H. M. FucHS 223 



sperm -suspension was made either so weak that "/^^ or less than 1 °/^ 

 (< 1 °/^), or so concentrated that 100 /_ of the eggs were fertilized in 

 each of the dishes, thus allowing of no comparisons. 



This applies to fertilization in the Echinoids and to cross-fertilization ' 

 in Gioiia. In almost every case, provided that none of the eggs are 

 immature or pathological, 100 /^ fertilizations can be obtained if a 

 sperm-suspension of sufficient concentration is used. With regard to 

 self-fertilization in Ciona the matter is simpler, for it is very seldom 

 that 100 7o of the eggs can be fertilized by sperm from the same 

 individual, however concentrated the suspension be. On the other 

 hand, a comparatively large number of cases occur in which no self- 

 fertilization at all can be obtained. 



The next important point preliminary to an investigation of this 

 nature is a determination of the degree of accuracy of the results. For 

 this purpose an extended series of trials was made, lasting over a month. 

 The object of these trial experiments was to discover the chief sources 

 of error and the ways of overcoming them, together with the extent to 

 which the numerical results were accurate. 



It was found that accuracy depends on four main factors, which are 

 as follows : 



(a) The mode of making up the sperm-suspension. 



This was found to be the largest source of error. If a suspension 

 be made by taking some of the sperm from the vas deferens, mixing it 

 with sea-water, and then stirring the liquid, the spermatozoa are by no 

 means evenly distributed through the water. In the case of a thick 

 suspension this is easily seen with the naked eye, but with the dilute 

 suspensions used in the cross-fertilization experiments it is not apparent. 

 If equal quantities of liquid are taken from such a suspension and added 

 to equal amounts of eggs, the percentages of the latter which are fer- 

 tilized are usually by no means the same : the error may even be as 

 high as 20 °/,^. A number of ways were tried of getting a more even 

 distribution of the spermatozoa in the water, such as continued stirring, 

 shaking the suspension in a tube, pouring once or twice from one dish 

 to another, filtering — but all with no certain success. The results could 

 not be depended upon. Eventually, however, it was found that by pouring 

 the liquid to and fro ten times at least, a very uniform suspension could 

 be obtained, so that equal amounts fertilized equal proportions of eggs. 



' Throughout this paper "cross-fertilization " means the fertilization of Ascidian eggs 

 by spermatozoa of another individual. 



