MECHANISM OF THE AMPULLARIIIXE. 343- 



negative ; but as the qnestion is very obscure and at the same 

 time connected with tiie whole question of the respiratory 

 meclianism, we may attempt to make the issue a little clearer. 



Bavay and Bouvier (2 a) imply that the siphon and I'espiratory 

 orifice form a continuous tube. If this is the case, the air 

 from the exterior passes down the siphon straight into the lung 

 and none of it is wasted in the pallial cavity. As to the details, 

 Bous'ier is the more explicit. After admitting that during the 

 act of siphonal breathing it is impossible to see what is going on 

 in the pallial cavity, he says that afterwards the ' plancher pul- 

 monaire ' becomes visible {i.e. by the modification of tlie shape of 

 the siphon and its base) and that it is " susceptible de faire saillie 

 en avant jusqu'au voisinage de la base du siphon." The chief 

 thing to notice is that during the actual process of siphonal 

 breathing it is impossible to see what is going on inside the pallial 

 cavity. It is a controversial point whether one is entitled to 

 assume that siphon and pulmonarj" orifice are continuous at a 

 given moment because a little while afterwards the latter is found 

 projecting towards the former! It is thus diiiicult to see on what 

 critical facts Bouvier and Bavay based their statements. As fai" 

 as A. vermifor I his is concerned, even if it is seized during siphonal 

 breathing and the pallial cavity foi"ced« open, the siphon resumes 

 its normal shape before the base and respiratory orifice come 

 into view. 



A j)riori one would be tempted to assume that the connection 

 in question is necessary. It would be physiologically advanta- 

 geous to have a means whereby the air is conveyed straight into- 

 the lung and not wasted in the pallial cavity. It might also be 

 desirable to keep air aAvay from the gill. But we may point out 

 («) that the siphon is not used at all in direct pulmonary breathing,, 

 so that the question of physiological economy does not seem to be 

 involved, and (h) that during siphonal respiration the mantle 

 cavity is kept tightly closed by the down-drawing of the mantle 

 folds, so that there is no risk of water entering the cavity and 

 finding its way into the lung, 



C. A comparison of the observations npoii Neotropical forms 

 with those published by Ilamanan(7)npon the Oi'\(intn\ Pachylabra 

 glohosa and by Bouvier and Fischer upon the Egyptian Lanistes 

 holteniana, suggests that the structural difierences upon which 

 Fischer (5) separated the American, Indian, and African forms 

 are correlated with difli"erences in the mode of respiration. 



Fischer divides the Ampullariidee into the following genera : — 



1. Ainpullaria s. s. including Marisa (=Ceratodes) : siphon 



very long. America. 



2. Pachylahra : siphon rudimentary. Africa, Asia, Oceania. 



3. Asolene : siphon absent. South America. 



4. Lanistes: (siphon?). Africa, Madagascar. 



5. Meladomas : (siphon ?). Africa. 



