The Relations between Marine Animai and Vegetable Life. 419 



concluded that none of thein singly can be beici to afford an exact 

 estimate as to the purity or otherwise of a water. Combined however, 

 the evidence they afford is mudi more trustworthy. The adoption 

 of a physiological test of the purity of a water is, I believe, a method 

 not hitherto employed. Whether it could be profitably used for test- 

 ing the purity of fresh waters, is somewhat doubtful. Probably for 

 testing drinking waters it would be of little avail, as ali of tbese 

 should be in such a high state of purity that the slight differences 

 actually occurring in them could effect but little in the rate of growtb 

 of au orgauism. For testing markedly impure waters, however, it 

 niight be found a valuable agent. In the present experiments, in 

 which Echinoid Plutei were made use of for the physiological test, 

 it may perhaps be objected that the larvai were never fed, and hence 

 had to subsist on the animai matter already present in the water, 

 and that therefore, the more of such matter a water contained, the 

 greater ought to be the size of the larvai. This criticism is probably 

 not valid, as most likely these larva? do not feed at ali up to the 

 eighth day or so of growtb, when they were killed and measured, 

 but subsist on the food material originally present in the ovum. Thus 

 Daveni'Ort has shown ' that frogs' embryos, though they increase about 

 sixfold in absolute weight during the first ten days of development, 

 contain at the end of this period less solid material than the originai 

 ova. The increase in weight seems to be entirely due to the im- 

 bibition of water. Again, the mere fact that the larva? grown in 

 pure open sea-water werg some 16% larger than those grown under 

 similar conditions in Aquarium water, shows there can be little in 

 this argument. 



S u m m a r y. 



The following are the chief conclusions^ arrived at in this paper: 

 Green weeds such as Ulva rapidly remove the free amraonia 

 from Aquarium tank water, but they slowly increase the organic 

 ammonia. Larvse grown in water thus purified are as a rule increased 

 in size. When grown in direct contact with the alga, they are 

 generally decreased in size, but an increased proportion of the fer- 

 tilised ova develop to larvai. 



1 Proc. Boston Soc. N. H. Voi. 28. 1897 pag. 73. 



2 An abstract of these results has been published in the Proc. R. Soc. 

 London Voi. 63 pag. 155. 



