D. J. SCOURFIELD ON THE EFHIPPIUM OF BOSMINA. 53 



by a comparison of Fig. 8 with Figs. 1 and 4, tliere is a rather 

 greater distance between the lateral bands and the back of the 

 shell in B. lilljehorgi {V) than there is in B. loiigirostris ; and more 

 important still, the lower ends of bands approach very much 

 nearer to the posterior ventral angle of the shell in the former 

 than in the latter species. I may mention here that the lateral 

 bands are not always so evenly curved as shown in the figures, 

 but that they sometimes exhibit rather abrupt bends — as if in 

 their formation the deposition of chitin had not followed one 

 line of cells the whole way, but had changed from one series 

 to another. 



As regards the function of the lateral bands, I would suggest 

 that, in addition to strengthening the ephippium, as they evidently 

 do from their position almost directly over the Qgg., they may also 

 help to keep the free edges of the valves more closely in contact 

 than might otherwise be the case. Such a result would certainly 

 follow if the ends of the bands, by their elasticity, possessed the 

 power of pressing inwards and carrying the free margins of 

 the valves with them. I do not know whether such a tendency 

 exists, but it is at least very probable. 



In addition to having the lateral bands very strongly marked, 

 B. lilljehorgi (?) also shows the thickening along the back as a 

 distinct ridge (Fig. 9). That the dorsal thickening is really in 

 the form of a sharply defined band can, however, be also seen in 

 B. longirostris, when views can be obtained either from the front 

 (Fig. 3) or back (Fig. 2). 



There is still one other point requiring elucidation, namely, 

 How is it that the ventral portions of the valves become detached 

 when the shell is to form an ephippium ? By very careful 

 scrutiny of a female carrying an ephippium and winter egg 

 (Figs. 4 and 8), it can be demonstrated that there is a line of 

 weakness, marked b}'- a faint doubly-contoured line on each valve, 

 already formed in the exact position where the anterior portion 

 will break away. This line of weakness can be developed into a 

 crack, at least for some portion of its length from the anterior 

 end, by applying pressure. The edges of the crack are quite 

 smooth. The line of weakness does not cross any of the ordinary 

 shell markings on the valve, but, from the relation of the latter 

 to one another on each side of the line, it is evident that a 

 considerable amount of alteration of the original shell sculpture 



