ig2i.] N. Annandai.e & R. B. S. Seweu, : Vivipara. 263 



organic connection inter se. The lumen of the tubule has no 

 special lining. At this stage ducts, perhaps of a temporary 

 nature, can be detected in sections, but they form mere gaps in 

 the epithelium, leading out from ill-defined spaces beneath it 

 (fig. 2, pi. iii). Later the gland-cells become greatly enlarged and 

 elongate and open direct on the external surface ; while tlie tubular 

 character of the gland disappears. 



The calcareous matter secreted by the calciferous glands is 

 apparently derived from concretions scattered through the connec- 

 tive tissue of the mantle and foot but congregated in large num- 

 bers immediatel}^ beneath the glands at times of active growth. 



The secretion of the nacreous layers, probably affected by 

 unicellular glands scattered over the whole of the upper part of the 

 mantle, is not discussed here. 



Function of the different parts of the Marginal Region in 

 reference to the Shell. 



We are now in a position to discuss the function of the edge 

 of the mantle in relation to the ornamentation of the shell. 

 The first structures in the soft parts to be considered in this con- 

 nection are the marginal processes. They are not organs of 

 secretion but, at any rate when hypertrophied as in V . oxytropis, 

 perhaps accessory breathing organs. The}-' are closely correlated 

 in position with both the periostracal sculpture, the colour-pattern 

 and the sculpture of the test. The connection between them and 

 the periostracal sculpture can be traced without diflftculty. They 

 mould this sculpture, apparently as erectile rather than muscular 

 organs. The horny matter that will form- the tliin outer cuticle of 

 the shell is poured in a liquid condition into the supramarginal 

 groove, in which it is kept in motion by the cilia of the epidermal 

 cells. It runs down the longitudinal grooves on the external sur- 

 face of the processes and by them is deposited on the edge of the 

 lip of the shell, over which they are retroverted as it consolidates. 

 The three primarj' rows of chaetae are thus formed by the three 

 primary processes, and in such forms as Vivipara dissimilis in which 

 there are more than three rows, those of the secondary rows by the 

 best developed of the secondary processes. The upright form of 

 the chaetae is due to the greater length of these processes. This 

 enables them to project well beyond the lip and be curled up over 

 it. The hooked tips of the chaetae are due to the fact that the 

 tips of the processes are curved at the moment of formation of the 

 chaetae. The fine subsidiary ridges of the periostracum, which 

 when first formed project horizontally from the edge of the lip as 

 fine hairs, are similarly produced bj^ the subsidiary processes, their 

 orientation being due to the fact that the moulding processes are 

 short and cannot be curved upwards over the lip. 



In those shells which like Vivipara bengalensis have a pattern 

 of dark spiral bands in the periostracum, the dark pigment is 

 also poured out along the grooves on the external surface of 



