84 Mr. R. Shelford oz 
Though the oviposition of this Cicindelid has not been 
actually observed, there can be little doubt but that the 
adult female perforates the woody tissue of the twig and 
deposits her egg in the central core of spongy pith. The 
larva has no organs adapted for boring through wood ; 
the mouth-parts are not very different from those of the 
larvee of Cicindela spp., the legs are modified merely for 
burrowing in relatively soft and non-resisting substances, 
and may well be compared with the legs of Coprid and 
Passalid beetles, of Gryllotalpa and of Panesthiid cock- 
roaches. As already stated, the burrows are made in the 
centre of twigs and the woody tissue of the twigs is not 
attacked; the larva on hatching out from the egg has 
merely to dig out the soft pith of the twig in order to form 
for itself a cylindrical burrow, and we may presume that 
the débris is expelled from the mouth of the burrow. 
As is well known, the adult females of all species of 
Collyris are furnished with a complex genital armature, 
which, however, has never been really adequately figured 
or described. If a dried specimen of C. emarginatus be 
examined with a simple lens the gonapophyses appear to 
consist of a pair of strongly chitinised crotchets pro- 
jecting beyond the last visible tergite and of a pair of 
short down-curved spines projecting beyond the last visible 
sternite. Each crotchet is made up of three stout 
hooks directed upwards and of a much smaller hook; in 
some specimens these hooks project considerably, in others 
they are withdrawn almost entirely into the abdominal 
cavity. When the dorsal integument is removed, 1t will be 
seen that the crotchets and spines are attached to a chitinous 
tube occupying the greater part of the abdominal cavity of 
the last three segments. The whole apparatus can be 
removed bodily from the insect and after boiling in caustic 
potash mounted and examined under the microscope, when 
it will be seen that the chitinous tube is a segmented 
structure (Plate III, fig. 8), the number of the segments 
being apparently four. I have not been able to make out 
in the first (¢.¢. the most proximal) segment the number 
of sclerites composing it, but the second segment is made 
up of two lateral sclerites which meet each other in the 
mid-dorsal and mid-ventral line, also of a large spoon- 
shaped sclerite which embraces the ventral half of this 
and the succeeding segments, runs backwards to the tip of 
the abdomen and bears on its posterior margin the two 
