100 DISPLACEMENT INTERFEROMETRY. 



of 7.05 cm. 2 in area, are enormously larger than the above reading, K = 

 ic" 12 . The result is in keeping with the enormously large: area at the sur- 

 face of separation of air and water in the diver and consequent effectiveness of 

 the solution-temperature mechanism which I have described in the preceding 

 paragraphs. Whether in case of tube I a condition of true diffusion has been 

 reached will have to be ascertained in the lapse of further years. Meanwhile, 

 the question whether a long, slender Cartesian diver may not obviate the dis- 

 crepancies hitherto encountered is worth consideration. 



