312 NATURAL SCIENCE. may. 



considered as ancestral to the Lemuroidea, it remains now to discover 

 to which group of the Mammalia they are related. I have endeavored 

 to prove that it is probably not correct to derive the lemurs from any 

 of the forms bearing unguliform terminal phalanges, and that the 

 Unguiculata is the group from which the lemurs have arisen. The 

 Insectivora are by general consent acknowledged to be the most 

 primitive group of the Mammalia, and it is in some generalised member 

 of this order now extinct, that we must look for the ancestor of the 

 Primates. 



Now the placentation of Tarsius, as shov/n by Hubrecht, closely 

 resembles that of the Insectivora, and both in the latter order and in 

 the true Primates the placental area is very limited in extent, and thus 

 differs widely from that of the lemurs, where the placenta is of the diffuse 

 type. In man the chorion extends in an early stage of development 

 around the whole embryo and is entirely villous. This represents 

 to a certain extent the diffuse stage of the placenta of the lemurs. 

 At any rate, on the criterion of the placenta, I see no objection to the 

 view that the lemurs are related to the apes. The diffuse form of 

 placenta of the lemurs has probably arisen from a non-deciduate 

 restricted type, which would be also the ancestral form of placenta of 

 the Anthropoidea. This is the view of the evolution of the placenta 

 in the Primates as stated by Balfour (4). 



In conclusion, the decidedly mixed character of Tarsius, relating 

 it on one hand to the apes and on the other by so many characters to the 

 lemurs, leads to the conclusion, that this genus approaches structurally 

 the common ancestral type from which apes and lemurs have descended. 

 If we do not grant this, then we are forced to the conclusion that the 

 type of placenta found in Tarsms, the straight colon, and the closure 

 behind and below of the orbit are all cases of parallelism, and that 

 Tarsius is not genetically related to the apes. I am aware of the fact 

 that it is quite the fashion of late to push the theory of parallelism to 

 the extreme, with the result that we find few types that fit exactly 

 in our phylogenetic series. Too much of this sort of thing would 

 rather bear against the theory of evolution, as we would have 

 innumerable parallel series, but few tending to convergence. I believe 

 the characters of Tarsius that are like those of the Anthropoidea 

 to be essential, and not adaptive. They prove that Tarsius is a true 

 connecting link, genetically related to both lemurs and apes. 



On palffiontological grounds I cannot accept Leche's dictum as 

 to the significance of the milk dentition in regard to the evolution of 

 the lemurs, and it appears most probable that the ancestral lemur was 

 a small insectivore-like animal, somewhat similar to Tupaia, an 

 arboreal type, which had canines of the normal upright form, and 

 larger than the incisors. In many of its characters this hypothetical 

 ancestral type would more closely resemble Tarsius than any other 

 living primate. I fully reahse the fact that Tarsius is quite specialised 

 in some respects ; but as a whole it is decidedly primitive. 



