STATISTICAL TREATMENT 



249 



in cleavage time being recorded for each of these conditions. Some of his 

 results are shown in Table 9, and if these results are plotted on arithlog 

 paper, the points for each of the different degrees of exposure fall essen- 

 tially on a straight line. If we take the logarithm of the delay in cleavage 

 and determine a linear relationship between these logarithms and the 

 insemination time, we have the following values of the constants for each 

 degree of exposure: 



While the values of a increase with additional exposure, the b values are 

 astonishingly constant. The standard error of these h constants is of the 

 order of 0.00010 and it is obvious that the values do not differ from each 

 other more than would be expected by simple sampling. If we treat 

 the entire series of observations simultaneously, we obtain as our value 

 of b, —0.00298, and using the previously derived values of a with this 

 value of b, we have for our equations representing the relationship between 

 the delay in cleavage and the time after irradiation of insemination, 



Exposure, min. 

 5 

 10 

 20 

 40 

 60 



Equation 



y 

 y 

 y 

 y 

 y 



1QlA68—.00i9ix 

 JQ1.3%-.00298x 

 JQ1.519-. 002981 

 JQ1.780-. 002981 

 JQl. 956-. 002981 



14.7(0.99316)^ 

 24.9(0.99316)^ 

 33.0(0.99316)- 

 60.3(0.99316)- 

 90.4(0.99316)- 



These equations are plotted on an arithlog scale in Fig. 6 and we see 

 that the agreement between theory and observation is excellent except 

 for the case of the 40-min. exposure. In this case the rate of decline seems 

 to be slightly greater than that given by the theoretical expression. 

 This divergence is not sufficient, however, to make us seriously question 

 our assumption that the rate of decline is constant for different degrees of 

 exposure. 



Turning to the interpretation of these equations, we see that the 

 general order of the delay in cleavage, as represented by the constant A, 

 increases with increasing exposure to radiation. But for each duration of 

 exposure the amount of retardation of cleavage, y, diminishes with delay in 

 insemination, x. This approach of the cleavage time toward the normal 



