606 THURAS: DENSITY MEASUREMENTS 



attained by adding weights to the hydrometer. Reggianni did 

 not determine the specific gravity of the sea water beyond the 

 fourth place of decimals. 



Dr. Nansen in 1896, while studying the salinity of the North 

 Polar Basin, developed a method similar to the second modifica- 

 tion of that of Pisati and Reggianni. A hydrometer of total 

 immersion, 3.5 cm. in diameter and 15 cm. long, of normal Jena 

 glass No. 16 m was used. The observations were made in a 

 Dewar cylinder, and equilibrium was attained by adding plati- 

 num weights to the float, the final adjustments being made by 

 slightly varying the temperature of the sample until the density 

 of the liquid was exactly equal to the density of the float. The 

 temperature of the sample was carefully measured by three 

 thermometers, the bulbs of which were placed in the upper, 

 middle, and lower strata of the sample. An accuracy of one in 

 the fifth decimal place was attained. Dr. Nansen makes the 

 following remarks in regard to this method: 



It is most astonishing that so little attention has hitherto been 



paid by oceanographers to this valuable method After a 



little practice I think that the method of those observations might be 

 so arranged that they should not take much more time than ordinary 

 determinations with the hydrometer, if the latter were to be made 

 fairly accurate. In my opinion, the method promises to become im- 

 portant for future oceanographic work, as it should enable us to de- 

 termine the density and salinity of the ocean with a considerably 

 higher degree of accuracy than has hitherto been reached. And it 

 excludes, even in the case of less skillful investigators, the chief source 

 of error, vitiating, under ordinary circumstances, all observations made 

 with the common hydrometer. 



T. W. Richards has also made a careful study of this method 

 and has obtained a much higher degree of accuracy. The so- 

 lution to be investigated is placed in a liter Erlenmeyer flask im- 

 mersed in an adequately stirred bath in which the temperature 

 can be kept constant to 0?001 C. He was able to measure 

 densities to a few units in the seventh decimal place. 



A. B. Lamb and R. E. Lee have also attained very high accu- 

 racies by using an electromagnetic attraction to obtain the 

 equilibrium condition. A piece of soft iron is enclosed in the 

 bulb of the sinker and by means of an electric current sent 



