HEAT PRODUCTION BY EGGS. 339 



Since this investigation was first projected others have made 

 studies of a somewhat similar character. Myerhof (6) measured 

 the heat production of segmenting Echinoderm eggs by means 

 of a Beckmann thermometer, using a large vacuum flask for his 

 calorimeter. Certain irregularities that occur in his curves may 

 be due to experimental error, but suggest changes in the rate of 

 heat production at different stages of development. Since then 

 Shearer (7) has carried out similar measurements in connection 

 with his work on the oxidation processes of Echinoderm eggs 

 and finds that the rate of heat production after fertilization is 

 constant for at least ten or twelve hours. He makes the state- 

 ment that readings were taken at fairly frequent intervals at 

 the commencement of the experiment, and at intervals of several 

 hours after that. In view of this statement it seemed advisable 

 to us to repeat the work making frequent readings and using 

 methods of higher precision. 



POSSIBLE SOURCES OF ERROR. 



In an investigation involving the measurement of such slight 

 temperature changes as are expected here care must be taken to 

 foresee and provide against all possible heat transfers into or out 

 from the experimental flasks or to know the magnitude of such 

 transfers. In any event such heat losses or gains should be 

 small as compared with the total heat production which it is 

 desired to measure. In our experiments the following possible 

 channels of heat transfer existed, and were checked: Conduction 

 to or from the water in the flasks 



(1) by the air in the mouth of the flask, 



(2) by means of the glass forming the neck of the flask, 



(3) along the main thermopile, 



(4) along the secondary thermopile, 



(5) along the tubes of the stirring apparatus, 



(6) along the fertilizing tube, 



(7) by changes in the temperature of the water surrounding the 



flasks. 



There are also possibilities of error arising from lack of care in 

 controlling the conditions environing the electrical apparatus. 

 Among these may be mentioned : 



23 



