142 INSECTIVOEA. 



the occipital region above it. The lambdoidal ridge is well marked, and a feeble 

 sagittal crest is prolonged forwards from it. The post-glenoid process is perforated 

 by a well-defined foramen close to its anterior or inferior margin, directed back- 

 wards, upwards, and slightly inwards, and on its inner margin there is a small, 

 inwardly projecting process. In Soriculus there is a similar foramen in the same 

 position, whereas in Crocidura and Facliyura it is always placed internally at the 

 base of the glenoid process. The condyle of the mandible consists of two distinct 

 articular facets separated from each other by a deep notch. The lower articular 

 surface is a transverse process projecting inwards, considerably beyond the upper 

 division, which is in a line with the coronoid process, and its facet looks downwards, 

 backwards, and inwards. The upper division is narrow, obliquely transverse, look- 

 ing backwards and inwards. It is separated by a very wide notch from the 

 coronoid. The lowermost of these surfaces is received into the articular concavity 

 defined by the post-glenoid process, and the uppermost is applied to the articular 

 surface occurring on the under aspect of the lateral ridge of the skull that corre- 

 sponds to the zygomatic ridge of the squamous, so that the summit of the coronoid is 

 on a level with the upper surface of the skull at the frontal depression. In Crocidura 

 this tendency to the division of the condyle shows itself to a certain degree, but not 

 to the marked extent that occurs in this form and in Anurosorex, while in Talpa 

 there is no trace of it. The palate projects behind the last molar, and is defined by 

 a well-marked, antero-posteriorly concave ridge, and there are no imperfections of 

 ossification. Looking at the skull sideways, there is a considerable depression in the 

 inter-orbital region, before which the facial portion is first shghtly convex and 

 then straight, flattened from above. 



The teeth, as in Nectogale, are white,^ and 28 in number, as in Crocidura, The 

 front incisors, rather widely apart at their bases, bend so much inwards and slightly 

 forwards that they touch each other, defining a triangular space between them. 

 They are curved forwards, downwards, and backwards, round at their bases, but with 

 sharp, slightly, laterally compressed points. Some little distance above the tips, on 

 the inner side of the teeth, there is a small process developed on each, as in Crossopus 

 and Sorex, but which is absent in Nectogale, and it is at this point that the two 

 teeth are in contact. The posterior basal process is not strongly developed, and is 

 smaller than any of the smaU lateral teeth. It is shortly conical, with flattened 

 sides to its crown, which has a longitudinal cutting ridge with a visible cingulum. 

 The lateral teeth are of moderate size, but with little vertical extension. They are 

 more or less oval from before backwards, and each has the cingulum strongly 

 marked for lateral teeth, and it is so prominent on the inner margin of the second 

 tooth as to produce the appearance of a cusp, the cingulum being separated 

 from the crown of the tooth by a deep groove. The crown is elongately oval and 



* Tomes, writing on the type Bpecimen of Crossopus Mmalayicus, Gray, says : " The teeth of this example are those 

 of a restricted Sorex, and I feel no hesitation in saying, after a careful examination, that they have been introduced by 

 the stuffer."— Ann. and Mag. Nat. Hist. 2nd ser. xvii (1856), p. 25. As the two spirit specimens, however, of this shrew 

 which have come under my observation have both pure white teeth, there can now be no doubt that Pr. Gray was 

 quite correct in describing the teeth of his type as white. 



