chap, ix.] FORMS OF SHELLS. 309 



then, we have another example of the danger of 

 arguing cb priori as to function from structure. 

 Physiology, like other branches of science, must pro- 

 ceed rather by observation and a posteriori argu- 

 ments. 



Nautilus is the only existing tetrabranchiate 

 Cephalopod, but to that division belonged a number 

 of extinct forms, whose shells are found fossil ; such 

 are the Ammonites, with shells like those of the 

 Nautilus, Gvroceras with discoidal, Trochoceras with 



' , 



spiral, and Baculites with straight shells. 



Among the Dibrancliiata, Spirula (Fig. 127; A), 

 whose body is so rare and whose shell so common, 

 alone has the shell coiled and divided into chambers ; 

 it is not, however, an external shell, like that of the 

 nautilus, but is internal. In the fossil Belemmite 

 (Fig. 127; B) the proximal end of the shell (phrag- 

 mocoiie) was divided into separate chambers, which 

 were connected by a siphuncle. The distal end of the 

 shell is dart-shaped and solid, and forms the so-called 

 giiarcl. In Sepia the shell is calcareous, straight, 

 flattened out for the greater part of its length, with 

 the apex only incompletely chambered. 



In the squids, such as Loligo (Fig. 127 ; c), the shell, 

 now ordinarily known as the pen, is merely horny, 

 and consists of a shaft with two wings; in the Octopus 

 the shell is lost ; its ally Argonauta (Fig. 1 27 ; D) 

 fashions for itself a shell which both morphologically 

 and physiologically is a different structure from those 

 we have hitherto been considering, for it is formed 

 by a pair of the arms and not by the mantle cells, and 

 is confined to the female, where it serves to carry the 

 fertilised ova. A structure, the origin of which is 

 unknown, but the function of which is likewise 

 incubatory, is that which is known as the float of the 

 Gastropod lanthina. 



The operculum is formed by the foot, is horny, 



