448 
MR. 0. THOMAS ON THE HOMOLOGIES AND 
Perameles, and Didelphys, and from which again we obtain, by the disappearance of 
pm 4 , that of Dasyurus and Sarcophilus, namely P.M. ^ ^ ^ 
Turning then to Myrmecobius, with its eight or nine cheek-teeth, of which only three 
are commonly reckoned as premolars, I examined several young specimens in order to 
see if this animal, like the abnormal Phascologale, had not really four premolars, as no 
exact observations on it had been published, and the usual determination rested 
merely on the shape of the teeth. In fact, even if no tooth-change could be found, it 
seemed possible enough that the small tooth commonly reckoned as the first molar 
should be really a persistent milk pm 4 , the permanent tooth corresponding to it having 
become aborted.* * This theory, however, I have been able to disprove by the exami¬ 
nation of such a jaw as is shown in Plate 27, fig. 9. Here we find that, of the whole set 
of cheek-teeth (premolars and molars), the first, second, fourth, and fifth are fully up, and 
all much on a level with one another, while, on the other hand, the third has scarcely 
penetrated the gum, and stands therefore far below the level of the rest. Hence we 
see that this third tooth, by its very lowness of position, or rather lateness of develop¬ 
ment, is certainly the true pm 4 , even without there being any milk-tooth above it; for, 
had it been pm 3 , it would have been at all stages, as with other animals, fully on a level 
with the teeth next in front of and behind it. Of a milk-tooth preceding this pm 4 I 
can find no trace whatever, although it is possible that, considering the very rudi¬ 
mentary degree of development in which the milk-tooth occurs in the Thylacine and 
Koala,! it may yet be discovered in specimens younger than have yet been examined. 
The correct formula of the cheek-teeth of Myrmecobius is thus P.M. ^ ‘ ^ ^ 
M. 
1 . 2 . 3 . 4.5 . 0 
1 . 2 . 3 . 4.5 . 6 
- ; and therefore, so far as regards the premolars, identical with that 
of Phascologale and Tliylacinus. 
As regards the incisors, two specimens of Myrmecobius in the Natural History 
Museum present the interesting anomaly of possessing four instead of only three lower 
incisors, the extra tooth being in each case clearly i 4 —a fact which proves what has 
generally been presumed to be the case, viz., that it is the fourth incisor that has 
disappeared in ordinary three-incisored Mammals. 
This completes the list of the living Dasyuridse to be referred to, but one fossil 
Marsupial, perhaps referable to the same family, and, although nearly the oldest- 
known Mammal, strongly resembling the modern Phascologale, has a dentition of so 
interesting a character as to call for special notice. This is the Triacanthodon serrula, 
described by Sir R. OwenJ from a single lower jaw found in the Mesozoic Purbeck 
* A process which, as noted below, p. 451, has taken place in certain Placentals. 
t Thomas, ‘ Zool. Soc. Proc.,’ 1887, p. 338. 
f ‘ Mammalia of the Mesozoic Formations’ (Palaeontographical Society), 1870 (pub. 1871), p. 72, PI. IN ., 
figs. 7 and 8. Mr. Lyhekker (‘ Catalogue Fossil Mammalia Brit. Mus.,’ vol. 5, p. 258) considers this fossil as 
not generically, or even specifically, separable from Triconodon mordax, Owen, described at the same time. 
