356 



M !:>>!;< i;. r. <;i,\/i:i:i;iH)K A\I T. c. KIT/.I-ATUICK 



the distance CD was measured. This distance should, of course, be constant for any 

 given tube, and experiment showed that it was nearly so. Let us put CD = c. 

 Then we have 



a 25 _ b ~ ' 

 ~ OA ~ b ' 



AD a 



~ + e 



Hence, 



Now, a is known with sufficient accuracy from the length and mass of the mercury 

 column, so that the above equation gives us 6. 

 The following Table gives the values observed : 



TABLE II. 



The mean value of 0, allowing for the number of observations in each tube, is 

 4l'45. The angle of contact between mercury and glass is usually given as 42, so 

 that the agreement is very good. 



Now, the volume of the spherical segment ACB is easily seen to be 



while that of the cylinder on AB as base and of height CD is 



TTO? 

 COS 



(1 - COS 6}. 



The correction, therefore, to be subtracted for each end of the tube from the whole 

 volume tra"l is the difference between these two, and this 



Tra 8 

 3 cos 3 6 



(1-3 sin 2 B + 2 sin 3 0} ; 



therefore, 



