EXTERNAL CONDITIONS 199 



by acting as a new and a very active ' growth-centre ' which, 

 of course, enters into competition for food-material with the 

 organs of the crab's own body. Secondly, however, he finds 

 that this effect is not uniform on all the organs. It is greatest 

 in those of most rapid inherent growth-potential (highest 

 growth-coefficient). Thirdly, he believes that normally excess 

 of growth of one part is associated with a ' drainage effect ' 

 upon the growth of other parts, and that when the growth 

 of the organs with high growth-coefficient is inhibited, as in 

 Sacculina, there is a releasing effect on those of low growth- 

 coefficient leading to an actual increase in their relative size. 

 It is by this means that he would account for the well-known 

 fact of the increase in the abdomen-width (approximation to 

 female type) in sacculinized male Crustacea. 



This third conclusion of his does not appear to be well 

 founded. It is in disagreement with the correlation we have 

 found between the presence of a highly heterogonic organ 

 (e.g. chela) and excess growth in the appendages posterior to 

 it ; and with the important fact discovered by G. Smith 

 (1906A), to which sufficient attention has not been paid, that 

 parasitized female crabs, even if Perkins be right in supposing 

 that their general growth is checked, show accelerated develop- 

 ment of the female secondary sexual characters of their abdo- 

 men. This last point, together with the fact that in males 

 which have recovered from the parasite, the gonad develops 

 ova, seems to me conclusive proof that the effect of the parasite 

 here is in major part a specifically sexual one, and not merely 

 a quantitative growth-effect. Be this as it may, Perkins 

 has shown that there does exist such a quantitative growth- 

 effect of Sacculina upon its host, and that in regard even to 

 appendages which show no obvious secondary sexual differ- 

 ences, it exerts a differential and graded influence ; and this 

 opens the door to all kinds of future work. 1 



Another influence upon differential growth is shown by 

 the work of Thiel (1926) upon the Lamellibranch mollusc 

 Sphaerium. He shows that in clean water with high oxygen 

 content, the form of the shell becomes more rounded than 

 in the dirty water of its normal habitat. This, however, 



1 In this connexion the specific and differential morphogenetic effect 

 of the parasitic worm Mermis upon its ant hosts should be noted, 

 although for the moment its relation to the growth-gradients of the 

 ant-body cannot be clearly envisaged. See Wheeler (1928), Vandel 

 (i93o). 



