158 A NATURALIST IN BORNEO 



The last ventral plate of the abdomen of Collyris emar- 

 ginata has on its free terminal margin a pair of little 

 decurved spines, and I believe that the function of 

 these is to guide the egg safely through the hole 

 bored in the wood on to the central cylinder of pith. 

 Without these guides the beetle would be very liable 

 to make a bad shot at the hole which she had bored, 

 and lay the egg on the outside of the twig, where it 

 would be exposed to all sorts of dangers, or it might 

 fall to the ground. It is interesting to observe that 

 these spines occur in other Cicinddidce which are 

 known to be arboreal. They are well developed in 

 Therates labiata from Amboina, which Wallace states 

 to be arboreal ; very minute and perhaps functionless 

 in some Bornean species of Therates which are more 

 terrestrial than arboreal in their habits, and they are 

 present in the arboreal Australian genus Distypsidera. 



When the larva of Collyris emarginata hatches out it 

 must proceed to form its burrow by digging out the 

 soft pith of the twig, and we find that the front legs 

 are well adapted for this purpose. In the larvae of 

 other Cicindelidce, which live in burrows in the soil, 

 the legs are long and slender, and act as stays to 

 prop the larvae up at the top of its lair ; but in 

 Collyris larvae the fore-legs are flattened and shortened 

 to form efficient digging instruments. The thigh or 

 femur is not much longer than the plate-like first 

 joint of the leg, the coxa: it is broadest at the distal 

 end, and is produced at one point into a strong, flat- 

 tened tooth, bearing in its turn smaller teeth. The 

 next two joints, the shank or tibia and the tarsus, 

 are extremely short, and have little teeth on the outer 

 aspect. By means of oar-like movements of these 



