592 REPORT— 1898. 



putrefying sea-urchins effected about ten times as much contamination as 

 did tlie various living animals examined. 



In the full paper are given numerous determinations of the specific 

 gravity of the aquarium water, and of the amounts of ammonia and 

 nitrites present ; also experiments on the effects of introducing various 

 salts into the water. To these it is unnecessary to refer here. 



During the last two or three months of my stay at Naples I was 

 working on another subject, 'The Relations between the Hybrid and Parent 

 Forms of Echinoid Larvse,' but I had also been making occasional experi- 

 ments on this subject whilst engaged on the research just described. I 

 also worked out some of the material obtained on my return to England. 

 An account of this work is in course of publication in the ' Phil. Trans. 

 Roy. Soc.', an abstract having already appeared in the ' Proc. Roy. Soc' 

 vol. 63, p. 228. The object of the reseai'ch was to determine systematically, 

 during the nine months' period I was working at Naples, the exact 

 relationship of structure and size existing between certain hybrid and 

 parent echinoid larval forms. Eight diiferent species of echinoids were 

 worked with, but the large number of observations were confined to 

 three of them. Upon the cross SphcHrechinus 9 Strongylocentrotus S 

 twenty-two experiments were made. The hybrids were most easily 

 obtained in the summer months, few or none of the ova being 

 cross- fertilised in the winter. It was found that the majority of the 

 hybrids obtained in May, June, and July were of an almost pure Spluvre- 

 chinus type, only a third or less of them being of an intermediate or 

 Strongylocentrotus type. In November, on the other hand, only about a 

 sixth were of the maternal and five-sixths of the paternal type. Finally, 

 in December and January, all the hybrid larv« were of the paternal type. 

 As regards the reciprocal cross of Strongylocentrotus ? and Sphcere- 

 chinus S a fair number of ova were cross-fertilised in April, May, and 

 June, but no plutei were obtained. In July and August, on the other 

 hand, 29 per cent, of the ova developed to eight days' plutei. In November 

 and December, with one exception, not only were no plutei obtained, but, 

 as a rule, not a single ovum was cross-fertilised. These extraordinary 

 variations in the capacity for cross-fertilisation seem to be due to varia- 

 tions in maturity which the sexual products undergo with change of 

 season. Thus in the summer months most of the Strongylocentrottis 

 individuals contain but very small quantities of ripe sexual products or 

 none at all. Also it was found that the normal plutei obtained in July 

 and August were 20 to 30 per cent, smaller than those obtained in April 

 and May, and also in November and December. At intermediate 

 times the larvce were of intermediate size. We see, therefore, that the 

 Strongylocentrotus $ Splwrechinus $ hybrid is only formed at the time 

 when the Strongylocentrotus ova have reached their minimum of maturity ; 

 whilst in the case of the reciprocal hybrid it follows that as the maturity 

 of the Strongylocentrotus sperm increases it is able to transform first a 

 portion and then the whole of the hybrid larvpe from the Sphcerechinus to 

 its own type. In other words, the characteristics of the hybrid offspring 

 depend directly on the relative degrees of maturity of the sexual pro- 

 ducts. 



On crossing Echinus ? with Strongylocentrotus $ it was found that 

 even more of the cross-fertilised ova developed to plutei than of the 

 directly fertilised ones, and also that these plutei were, on an average, 

 8 per cent, larger than the pure parental larval forms. In the reciprocal 



