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

species 90 adult skulls seen having this number of incisors and three cases of abnormal number of incisors. Of these the first two must not be reckoned in estimating the percentage of abnormalities in a promiscuous sample, for Mr Thomas, who kindly shewed me these specimens, informs me that they were preserved and brought to the Museum expressly as abnormalities. The existence of these variations is nevertheless particularly interesting in connexion with the exceptional number of incisors normal in DidKlpJtys.

363. Didelphys marsupialis : in right upper jaw six incisors ; left upper jaw and the whole lower jaw missing. B. M., 92. 11. 3. 28. 36-t. Another specimen has on the right side |- as usual, but on the left i -|. It appears that 71 and 7- of the two sides correspond, but on the left side three very similar teeth stand in series behind p. B. M., 92. 365. D. turner! (= crassicaudata), Demerara. A single specimen of DASYURID.E : incisors normally ~; of genera other than ius, 63 normal skulls seen.

Dasyurus sp., having only two incisors in left lower jaw; right lower jaw normal, upper jaws missing [doubtful case]. B. M., 250. II. Right and left profiles of the two jaws of No. 367. (Premaxillary teeth alone shewn.) 5 skulls without Imver jaws, and 1 lower jaw without skull; abnormal- '2. as follows : *3GG. A ynung skull having in the upper jaw on the left side i Fu. M! l.i two teeth, l)«-th apparently in place of left j^, making *3(i7. A specimen having four incisors in the right lower jaw, the 1. ft IH innormal. IVrhaps the two hindmost of the four repre- - nt thi' thud lower incisor of the left side in the way suggested In the dotted lines iii the figure (Fig. 62, II.). B. M., 314, b.

PIIAI.AM.I I.ID.K : incisors (neglecting "intermediate" teeth of lower jaw) normally £; this seen in 209 skulls of various genera and spec 3(J8. Fhalanger orlentalis, Solomon Islands : left f as an imperfectly double tooth, Laving two sub-cylindrical crowns and only one root (Fig. 63). The two crowns incisors and canines. The separate figure shews the left i3 extracted. stand in the same transverse plane, the one being internal to the other and rather smaller than it. Lower jaw missing. 13. M., 1936, c. [Two other skulls from same locality normal.]

369. ¥• maculatus, Port Moresby: only tico incisors on each side in the upper jaw. The centrals, ^, of each side, are in place; externally to them there is on each side an alveolus for a tooth, which, judging from the size of the alveolus, was probably F- Immediately behind these alveoli the canines follow on each side. In this case it uiav In- s.iM that the missing teeth are ^ in all probability. Lower jaw normal. B. M., T'.i. :',. :,. H. (2) Premolars, and the " fnteriin-il/ute" teeth (in the lower jaw) of

The evidence here offered relates to the following genera:— J'/Hi/niiifcr. Triclinxiiriis, J'xcui/ocliirus, Petauroides, Dactylopsila and rcfinn-ii*. I'.efmv sjieaking of the variations seen, a few words are needed in explanation of the nomenclature adopted. In these forms there is only one tooth having a milk-predecessor, and in all the genera here referred to this is a distinct and recognizable tooth, with a chisel-shaped crown. Following Thomas' system I shall call this tooth p* throughout. This name is used as being well understood and convenient, but without any intention of subscribing to the principles of homology upon which the system of nomenclature is based.

In Thomas' paper1 a careful and well-considered attempt was made to bring these anterior teeth into a formal scheme of homologies, and though the application of this method to the teeth of the lower jaw was avowedly tentative, yet at first sight the results in the case of the upper teeth were fairly satisfactory. Nevertheless it appears to me that in view of the facts of Variation about to be related, the system elaborated by Thomas breaks down ; not because there is any other system which can claim to supersede it, but because the phenomena are not capable of this kind of treatment. To anyone who will carefully study the examples given in the following pages, especially those relating to the genus Phalanger, it will, I think, become evident that it is not possible to apply any scheme based on the conception that each tooth has an individual Homology which is consistently respected in Variation.

The evidence concerns first the premolars of the upper jaw, and secondly the lower " intermediate " teeth. Inasmuch as in several of the cases there was Variation in both these groups of teeth, the evidence relating to them cannot well be separated. As regards the upper teeth, all the cases of importance occurred in Phalanger and Trichosurus, and owing to the similarity between the dentitions of these two genera it is not difficult to employ terms which shall be distinctive, though the question of the homologies of the teeth go unanswered. In all the forms concerned there are three upper incisors, and the tooth immediately succeeding them will be called the canine, though its position and form differ greatly in the various genera ; for while in Phalanger and Trichosurus it is a large caiiiiiiform tooth placed on the suture between premaxilla and maxilla, in Pseudochirus, for instance, it is proportionally smaller and stands in the maxilla at some distance behind the suture.

Upper jaw. As already stated, the large preuiolar having a milkpredecessor will be called p*. In Trichosurus between the canine and p4 there is usually one large tooth, in shape and size much like the canine : this tooth will be called F1 as Thomas proposed. Though when present it is large, it is not rarely absent altogether (v. infra}. In Phalanger there is a similar p^, though of somewhat smaller size ; but besides 1^_ there is usually another premolar, a small tooth, placed between p^_ and P4. On Thomas' system this is F! and for purposes of description the name will be used here. In the left upper jaw of the skull shewn in Fig. 65 C, /, F1 and £» are

shewn in the ordinary state. Lastly, in Pseudochirus behind the canine th.-i'- i- .i v.-i y small tooth, presumably jj\ and between it and ;/ a tooth of i;i md size, presumably yA Low i- jaw. la the front of the lower jaw there is on each side one long inci-or. Between it and the tooth corresponding to y/ of the upper jaw there are several small or "intermediate" teeth, whose number varies greatly throughout the group. Thomas has made a provisional attempt to find homologies for these small teeth, but in view of the facts of their Variation it seems impossible to attribute indi\ idiiality to them and they will therefore be here merely numbered from before backwards.

Phalanger orientalis. In this species evidence will be nt'l'i -I'd to prove the following facts:— ( 1 ) That between p^_ and p^ there may be two small teeth, one or both of which may perhaps represent /?:i (Fig. 65). (ii) That between p*_ and the small p'A there may be a Ian;.1 tooth (Fig. G-i, C), like the p* of Pseudochirus. (4) That in case of absence of p3 , Pl may be near to X (Fig. 64, A). ( ."> > That between the canine and p^ there may be on one side the usual large pl , but on the other two teeth, evenly spaced, each of about the proportions of X (Fig. 64, B^.

(6) That in the lower jaw the number of intermediate teeth may vary from none to rive, three being the most usual number. *372. Specimen having left side normal, one small premolar standing be- t ween 7^' and F1. In the right upper jaw p^ is normal and stands at the same level as left ^ ; p^ is also normal in size, form and position (Fig. i',J Ci. In front of /'4 however there is a two-rooted tooth (marked y in the figure) having somewhat the same shape as p4, but about yji'ds the si/.e. This tooth has not the form of the milk-predecessor of p^. \ small peg-like tooth (.u in the figure) matching the small premolar ("P3") is also present, but is crowded out of the arcade and stands internal to the tooth //. The lower jaw has three intermediate teeth on each side. B. M., 1780, f. The form and position of the tooth // suggest a comparison with the arrangement in Pseudocltirus, in which "/'•: " is in a very similar condition. In Fig. 64, D, a profile of Pscn<l<>- chirnx is shewn, the dotted lines indicating the comparison suggested. It will thus be seen that if the tooth y corresponds to y/ of Pseudo- f/iiriix, the tooth .<• then has no correspondent.

*37.S. Specimen (var. breviceps, Solomon Islands) having in right upper jaw pl and p* but no "j>3": in left upper jaw pl stands at a level anterior to that of right pl, and a small peg-shaped tooth, "/>3," is present close to and almost touching p4. (Fig. ('•4, A) Lower jaw, right side, two intermediate teeth, of which the posterior stands internal to ~p* ; right side three intermediate *374. Specimen (var. breviceps $ , Duke of York I.) having in rt. upper jaw p1 and p4, but no "/)3": in left upper jaw there are two teeth of the size and shape of pl (Fig. 04, B), one of them

B. P. orientalis, No. 374 : no right "p3"; on left, two teeth both like pl, in symmetry approximately balancing right p1. Below are the right and left profiles of the upper jaws of this skull. C. P. orientalis, No. 372. The left side normal, lettered on THOMAS'S system. Eight side described in text. Below is a profile of right side. D. Pseudochinis, profile of normal upper teeth from right side enlarged to compare with C. Teeth lettered on THOMAS'S system.

being at a level anterior to right pl and the other posterior to it (see figures). On neither side is there any tooth having the *.'>7.~). SjM'cimi-n li.-iving two small premolars on each side between p^_ and £. Tintwo teeth are very small and sharply pointed (Fig. 65). In II. and III. Kight and left lower jaws as far as p4. IV. The same in surface view. the lower jaw there are on the right side five intermediate teeth betwi-i-n the inri-or and //<. and on the left side four such teeth (instead

i >f' the occurrence of small Premolars n ml Intermediate" f<'<-t/i in Phalanger orientalis Phalanger maenlatus. St;iti-tic< as to tlie absence of tin- Mnall ";>:t, " and as to the number of the "intermediate" ti«-th. may conveniently lie ^'ivi'ii to^'i'tln'1- in tabular ft inn. The species has u wide diMril>ution and is by TIHOIAS divided into a larger var. typiciix, and u smaller eastern var. lu'< r/i'c/ix. In the latter the small ]>'•' is usually absent. Tlie I.evdi n specimen^ an' not thus divided by JKNTIXK, and in order to include tinstatistics Kiveii la him (I. <-.) the distinction into two races is riot followed in the table.

When present, "//:l" generally stands at an even distance from yi1 and y<4. as in the left side of Pig. M. (', and n«t as in the left side of Fig. 64, A. The positions of the intermediate teeth are most various, sometimes they are evenly spaced out between pl and p4, but sometimes they are crowded together. The teeth in corresponding ordinal positions do not always stand at the same levels ou the two sides. *377. Phalanger maculatus: in this species the small premolar ("p3") between upper p1 and p4 is generally absent, and in the lower jaw there are usually only two "intermediate" teeth. The following table shews the variations seen in 58 skulls and 7 lower jaws wanting skulls (including 43 l Leydeu skulls described by JENTINK, 7. c.).

1 Not including the case, Leyd. Mus., 153 (Jentink, /. c., p. 91), in which the " small" upper premolar is stated to be absent as an abnormality. As p3 is usually absent in the species, probably this refers to p1. The above includes six Rkulls from Waigiu, the individual peculiarities of The great variability of these skulls from the island of Waigiu is very remarkable. The 4 Leyclen specimens were described by JEX- TINK'. In one of these there was besides no left upper 2nd molar, which was entirely absent without trace, leaving a diastema between ///' and w3. In connexion with the variations of the dentition of P. iitin-nlatus in Waigiu the following singular circumstance should be mentioned. In all other localities the male P. iu<i<-i<l<ttnx alone is sjiut ted with white, the female being without spots, but in Waigiu the females are spotted like the males2. This curious fact was first noticed by JKXTIXK (f. c., p. 111).

In the other species of Phalanger no case of special importance met with ; but since in P. umimts p] is normally (4 skulls seen) two-rooted, it may be of interest to note that such a two-rooted condition of pl was seen on both sides as a variation in P. ornatus, B. M., 1317, b (2 other specimens having single-rooted p1). *378. Trichosurus vulpecula ( Phnlnngistn milpina). The typical form of this species is Australian, while the large variety, fuliginosa, is peculiar to Tasmania. In the typical form no instance of absence of P^ seen in 17 specimens examined. All possessed this tooth on each side, and though varying a good deal in size, it was in every case wellformed and functional, never being in a condition which could be called rudimentary.

Of the Tasnianian variety ful'Hjinosa, 18 specimens (8 in B. M., 10 in C. .M.) were examined. Nevertheless in every case in which this tooth is present it is a large tooth of about the size of the canines. In one case ^ is two-rooted on each side, as (THOMAS, Cat. Marwp., p. ISO) in the Celebesian Phalanger wainv*. C. M., 14 a, Hobart Town, Tasmania. Of the "intermediate" teeth in lower jaw one only is usually present, being I The small premolar was accidentally described in the paper referred to as being between tin- ,-,/;,/>„• and p4, instead of 1,( t\\r, n the anterior premolar aud /*4 JKNTINK, in tin.

- Coini.ai. the converse case of Hepialus humuli (tlu- Glmst Moth), of which. II """ ' localities, the males are «-lear \\liite ,-ui.l the females are li"ht yellowbrown with spots; l,,it in the Shetland Islands tlumales are like the females close to the large incisor. In two cases (C. M., 15 17 and h, prob. both Australian) there are two intermediate teeth, one near the incisor, the other near ^. 379. Pseudochirus. Of various species 2!) skulls shew no numerical variation in upper series. The number of "intermediate" teeth in lower jaw is very variable, '2 on each side being the most frequent, but 1 and 3 being also common. P. peregrinus, Upper Hunter K., B. M., 41, 1182, has 2 intermediate teeth in left lower jaw, but on the right side one partially double intermediate tooth. (See also No. 371.)

Petaurus : 25 skulls shew no numerical variation in upper series. In this genus the number of small teeth in the lower jaw is remarkably constant. In addition to ^ there were 3 small teeth on each side in 380. all cases seen except two, viz. — P. breviceps var. papuanus (8 normals): right side normal; left lower jaw has 4 teeth besides J74 (Fig. FIG. 66. Petaurus breviceps, No. 380. Lower jaws in profile: on right side three intermediate teeth, on left side four.

381. Another specimen has, in addition to p*,four small teeth in each lower jaw. There is a small diastema between the 3rd and 4th. B. M., 42. 5. 26. 1. [no skull]. 382. Dactylopsila trivirgata : 3 specimens have upper series normal. In addition one has an extra tooth in left upper jaw between F1 and canine. This tooth somewhat resembles but is rather smaller than the canine, near and slightly internal to which it stands [? reduplicated canine]. B. M., 1197, d.

Thylacinus, 19 normals; Sarcophilus, 9 normals, no abnormal known to me. 383. D. geoffroyi : specimen in which p4 in right lower jaw has its crown partially divided into two, the plane of division being at right angles to the jaw. C. M., 39, «. D. maculatus, Tasmania, having a supernumerary molar in left upper jaw, and on both sides in tinlower jaw. The fourth molar in the upper jaws is increased in size in a remarkable manner < Fig. b'7, B and C).

This case requires detailed description. In Fig. 67, A, a normal right upper jaw is shewn. It belongs to a specimen consiileralily larger than the abnormal one, but the latter, Mr Thomas tells me, is a good deal smaller than the normal size of the species. In the normal there are two small premol.iis < /,' and j* of Thomas), and behind these, four molars. The molars increase in size from the first to the third, which is by far the largest. Behind the third is the fourth molar, which i- much smaller than the others, having the peculiar flattened form -hewn in the Figure 67, A.

FIG. 67. A. Right upper jaw of normal Dusyunts maculatus (shewn as far as the canine) for comparison with the variety. (N.B. The latter is considerably smaller.) <>n comparing the abnormal skull with a normal one it is seen that the two pivnmlars and first three molars on each side are unchanged. IVhind the third molar on the right side there is a single tooth; but this, instead of being a thin tooth like normal »'', is considerably larger and the longitudinal measureni' ni in the line of tile jaw is not very much less than the transverse measurement. In the right upper jaw therefore the iiim/licr of the teeth is unchanged.

On the left side, behind the third molar, there is a square tooth (/"') of good size, about equal in bulk to half »' , while behind this again there is another tooth, m^, which is a thin and small tooth having nearly the form and size of normal w*. The lower series is alike on both sides, each having an extra molar behind w* (Fig. 67, C). The two extra teeth are well formed, being as long but not quite so thick as m*. B. M., 41, 12, '1, 3. In Cat. Marsup. Brit. Mus., 1888, p. 265, note, THOMAS refers to this skull, and describes it as an instance of an additional molar inserted between in* and in* on the left side above and on both sides below. This view is of course based on the resemblance that the extra ?«_5 of the left side bears to a normal m? and on the fact that the left «^ is like no tooth normally present. In the light however of what has been seen in other cases of supernumerary molars a simpler view is possible. For in cases in which a supernumerary molar is developed behind a molar which is normally a small tooth, the latter is frequently larger than its normal size. In the present case it appears that on the right side ™4 has been thus raised from a small tooth to be a tooth of fair size, while on the left side the change has gone further, and not only is «^_ promoted still more, but a supernumerary ™?_ is developed as well. It is interesting to note that this «^ is a small tooth, very like normal ?^4, and it thus may be said to be beginning at the stage which m^_ generally reaches. In the lower jaw n& is added without marked change in m4" ; for m4 is normally a large tooth and has, as it were, no arrears to be made up. Mr Thomas, to whom I am indebted for having first called my attention to this remarkable case, allows me to say that he is prepared to accept the view here suggested.

Phase ologale. In the upper jaw normally 3 premolars, by Thomas reckoned as p\ p3 and p*. Between the first and second ("p3") there is sometimes, but not always, a small space, and in the following case a supernumerary tooth was present in this position. In the lower jaw ]T4 is often small and may be absent. As Thomas has observed, the size of p4 in the upper and lower jaws maintains a FIG. 68. Phascologale dorsalis, No. 386. Teeth of left upper jaw from canine to first molar; below, the teeth of the right side reversed (after THOMAS).

fairly regular correspondence. Within the limits of one species p* may 387. shew great variation : for instance, of Phascologale flavipes 7 -pecimens were seen : in 1 JT1 was absent, in 2 it was small, in - mode- i-atf, and in '2 it walarge. 388. Didelphys: 7!l specimens normal. One specimen alone, D. lanigera, Colombia. M.. IT.'W, /), was abnormal, having no m* in either upper or lower jaws. D. opossum (one specimen, B. M. ) had right m^ larger than the left.

Tinfollowing evidence relates to the genera Bettongia, Pot on m* :ind Lui/orcJtestes. Tn these forms the molars are normally four in each ju\v. As THOMAS observes (Oat. Marsup. Brit, ^fus., p. 10~>, /,<,/• ), in /-'- //mii/ia cases of tifth molar occur, but on the other hand cases of noneruption of m4 occur also. The variations seen in the three genera tin-re is a small empty crypt behind m3, and on right side behind this M-MJII there is a minute tubercular tooth not represented on the other