48 Dr. T. Williams on the Mechanism of Aquatic 



larger the ventral or common opening of the mantle, the less is 

 the necessity for the lower or extra-branchial siphon. If, on the 

 contrary, the leaves of the mantle be fused at their borders all 

 round, a well-developed siphon is absolutely required. This cavity 

 must have a free and ready communication with the exterior. 

 If this communication is not established in one mode, it must in 

 another. A siphon is a necessary provision if the mantle be 

 closed ; if open, it is only supplementary. In the former case, 

 everything fluid and solid which enters the pallial cavity must 

 pass through the extra-branchial siphon. It can gain the cham- 

 ber through no other source. In the latter, the siphon is only 

 incidentally and occasionally used. The great bulk of water 

 drawn into the cavity rushes in through the ventral and pedal 

 openings. That which, alike solid and fluid, is returned un- 

 used from this cavity, is indiscriminately jerked out by muscular 

 action through any of the mantual openings. If the pellet of 

 saud be situated near the opening of the siphon of this cavity, 

 at the moment when it receives the impulse of ejectment, it escapes 

 through the inhalent or extra-branchial siphon (PI. I. fig. 6, b). 

 If, on the contrary, it be placed at the other end of the chamber, 

 it will be driven out either through the ventral or pedal gape. 

 The orifice and direction in which refused jets of water take 

 place from this cavity are contingent upon the position which 

 the rejected portion may have previously occupied in the cavity. 

 Upon this important point neither Mr. Clark nor Mr. Hancock 

 are clearly informed. Mr. Clark is correct in stating that the 

 ingress of the water into the great mantual or extra-branchial 

 cavity is due, not to the invisible agency of vibratile epithelium 

 on the lining membrane of the* siphon, but to the diastolic sepa- 

 ration of the valves. Mr. Hancock is undoubtedly in error in 

 stating that the water entering this cavity is drawn in by cilia 

 of the siphon. The microscope disproves completely the asser- 

 tion that the internal lining membrane of the inhalent or extra- 

 branchial siphon is the scene of ciliated epithelium. In uo 

 single instance of the numerous siphonal species examined by 

 the author, could cilia be discovered in this situation. If at 

 this place cilia do not exist, it admits of no dispute, that the 

 occasional inward-tending current which reaches the cavity 

 through this siphon cannot be due to the instrumentality of 

 cilia, at this point at least. Mr. Clark is unquestionably wrong 

 in supposing that, because now and then the inhalent or extra- 

 branchial siphon, and the ventral and pedal openings of the mantle 

 emit a jet of water and solid pellets, this cavity is therefore inde- 

 pendent, that the " siphons therefore do not communicate," and 

 that therefore the ingress and egress of the water designed for 

 respiration take place through the' same orifices. Every one of 



