No. 2304. REPTILIAN CHARACTERS IN MAMMALS— WORTMAN. 7 



winglike appearance to the anterior part or edge of the orbital cavity 

 not seen elsewhere among mammals. 



Of the 30 or more skulls of this species in the collection, all of them 

 without exception show the presence of these bones, where they have 

 not been scraped away or detached in the course of preparation. 

 Even in those skulls in v/hich they have been lost, the thickened and 

 roughened edge of the crista, however, gives ample evidence of their 

 having been present in the fresh state, and I conclude, therefore, that 

 their occurrence is a constant and well-marked feature of this species. 

 Of the 20 specimens out of 32 in which these ossicles are preserved 

 there appears to be little variation, either in the size or position 

 which they occupy on the edge of the orbit. 



At least one of these ossicles is found in the same position on the 

 border of the lachrymal, as just described in Cercodenus sultana, in 

 one of the three skulls of Petrodromus tetrad actylus in the collections; 

 but there can be little doubt from appearances that it was originally 

 present in all three cases, but has been subsequently lost in prepara- 

 tion. I am unable to say just what degree of constancy it has in the 

 skull of this species, but it is not altogether unlikely that it is very 

 generally present. In a like manner in a single sloill of Macroscelides, 

 there is one of these ossicles present upon one side with evidence that 

 it has been lost from the other. In the remaining genera, Nasilio 

 and Elepliantulus, these ossicles are occasionally present, but it is 

 always as a very thin and weak spicule of bone. Just to what extent 

 its absence m^ay be due to faulty preparation I am unable to say, 

 but I am of \h.e opinion that careful investigation will sliow its fre- 

 quent presence. 



None of the preceding genera exhibits any of the remarkable 

 characters of the jugal arch of RhyncJiocyon, above described. The 

 skull is high and narrow in front of the orbits, the nasals are long and 

 narrow, and the premaxillae send long-pointed processes backwards 

 between the nasals and maxiUaries. There is no postorbital process 

 of either frontal or parietal above, and but a very faint indication 

 of a postorbital process of the jugal arch, which is weak and slender. 

 In the lower jaw the ascending portion of the ramus rises more 

 abruptly, but the condyle is placed high above the tooth row, and 

 the coronoid is small as in Kfiijncliocyon. 



SkvII of TiqKtia. — In the species of this genus the skull has a very 

 considerable interorbital width, as in RhyncJiocyon; the snout is 

 long and pointed, but the brain case is much more capacious, and as 

 a consequence the upper part of the cranium is more rounded. The 

 postorbital process above (fig. 3), while not as primitive as that of 

 Rni/ncJiocyon, is, however, in a stage of development not far removed 

 from it; it apparently arises from the frontal, as a long slender strip 

 of bone, is directed backwards and outwards, being closelj^ applied 



