8 4 



A. RICHARDS. 



A control, 1 whether part of the same cluster or another cluster 

 in the same stage of development, may be started at the time the 

 exposure is made. If the control is observed after an interval 

 of several hours, it is found ahead of the radiated eggs in develop- 

 ment. If, however, the control as well as the radiated eggs be 

 observed at more frequent intervals, quite a different state of 

 affairs is to be seen. During the first two or three divisions 

 after the exposure the radiated eggs are ahead of the control. 

 They gradually get slower, however, as already explained, while 

 control maintains its normal course. Not only do the radiated 



\ 



\ 



FIG. 2. 



eggs slow down to the normal rate so that for a very short time 

 both they and the control progress together, but presently they 

 become even slower. Usually by the time the twenty-four cell 

 stage is reached, if the exposure was during the first or second 

 cleavage, the control has more than passed the radiated eggs in 

 the degree of its development. This is explainable on the basis 

 of the observations previously noted. First the divisions of the 

 eggs are stimulated by the exposure, during which time they get 

 ahead of the normal eggs; then the depression phase sets in and 



1 Compare experiments (i) and (7). References to experiments in point might 

 be multiplied greatly, but those given are thought to be sufficient to illustrate the 

 principles set forth. 



