180 CAENIVORA. 



tip being abraded. The dark-coloured example is 13-30 incbes from the muzzle to 

 the root of the tail, and the tail is 9-20 inches long. 



In the type, the sutures of the skull have all disappeared, and yet the orbit 

 is incomplete ; but, from the circumstances in which the animal lived, it would be 

 premature to accept the character of the orbit as determined by the skull of this 

 captive Mungoose. The skull has a strong resemblance to the skull of S. palUdus, 

 to which and H. mirojnmctatus it is very closely allied, and for which it might even 

 be mistaken if the orbits were complete. The notched character of the posterior 

 border of the palate is only a modification of the arched form which occurs in 

 this genus, and which may be traced in a series of specimens leading into and ending 

 in a straight, transverse border as in S. cmropunctatus. But the form of the posterior 

 border of the palate is so variable that I would hesitate to resort to it even for 

 specific characters. The last inferior molar is markedly tricuspidate. 



The vertebral formula is C. 7, D. 13, L. 7, S. 3, C. 26 = 56. The axis is 

 strongly ridged inferiorly, the ridge terminating posteriorly in a tubercle which is 

 partially divided in the succeeding vertebrae, but both cease on the sixth segment. 

 The spinous processes of the dorsal vertebrae are proportionally longer and more 

 backwardly directed than in the much smaller H. aitropunctatus. Anapophyses 

 begin on the tenth and metapopliyses on the eleventh dorsal vertebra, the former 

 disappearing on the penultimate lumbar. In the caudals, the transverse processes 

 become gradually transferred from the middle to the posterior ends of the bodies, 

 which latter position they occupy in the sixth vertebra, on which the secondary 

 transverse process makes its appearance, finding its fullest development in the 

 tenth caudal and disappearing later than the true transverse processes. The latter 

 are strong and well developed to the sixth, and bear a small process on their upper 

 surface, near their free ends, which first shows itself on the third caudal, and gradu- 

 ally approaches the base of the transverse process as it shortens, till at last on the 

 eighth caudal it is transferred to the posterior end of the centrum and occupies 

 the position of an anapophysis and can be detected to the tenth vertebra. The 

 zygapophysial facets cease on the sixth, but processes serially homologous with 

 them can be detected as far as the seventeenth vertebra. The neural canal ceases at 

 the sixth segment. Haemapophysial nodules begin on the seventh caudal, and occur 

 to near the extremity of the tail, but it is probable that the anterior chevron bones 

 have been lost in the preparation of the skeleton. The sternum has seven pieces, in 

 addition to the manubrium which is essentially Herpestine, but with the anterior 

 projection not so long as in H. smithii. The limb bones are generically the same 

 as in other Uerpestes. 



This species, as far as is known, is peculiar to Ceylon. 



