DAVIS: BIONOMICAL CONSIDERATIONS IN GASTROPOD EVOLUTION. 6 1 



Since the branchial cavity was not very wide, the excreta would tend to 

 soil the gills to some extent, especially in forms which came more or less out 

 of the water, or were left uncovered by the tide, so that the respiratory 

 streams — and consecjuently the flushing of the branchial cavity — were 

 temporarily susp'^nded. The left side 1 eing predominatingly excretory the gill 

 of that side would raturally be more soiled than its fellow, which would acquire 

 added importance in this way. If now the l)ranchial cavity were shifted 

 even to a very small extent up the right side of the animal, the more 

 important right ctenidium would be correspondingly raised, and thus freed 

 to some extent from ])ollution. The consequent loss of s)mmetry would be 

 of no great moment to a form cree|)ing out of water or, generally, over rocks 

 and weed. 



An upwar(1 .^-liitting to tlie right would also be of advantage in another 

 wav. I'or a svnniietrical postero-dorsal visceral hump would tend to lag 

 behind, bringing the edge of the shell down on the chitinized upper surface 

 of the hind foot, and thus partly or completely closing the opening of the 

 branchial cavitv. \\'e find therefore those variations were selected which 

 have ultimately raised the branchial cavity up the right side to the antero- 

 dorsal region of the body. 



The earliest Molluscs undoubtedly clung to the rocks as a protection 

 against the wash of the tide and the buffeting of the waves, but this became 

 more difficult as the evolving Gastropods began to wander actively over reefs 

 and boulders. Under these circumstances the habit seems to have gradually 

 become more marked of retracting as much of the b')dy as possible into the 

 cap-like or slightly symmetrically coiled shell. The head would be first 

 drawn in, to be followed by the front part of the toot, but the posterior 

 extension of that organ would be too long to be simplv drawn in, and would 

 therefore be flexed on itself, its chitinized dorsal surface thus coming to 

 lie in the mouth of the shell. From this stage onward the gradual specializa- 

 tion of the chitinized area into an o[)i-rculum must neces.sarily have foUoweil, 

 and it would see-m llial we are justified in rrgarding this structure as a very 

 ancient one. 



We know that the shifting of the branchial cavity profoundly modified 

 internal arrangements, and greatly disturbed their original syunnetrv, so it is 

 possiltie that the external syr.iuietry was simultaneously affected. As the 

 branchial ca\ity was shifted in a counter-clockwise direction, it is likely 

 enough that at Ica^t the outer whorl of the slightly coiled shell ac(]uired 

 somewhat of a tilt to the left. Ivxternal sytnnietry, howexer, would ]M-ol)abI\- 

 not be greall}' modified at this stage, as the still important clinging 

 and probably surface-lensi(jn creeping habits would militate againt largi- 

 alteiations of the kind. 



The early Gastropod was attachetl to the ca[) shajjed shell by means of 

 a muscle taking (jrigin in the apex of the latter. I!ut as the branchial cavity 

 shifted and many of the viscera moved or twisted round with it, ihiN inu>ele 



