C— GEOLOGY 75 



have undergone no change since their first appearance. An outstand- 

 ing example of this has been claimed by Garstang (1928) from among 

 the gastropods. In some of these, whilst the organism is still embryonic, 

 the visceral hump with its shell rotates rapidly in relation to the rest 

 of the body through nearly 180° in only a few hours. Though, for the 

 sake of discussion, I shall use Garstang's example I am not convinced 

 that he is right in concluding that this phylogenetic twist arose primevally 

 in the larval stage. The twist exhibited by the embryonic gastropod is 

 not more striking than is the contortion experienced by the equally young 

 starfish. Starting from a condition of attachment by the preoral lobe 

 and an outward attitude of the mouth, this passed rapidly to one 

 of free movement with the mouth facing downwards. Bather (1915) 

 has shown that the fully grown edrioasteroid of early paleozoic times 

 must have been equally at home in either position. The Edrioasteroidea 

 therefore provided a stock through which such freely moving forms as 

 starfish and sea urchins could have been derived from the more primitive 

 attached ancestors. When one watches a gastropod, as it crawls along, 

 rotating its visceral hump and shell from this position to that, he cannot 

 deny to its primeval ancestors an equal flexibility, and to those that 

 most frequently rotated the mouth of the mantle cavity into a forward 

 position possibly a greater selective value. Thus with gastropods, as 

 with edrioasteroids and their descendants, a bodily position which at first 

 was assumed by adults as a temporary convenience may have become 

 stabilised into a permanency. 



Consideration of some well-known facts among fossils brings to light 

 other possible examples of the coenogenetic origin of new characters 

 which have influenced subsequent histor}'. Thus in the oysters and in 

 forms derived from them the process of cementation of the shell to other 

 objects is confined to early life. It must, in all probability, have 

 originated at about the close of the embryonic phase and remained with 

 varying degrees of persistence into early stages of the neanic phase, but 

 rarely if ever into later life. Here as with the torsion of the visceral hump 

 of the gastropod the change was coenogenetic, but it has brought in its 

 train, or opened the way for, series of other changes such as the marked 

 variability of form in the oysters and various degrees of coiling in Gryphcea 

 and Exogyra. Nevertheless, even in such an advanced form as Gryphcea 

 incurva this attached stage is preceded in development by the prodisso- 

 conch which exhibits primitive or palaeomorphic characters. These 

 observations appear to be equally applicable to other organisms which 

 have become attached by other means such as a byssus or by spines. In 

 each case the character that has been introduced coenogenetically does 

 not displace but plays its part along with those which belong to the 

 ancient category. 



Proterogenesis {the extension of nezv characters from early to late 

 stages of development): 



The coenogenetically introduced characters discussed above are either 

 confined to the early stages of development or, if they last into adult life, 



