142 On Sjnny Rats from South-eastern Brazil. 



myosuros was described. Tlie names leptosoma and cinna- 

 ruomeus were mere renamings of myosuros. If, however, 

 Lagoa Santa specimens siiould ultimately prove different 

 from those of Bah la — and perhaps tiiey are browner and less 

 rufous, though the indifferent material does not suffice to 

 prove it, — they should bear the name of eler/ans, Lund, with 

 synonym fid iyinos us, leaving setosus for the Bahian animal. 



The type of E. albispinus, I. Geoff., came from Deos 

 Island (=Madre de Dies), Bay of Bahia. Its skull is in the 

 Paris Museum (No. A. 7669) ajid is practically perfect. 

 The two specimens (327/2, 327/3) from Geneva, which were 

 among those referred to by Pictet * as being true albispinus^ 

 also show clearly the characters of the species. 



Finally, the Museum contains a fine series of an allied form 

 obtained by M. Robert at Lamarao, also in Bahia, but in the 

 liighlands of the " sertao" further to the north. It is on this 

 series that I have been able to observe the various cha- 

 racters of the subgenus Trinomys. The form may be briefly 

 described as follows : — 



Proechimys alhispinus sertonius, subsp. n. 



Size about as in albis^nnus. General colour above lined 

 brown; the fore back with buffy hairs which show through 

 on the sui face; the hinder back blackish brown, this colour 

 arising from the dark ends of the spines. Sides not more 

 buffy or rufous than back — in fact, less so; while the type of 

 albupinxis was stated to have strongly buffy sides, such as 

 are found in old specimens of setosus, as has also the normal 

 coloured Pictet^specimen received from Geneva, the other 

 being an albino. Sides of body, rump, and thighs with 

 numerous prominent white-ended spines. Under surface, 

 h;inds, and feet white. Tail dark brown, nearly black, for its 

 whole length above; whitish below; not pencilled. 



Skull short and squat, with broad muzzle ; the breadth 

 between the two lacrymal bones decidedly greater than in 

 true albispinus. Supraorbital ridges well marked, but not 

 extending on to parietals. Palatal foramina short, fusiform. 

 Palatal notch very narrow, acute-angled, reaching forwards 

 to the level of the front edge of m^. liamular processes of 

 ]itervgoids narrow, but not absolutely linear. Bullae rather 

 small. 



Incisors more proodont than in other members of the 

 group, the index of the type 93°, and in some specimens 

 attaining 96°; that of the type of albispinus 86° and of the 

 two Geneva specimens 86°-87°. 



* Anim. Nouv. GeneA'. p. 2 (1841). 



