52 SOT.ENODON PARADOXUS. 



The muscular anatomy of the two species is essentially similar although in 

 case of the Cuban Solenodon, our knowledge is still somewhat imperfect, and 

 rests almost wholly on the account by Dobson. This author fails to describe 

 in S. cubanus a stylo-hyoid, though it may be present as in »S. paradoxus. Other 

 differences found in the latter as compared with Dobson's description of S. 

 cubanus are: a single head to the biceps instead of two; a second head to the 

 gluteus maximus arising from the ilium; slight differences in the insertion of 

 the adductor longus; the union of the soleus with gastrocnemius. .Additional 

 peculiarities are the insertion of tibialis anticus upon the entocuneiform instead 

 of on the first metatarsal; the insertion of peroneus longus upon metatarsals 

 1 and 5; the insertion of tibialis posticus upon the os calcis instead of upon the 

 naviculare as in S. aibanus and other allied genera. 



Osteological differences beyond tliose emphasized l)y ])revious writers who 

 have comjiared the skulls only, are: 16 instead of 15 dorsal vertebrae and thus 

 an additional ril) in .S. paradoxus; one less segment to the sternum and a differ- 

 ently shajied xiphisternunr, the separation of the radiale and the intermedium, 

 which are said by Dobson to be fused in S. cubanus; the distal fusion of the 

 tibia and fibula, described as separate in S. cubanus. The proximal expansion 

 of the entocuneiforme to articulate ventrally with the anterior end of the os calcis 

 is also a peculiarity not previously noted, though it may be similarly developed 

 in the Cuban animal. The number of sacrals is jirobalily the same in both 

 species, since the fifth sacral figured by Peters for the latter, seems to be a fused 

 caudal. 



Other differences are noted in the visceral anatomy, such as the number of 

 palatal rugae, which are normally eight in number, whereas nine are figiu'ed by 

 Peters for .S. cubanus; the presence of a rounded azygos lobe to the right lung 

 is not noted by Peters, and his figures of the stomach and pancreas show slight 

 differences that may be in part due to poor preservation. The Cuban Solenodon 

 appears further to differ from S. parado.uis in that the fifth cervical nerve enters 

 as a major element int( > the brachial plexus. The jilexusos, however, are subject 

 to some slight variation, and additional specimens might show that this is not a 

 constant difference. 



Among the less specialized Insectivora, characterized by tritubercular 

 molars, the two species of Solenodon possess characters that abundantly warrant 

 their separation as a family, as has been emphasized by previous writers. Un- 

 doubtedly their position is near the Centetidae, from which they differ in a 

 number of specialized characters, as the position of the penis, the chfferentiation 



