FROM THE PHILIPPINE ISLANDS. 397 



pterygoid processes large, triangular, projecting far downward, in fact below the level 

 of the tips of the incisors ; external processes practically or quite obsolete, so that 

 there are in this animal no enclosed pterygoid fossce. Bullae small, but not of abnormal 

 structure. 



Lower jaw exceedingly low, slender, and little curved. Coronoid processes very fine, 

 slanted backwards. 



Teeth (PL XXXV. fig. 7) extraordinarily reduced, the dental armature in 

 JRhynchomys being less in proportion to the size of the animal than in any other 

 Rodent, perhaps even — apart from the Cetacea — than in any other toothed mammal. 

 Incisors white above, pale yellow below; the upper ones not grooved, very short, 

 narrow, slender, and forming the arc of a very small circle, so that their roots come 

 opposite the anterior end of the palatal foramina, and the chord of the circle they 

 describe is barely more than a third of the diastema. Molars f-, so minute that it is 

 difficult to understand of what use they can be to the animal; to. 1 oval, flat-crowned, or 

 with low indistinct cusps, but without quite young specimens it is impossible to make 

 out for certain whether there is any true cuspidate structure ; m. 2 about half the size 

 of to. 1 and a shorter oval in outline. 



Lower incisors very slender, and, owing to the oblique set of their enamel-covered 

 faces, they wear to an unusually fine point, as sharp as a needle. Lower molars 

 very similar both in size and shape to those of the upper jaw. In one specimen 

 the posterior lower tooth is altogether absent on both sides, so that there is only one 

 molar present ; this fact shows strikingly the tendency there is to a progressive 

 reduction of the molar teeth. 



Type. Bhynchcmys soricoides. 



This most remarkable genus, by its peculiar Shrew-like appearance, reduced teeth, 

 elongate muzzle, and other characters, seems at first sight perfectly isolated from any 

 other known group, and in my recent classification of Rodents it was made the type of 

 a subfamily of the Muridse, the Rhynchomyinse 1 . But among the Rodents then 

 included in the Murinse — on account of its strictly murine molars — there occurs the 

 genus Echiothrioc 2 , a native of Celebes, which also has an elongated snout. This 

 animal, of which the Museum now possesses two perfect specimens from N. Celebes, 



1 P. Z. S. 1896, p. 1017. 



2 Described by Gray, P. Z. S. 1867, ibs correct locality determined by Jentink, Notes Leyd. Mus. v. p. 177 

 (1883) ; renamed by me Crawothrix, Ann. & Mag. N. H. [6] xviii. p. 216 (1896). As I have now joined those 

 who think that names should be retained as originally spelt, whether classically right or wrong (except in the 

 case of obvious misprints), I am now prepared to consider that Peters's Echinothrix of 1853 does not preoccupy 

 Gray's Echiothrix of 1867, and therefore again recognize the latter term. Those who are not of this opinion 

 must call it Crawothrix. That the missing out of the letter n is not a misprint is shown by Gray having 

 written on the type skin what appears to be " Echithrix," might be " Echiothrix," but is certainly not 

 " EchinothrLv." 



