176 



Mr. H. Ohshima, Reversal of 



newt ISTecturus are at first all alike, but that particular one will develop 

 into a sacral vertebra which happens to come in contact with the pelvic 

 girdle — and this may occur at slightly different levels on the two sides 

 of the animal. Unless the right and left anterior coelomic sacs of 

 Echinoderm larvae had been originally alike in their potentialities, it is 

 difficult to see how the powers of one could have been transferred to the 

 other, as they do not form parts of a series. 



In the paper which I communicated to the Society in 1918, I gave 

 reasons for believing that one means of procuring the growth of two 

 hydrocoeles was to subject the growing larvae to the action of hypertonic 

 water for a period of about a week. This result was obtained during the 

 summer of 1917. During the summer of 1919 I repeated the experiments, 

 and obtained results so entirely confirmatory of those obtained in 1917 that 

 I did not think it worth while to publish them. 



During the spring and summer of 1920, when Dr. Ohshima worked in 

 my laboratory, attempts to procure similar results failed: we did, indeed, 

 obtain a few larvae with double hydrocoeles, but these were found mostly 

 in cultures vi^hich had not been treated with hypertonic water at all. 



Dr. Ohshima is naturally sceptical as to my diagnosis of the cause of the 

 appearance of the double hydrocoele. I have therefore to offer the following 

 suggestions as to the reason for the divergence of the results of 1920 from 

 those of 1917 :— 



(1) The experiments of 1917 were carried out in natural sea water, which 

 up till that time was sold — as in pre-War days by the Great Eastern 

 Railway Company — at a cheap rate in London. 



After the spring of 1917 this service was stopped, and for further experi- 

 ments we had to rely on artificially prepared sea water, which is not so 

 good for the growth of the larvte. 



(2) Dr. Ohshima's work is based on the results of two cultures, one 

 obtained during May and one during June. 



The May culture was not a very vigorous one — in other words, the 

 proportion of ripe viable eggs shaken out was small compared to the mass 

 of unripe eggs, and the early development was not very vigorous, though 

 ultimately quite a considerable number of larvae completed their meta- 

 morphoses. 



I suggest that in the case of this culture the reaction to hypertonic water, 

 which I noted in the vigorous culture of 1917 did not take place. 



(3) On the other hand, the June culture was an extremely vigorous one. 

 When I examined the larvae of that portion of it which had been exposed 

 to the action of hypertonic water, a large proportion showed signs of the 



