108 MICROTIN^ 



over the actual summits of the posterior loop and five closed 

 triangles or not is difficult to say. If enamel does occur on the 

 apices of these parts it must be very thin indeed ; personally I 

 am inclined to think that it is absent and that the teeth of 

 Arvicola (and of many other voles) afford a parallel to the 

 Murinse, in which Hensel ^ long ago observed that apical enamel is 

 lacking. The reduction of the apical enamel is a specialization, 

 the significance of which will become apparent a little later. 

 With the apical enamel in such a reduced condition one could 

 hardly expect to find any of the coronal tubercles very clearly 

 differentiated, and those of the middle row have indeed become 

 intimately blended with those of the outer and inner rows, so that 

 no obviously independent and therefore readily recognizable 

 tubercles of the middle row remain. Nevertheless, traces of this 

 row are to be seen in the form of the closed triangles and it would 

 appear that the bases of both the inner and the outer triangles 

 are formed by the median tubercles ; the junction of the median 

 tubercle with either the outer or the inner tubercle which forms the 

 peripheral part of each salient angle is marked by a slight con- 

 striction; that constriction in the unworn or slightly worn m^ 

 affects both the posterior and anterior sides of each salient angle, 

 forming on each a vertical furrow; the furrow dies out rapidly 

 on the anterior surface, but on the posterior surface it frequently 

 persists through all stages of growth until the animal becomes 

 aged (Fig. 59). A much clearer, less disputable memorial of the 

 primitive brachyodont and tubercular crown is afforded, however, 

 by a series of four transverse notches or passes, crossing the 

 isthmuses that connect the posterior loop and the first four tri- 

 angles one with another, thus pixtting the outer and the inner 

 re-entrant folds in open though narrow connection with each other. 

 These passes stand in relation to the rest of the tooth exactly as 

 " wind gaps " or " hanging- valleys " do to a mountain range. 



The summit of the anterior loop is formed by a large irregular 

 postero-median tubercular eminence from which a number of 

 furrows or little valleys radiate, descending obliquely to the outer, 

 anterior, and inner borders of the loop. Three of these valleys, 

 one external, one internal, and one median and longitudinal, are 

 comparatively large and are very constantly present ; of them the 

 median valley is the largest and most persistent, wearing down 

 to form an ephemeral enamel islet which is shown in Forsyth 

 Major's figures. The valleys are separated from each other 

 by crests or digitations, which appear as peripheral prolongations 

 from the posterior tubercular mass. As one would expect, these 

 vanishing elements of the tooth show, as regards their details, a 

 not inconsiderable amount of variation in different individuals or 

 species. Thus in an unworn m^ of the Pleistocene A. abbotti 

 (Fig. 58c) the whole tubercular cap of the loop is cut up by the three 

 valleys into four elevated crests, whereas in a recent Arvicola 

 ^ Hensel, Zeitschr. deutsch. geol. Ges., 8, p. 283, 1856. 



I 



