No. 410.] THE MAMMALIAN SUBCLASSES. 133 
page 365 of his paper he speaks of a fundamental correspond- 
ence between the artiodactyl ungulates and the monotremes in 
this respect. 
Here we may notice the opinion of Wilson and Hill (97, p. 
579): “ We do not believe that in the long run it will be found 
possible to maintain the essential dissimilarity of the mam- 
mary glandular organ in monotremes." This appears to be 
the more appropriate attitude. 
With regard to the results of Osborn, Hubrecht quotes, 
among other statements, the following one, which he inter- 
prets as indicating polyphyletic origin: ** All stem mammals 
were related in their double succession (of teeth), in their 
dental formula, and in their primitive molar form. These 
features point not to a succession, but to a unity of ancestry 
of the monotremes, marsupials, and placentals." ! 
This involves the broader question as to how convergence 
of type ought to be interpreted in the Mammalia; for, in 
addition to the results of Osborn, those of Wilson and Hill 
(97) confirm convergence in dentition, those of Gegenbaur 
(86) and Klaatsch (92), convergence in the mammary glands, 
those of Klaatsch ('92, 95), finally, convergence in the acces- 
sory mammary apparatus. All of these suggest unity of 
ancestry. 
If there were no convergence of type in the structures 
under consideration, Huxley's hypothesis would immediately 
become untenable, because it would then be necessary to 
assume a polyphyletic origin. But if we admit convergence, 
the chances are equally divided between genetic succession and 
polyphyletic origin. In order to prove the latter, however, 1t 
would be necessary to trace the converging structures back 
to a pre-mammalian stage. In other words, it would be neces- 
sary to prove that in progressive development certain struc- 
tures diverged in character before the mammalian condition was 
instituted ; because if any structure diverged from a type which 
was still mammalian, the latter would be regarded as also pro- 
totherian. None of the above-named structures have as yet 
been traced to a non-mammalian type. 
1 Osborn ('93, p. 20). 
