422 'MR. R. I. POCOCK ON THE EXTBJINAL 



Xoies on the Families and Siobfamilies. 



Family Hystricid^, 



Since Lyon (Proc. U.S. Nat. Mus. xxxii. p. 576, 1907) has com- 

 paratively recently classified the Hystricida3 into the two sub- 

 families Atherurinfe {Atherurus, Trichys) and Hystricinee. 

 {Thecurus, Aca7ithion, Ilystrix), mainly by the structure of the 

 tail, I need not refer to these animals further beyond expressing 

 my complete concurrence \iith his opinion as to the systematic 

 value of the characters he uses as a basis for the separation of 

 the two subfamilies. 



Family EREXHizoNTiDiE. 



By their external characters the three genera {Chcetomys, 

 Coendu, Ereihizon) composing this well-marked family fall into 

 two groups, the first represented by Erethizon, the second by 

 Coendu and Chct^tomys. By cranial and dental characters, how- 

 ever, Mr. Thomas separated Chcetomys from the others as repre- 

 senting a special subfamily Chsetomyinpe, and assigned Coendu 

 and Erethizon to the Erethizontinse. Adopting the separation of 

 Chcetomys from Coendu, I think the latter should be similarly 

 separated from Erethizon and the family divided into three 

 subfamilies, which may be diagnosed as follows * : — 



a. Tail Ions, C3'lindrical, comparativelj' slender, and prehensile; 

 hind feet with a large moveable lobe on the inner side and 

 minute hallux ; fore feet also expanded on the inner side ; 

 nostrils widelj^ separated; ear specialised, with large antitragus. 

 a' . Body covered with flexible spine-like waved bristles ; skull 

 with small orbit circumscribed by frontal and jugal post- 

 orbital processes; jugal arch very deep; palate, tooth-rows, 



and symphysis of mandible long ChcetomyincB. 



b' . Body armed with comparativelj' stout unwaved spines; 

 orbit large, no postorbital processes ; jugal arch not deep ; 

 palate, tooth-rows, and mandibular symphysis much 



shorter , Coendina, 



h. Tail short, thick, not prehensile ; hind foot with at most a small 

 inner lobe, but with well-developed hallux ; fore feet without 

 expansion on inner side ; nostrils with narrow septum ; ear 

 simple Eretliizontina. 



At present each of these subfamilies contains a single genus ; 

 but it seems to me to be probable that the first modern syste- 

 matist who has adequate material of Coendu, and the time to 

 devote to the study, will find characters justifying the separation 

 of that genus into two or more genera. There appear to me to 

 be no good reasons for thinking the Erethizontidfe especially 

 related to the Hystricidte — a conclusion to which other authors 

 have come. The Erethizontidte are, I think, probably very 



* The external characters, apart from the ear, here made use of were long ago 

 pointed out by Waterhouse. It is, however, merely a guess on my part that the 

 nostrils and ears of Chcetomys are like those of Coendu. 



