( 562 ) 



a horizontal luigle, not a vertical oue, as for instance in Locusts or Carabiils, this 

 applying more especially to the hind legs. Both legs, however, can assume an 

 obliquely upright position, more particularly in young specimens with small elytra 

 and narrow abdomen and in the long-legged American Pohjctenidac. The upper 

 surface of the horizontal hind leg of fig. In (PI. XIV.) corresponds to the inner 

 surface of the Locust leg. The small claw, in a view from above, therefore, lies 

 on top of the large one. The most asymmetrical claws have the base produced 

 into a large touth (Pl.XIIL fig. 7), which is only indicated in the less asymmetrical 

 claws (PI. XIV. fig. 13) and in the symmetrical ones. Pro.ximally to the claws, 

 on the ventral side, the extension-plate projects from the third segment, and 

 near the apex of this plate there is a pair of thin bristles as in most other 

 insects. In some species the extension-plate is supported by a projection from 

 the segment which is triangular in a lateral aspect (PI. XIII. fig. 7). In the 

 American form and one of the African ones (PL XIV. fig. 13) the sole of the 

 apical segment is armed with very stout short spines, which presumably have 

 the function of taking hold of the hair of the host, which is accomplished by 

 the basal projection of the claws of the other species (PI. XIII. fig. 7), whose tarsus 

 bears only small spines in the place where those large ones are situated. The 

 spines are not present on both sides of the segment, but form a single row, which 

 is placed near the outer (or under) side, i.e. the side of the larger claw. 



The false segmentation of the mid and hind tibiae is most pronounced in 

 the American forms and least developed in the remarkable African species depicted 

 ou PI. XiV. In this latter species there is a slight indication of a division near 

 the base and only one other pseudo-joint, which is placed in the centre (mid 

 tibia) or beyond (hind tibia). A test of the movability of the pseudo-segmented 

 tibiae of alcohol specimens proves them to be very flexible, but shows at the 

 same time that the segmentation is only partial or superficial. 



One of the most striking features of the Folijctenidae is the reduction which 

 obtains in the forelegs throughout the family, more especially in the tarsus. As 

 in the other tarsi, the American forms are also here less modified than the 

 Old- "World species. There are always three segments to the fore tarsi of adult 

 specimens. While, however, in the American forms there is no difficulty in 

 finding the parts homologous to those of the other tarsi, the bristles placed at 

 the apex of the third segment and the claws are so modified in the species of 

 the eastern hemisphere that it is not always easy to say which is claw and which 

 bristle. The end-segment of the fore tarsus of an Old-World species (PI. XIII. 

 tig. 9) bears indeed little resemblance to that of the other tarsi (PI. XIII. fig. 7). 

 The two claws of this segment are small and slender, and can only be distinguished 

 from the thick bristles placed near them by using a higher magnification, the larger 

 of the two claws being generally curved and twisted, but sometimes obtuse. 



We do uot know what the function of the modified fore leg is, but we may 

 guess at it from its structure. Three points strike us as being suggestive. The 

 first is that the anterior femur is large, being in proportion to its length wider 

 than the other femora, and carries a considerable muscle-power, which must have 

 its use. Secondly, in consequence of the shortness of the tibia and tarsus, which 

 move against the femur like the blade of a clasp-knife against the haiuUf, the 

 power of the femoral muscles is rendered more effective in a downward movement 

 of the tibia than if the til)ia and tarsi were long. Thirdly, we find in all males 

 at the apex of the tibia on the true ventral side a membranous lobe which bears 



