REVIEWS. 103 



burrow ; it selects the dead branches of the common bramble; with little labour 

 the parent bee removes the pith, usually to the length of from five to six inches ; 

 at the end she deposits the requisite quantity of food, which she closes in with a 

 substance resembling masticated leaves evidently vegetable matter ; she usually 

 forms five or six cells in one bramble-stick. The bee does not extract the whole 

 of the pith, but alternately widens and contracts the diameter of the tube, each 

 contraction marking the end of a cell. The egg is deposited on the food imme- 

 diately before closing up the cell ; it is white, oblong, and about the size and shape 

 of a caraway-seed. The larva is hatched in about eight days, and feeds about ten 

 or twelve, when it is full-grown ; it then spins a thin silken covering, and remains 

 in an inactive state until the following spring, when it undergoes its transforma- 

 tions, and appears usually in the month of June. 



" Osmia hirta burrows in wood, seldom in any other material ; the same habit 

 will be observed in Osmia aenea ; but I have observed this bee more than once 

 constructing its burrow in the mortar of walls, and sometimes in hard sandbanks. 

 Osmia aurulenta and O. bicolor are bees which commonly burrow in banks, the 

 latter being very abundant, and forming colonies in some situations , but although it 

 appears to be the natural habit of these species to construct tunnels in hard banks, 

 with great labour and untiring perseverance, still we find them at times exhibiting 

 an amount of sagacity, and a degree of knowledge, that at once dispels the idea of 

 their actions being the result of a mere blind instinct, impelling them in one un- 

 deviating course. A moment's consideration will suffice to call to mind many 

 tunnels and tubes ready formed, which would appear to be admirably adapted for 

 the purposes of the bee for instance, the straws of a thatch, and many reeds ; 

 and what could be more admirably adapted to their requirements than the tubes of 

 many shells ? So thinks the bee. 0. aurulenta and 0. bicolor both select the 

 shells of Helix hortensis and H. nemoralis ; the shells of these snails are, of course, 

 very abundant, and lie half-hidden beneath grass, mosses, and plants ; the bees 

 finding them in such situations, dispense with their accustomed labour, and take 

 possession of the deserted shells. The number of cells varies according to the 

 length of the whorl of the shell selected, the usual number being four, but in some 

 instances they construct five or six, commencing at the end of the whorl ; a suit- 

 able supply of pollen and honey is collected, an egg deposited, and a partition 

 formed of abraded vegetable matter ; the process is repeated until the requisite 

 number is formed, when the whole is most carefully protected by closing up the 

 entrance with small pellets of clay, sticks, and pebbles ; these are firmly cemented 

 together with some glutinous matter, and the bee has finished her task. 



" We will now observe the intelligence of the bee under different circumstances. 

 She has selected the adult shell of Helix aspersa ; the whorl of this species is much 

 larger in diameter than that of H. nemoralis or H. hortensis too wide, in fact, 

 for a single cell ; our little architect, never at a loss, readily adapts it to her pur- 

 pose by forming two cells side by side, and as she advances towards the entrance 

 of the whorl, it becomes too wide even for this contrivance ; here let us admire the 

 ingenuity of this little creature ; she constructs a couple of cells transversely ! And 

 this is the little animal which has been so blindly slandered as being a mere 

 machine. 



u I will take this opportunity of correcting a very widely- diffused error, which 

 appears to have originated with Reaumur ; or, if his account of the development of 

 Xylocopa be correct, it differs from that of every wood-boring bee which inhabits 

 this country. He says 1 When the larva assumes the pupa, it is placed in its 

 cell with its head downwards ; a very wise precaution, for thus it is prevented, 

 when it has attained to its perfect state and is eager to emerge into day, from 

 making its way out upwards, and disturbing the tenants of the superincumbent 

 cells, who, being of later date each than its neighbour below stairs, are not yet 

 quite ready to go into public.' Mr. Kirby also quotes from a letter by the Kev. 

 George Ashby, who, after describing the nest of Megachile centuncularis, says 

 4 The lowest and first born passes out through the bottom of its own (lowest) cell, 

 and so escapes without disturbing the rest, who are not yet ready to emigrate.' 

 All such conclusions originate in conjecture. In the case of Osmia aurulenta con- 

 structing her cells in the spiral tube of a snail's shell, where is the possibility of 



