HABITATIONS OF INSECTS. 283 



that the first egg will become a grub, and consequently a perfect bee, 

 many days before the last. What then becomes of it? you will ask. It 

 is impossible that it should make its escape through eleven superincumbent 

 cells without destroying the immature tenants ; and it seems equally 

 impossible that it should remain patiently in confinement below them until 

 they are all disclosed. This dilemma our heaven-tought architect has 

 provided against. With forethought never enough to be admired she has 

 not constructed het tunnel with one opening only, but at the further end 

 has pierced another orifice, a kind of back door, through which the insects 

 produced by the first-laid eggs successively emerge into day. In fact, all 

 the young bees, even the uppermost, go out by this road ; for, by an 

 exquisite instinct, each grub, when about to become a pupa, places itself 

 in its cell with its head downwards, and thus is necessitated, when arrived 

 at its last state, to pierce its cell in this direction.^ 



Ceratlna alhilahris of Spinola, who has given an interesting account of 

 its manners, forms its cell upon the general plan of the bee just described, 

 but, more economical of labor, chooses a branch of briar or bramble, in 

 the pith of which she excavates a canal about a foot long, and one 

 line, or sometimes more, in diameter, with from eight to twelve cells sepa- 

 rated from each other by partitions of particles of pith glued together^ ; 

 and from the dead sticks of the same plants, in which they had formed 

 their cells in a similar way, MM. Dufour and Perris have bred in the sandy 

 district of the Landes in the south-west of France not fewer than twelve 

 distinct species of wild bees and other Hymenoptera, namely, four species 

 of Osmia, two of Ceratina, three of Odynerus, two of Solenius, and 

 Trypoxylon Jigulus, besides fifteen species of parasitic Hymenoptera of 

 the genera Stelis, Prosopis, Ichneumon, Chrysis, &£c., making in all twenty- 

 seven species of hymenopterous insects obtained from this prolific habitat, 

 for which, too, they were indebted for very rare insects, which they had 

 never before met with.^ Mr. Thwaites has been also very successful in 

 obtaining Hymenoptera from this source, having bred from dead bramble 

 sticks found near Bristol Hylceus annuloris and a new species, Ceratina 

 albilahris Sp. cyanea K., Osmia leucomelana, Epipone levipes, Cemonus 

 unicolor, Spilomena Troglodytes, a new species of Trypoxylon, and an 

 unascertained one of Cladius, besides seven species of parasitic Hyme- 

 noptera, including Stelis minuta, Chrysis cyanea, Hedychrum aiiratum, 

 Cryptus hellosus, and three other Ichneumonidae, in all, sixteen species. — 

 Crabro tibialis, which M. Perris says is parasitic on Hymenoptera residing 

 in bramble-sticks (Ann. Soc. Ent. de France, ix. 407.), has been also 

 found in this habitat near Bristol by Thomas Lighton, Esq. ' 



Such are the curious habitations of the carpenter bees and their ana- 

 logues. Next I shall introduce you to the not less interesting structures of 

 another group of bees, which carry on the trade of m,asons {Megachile 

 muraria), building their solid houses solely of artificial stone. The first 

 step of the mother bee is to fix upon a proper situation for the future man- 

 sion of her offspring. For this she usually selects an angle, sheltered by 

 any projection, on the south side of a stone wall. Her next care is to 

 provide materials for the structure. The chief of these is sand, which she 



' Reaum. vi. 39—52. Mm. Ap. Angl. i. 189. Aphis. **. a. 2. /?. 



- Ann. du Mus. x. 236. > Ann. Soc. Ent. de France, ix. 1 — 53. 



