1915] Michael, ct al. ; IlinlrofjrapJiic Ii'rconh of Scripps Institution 29 



In using this instrument tlie tcmperatuix' cf tlie enclosed sample 

 is taken by means of a standard thermometer whose bulb is so im- 

 mersed as to give the temperature at the center of the pycnometer. 

 The thermometer is graduated into divisions of one-tenth of a degree 

 C. and the tejnperature read to the nearest 0?02. When the reading 

 becomes constant the thermometer is removed and enough more of the 

 sample, usually 2 cc. is added to fill the bottle. The stopper is then 

 carefully inserted and the e.xeess of water forced through its top is 

 immediately wiped off with the thumb. When the outside of the 

 pycnometer is thoroughly dried, its apparent weight is measured to 

 the nearest milligram and the specific gravity found from this weight, 

 corrections for buoyancy being made. 



Each pycnometer is carefully calibrated from observations made 

 on distilled water. The calibration formula relative to one of the 

 100 CO. instruments was. in this way. found to be 



\\ = 100.393+0.00282 {t°—20° ) 



where Ft is the volume of the instrument in cubic centimeters at the 

 temperature t. The accuracy of this formula is shown below, where 

 the volumes computed b.A- its use are compared to those obtained Ijy 

 direct measurement at the particular temperatures in question : 



These results show that the volume computed from the formula 

 differs from each measured value by less than 0.001 cc, while the 

 average is about half a.s much. The above calibration formula, there- 

 fore, gives the volume of the pycnometer within a relative error of 

 1 in 200,000. 



From this formula, together with the weight of the pycnometer 



itself, the specific gravity. S — . is computed as follows : 

 4 



S - :== 1.023667— 0.0000288 (6°— 20°)+ (^—131.90) 



4 ■ ' ' ' 100.293 



