit the condition of a Minor Symmetry within the limits of its own series of digits. Not only may each limb geometrically balance the limb of the other side

luit its own external parts may more or less balance its own internal parts. This relation diff.-is greatly in different animals, the Minor Symmetry b.-ing nearly complete in the Artiodactyles and in ih. Horse, hut much less so in the human maims and pes, &c. 'I'll.- matter now for consideration is the influence or consequences of ill-- existence of this symmetry in the Meristic Variation of digits ; and conversely the light which the observed phenomena of Variation throw on the nature of that relation of symmetry. It will be seen that in some points the two halves of a bilaterally symmetrical limb behave just as do the two halves of the bilaterally symmetrical trunk, while in other points their manner of Variation is different.

Thus, the digit III of the Horse may divide into two halves related to each other as images, bearing hoofs flattened on their ntljncent edges: that is to say, the two resulting parts are formed ii-.t as copies of the undivided digit, but as halves of it, a condition m-vei seen in division occurring anywhere but in the middle line of a bilateral Symmetry. In the syndactyle teet of the Pig or the Ox the converse phenomenon exists; 'for the digits III and IV, which normally stand a- images of each other, are here wholly or in part compounded to form a digit to which the uncompounded digits are related as halves.