I92I.] N. Annandale & R. B. S. SE\vei<l : Vivipara. 249 



also probablj' in Margarya, in which, however, I have examined 

 onl}' adult individuals. In the first three genera, and probably in 

 all Viviparidae, it bears at this period three distinct finger-shaped 

 processes,' one situated just above the snout near the middle of 

 the edge, the other two to the right of this point. The left process, 

 which marks a very important point in the orientation of the 

 ornamentation of the shell, I shall call the peripheral process. It 

 moves along, in the expansion of the animal, under the most pro- 

 minent line of the bodj'-whorl of the shell and is usually, but not 

 always, a little longer than the others. The two processes to the 

 right of the peripheral ])rocess ma}' be called the first and the second 

 process, the former Ij'ing the furthest to the right. These three 

 processes correspond in position with the three primarj' rows of 

 chaetae on the young shell (p. 250, fig. 13) and bear the same nota- 

 tion in m^- figures. The peripheral process, though usuallj- the most 

 conspicuous and the most important m the future history of the 

 shell, is morphologically the youngest, while the first process, the 

 least important of the three from this point of view, is the oldest. 

 These three processes I call the three primary processes. Thev 

 were first observed and figured in the embryo of Vivipara vivipara 

 by Leydig,^ but are omitted in the figures of more recent authors. 

 I find them just as well developed in the fully formed embryo of 

 the European V. contecta as they are in Indian species. 



In most other Viviparidae examined, at least traces of other 

 processes between and to the left of the primary three can be de- 

 tected at the same period. In V. bengalensis (fig. 12) they are 

 small and inconspicuous, but in V. dissiniilis, another common 

 Indian species, four secondary PROCESSES can be easily detected 

 in fresh material, two of them being longer than the other two. 

 The two longest secondary processes are situated one immediately 

 to the right of the second primary process and the other to the 

 right of the third. They correspond in position with the two secon- 

 dary rows of chaetae on the embryonic shell (p. 248, fig. 12). The 

 condition is similar in V. oxytropis, but in Lecythoconcha lecythis 

 only the three primary processes can be detected as such, even in 

 fresh material. 



Even the seven processes of V. dissimilis and V. oxytropis are 

 not all that actually exist, for between each pair associated witli 

 lines of chaetae two or three other minute projections occur, but 

 can only be detected as projections if the mantle be examined in a 

 fulh' expanded condition. These minute or tertiary processes 

 correspond in position with the minute serrated ridges on the peri- 

 ostracum of the embryonic shell (p. 245, fig. 10). Both they and the 

 secondary processes are probably present and functional, though 

 often difficult to detect, in all Viviparidae. 



' It mav be fixed in this condition by being subjected to gentle pressure be- 

 tween two glass slides as soon as it is removed, and treated with corrosive acetic 

 solution while under pressure. 



- I.eydig, Zeits.f. wiis. Zool. II, pi. xi. fig- 16 (1^5")- 



