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

On the whole, then, it seems satisfactorily shewn that the tailless form is connected by intermediate stages with the fully-tailed .-1. xalina, and that this transition is at all events partly connected with the degrees of salinity of the water in which it lives. Almost each locality has its own pattern of Arteinia, which differs from those of other localities in shades of colour, in average size, or in robustness, and in the average number of spines on the swimming feet, but none of these differences seem to be especially connected with the degree of salinity.

Passing now to the question of the distinctness of A. niilhiniiirnii, it seems clear tlnit. ke said, it should never have been considered a distinct species. The the Unless tail, which is now seen to be one of degree, does not differ - lorily, and, as Schmankewitsch found, it is to be seen swimming \utli tin i individuals. It has never been shewn that there is a male A. mil- \utb distinctive sexual characters, and among the Branchiopoda the -cxual characters of the second antenna in the male are most strikingly

ictiv* "i tinseveral forms. While being in no sense desirous of disparaging •uankewitsch's very interesting observation, I think it is misleading ihe the change effected as a transformation of one species into another. bach himself expressly said that he did not so consider it, and it is t inn nf th«' division of the 8th post-abdominal segment of Artemia, ii mi dilution of the water, directly concerns the subject of Meristic :. A- to tinflirts, there is no doubt that the tail of Branch ipu.f appears to

in. 'iii- i.esidt.-s the two which bear the external generative in all. niiif, \vlnliin the commonest forms of A. mlina there are only nt-; and that the difference lies in the fact that in the long t'li:. . [. galina there is generally no appearance of division. But as ' -h<\\n. tli«- la-t apparent division in Branchipits is of a different • I'l.. m that of the other alidmninal -• ' . This is indeed easily seen .on i-, \. :roin that of the other .li\ isious. It appears, in fact, to be

I. annulatioii thai -illation. In longitudinal sections file distinction -u. -li a di\i-i- • iin^' to Schmankewit-eh. appears in the third in a <-. i. nuniliei, and tln-.-e are not hv any means from the most dilute v.at«T.- iilon. . • ih. in lieiiiL' from wai real e..nrentratinn. For instance, .menin XXIX. LI, XXXVII. XXXIX and XIV, all have no trace of sneh •ion of this appearance of division to the salinity of the water is a constant one. Lastly, It tinrelation of Art.min t.. Branc/iipus, Schmankewitsch has

in:iinta:in d that the di\isi,,n of thila-t al.doininal Moment i- tinonly struetuial ill\ ditTei. ntiatinj.' llnnn-tiiinif. Cl.au- (Lc.) pointed out that there are many otin-r pointol ditT.ieii.-e. and that the suppo.-ed di\i-ion is not a struetuial •I' i l-ii haany t-tiu<-ture at all i . -em). M at leaf-like second anteiinsB imi • ttmia bear on the second antenna a knob, which is possibly the repre- ' '••in he ma, i une the charai BrancAtpiw Schaffer, depends

. nt 11 el \ on the acceptance ol s.-hmanke\\it-eirpriterion <.t thai genus, which is set i inhabiting uat.i- ,,f dilt, i, nt salinity and composition. It is not a little surprising that the animals living in No. XIV, for example, are scarcely distinguishable from those in No. XXIX, though the water in the latter was so strongly alkaline as to feel soapy. The conditions of animal life in these two waters must surely be very different, and yet no visible effect is produced. It is of course certain that there are great differences in the physiology of these forms, for, as I have often seen, animals (Copepoda, Cladocera, &c.) transferred from one water to another of materially different composition, die in a few minutes, though the second water may be inhabited by the same species; but in visible structure, the differences are for the most part trifling and equivocal.

'I'm: Men-tic Vai-iati..us of the vertebral column constitute a jed ..I some complexity. In considering them it must be remembered that numerical change may be brought about in the of vertebrae by \\\» different processes: first, by Variation in tintotal Qumber of segments composing the whole column, in which the variation i- truly Men-tic ; and second 1\ by Variation in tinnumber or ordinal position of tinvertebra* comprised in one or mopregions of tincolumn, not neces-arily involving change in th«- total number of segments forming the whole series, and in this the \ariation i- Homo-otic. Though Eomceotic Variation is often a-.-oriated \\ith change in the total number of segments, fi-oni tinnatmv of the case it is rarely possible in any given in-tanec to di>tinguish clearly whether >udi change has occurred or DOt. Thiari-es largely from the tad 1 hat while to find the

il numbet of vertebrae it is necessary to kn<>\\ the exact number of caudal \ei-tebr;i', in man\ specimens the>e arc incomplete, and •i if present their number cannot often be ^iven with confideiice. For then reasons the chief interest of this section of the fartarises in connexion with Honiu-otic Variation, and the modes in \\hich it OCCUTS; but it must be constantly borne in mind that in almost ;my gi\en case there may be Mcristic Variation also, though

True M eristic \'ariation, that is to >,-i\. change in the total number of .-• ur"n m- conii-o-inL; the \\hole column. ma\ neverthele-s be j.lainly iec,,UMii/ed in <-, itain animal-. Among -oine of the lower vertebrates, Fishes and Snakes, for example, the range of such Variation may be very great. Among Mammals the following may be given as an example of considerable Variation in the number of praesacral vertebras in a wild animal, and such evidence may be multiplied indefinitely.

6. Man. The simplest form of true Meristic Variation in the total number of vertebrae may occur in Man by the formation of an extra coccygeal vertebra, making five coccygeals in addition to five sacrals, i.e. ten pelvic vertebras in all. Instances of this are rare (SxRUTHERS), though in many tailed forms such Variation is common. Two cases, in both of which the sixth piece (1st coccygeal) was partially ankylosed to the sacrum, are fully described by STRUTHERS, J., Journ. Anat. Phys., 1875, pp. 93 — 96.

In the presence of cases like that last given, there is a strong suggestion that the number of vertebrae has been increased by simple addition of a new segment behind, after the fashion of a growing worm : the variation of vertebrae thus seems a simple thing. But there is evidence of other kinds which plainly shews this view of the matter to be quite inadequate. Some of these facts may now be offered, and in them we meet a class of fact which will again and again recur in other parts of the study of Repeated Parts.

*7. Python tigris1. This is a case of great importance as illustrating several phenomena of Meristic Division. In a skeleton of Python in the Mus. Coll. Surg., No. 602, the following peculiarities of structure are to be seen. Up to the 147th inclusive the vertebras are normal, each having a pair of transverse processes and a 1 This and the following cases of Pehnnis and C imoliasaurus are discussed by BAUE, G., Jour, of Morph., iv. 1891, p. 333.

pair of ribs. The appearance of the next vertebra is shewn in the ti_rui' < Kr4. K), I.). Anteriorly, and as far as the level of the I" '-!• -ri'»r surface of the transverse processes, it is normal, save that n- neural spine is rather small from before backwards. The transverse processes bear a pair of normal ribs. But behind this pair of transverse processes the parts, so to speak, begin again rising in into a neural spine, and growing outwards into a second pair of transverse processes, with a second pair of normal ribs. Posteriori \ a_';iin the parts are normal. This specimen is described in th«- Catalogue of 1858, as '14-Sth and 149th vertebrae anky- l..-.-d," but upon a little reflexion it will be seen that this account mi sseutial point. For the bone is not two vertebra?

-imply j'-iii'-d together as bones may be after inflammation or the \ l>ut it is two vertebra.- whose adjacent parts are not formed, and between which the process of Division has been imperfect. With more reason it may be spoken of as one vertebra partly divided into two, but this description also scarcely recognizes the real nature of the phenomenon. Further on, in the same specimen, at the 166th vertebra, there is an even more interesting variation. This vertebra is represented in Fig. 10, II. As there seen, it is normal on the left side, bearing one transverse process and one rib, while on the right side there are two complete transverse processes and two ribs. The 185th vertebra is also in exactly the same condition, being double on the right side and single on the left.

8. Python sebae : a precisely similar case (Brussels Museum, No. 87, I. (Jr.), in which the 195th vertebra is single on the right side and double with two ribs on the left, is described by ALBREGHT, P., Bull MILS. Nat. Hist. Belg., 1883, n. p. 21, Plate II. 9. Python sp. : a precisely similar case of duplicity in the 168th vertebra, on the left side, in a mounted skeleton in the Canib. Univ. Mus. It is to be especially noticed that in each of these four cases of lateral duplicity, the degree to which the process of reduplication has gone on is the same.

10. Pelamis bicolor [ = Hydropkis~\. The 212th vertebra simple on the left side, and double on the right. It bears one rib on the left side and two ribs on the right side. Yale Univ. Mus., No. 763. BAUR, G., Jour, of Morph., iv. 1891, p. 333. 11. Cimoliasaurus plicatus (a Plesiosaur). " Centrum of a small and malformed cervical vertebra from the Oxford Clay near Oxford. This specimen is immature, and on one side is divided into two portions, each with its distinct costal facet." LYDEKKER, R., Cat. Fossil Rept. and Amph. in Brit. Mus., Pt. II. 1889, p. 238, No.

A case somewhat similar to the above is recorded in the Rabbit by BLAND SUTTON, Trans. Path. Soc., XLI., 1890, p. 341. See also certain cases of a somewhat comparable variation in Man, considered in connexion with the variations of Bilateral 12. Man. Partial division of ribs is more common than that of vertebra?. Five cases are given by STRUTHERS. 1. Fourth rib becoming broadband bifurcated in front. Male, aged 93. From about middle of shaft these ribs gradually increase in length from 7 lines to li inch on the left side, 1^ on right. They then fork, the left l;y inch, the right 1 inch from where they join their cartilages. Cartilage of right forks close to rib, enclosing a space which admits little finger ; cartilage of left lost, but the diverging bony divisions, each of good breadth for a rib (6 to 7 lines) enclose an intercostal space 1^ inch long, attaining a breadth of f inch, which was probably continued forwards by the division of the

cartilage "i- by two cartilages. The cartilage of the left 7th rib is als.. double I'm- 1^ inch, all the others are normal. 2. LI -ft fourth rili becoming very broad and bifurcating in front; two large spaces, on.- in t IK- b.iiie, one at the bifurcation, o. Left fourth rib becoming broad t.i \\ards sternal end, where it joins bifurcated cartilage. In til-— three cases the division affected the 4th rib. Three others an in \\hich the rib affected was probably the 4th or oth.

I -idt •- these cases of obviously M-ri-tic Variation, there are maiiN \\hidi are r.uiibined with Homoeosis so as to produce far anatomical divergence. Though in some of these examples tin-re ma\ be change in the total number of vertebrashewing thai true Meristic change has occurred, they cannot well be treated apart h»m tin- m-. re di-tinctly Homoeotic cases. an.! tinfollowing account in the tir-t instance relates chiefly to them. r.etoie considering the details of such \ariations in vertebra-1,

it may be useful to d, -senlie briefly the ordinary system of oedat'lle \\hich is here followed. I 1 1 ! 1'ea I i I Ig t ll JS M ll ij ect it IS 1111- -ible to employ a terminology \\hich docs not seem to imply • ptance of the \ie\\ that 1 1 1 e iv is a true homology between the individual \eiiel, ra of t\\o >pin.-> containing different total numberfor all the nomenclature of Comparative Anatomy is devised on thi> hypothesis. This ditHculty is especially felt in regard to vertebrae, and at this point 11 should l,e expressly stated that ill using the ordinary terms Mo Midi assent is intended. This matter has already bee,, lefenvd to m Section \'l. of the Introduction, and will lie dis, 'iis.,., | in relation to the facts to be given.

i'-"ii in th" rilj given to it in human anatomy, to mean those vertebra The characters thus defined andistributed among the several vcit.-braacc,,rilin- to their ordinal positions Among mammals th.- number of vertebra \\hi.-h develop the characters of each region, though differing widely in different classificatory divisions, are as a rule maintained with some constancy within the limits of those divisions, which may be species, genera or larger groups, so that vertebral formulae are often of diagnostic importance. Changes in the numbers of vertebrae composing the several regions must therefore have been an important factor in the evolution of the different forms.

Homoeotic Variation in the spinal column consists in the assumption by one or more vertebra? of a structure which in the type is proper to vertebras in a different ordinal position in the series. Examples of this are seen in the case of the development of ribs on a vertebra which by its ordinal position should be lumbar ; or in the occurrence of a vertebra, normally lumbar, in the likeness of a sacral vertebra, having its transverse processes modified to support the pelvic girdle, &c. Variations of this kind have one character in common, which though at first sight obvious, will help us in interpreting certain other cases of Homoeosis. In all cases of development of a vertebra normally belonging to one region, in the likeness of a vertebra of another region, this change always takes place in vertebrae adjacent to the region whose form is assumed. For example, if one vertebra, normally cervical, bears ribs, it is always the last cervical ; if two cervicals bear ribs, they are the last two, and so on. No gaps are left.

Homoeotic Variation in the spinal column may occur by the assumption of (1) dorsal characters by a vertebra in the ordinal position of a cervical, (2) lumbar characters by a vertebra in the ordinal position of a dorsal, (3) sacral characters by a vertebra in the ordinal position of a lumbar, (4) coccygeal characters by a vertebra in the ordinal position of a sacral, or by the reverse of any of these. Since almost any of these changes may occur either alone or in conjunction with any of the others, it is not possible to group cases of such Homoeosis under these heads, but the consideration of the more complex cases will be made easier if simple examples of each class are first described as seen in Man.

The chief character distinguishing dorsal vertebrae is the possession of moveable ribs. This chai'acter may to a greater or less extent be assumed by cervicals. 13. i -es of the development of ribs on the 6th cervical seem i.. be • Ktremely rare. One is given by STRUTHERS in a young f. Theribs were present as rudiments only, being the same on both sides in the Gth vertebra, and on the left side in the 7th. Each of these rib-elements was ^ inch long. In the lith the libs rested on the body of the vertebra, but in the 7th the nt. .lid not reach so far. Full details, q. v., STRUTHERS, /. A nut. Phyt ^7:i. p. 32.

1 i \i--al ribs on the 7th vertebra are comparatively common, 14 -"in. 'times moveable and sometimes fixed. The literature of of -H. -h ribs, occurring in 45 bodies, being all that were known t«. him in literature or seen by himself. In addition to these 12 tinresults of an analysis of these cases are important to the study \ • »rding t" the decree of completeness with which the cervical nlandeveloped, (•'inr.i.it divided them into four classes l.

I. Lowest development Cervical rib not reaching beyond the tran-\er-e pi< ..... -re-ponding to the vertebral end of a true rili with r"/'ttii/ii//t and tubercidum, Bud articulating by both of them. //<//•' f'n-in. '1. High* development. ( 'ervieal rib reaching beyond the tra sa f..r a greater i.r le— extent, either ending freely ami joining the cartilage of the first true rib either l>y its cartila- QOUfl • ml "i- by a ligament continued from this, linrcxt fnrm.

k Complete development. Cei-vical ril» re>cnil)ling a true rib, ha\in^ a carl generally for a greater or less part of its length united with the cartilage of the first true rib) connecting it with the .-termini. /., M rare form. Qruber states, as the result of an analysis of 47 eases, that the third of ti - i- \rry rare, that the s.-c.md condition is the '•••nimon on,', and that the fourth or complete condition is commoner than the tir-t or lea-t >tate of dc\ .•lojniient , which is also rare. of Stnitli- i^ caa - the majority seem to lielong bo Gmber's second da--, \\hile that on the |.-|'t side in Struthei- Case 4 must have appioarhe,) ('la— Land that on the left side in Case 10 belonged

Two ti-atuivs in this evidence are ot 'e-jiecial consequence: first that the variation is more common on both sides than on one side ; secondly, that it is not in its lowest development that it is most frequent, but rather in a condition of moderate completeness, having the proper parts of a true rib. 14. Reduction of ribs in the first dorsal is described by Struthers in a specimen in the Path. Mus. of Vienna. " The whole of the cervical vertebra being present 1 there is no doubt as to the case being one of imperfect first rib. On left side rib goes about f round, and articulates with a process of the second rib. On right side it joins second rib at from |- to 1 inch beyond tubercle, but again projects as a curved process where the subclavian artery has passed over it. The maimbrium sterni first receives a broad cartilage, as if from one rib only, and secondly a cartilage at the junction of the mauubrium and body which is the cartilage of the third thoracic rib." STRUTHERS,/. Atiat. Phys., 1875, p. 47, Note. (See also Nos. 24 and 25.)

15. (a) From dorsal towards lumbar type. The characters chiefly distinguishing dorsal vertebras from lumbars are the presence of ribs attached to the former, and of long, flat transverse processes in the latter. Secondly, the articular processes of lumbar vertebrae generally differ from those of most of the dorsal series, each pair of articular surfaces facing inwards and outwards respectively instead of upwards and downwards as they do in the dorsal region. The transition from the one type of process to the other, in passing down the column, is generally an abrupt and not a gradual one. In Man it occurs between the 12th dorsal and 1st lumbar, but in most Mammals it takes place more or less in front of the last dorsal, leaving several dorsal vertebras with articular processes of the lumbar type. (STRUTHERS, 1. c., p. 59.)