ON THE USE OF THE HYDROMETER OF TOTAL IMMERSION. 



The following table will give an idea of the accuracy of 

 the determinations. Three different samples of sea-water in 5 

 litre bottles for controlling determinations of the hydrometers 

 among themselves, were each determined several times. The 

 first sample was determined in October, 1900, the second in 

 March, 1901, and the third in April, 1901. 



No. of 

 Sample 



Date 



17-5 -0. 

 *-> 17-5° C. 



Difference 



1 



Oct. 29, 1900 

 Oct.. 30, 1900 



1-026690 

 26692 



26687 

 26689 



> 0-000002 



> 5 



> 2 



2 



March 5, 1901 

 March 7, 1901 



1-026826 



26825 

 26822 



> 0-000001 



> 3 



3 



Apr. 17, 1901 



1-025914 

 25910 

 25913 



> 0-000004 



> 3 







Mean 



0-000003 



The greatest difference between two determinations of the 

 same sample is consequently 0000005. The 5 decimal places 

 of the specific gravity of the samples is thus determined with 

 absolute accuracy. These determinations were made in exactly 

 the same manner as the daily determinations of the samples 

 from the cruise of the "Michael Sars". The difference that may 

 appear between the 1st determination of these and the 2nd, is 

 due, as already mentioned, to evaporation, which will naturally 

 have a far greater influence upon a sample with a volume of 

 500 cub.cm. than upon a sample with a volume of 5000 cub. cm. 



In conclusion, a number of instances are given of the 

 agreement between the 1st and 2nd determinations of the samples 

 from the cruise of the "Michael Sars". 



