209 



but, in view of Verhoeff s recent contentions, a more minute account 

 of this much-discussed problem may not be out of place here. 



The observations were made on G/onicris margincda that had been 

 accustomed to a good light for many months before the breeding period, 

 some two hundred, females and about twice as many males being kept 

 in a series of glass vivaria with different kinds of soil on the bottom. 

 The whole procedure of capsule-making was watched several times with 

 the aid of a lens. This is quite necessary as the delicate nature of the 

 operation will show. 



The results entirely bear out the statements of Humbert and 

 von Schlechtendahl that the nests of Glomeris and Polydestnus are 

 l)uilt entirely of material derived from the intestine. This is also true 

 of Brachydesmus siiperus and of several Julids examined, with the re- 

 servation that in the latter case the excrement is used to line an under- 

 ground space previously hollowed out by the female. During the laying 

 period the female eats indiscriminately not only the staple food material, 

 but the underlying soil of whatever nature it may happen to be. Exa- 

 mination of the intestinal contents of a laying female also shows that it 

 consists largely of earth and varies with the nature of the substratum, 

 whereas at other times of the year the intestine contains nothing but un- 

 digested vegetable particles. Moreover, the egg-capsule is found to con- 

 sist of earth and a varying quantity of woody tissue bound together and 

 hardened by a glandular secretion; there is none of that difference in 

 consistency between the inside and outside which would be expected 

 from vom Rath 's account. 



According to my observations, the method of procedure is as fol- 

 lows. The female rolls over on its back, the head and tail being more 

 or less approximated. As soon as the egg appears through the vulva, 

 it is taken up by the neighbouring legs and passed back to the anal re- 

 gion. There it is held by cooperation of the posterior legs, and the 

 rectum everted underneath it bearing a blob of wet excrement on which 

 the egg is placed as on a pad. This soon dries and forms the foundation 

 round and over which the whole of the rest of the capsule is built. By 

 moulding movements of the very mobile rectal walls it is made into a 

 concavo-convex disc, the convexity being towards the egg as shown in 

 Verhoeff 's figure. It is thus seen that the egg is glued to one end of 

 the capsule at the first moment of formation. This end may usually be 

 recognised in finished capsules by its concave and truncated shape, 

 especially in thinwalled ones. If, however, the thin first wall is submit- 

 ted to much subsequent plastering, as often happens, the truncation may 

 be obliterated. The rest of the capsule-wall is built in such a way as to 

 leave a space between it and the egg. AVhen the basal disc is firm it is 



14 



