Echinoderm Larcw ivith Two Water- Vascular Systems. 341 



Discussion of the Eesults of the Experiments. 



We may say, then, that, as the result of experiments carried on for four 

 years, two points have been determined, viz. : (1) that if the larvae of Echimis 

 iniliaris be exposed to the action of hypertonic sea-water for a week com- 

 mencing with the fourth day of development, many of them will develop on 

 the right side as well as on the left a hydroccele or water-vascular rudi- 

 ment, and that in connection with this second hydrocoele all the structures 

 which normally develop in connection with the left hydrocoele may make 

 their appearance, i.e., spines, tentacles and dental sacs ; (2) that if the larvae 

 of Echinus miliaris be starved during the first week of their existence, and 

 then placed under favourable conditions both as to food and space, they will 

 continue their development, but many of them will be devoid of both 

 pedicellariae and hydrocoele, but will have in place of both a group of pointed 

 spines on each side. Such larvae will, in the majority of cases, be devoid of 

 madreporic pore and axial sinus, but will possess a well-developed madreporic 

 vesicle. 



If we confine ourselves for the moment to the first of these results,, 

 we cannot help recalling that the result of using hypertonic water on the 

 unfertilised eggs of other Echinoidea is to cause some of them to develop 

 into larvfe without the aid of spermatozoa. 



Although Loeb, in his improved method, employed butyric acid to 

 stimulate unfertilised eggs to development, and only used hypertonic water 

 as a subsequent treatment, in order, as he supposed, to correct the cytolytic 

 tendency which had been produced by the acid, yet, in his earlier experi- 

 ments (10), he found that hypertonic water alone was able to produce 

 parthenogenetic development. 



If, however, we make this comparison, another thought is inevitably 

 suggested. Just as the process of development of an unfertilised egg is not 

 something foreign to its nature, which has been forced on it by the 

 hypertonic water, but consists merely in rendering active a latent 

 potentiality, so we cannot assume that the formation of a right hydrocoele 

 is a new accomplishment forced on the larva ; it must be merely the 

 rendering actual of a potentiality which exists in the normal larva, but 

 which usually remains in abeyance. That this is so is supported by the 

 occurrence of a single specimen showing the double hydrocoele in a culture 

 reared in normal sea-water which contained hundreds of larvae. 



The most obvious explanation of this latent potentiality is to assume that 

 the Echinoderms are descended from some type of animal which had a 

 hydroccBledike structure on each side. Such a hypothetical ancestor is 



2 E 2 



