444 
MR, 0. THOMAS OH THE HOMOLOGIES AND 
the study of which I have been led on to form a theory on the evolution and succession 
of the teeth applicable to the Mammalia in general. This theory attempts to bridge 
over the existing gap between the Metatheria and Eutheria, and to shew how the 
teeth of the one may have passed into those of the other. 
Before commencing, I must express my sincere thanks to Mr. B. Lydekker, to 
whose extensive knowledge of Mammalian Palaeontology and its literature I am largely 
indebted, and with whom every point in the present paper has been fully discussed—a 
sifting process which has, I hope, eliminated some of the unsound conclusions to which 
I might have otherwise come. Throughout the course of my work he has taken con¬ 
siderable trouble in obtaining information on various points, and this he has at all 
times freely communicated to me. I feel, therefore, that he should be credited with a 
very large share in the results, whatever their value may be, that are put forward in 
the present paper. I must also record the obligations I am under to Professor Flower 
himself, whom I have consulted on several points, and who has freely given me the 
benefit of his large knowledge on the subject. 
To confine our attention, first, to the Dasyuridse. In this family we find the greatest 
amount of variation in the extent to which the change of teeth takes place, some 
species having a well-developed successional tooth preceded by an equally well- 
developed milk-molar, while, on the other hand, others have no successional tooth at 
all, either in the milk or permanent stage. In this family also occurs Myrmecobius, 
remarkable for being the only known heteroclont Mammal normally possessing more 
than four true molars. 
This family has also another and more vital interest for the evolutionist, arising from 
the presumption that it was in all probability the family in which the change from 
Metatherian to Eutherian occurred. This presumption is based partly on the very 
generalised character of the family as a whole compared to the other and more 
specialised groups of Marsupials, but chiefly on the strikingly exact resemblance 
existing between the structure of the teeth of many of its members and that found 
both in certain of the “ Creodonta ” or Carnivora Primigenia, among which it is 
generally supposed that the direct ancestors of the modern Carnivora should be 
sought for,'" and also in many of the more generalised Insectivora, whose claims to the 
parentage of other Placentals have been advocated by Professors Huxley,! Parker, j 
and others. 
The family Dasyuridse consists of the following genera, the respective numbers of 
their premolars and molars being placed after each :— 
* See R. Lydekker, ‘ Catalogue of the Fossil Mammalia in the British Museum,’ Part 5, 1887, pp. 26 
(footnote) and 307. 
t ‘Zool. Soc. Proc.,’ 1880, p. 657 and elsewhere. 
+ ‘Phil. Trans.,’ 1885, p. 268. ‘Mammalian Descent,’ 1885, p. 125 et seq. 
