208 MODERN CLASSIFICATION OF INSECTS. 
which it first’ stings to death. The account of the proceedings of 
this species was communicated to the latter by M. Cossigni, who had 
observed it in the Isle of Bourbon. 
The second subfamily Pompilides has the collar either transversely 
or longitudinally square, with the abdomen more or less oval, and at- 
tached to the thorax by a very short peduncle. The legs are very 
long. The fore wings have two or three perfect submarginal cells, 
and another commenced at the tip of the wings. The maxillary 
palpi are more robust than in Ammophila ( fg. 83. 17.), and the labial 
lobes much shorter (fig. 83. 18. Pompilus). These sand-wasps are 
amongst the most ferocious of the insect tribes, and some of the large 
exotic species of Pepsis are the largest of known Hymenoptera; many 
of them have their wings black, or varied, and shining with the most 
brilliant metallic tints. The species of the British genus Pompilus 
are exceedingly active, running amongst grass, &c. in hot sandy situ- 
ations with great agility, and with their wings in constant agitation: 
their long legs give them much the appearance of spiders ; and, indeed, 
some of the species provision their cells (which they burrow in the 
sand) with these insects, which they first sting to death. P. niger, how- 
ever, attacks a small Lepidopterous larva. (Shuckard, Hssay, p. 52.) 
Some of the species of this genus have the legs unarmed, as P. bifas- 
ciatus and petiolatus: these Mr. Shuckard considers to nidificate in 
wood (Trans. Ent. Soc. vol.i. p.56.); although, never having ob- 
served any species with these habits, he considers it would necessarily 
be such as are without the ciliation of the tarsi. St. Fargeau also 
states that some species make their nests in burrows, which they dis- 
cover already formed in wood, without, however, naming such species. 
We have already seen that the absence of ciliz does not necessarily 
indicate wood burrowing habits, according to Mr. Shuckard’s theory, 
neither does it indicate parasitic habits in these insects, according to 
St. Fargeau’s, as I have proved by observing the proceedings of P. 
petiolatus in the act of carrying its prey, which consisted of a large 
dead spider *; and ofa species allied to P. petiolatus, which I captured 
in the act of burrowing in the sand of the Drachenfels, although fur- 
nished with simple tarsi. (Westwood, in Ann. Soc. Ent. de France, 
1836, p. 300.) 
* T have more recently observed P. petiolatus in the act of forming its cell in the 
chinks of the walls of my residence at Hammersmith. It employs a particular kind 
of clayey soil, which it carefully selects for this purpose. 
