ON THE ARGONAUT. (51 



into the bottom of the shell, that the poulp thus holds itself 

 upside down during its ascension ? If, on the contrary, it 

 ascended with the keel downwards, this air could not fail 'to 

 escape, and it would .then be necessary for the animal to use 

 its organs of " refoulement" to remedy this loss. This remark 

 will perhaps appear strange to some persons ; but it is cer- 

 tain that many of the Molhisca and Acalepha ascend on this 

 principle : and we have many times seen them leave at the 

 surface of the sea, the bubble of air which they had undoubt- 

 ly obtained at the bottom by means of some peculiar faculty. 



Examination of the arguments which have been presented 

 in favour of one or the other opinion. — M. de Blainville, in 

 his interesting letter, has advanced a series of arguments in 

 favour of parasitism, to the greater part of which our preced- 

 ing observations apply. But there are some still remaining, 

 to which we have objections to oppose; such as, for exam- 

 ple, his fourth argument, which is drawn from the absence of 

 organic attachment between the shell and the animal, a cir- 

 cumstance tending to indicate that the two are foreign to each 

 other. 



We quite agree with those naturalists who recognise this 

 want of attachment ; in fact there is no other connection be- 

 tween the shell and the animal, than that of contact, and this 

 argument has always been regarded as one of the most valid : 

 we can, how^ever, meet it by another, namely, that the true 

 constructor of the shell, supposing it not to be the poulp, did 

 not adhere to it either ; since, contrary to what we observe in 

 other shells, there are upon the argonaut no traces of attach- 

 ment, or, in fact, of any muscular impression. This remark, 

 moreover, is not our own ; it was made to us by Cuvier ten 

 years ago, in a conversation with him on this subject. 



To this observation it has sometimes been objected, that 

 the argonaut is evidently an internal shell. We confess that 

 we should have some difficulty in picturing to ourselves an 

 internal shell of this description, — so ^^enroulee", so symme- 

 trical and diaphanous, and possessing ribs and tubercles, and, 

 in short, so little analogous to all internal shells, whatever 

 may be the order of Molhisca to which they belong. 



An attempt has also been made to approximate this shell 

 to the genus Atlanta, which attempt rests upon some recital 

 of an inhabitant of the island of Amboina, of which however 

 we have nothing but a completely anecdotal report. But in 

 the genus Atlanta the animal is united to the shell by a very 

 evident attachment, and upon the whole there exists no rela- 

 tion between the argonauts to the Atlanta on the one hand, or 

 to the Carinaria on the other ; for these two nucleobranchiate 



