BY T. THOMSON FLYNN. 543 



Conclusions to be drawn from the above facts. 



The question now presents itself: — What conclusions can be drawn from 

 the above facts as to the phylogeny of the marsupial group? We can at once 

 rule out of court Hubrecht's suggestion (1908) that the marsupials are a group 

 derived from the Monodelphia. The death-blow was dealt to this theory by Hill 

 in his brilliant and forceful essay of 1910 and it does not seem possible that it 

 should ever again be seriously advocated. 



The present conception of the interrelationships of the three mammalian 

 subclasses rests mainly on the work of Huxley. In his paper of 1880, in apply- 

 ing the law of evolution to the problem of mammalian descent, Huxley came 

 to the conclusion that, starting with a hypothetical pro-mammalian group as a 

 basis, the evolutionary trend which resulted in the higher mammals (so-called 

 Placentalia) was a continuous one, but at two periods, groups were formed, Pro- 

 tptheria and Metatheria, which were prototypal to the Monotremata and Mar- 

 supialia respectively. Huxley's group Metatheria was definitely defined as being 

 aplacental. 



Gregory (1910) has ably and exhaustively reviewed the relationships of 

 these groups and Bensley (1903) has emphasised the complexity of the problem 

 of their definition. 



There exist two diametrically opposed views as to the phylogeny of the 

 Marsupialia. On the one hand, there are those who believe that the immediate 

 ancestors of Marsupials were aplacental and that the presence of an allan- 

 toplaoenta in Perameles is an example of homoplasy or convergent development. 

 The modem exponent of this theory is Bensley (1903). On the other hand, 

 Wilson and Hill (1897), Hill (1897-9) and DoUo (1899) hold the opposite 

 opinion that the marsupials came from a placental stock. 



The facts which I have been able to adduce in this and other papers can 

 leave no doubt as to which view is right. The absolute agreement, even to 

 minute details, of allantoplacental preparation and formation in Perameles with 

 the phenomena occurring in Monodelphia show with the utmost certainty that 

 this placenta in Perameles is no independently acquired organ. And the value 

 of this conclusion is strengthened by the evidence afforded by changes in the 

 pregTiant uterus of Pseudochirus and Dasyurus. We can take it, then, that the 

 available evidence shows that the marsupials are placental in origin and that the 

 present aplacental condition of nearly all marsupials represents a condition of 

 degeneration from the ancestral complex placental condition. 



Huxley's plan of the relationship of the mammalian subclasses as indicated in 

 his text-figiire needs, therefore, certain modification. His metatherian group be- 

 ing definitely defined as aplacental, cannot be regarded as immediately ancestral 

 to the Marsupialia and the term Placentalia as applied to the Monodelphia loses 

 its sig-nificance. 



My own views as to these relationships are expressed in the accompanjang 

 diagi'am. 



iMonotremata Marsupialia Monodelphia 



Protoplacentalia 



Promammalia 



