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

*293. H. nipalensis £ : only four incisoi^s in lower jaws. This is a remarkably clean and sound skull. The four incisors stand close together, filling up the whole space between the two lower canines. There is no reasonable doubt that only four lower incisors have been present. It is difficult to see that any of the four incisors exactly corresponds with any of the normal teeth; for while the two lateral teeth are of about the same size as normal J-, they have a different position, arising from the outer sides of the jaw, slightly in front of the roots of the canines, whereas normal 7* arises internal to the other incisors. To what extent the alteration in position is correlated with the change in number cannot be affirmed. B. M. 146, m.

*294. **• persicus : only four incisors in lower jaw. Judging from general appearances it seemed that 7"' was missing from both sides. The teeth stand in a close series between the canines, which are nearer together than in normal specimens. The consequence of this to the arrangement of the bite is curious. The left lower canine bites in its normal place, between the upper canine and &; but the right lower canine bites in front of the upper P, which is displaced backwards

towards ili.- ri-ht np ]i.-r canine. The whole .anterior part of the lower jaw is thus twisted ;i little towards the left side. these two ili -finite cases of absence of incisors, in the following instances the, , |,n -imii'tiun thiit the absence was due to variation, but a definite state- In tingreat majority of both Asiatic and African species of Herbhe ant. -i in) premolar (//) is normally present in both jaws, and in these species ''• <-ases of absent y were seen. When present it i> a tooth of small but still considerable size. It appeared from tinspecimens that / in the species //. yracilis (Africa generally), and IK. tli i •' and y,! in //. ijalura (E. Africa) are commonly absent. As in other cases ft' ali.M-nce nf ti-i-th tinijuestion arises whether the absence is due t<> age or accident, or on the other hand to original deticiencv. This question cannot be definitely answered, but some considerations touching it should be mentioned.

First, as has been said, the tnoth when present is of moderate size: though small, it is ijuite large enough to be functional, and is in no e rudimentary. In his synopsis of the genera, THOMAS* says of Hi I'IH *tes, " Premolars [ (if only 3 in either jaw, a diastema always present)." There is however no reason for supposing that the presence or absence dt' //' is d. termined by chance. From the fact that a tooth is small, it bv no means follows that it is often lost. To any one handling large numbers of skulls, instances of the contrary must lie familiar. A case in the Otters well illustrates this point. In Lutra riili/u t-i* upper />' i- a small toot h. and from its singular position internal to the canine, it might be supposed that the development of the canine might easily push it out; yet in 41 skulls of Lutra vulgaris, only I case of absence of p_' was seen. Of L. cinerea on the contraiy six skulls are without 7^ ; but as in two young skulls it is present on both side-, thenis thus a strong presumption that in this species the tooth is lost with maturity. The frequent absence in the one species and the constant presence in the others points to a difference in organi/ation li.-i \\een them. When //' is missing in a skull, though we annot entitled to infer that it has not been present, still the fact of its pr.-M-iH-e in one case and of its absence in another is on the face of it an indication that liet ween the two there is a difference or Variation, but whether tin- Variation lay in the number of teeth originally formed or in the mode in which they were affected by subsequent growl h i- im.-.-rtain. In the specimens to be described the absence of ]>l in certain individuals or species is no less detinite than its presence in the others, and that which i^ a variation in one species will be seen to be the rule iii others.

A regards the presence of y/' the specimens thus make a progress! ve series. .M,,st species ha\ing /.}, but //.'. as ,-i variation; //. i/rai-ifi'n (and also as a variation; and lastly //. yalera having p| normally but shewing a case of p -= and another of p - ^ . Lastly, all specimens of o — 4 4 — o Of species commonly having p\, 91 such skulls and the following cases of absence of p1 were seen: 290. H. gracilis on the contrary shewed p% in 8 specimens, pi present both sides once, B. M., 789, a ; left Jt absent once. B. M., 789, b. H. pulverulentus : p$ in 2 specimens.

297. H. galera : p§ in 7 skulls, one being quite young: jj1 is present in all four places, in one young skull making p\, B. M., 148, d; pi and p^_ both present and well developed on right side in old skull. On the left there is ample room for them. 15. M., 79, a, P\ present on both sides and alveolus for pi on right. B. M., 148, I. 298. H. microcephalus : on right side two teeth like p^, crowded together, others normal. Leyd. M. Compare Rhinogale melleri (an

African Mungoose) of which only known skull (in B. M.) hasp 7— -.. The appearance here is that a tooth unlike and rather larger than p^ stands in front of it on each side (see THOMAS, /.c., pp. 62 and 84). Taken together the two following cases are important as illustrating the difficulty of drawing any sharp distinction between cases of duplicates of particular teeth and cases of extra teeth in series. They should be read in connexion with the cases of F. domestica (No. 286), Helictis orientalis (No. 312), Vison horsfieldii (No. 311), Ommatophoca rossii (No. 320), Phoca grcenlandica (No. 324), &c.

*299. Herpestes gracilis : supernumerary tooth in right lower jaw (Fig. 50). On comparing the teeth of this specimen with those of FIG. 50. Right lower jaw of Herpextes gracilis, No. 299. View from labial side ; ground-plan of the jaw ; separate view of the tooth + . (7, the canine. other Herpestes in which p is present it is quite certain that no tooth in the abnormal jaw corresponds with p^. The foremost of its premolars on both sides clearly has the form of ^. The next teeth have the correct form of p5. In the left lower jaw the next tooth is^; but

on tin- i-iu'lit -idiimmediately in succession to ^ but slightly within the an-adc i-, ;iniither tooth (marked + in the figure), which is very nearly a • •"]>>• ofy,>. tin. ugh a little smaller. On the outside of the jaw and lii-liiiid thi.s to.ith is a normal ;A From its singular position outside tli.- series, this tooth might easily be taken for a supernumerary one though it> f.ii in clearly shews it to be a natural p* displaced, while two t.-t-fh having the form of j7s stand in succession. B. M., 63. 7. 7. 18. (in. -mi,. ned by THOMAS, P. Z. S., 1882, p. 62).

H. ichneumon (Andalusia): in one of the upper jaws between :in<l internal to }^ and /j_3 is a 3-rooted tooth (not a milk-tooth) which in >i/f and shape is about intermediate between r and p\ Leyd. M. Tin.- only cases of noticeable variation in molars were both in t In- same species, Crossarclnis zebra. Of this species six skulls were n, four normal, and also the two following, the first being a case of extra molar on each side, the next a case of increase in size and cninpN-x variation in m?, on the left side suggesting a partial division of this tool li.

'301. Crossarchus zebra: small but well-formed additional molar in uppi-r ja\\ .,11 each side, making pf, ??i|. (Fig. 51, III.) Teeth unfortunately all much worn, so that it is not possible to determine whrtluT any of the molars differ from their normal forms in correlation with the existence of these extra teeth ; but as far as size is concerned, thenwas no sign of such change, ^ and m?_ being of th.- ii-ual >j/e. B. M., 73. 2. 24. 18 mentioned by THOMAS, P. Z. X. l.ssi>, pp. 01 and 89.

II. A Hoi in;il sjurinicn, right upper jaw. III. Upper jaw of No. 301. 302. C. zebra: all teeth normal except second molars in the upper jaws on each side, which depart from the normal in the following manner. Right >^ has a small extra cusp (Fig. 51, I.) on its outer side, making four instead of three as usual (cp. figure of normal, Fig. 51, II.). The left ^ is very extraordinary. It is rather less than twice the size of its fellow of the other side (Fig. 51, 1.). The crown is of an irregularly elliptical form, the long axis being oblique. The posterior and anterior faces are marked by a shallow groove, giving an appearance of imperfect division into two teeth. The total number of cusps is greater than twice that borne by the other, but from the irregularity of the surface it is not possible to speak more precisely. For fear of injury the tooth was not extracted, so that the number of roots cannot be specified. B. M., 82. 5. 26. 1.

The evidence of dental Variation in this family is at present too small in amount to be of much value. It is chiefly interesting in so far as it relates to cases of the occurrence in one genus or sub-family, of a formula characteristic of another. Variations of this class, consisting in the presence of or absence of the anterior premolar or last molar, are in some of the forms very common. As will be suggested in the next chapter, some of these, for example, the variations in p1 in the Badger, have a certain importance as giving some measure of the magnitude which a tooth may have when the species is, as it were, oscillating between the possession and loss of the tooth in question.

Amongst Mustelidse there were two cases of supernumerary large premolars, probably reduplicatory. Ittustela (Martens), normally p|, m&. Seen in adult skulls of various species (M. 2Min<mti, martes,foina, zibettina, flavigula, americana), 62: also the following: Putorius (Weasels. Stoats, Ferrets and Polecats), normally p%, mL Seen in adult skulls of various species (P. vulfjuria, erminea, brasiliensis = xanthotjenys, fcetidus = eversmanni = sarmaticus, lutreola, nudipes, &c.), 105: also the following: 3Q5 " P. erminea: 1. ^ absent, B. M., 43. 5. 27. 11. On the other hand, P. foetidus, ' B. M., 192 *, has rt. y as a two-rooted tooth, standing in a plane at right angles to the arcade.

1 Totals of normal skulls refer to Brit. Mus. and Cambridge If"*, only. In 1'* skulls HKXSEL found the following variations in premolars, the molars being always m\- Taken tn^rihntherefore there were 12 normals with p§, 16 cases of or less reduction, and 2 cases of increase. HEXSEL', Saugetldere 308. ConepatuB: /''.m|. 1'J specimens. Con<-i><Hn* is the S. American representative .it' M,-j>liitiK, iiinl normally differs from it in having one premolar less in upper jaw. This tnotli is sometimes present as a minute tooth making p§. Sometimes on tlir i-ontrary there is a premolar less in the lower jaw, giving jjg. COUES, i'ur-lx.irinii .liiiinnl.-i i if .V. .-I/id-;-., p. 1'.I2 and X»tr.

In addition to the 12 normals mentioned two cases of />§ were seen, viz. C. mapurito, 11. M., KM. 11. •_':,. s. and C. chilensis. 1'.. M. S29, a. In the former anterior premolar is of ^oml si/.e, hut in the latter it is very rudimentary. th.-r t-asc mentioned by BAIBD, Miunm. uf X. Am/'r., p. *309. Meles : .•..iiiiimnly /yj, ///.',. In M. taxus, the common Badger, // is t'lvijiii'iit ly ahsi'iit from one or more places. Of 36 skulls only 16 lui'l /-' in all jaws, 7 have it in each lower jaw and '1 had no such tooth in .-it her jaw. In remaining cases it was sometimes absent on right, sometimes on left, sometimes from above and sometimes from below. Some of these ca:-.-- iu.iv I >e due to senile changes but this was certainly not BO in all. Ali-em-e from lower jaw seems the most common. HK\SI.I.. Mm-fitt. .lulu-It., 1S7«), y. p. :>.-)d.

< if yen us .!/. /- x the following were seen by myself. + means presence, — absence of //. Tinnumbers ^iveii l>y Heiisel are the totals of ji + m, but he states that the viii i.'itiuii always coin-eiii. .1 the small anterior pu-molars next the canines. 310. Lutra. The Otters for the most part have p±, m|. The anterior premolar of the upper jaw is a small tooth standing internal to the canine, but in the common Otter its presence is most constant. In the Oriental L. cinerea, and the Neo-tropical L.felina on the contrary this tooth appears to be more frequently absent than present. The following table gives the results of examination of a series of skulls.

In L. cinerea ( = leptonyx) the absence of p^ is associated with a more forward position of p^, of which the anterior border is then level with the posterior border of the canines1. Putorius (labelled " Vison Horsfieldii"): at the place in which tin- rL'ht lower posterior premolar ("jT* ") should stand there are two teeth at the same level. They are almost identical, but the inner FIG. 52. Putorius, No. 311, right lower jaw, ground-plan of teeth and profile vit v. - ,,( two teeth at the same level. Upper figure is the internal tooth.

Helictis orientalis, Java : having supernumerary two-rooted toe. Hi internal to and between P2 and 7^. This extra tooth is almost a copy of /<; (Fig. 53). B. M., 824, a. I i>.. '>'••>. Ili-Hrt i* urii'iitiilis, No. 312. Surface view of upper jaw and a representation of the ri^ht upper teeth as seen from inside. 313. Putorius: HF.NSEF,, M, „•]>!,. Jahrb., \. 1S7!», p. 540, states that he h.-is several skulls of Fcetorin* i>nt<>rius with an extra upper molar on one ^iiic iii ;i rudimentary condition. GIEBEL, Bronn's Kl. ». Onl., p. l^'i, T.-if. xv. figs. 1, 2 and 3, figures a spcriiueu of " Piiforiitu fi/jms" having . -i fairly well developed extra upper molar on each side making ///.: instead of m^. Pn.hahly both these accounts refer to P. foetidus.

316. Meles taxus has normally in\. Skull from Quarternary diluvium of Westeregeln has small alveolus behind right "^ and left m-. Another fossil skull has »i|. NEHRING, Arch. f. Anthrop. x. p. 20. [1 Small alveolus behind left ^ in B. M., 211, //.] 317. Lutra: case of absence of jwi ; Blustela : ^2 may be absent. HENSEL, /. c. With reference to dental Variation in Otariidse and Phocidee there is a considerable quantity of evidence. In some of the species the frequency of abnormalities is remarkably great. Among the most interesting examples are two cases of reduction in the number of incisors, both occurring in Phoca barbata. These cases are especially important in connexion with the fact that the Seals are exceptional among Carnivores in having a number of incisors other than |, and that among the different sub-families of Seals there is diversity in this respect.

Taken together, the cases of Variation in the premolars and molars of Seals illustrate nearly all the principles observed in the numerical Variation of teeth. In both premolars and molars there are examples of the replacement of one tooth by two, and in some of these the resulting teeth stand in series while in others they do not. Besides these there are numerous instances of extra premolars and molars belonging to various categories. As regards the frequency of extra teeth in Seals it may be mentioned that of Phocidse 139 normals were seen, and 11 cases of supernumerary teeth ; of Otariidse 121 normals and 5 cases of supernumerary teeth.

From the simplicity of the normal dentition and from the diversity of the variations presented, the evidence as to the teeth of Seals may conveniently be studied by those who are interested in the phenomena of Variation without special knowledge of the subject of mammalian dentition. It will be remembered that of Phocidse the sub-family Phocinse (like Otariidse) has normally t|, while the Monachinse have i\ and the Cystophorinse i |. Of Phocinse of various genera and species 105 skulls having if were seen, and in addition the two following.

*318. Phoca barbata. Greenland: skull having i^- on both sides (Fig. 54). This skull is a particularly good one and is neither very old nor very young. The teeth stand regularly together and there is no lacuna between them. There is no reasonable doubt that an incisor is absent from each side of each jaw. The shape of the premaxillais different from that seen in other specimens of Phoca, and, doubtless in correlation with the absence of the two upper

incisors, the width of the premaxilla? is considerably less than in -p. cimens having the normal dentition. B. M., 90. 8. 1. 6. 310. P- barbata: in left upper jaw are three normal incisors; but on the right sidethe incisors have been lost. The alveoli, however, shew plainly that only two incisors had been present. Of these tinouter one in size agrees with &, being a large alveolus equal to that "| , df the other side, but the second alveolus, occupying the place of j7"1 and /-, is also a large alveolus, scarcely smaller than that for P. It appears therefore that in this specimen a single large

t oot h st 1 in place of il and i~. A lower jaw placed with this skull was normal, but it was not certain that it belonged to the skull. Normal urninf/einent. In Phocid;e there are normally rive teeth behind the canines in each jaw, and according to the received accounts, of these teeth 4 are preinolars and one is a molar, giving /> ;}, in j, Tin1 Otariida1 on the other hand have generally p^, m^, but Imtli the two upper molars stand at a level behind that of the lower molar, so that the posterior molar, ^ is placed so far back that it meets no tooth in the lower jaw. Some of the ( Mariid;e. however, as 0. cul (/'<>riii<nt<i, do not possess such a posterior tooth, and have only m\. 0. stelleri is peculiar in the fact that it also has only one upper molar, but this tooth is separated by a large diasteina from y1, and stands in the position characteristic '"' of the other ( Hariida'. Hence it may be supposed that 'w1 i- really aliment while m* is present.

Amongst the eases will be found some of the presence in Phoeid;e, especially Halichoerus, of an extra molar placed in the usual position ,,(' 1,1^ in the < Mariidii'. But lest any one should think it manifest that this is an example of Reversion to the Otarian condition, at tent ion is called to cases of such an extra molar in the Otariida1 also. Similarly there are instances of absent molar in those OtariidsB which have /// j, leaving m ] : and of these cases one occurs in such a way as to leave the peculiar diasteina between

£* and the molar, referred to above as characteristic of 0. stelleri (see No. 342). The cases are grouped in an arbitrary collocation, according as it seemed desirable that particular variations should be studied together. In the sections dealing with premolars, Phocida3 are not separated from Otariidse. *320. Ommatophoca rossii, an Antarctic Seal. Of this form only two skulls are known, both in the British Museum. One of these (B. M., 324, 6.) has the arrangement usually found in PhocidaB, namely, five teeth behind the canines in each jaw, giving the formula

(on the analogy of other Seals p%, m}}. The other specimen is exceedingly remarkable (Fig. 55). In it the incisors and canines are as in the first specimen, but the first tooth behind the canines on both sides in the lower jaw and on the right side in the upper jaw, has a very peculiar form, having a deep groove passing over the whole length of the tooth, on its outer and inner sides. These grooves extend from the tip of the root along both sides of the crown, and thus imperfectly divide each tooth into an anterior and

a posterior halt'. The cusp of each tooth is also divided by the grooves -.» as to form two small cusps. Each of these teeth is theref-re an imperfectly double structure, and maybe described as bein:_r ju-t halfway between a single tooth and two teeth. These are shewn in Fig. 56. l'ii,. ."•(>. (iiniiHitojihnca ros.iii. No. 320. The anterior premolars of upper and ]n\viT jaws from the side. (The left lower and right upper teeth were not On the left side in the upper jaw, as the vis-a-vis to one of these double teeth, there are actually two complete teeth, of very -imilar but not identical form, as shewn in Fig. 56. Each stands in a distinct alveolus, the two being separated by a bridge of bone. The denial formula of this skull, taken as it stands, is therefore

:\-l\. Cystophora cristata : internal to and slightly in front of// on each side in the upper jaw> is an extra tooth. These extra teeth are alike in t'i>nii but are rather smaller than P^. C. M., 895. ;;o-> Cystophora cristata (label, Phoca cristata): internal to ri;//tt upper ,"! is an alveolus for a small one-rooted tooth. In the corresponding situation in the left /<>/>•,,- jaw there is such an extra tooth in Zalophus lobatus ( = 0tdi-i>i lultot,,)-. left ^ smaller than right 7*1, and between the canine and the left i^_ there is a supernumerary tooth, smaller than left /<'. (The same skull has another extra tooth outside and between jr' and j"4, see below No. 333.) Leyd. ^f.