1889.] MR. O. THOMAS ON A NEW MUNGOOSE. 623 



group. Tlie premolars are all rather longer both vertically and 

 horizontalljs and the cusps higher and more distinct. ^, t^, and m^ 

 are on the whole similar to those of the allied species in their form 

 and relative proportions, but are all markedly larger (compare the 

 dimensions given below with those on p. 78 of my former paper). 



But it is by the characters of the lower molars that the new species 

 may be most readily recognized. As to size simply, the length of the 

 two molars combined is in JFT. grandis 17 mm., while in the largest 

 of a considerable series of H. albkaudatus this combined length only 

 attains to 14-4 mm., its ordinary amount being abo\it 13 mm. In 

 structure, as will be seen by Plate LXII. figs. 4 and 5, these teeth in 

 H. grandis are more complicated than in the older known form ; 

 in ml there is not so much difference, except that the cusps and 

 hollows are more marked, and the ridge running round the posterior 

 half of the tooth, or " talon," is much sharper and better defined. In 

 n?, firstly, the two antero-internal cusps, the paracone and metacone 

 of Mr. Oshoru's nomenclature of tooth-cusps\ which have coalesced 

 in H. albkaudatus, are sharply and distinctly separated from one 

 another, so that the primitive anterior triangle forming the blade of 

 the tooth is as well defined as in m^ ; secondly, in the talon, the extra 

 median external cusp characteristic of H. alhicaudatus (see p. 76, 

 fig. 1 of the nionograi)h) is duplicated in H. grandis, being supple- 

 mented by a second cusp on its internal slope ; then the posterior 

 edge of tlve talon is more developed, crenulated, and with its centre 

 sharply and prominently notched in the middle line. As a result 

 of this increase in complexity, the size of m'^ as compared to m^ is 

 much increased ; for while in H. alhicaudatus its length is never more 

 than from 80 to 83 per cent, of that of the latter tooth, in H. grandis 

 the two teeth are of i)racticajly the same length, m- being no less 

 than 96 per cent, as long as m^ 



In view now of the extreme constancy of the teeth in the present 

 group, both in size and structure, I feel that it would be impossible 

 to refer the Cambridge skeleton to H. alhicaudatus, and can only 

 describe it as new, trusting at the same time that its external cha- 

 racters will not long remain unknown. 



As will be seen from the figure (Plate LXII. fig. 1), the skull of 

 H. grandis is characterized by its slenderness and by the great length 

 of its facial as compared to its cranial portion. 



In the skeleton both H. alhicaudatus and H. grandis are remark- 

 able for not possessing an entepicondyloid foramen to the humerus, 

 the bony bar usually closing in this foramen being unossified. All 

 other true Mungooses have this foramen closed in by bone, with the 

 exception of the aberrant genera Galidiu, Galidictis, and sometimes 

 Suricata. This fact, combined with the addition of a second allied 

 species, gives increased validity to the group " Ichneumia," recog- 

 nized as a subgenus in my monograph, but which may possibly in 

 the future have to be admitted as a distinct genus. 



Dimensions of the type specimen : — 



Skull. Length (from back of condyle to gnathion) 112 mm. ; basal 

 ' Amer.Nat. 1888,p. 1072. 



