456 
MR. O. TIIOMAS ON THE HOMOLOGIES AND 
appears to have been wholly in the number of the true molars, and not of the pre 
molars, as is particularly well illustrated by the early representatives of the many¬ 
toothed Odoutocete Cetacea,* and also by the great majority of the Mesozoic Mam¬ 
malia, many of which,t while still having only four premolars, have from five to eight 
true molars. In fact, the earliest forms seem to have normally had a considerable 
number of molars, but these were soon reduced down to four in the Metatheria, and 
later to three or less in the Eutheria, leaving a few isolated forms, J with larger 
numbers as remnants, retaining what the common ancestors of all had once possessed. 
Two only of the many described genera of these ancient forms have been said to have 
more than four premolars, namely, Amphitherium and Amphilestes, each described as 
having six premolars and six molars § ; but, as even this determination is disputed,! 
and in any case rests entirely on the form of the teeth, and not on a knowledge of 
their manner of changing, it can hardly be said to prove incorrect the practically 
universal rule that the typical number of premolars is, and has always been, at least 
since middle Mesozoic times, four, and four only.^j 
Whether the teeth in the first stage of all were rooted or rootless is very doubtful, 
but the probability seems to be on the w r hole that they were simple conical teeth, 
rootless for part, if not the whole, of the animal’s life, and possibly not unlike those 
now possessed by some of the Dasypoclidfe.'"* 
This first stage in the Mammalian dental series would probably very fairly repre¬ 
sent, so far as can be judged, the dentition of the Prototheria, the toothed and gene¬ 
ralised ancestors of the living, and now highly specialised, Monotremata, wdiich 
there is every presumption for believing, as Professor Flower has pointed out, ft were 
for some time both Homodont and Monophyodont. 
If we now combine these diagrams and tabulate the yarious processes and examples 
already described, we obtain (Plate 28 , I. to IX.) a complete diagram of the tooth 
* E.g., Squalodon, which lias tlie following formula :—I. -| C. 4 P.M. % M. -f X 2 = 60. Flower, ‘ Encycl. 
Brit.’ 9th edit., Article “ Mammalia.” 
t E.g., Achyrodon manus, Stylodon pusillus, &c. See Owen, ‘Mesozoic Mammalia’ (Palaeontogr. 
Society, 1870), 1871. Plate 2, figs. 6, 14, and 17. 
t E.g., Myrmecobius, Bettongia, which has not infrequently five true molars, and the many-toothed 
Edentates. 
§ Owen, loc. cit., Plate 1, figs. 21-23. 
|| Lydekker, ‘ Cat. Foss. Mamm. Brit. Mus.,’ Part 5, 1887, p. 271, footnote. 
A single instance is, however, known to me of a true Heterodont and Diphyodont Mammal with five 
premolars, namely, Rhinogale melleri, Gray, a member of the Herpestine section of the Yiverridae. Of 
this remarkable species, however, only one skull is as yet known (figured ‘ Zool. Soc. Proc.,’ 1864, p. 574), 
so that no positive deductions can be drawn from it. It may he either that the supernumerary premolar 
is a mere accidental duplication of one of the other premolars, or that one of the milk-premolars has been 
retained in position, but these points can only be settled by the examination of further specimens of the 
species. 
** See Baume, op. cit., p. 152, &c. 
tf ‘ Odontol. Soc. Trans.,’ vol. 3, 1871, p. 221. 
