1892.] NEW YORK ACADEMY OF SCIENCES. 51 



the carnivorous marsupial Dasyurus, in the single instance of 

 -3Iyoxt(s among Rodentia, in certain Edentates, Tardigrada and 

 Manidfe, and notably in certain members of the carnivorous 

 groups of Arctoidea. 



In taking a general view of the Carnivora, it appears that in 

 respect to the structure of this portion of the alimentary tract, 

 as well as in reference to other features, the Cyuoidea, including 

 the dogs, the wolves, jackals and foxes, form a well-marked 

 central group, with highly developed and convoluted ciieca, from 

 which on the one hand the Ailuroidea, including cats, civets and 

 hyenas, depart, with caecum uniformly present, but short and 

 markedly pointed at the termination, suggesting the degenera- 

 tion of a formerly more develo^Ded structure, while on the other 

 the Arctoidea, bears, weasels and raccoons, constitute a series 

 bound together by many common fundamental peculiarities of 

 structure, and presenting in many members of the group a 

 complete or nearly complete absence of a caecal appendage. In 

 the typical Ursidte the absence of the ctecum appears to be the 

 rule. 



Among the Procyonidee, Na!<ua has long been known to pre- 

 sent the same peculiarity. The addition of Procyon lotor to 

 the list of forms devoid of a caecum has, to my knowledge, not 

 been made before, although the close relations existing in other 

 respects between the subfamilies of Procyonina3 and Nasuinae 

 would render the agreement in this particular not unexpected. 



As regards the remaining families of Cercoleptidae and Ailuri- 

 dae, no data are at hand. 



In the group of musteliform Arctoidea the absence of the 

 caecum in Mustela has been noted. It is among the subfamilies 

 of the Mustelae that further investigations should reveal inter- 

 esting forms, for here in other respects deviations from the 

 Arctoidean type are met with, as the transition from the j^lauti- 

 grade Galictis to the subplantigrade Gulo and the completely 

 digitigrade weasels and martens. 



Surely, the general impression gained from a bird's-eye view 

 of the mammalian alimentary canal, which would assign to 

 herbivora a complicated caecal apparatus, and reduce the same 

 to the simpler forms in Carnivora, must become modified when 

 considering the marked exceptions prevailing in the group 

 under discussion, where with a combination of frugivorous and 

 carnivorous habits, the ileo-colon presents this complete reduc- 

 tion, nor could we desire a better illustration of the truth that 

 other factors besides use influence and determine structural 

 peculiarities. The persistence of a piimitive ancestral type in 

 •one, and successive modifications of the same in other directions. 



