68 Mr. O. Thomas on the Rats usually 
nized Lemniscomys as a genus for the “‘ striped or spotted ” 
forms, but, working only from E. African material, he has 
exaggerated the degree of difference occurring in the 
zygomatic plate of the skull, nor has he mentioned the 
pumilio group, which in that respect agrees with Arvicanthis. 
A careful examination of the Museum material shows 
that three quite distinct genera may be distinguished, whose 
characteristics are as follows :— 
ARVICANTHIS, Less. 
Syn. Lsomys, Sund. 
Type. A. niloticus (Mus niloticus, Geoff.). 
Back generally without stripes, but sometimes with a 
single mesial one, not very sharply defined. Fifth finger 
normal in length, with a distinct sharply pointed claw. 
Skull stout and heavy. Anterior zygomatic plate high, 
its top at about two-thirds the height of the muzzle, its 
anterior edge not or little cut back below. 
Teeth stoutest of the group. Molars heavy, simple, the 
laminze transverse and little folded ; in m*® the main lamina 
is slightly slanted, thicker externally, but not folded on 
itself. 
These are the ordinary members of the “ niloticus” group, 
known by their heavy form and generally unstriped colora- 
tion. A paper on their species and subspecies has been 
recently published by Mr. Dollman *. 
Lemniscomys, Trouess. 
Type. L. barbarus (Mus barbarus, Linn.). 
Body striped and often spotted as well, the median line 
of the back always black. Fifth finger shortened, with a 
short nail instead of a claw. 
Skull rather more lightly built than in Arvicanthis. 
Anterior zygomatic plate low, the top only about half the 
height of the muzzle, its front edge generally undercut and 
concave, sometimes sharply so. 
Molars smaller than in Arvicanthis, their laminz less 
simply transverse, more bent backward internally. Main 
lamina of m® strongly bent back on itself externally ; cleft 
* Ann. & Mag. Nat. Hist. (8) vii. p. 584 (1911). In this paper 
A. rex, Thos., is put aside ‘as its affinities are very doubtful.” In this 
conclusion Mr. Dollman’s opinion is confirmed by my present observa- 
tions, for the type of vex, unlike all members of true Arvicanthis, proves 
to have a nail instead of a claw on the fifth finger. It may be pro- 
visionally considered as a giant member of Desmomys, Thos. 
