250 REV. o. p. CAMBRIDGE ON A [May 7, 



characters as the form of the palate-ridges and the number of the 

 mammas, have remained quite unaffected during all the changes that 

 the rest of the animal has undergone. 



A parallel case, but one in which the differences between the two 

 are by no means so strongly marked, is that of the raie Floridan 

 Neofber \ in its relationsfiip to the common and widely-spread 

 North-American Fiber. 



But the question next arises as to which of the Murines Xeromys 

 itself is most allied ; but here the very high specialization of its 

 teeth presents the same difficvilty as in the case of Hydromys, so that 

 in this respect the discovery of Xeromys hardly helps us at all. The 

 slight differences between the teeth of the two genera ])rove that the 

 almost continuous walls round the lobes of the molars of Hydromys 

 were formerly cusps, as in other Murines ; but although this leads 

 directly towards Mus, it leads equally directly towards nearly all the 

 other members of the family. In fact one cannot say with absolute 

 certainty that the teeth are more nearly allied to those of Mus tlian 

 to those of Urotnys, Hapalotis, Gerbillus, or even Criceius itself; 

 and we must therefore be content to wait in the hope tliat more of 

 the missing links, either fossil or recent, may yet turn up, and that 

 then a more enlightened study of larger material may tend to elac - 

 date this most interesting question. In any case we must be thankful 

 that by the preservation of the apparently common-looking little 

 Xeromys myoides, so important an advance in onr knowledge of 

 the ancestry of Hydromys has been made practicable. 



EXPLANATION OF PLATE XXIX. 



Figs. 1-4. SknW o( Xcwmi/s myoides. 

 5. Palate-ridges of ditto. 

 (j. Anterior zygoma-root of ditto. 

 7. Anterior zygoma-root of Hydromys chrysogaster. 

 8, 9. Ear and right hind ibot of Xeromys myoides. 

 10-12. Left upper and lower molars of ditto. 



2. On a new Tree Trap-door Spicier from Brazil. 

 By the Rev. O. P. Cambridge, M.A., F.R.S., C.M.Z.S., &c. 



[Received April 10, 1889.] 



Class ARACHNIDA. 



Order Araneidea. 



Fam. Theraphosid^, 



Gen. nov. Dendricon. 



Dendricon rastratum, sp.n. 



This genus is evidently nearly allied to Moggridgea, Cambr., but 



the presence on the faices of a strong rake-hke group of spines near 



the Jjase of the fang, and a difference in the form of the maxillfe and 



labium, lead me to conclude that it is certainly distinct from that 



' See True, P. U. S. Nat. Mus. vii. p. 170 (1884). 



