210 



turned over through nearly a right angle. At the same time the rectum 

 is drawn in and again everted with a fresh blob of excrement. This is 

 added to the basal disc at the angle named. This process is repeated 

 round the rim of the disc until the egg in the middle is hedged in by a 

 low w^all. The height of this wall is increased by an exactly similar pro- 

 cess, fresh pieces being made to overlap the preceding ones like tiles in 

 the roofing of a house. By appropriate gradation of angles the rounded 

 end of the capsule is managed in the same way, and if it is taken from 

 the grasp of the female at this stage, traces of the overlapping may be 

 seen all over it. Usually, however, the whole is plastered over and over 

 again in an irregular manner, the thickness of the wall depending on 

 the extent to which this is done, which, in turn, perhaps depends on the 

 intestinal content at the time being. This secondary plastering is that 

 described by vom Rath in the excerpt (j^uoted by Verhoeff. It is 

 done much more rapidly than the actual building which is slow and 

 very delicate work. It should be mentioned that the blob of excrement 

 as first extruded appears largely liquid, but before use much of the li- 

 quid may be seen to filter back into the rectum. 



As to the twin chambers, I have only seen the actual building of 

 the latter half of the second chamber. Though this was done in exactly 

 the same way as the single chambers, the observation threw no light on 

 the cause of twin forms, and I am only able to offer an explanation 

 obtained by indirect means. The fact that for three years I had not 

 found a single double-chambered capsule among many thousands of 

 eggs laid in my vessels, whereas I found them commonly out of doors, 

 led me to suspect that I was not providing natural conditions in the vi- 

 varia. I noticed also that in nature the females receded into the loose 

 earth to lay, while in captivity the laying was done on the surface. I, 

 accordingly, gave them a greater depth of looser soil wàtli the result 

 that the double capsules soon began to appear in fair proportions, the 

 number increasing towards the end of the laying season. From this I 

 concluded that double-chambered capsules are only produced by females 

 lying in among a number of capsules previously made. One of these 

 might easily roll into the sphere of activity in the anal region , and, 

 sticking to the capsule actually in process of formation, be incorporated 

 with it so as to form a twin. The double capsule which I saw^ being 

 made was »biskuitförmig« at first, but this shape was obscured by the 

 secondary plastering. This view of the origin of twin forms is corro- 

 borated by the fact that the larvae they contain are never of the same 

 age and may differ by several days. The empty chamber figured by 

 Verhoeff may have been one from which the older larva had departed. 



Though the foregoing account deals with Glomeris marginata, it 



