lUlf^.J F. II. liKAVELV: I'assaliddf of the World. 3 



he situated on the inner side of th.iu. That tliev still mark the lateral extremities of the 

 free surface of the clypeus in these cases is shown by the fact that they are joined by the 

 sides of the opposed surface of the clypeus. The accompanying diagrams (fig. i) will 

 help to illustrate the various stages in the suppression of the clypeus. With them may be 

 compared figs. V, 1 1-15 ami vi, 6-i i on pp. 34 and 45, illu-strating in a less diagrammatic 

 fashion its suppression in Vencs and Plati/rrrres and in PaxUlus re.spectively. 



The presence, in most .\merican and African as well as Indo-Australian Passaliilae. of 

 definite frontal ridges and of central, inner and outer tubercles is .sufficient indication of the 

 morphological importance of these structures. The outer tubercles, as has ju.st been shown, 

 are formed m the more primitive species by a fusion of the frons and clypeus about the 

 posterior angles of the latter. The central tubercle is normally situated at the angle frumed 

 bv the union of the two frontal ridges which appear, from the evidence afforded by certain 

 Indo-Australian species, to mark the position of part of the line .separating; the frons from 

 the vertex. The inner tubercles appear to mark the point at which this line, usually quite 

 imaginary for a short distance beyond this point, bends more or less back.vards towards a 

 .suture a little to the inner side of the supraorbital ridges, in which it commonly ends (.see 

 Gravely, 1914c, p. 184). This suture is ea.sily seen in imperfectly hardened specimens of 

 various groups, but is less distinct in those which have become thoroughly hard and black. 



The homologies of the tubercles found in genera such as Leptauliu- and Tarquinins, in 

 which more than one pair of tubercles is situated on the anterior margin of the head, with 

 those found in genera with only one pair so situated, are more difficult to determine than 

 appeared to be the case when Indo-Australian forms only were under consideration. 



A comparison of Tanj'iinnis with Pseudefisphenus leaves little room for doubt that the 

 more mediallv situated pair of tubercles in the former is homologuous with the pair of inner 

 tubercles of the latter (.see Gravely, 1914c, pp- 328-329), and in the ab,sence of any evidence 

 to the contrary it would be natural to assume the same homologies for the tubercles of 

 Leptaulax, and of genera from the Ethiopian Region with similarly arranged tubercles. 



But in the case of Ethiopian genera there is strong evidence in favour of different 

 homologies. This is best illustrated by reference to the group characteristic of Madagascar, a 

 group which is separated from the group found on tlie mainland of Africa by the jjossession 

 of a pair of tubercles situated on the frons in the angle made by the fronto-vertical suture 

 with the anterior margin of the head (see below pp. 68-69). 



The most primitive of the Malaga.sy genera appears to be a new one descril>ed below 

 (pp. 69-70, fig. viii. 1) under the name MolagasnlKS. In this genus the free surface of the 

 clypeus, though almost vertical, is fidly exposed. The outer tubercles, with a pair of well 

 developed inner tubercles close behind them, project above it and are separate:! by a space 

 equal to scarcelv as nmch as two-thirds of its breadth. In Soloioci/chis. the next genus of 

 the series, the inner tubercles are also well developed and are situated further back from 

 the outer tubercles: and an additional ]iair of marginal tubercles is present immediately 

 above the ends of the clypeus. The outer tubercles, as pointed out above (p. 2). are formed 

 in the first instance about the posterior angles of the clypeus ; and their closer approximation 

 one towards the other in forms such as MuhufasnlKS leaves space in wliich the tendency 

 towards tubercle-formation at this point can manifest itself over again. Presumably, therefore. 



D 2 



