244 FRANK R. LILLIE. 



water in which it was shed." " When a small quantity of sperm 

 was mixed with a large quantity of sea -water, the duration of 

 vitality of the spermatozoa is short, but when the converse pro- 

 portions are used, it is greatly lengthened." " By taking sperm 

 from a sea-urchin and mixing it in different vessels with different 

 quantities of sea-water, one obtains sets of spermatozoa, which will 

 retain their vitality for a rising series of terms, e. g., 8, 12, 16, 24, 

 48 and 72 hours. For the longest term, the proportion of sperm- 

 atic fluid to sea-water should be not less than i to 10." 



Gemmill was thus dealing with the same phenomenon with 

 which we are concerned. He gives, however, no exact quanti- 

 tative data and relatively few experiments were performed. He 

 attributes the results to (i) greater activity of the spermatozoa, 

 and consequent earlier exhaustion in the more dilute suspensions 

 and (2) to dilution of the "spermatic fluid" by which he sup- 

 poses the spermatozoa to be nourished. 



5. Other Factors in the Fertilizing Power of Sperm Suspensions. 



In the large number of experiments carried out to test the 

 fertilizing power of sperm suspensions the general form of the 

 curves is remarkably constant. Some, however, are quite ir- 

 regular, and it was never possible to get exactly the same curve 

 in the repetition of any experiment. A few of the irregularities 

 may conceivably be due to error, as for instance the accidental 

 presence of some toxic substance in one of the dishes of a series, 

 though painstaking care was used to avoid such sources of error. 

 The failure to obtain exactly the same curve in different experi- 

 ments is no doubt also due in part to the natural variability of dif- 

 ferent lots of eggs and sperm. 



In an attempt to discover the sources of variation and error, 

 the effect of egg concentration, i. e., the absolute quantity of 

 eggs in a given bulk of a sperm suspension of given strength was 

 tested. On the whole the effect of egg-concentration was found 

 to be relatively small within so wide a range that it cannot be 

 regarded as a large factor in the variability of the curves ; because 

 the egg-concentration of the curves was always below the point 

 where it was demonstrably a limiting factor. Tests were made 

 of sperm suspensions ranging from 1/62.5 per cent, to 1/8,000 



