342 



CHARLES G. ROGERS AND KENNETH S. COLE. 



any considerable length of time. By the more constant stirring 

 the eggs would at all times be able to get their needed supply of 

 oxygen and to get rid of carbon dioxide. Work previously done 

 by. one of us (R.) had shown that stirring the water in the 

 experimental flasks once every two or three minutes by means of 

 an ordinary pipette was sufficient to allow the eggs to go through 

 a normal cleavage. Such stirring was not sufficient to ensure 

 that the water in the flask would be of uniform temperature 

 throughout. It seemed necessary, therefore, to devise some 

 automatic stirring device which would keep the water thoroughly 

 stirred at all times during the course of an experiment. As a 

 matter of interest it was found during the course of the experi- 

 mentation that a failure of the stirring apparatus for as much as 

 three minutes could be detected by a marked variation in the 

 galvanometer readings. 



j 



j 



FIG. 2. Stirring apparatus. Details described in text. 



The stirring was accomplished as follows: Saturated air was 

 bubbled from the jets (c, c) at the bottoms of the flasks, by 



