32 THE VOYAGE OF H.M.S. CHALLENGER. 



On the middle of it is a tubercle whose apex looks backwards and downwards, and is 

 pierced by a somewhat transverse perforation. All the species, and both sexes, are fur- 

 nished with this tubercle, though in some specimens it is scarcely visible, from the seg- 

 ment beina; retracted below the mesosternum. 



What may be the nature of this tubercle, and its use, must remain uncertain until ob- 

 servations have been made on the living animal. There is nothing, so far as I am aware, 

 corresponding to it in any other true insect. It is possible that its homologies may be 

 found in the "ventral tube" of the CoUembola. If such be the case the use may be to 

 secure attachment to the animals on whose juices Halohates dowhthsfi ieeds, but it is some- 

 what idle to speculate on the origin and use till actual observations have thrown some light 

 on the nature of the organ. (See note in the Appendix.) 



The sixth segment is longer (sometimes much longer) than the others, and the sides 

 are produced obliquely backwards to clasp the sides of the first genital segment. In all 

 the seo-ments the hind margins on the ventral surface are nearer the base of the abdomen 

 than the hind mars;ins on the dorsal surface. 



o 



TJie Ahdomen of the Female. 



On the dorsal surface it is, on the whole, similar to the male abdomen, with the 

 exception of the differences in the structure of the pleuron mentioned in the description 

 of the thorax. 



On the ventral surface (PL I. fig. 7 ? ah. h.) all the six segments are ring-like. The 

 sixth is prolonged a little at the sides. The first segment is furnished with a tubercle 

 as in the male. 



The Genital Segments of the Male. 



The First Segment is, on the dorsal surface, ring-like, and similar to the last abdominal 

 segment, but not so broad. Below it is nearly as long as (or even sometimes longer 

 than) the whole of the abdominal segments taken together, whUe in shape it is trans- 

 versely oblong or nearly square. 



The Second Segment is almost hidden, both above and below, by the first, the only 

 parts usually visible being a small, crescent-shaped, or semilunar plate above, while be- 

 low all that can be seen is a long horn-like process on each side. 



On dissection the true form of the segment is seen to be as follows : — 



Above (PI. III. figs. 20 and 21), transversely oblong, about four times as broad as long ; 

 hind margin much prolonged in the middle, the prolongation (the apex of which is the 

 only part visible without dissection) presenting a rounded outline posteriorly. 



Below, the middle of the hind margin is concave, giving rise near each side to a long 

 horn-like process which lies along the side of the next segment, and reaches to or beyond 



