November 19, 1915] 



SCIENCE 



733 



between this temperature and 60 degrees, mul- 

 tiply again by the correction factor taken from 

 the table opposite the observed temperature and 

 add the resulting product to the density read- 

 ing as observed on the instrument. The table 

 of corrections is reproduced herewith. 



One frequently finds density expressed as sp. 

 gr. 15° C./15° C. This value differs so slightly 

 from sp. gr. 60°r./60° F. that in the follow- 

 ing discussion the treatment applied to sp. 

 gr. 15° C./15° 0. may be understood to apply 

 to sp. gr. 60° F./60° F. without error suffi- 

 ciently large to interfere with the usefulness 

 of any biological results in which they may 

 play a part. 



More often, however, the density is ex- 

 pressed in terms of 15° C. referred to the 

 temperature of maximum density, sp. gr. 

 15° C./4° C. Again by the help of a correction 

 constant determined by Dr. O. H. Tittman sp. 

 gr. 60° F./60° F. may be reduced to sp. gr. 

 15° C./4° C. To make this correction 0.00082 

 is subtracted from the value expressed as 

 60° F./60° F. 



When the density of a sample has been re- 

 duced to 15° C./4° C. it is possible by means 

 of Petterson's* determinations to ascertain the 

 corresponding quantity of salt in the water. 

 This tabulation gives the specific gravity read- 

 ings both as sp. gr. 15° C./15° C. and as sp. gr. 

 15° C./4° 0. of a series of sea-water samples 

 of different density and the corresponding num- 

 ber of grams of total salts per liter. This 

 latter value was determined by the silver titra- 

 tion method of Forchhammer.^ Since the sea- 

 water varies in concentration rather in the 

 proportion of salts present, the accurate deter- 

 mination of any one constituent should by 

 simple calculation give the total salts. Since 

 CI is present in relatively large quantity and 

 can be determined by titration with silver 

 nitrate to a very great degree of accuracy, it 



^Petterson, Otto, "A Eeview of Swedish Hydro- 

 graphie Eesearch in the Baltic and North Seas, ' ' 

 Scottish Geographical Magazine, 10, pp. 296-299, 

 1894. 



5 Forohhammer, G., "Om Sovandets Bestanddele 

 og deres Pordeling i Havet," Kjobenbavn, 1859. 

 Engl, trans. Philosophical Transactions, 1865. 



is most often used for this purpose. It has 

 been found by Dittmar^ that the ratio of salt 

 to chlorine 1.8058 applies to all oceanic waters. 

 Petterson found a slightly greater value. He 

 carefully determined by titration the salt con- 

 tent of the samples just referred to and ob- 

 tained a series of values which by interpolation 

 can be used in determining either salt content 

 from sp. gr. or sp. gr. from salt content. 

 Petterson's determinations are here given. 



TABLE COKTAINING THE RELATIONS OF CHLORINE, 

 SALT, AND SPECIFIC GRAVITY 



Since it is somewhat awkward to interpolate 

 observed values into this series, the writer has 

 used the data as the basis of a generalized 

 scheme by means of which equivalents can be 

 promptly and easily found over a range of 

 variation of salt content somewhat greater than 

 has been observed at Woods Hole.^ By plot- 

 ting all of Petterson's values and by prolonging 

 the curves, the range of the table may be 

 greatly increased though somewhat at the ex- 

 pense of accuracy if an extension to either 

 great concentration or great dilution is at- 

 tempted. 



In the accompanying diagram (Fig. 1) the 



sDittmar, Challenger Reports, Physics and 

 Chemistry, Vol. 1, 39. 



' Sumner, Francis B., Osburn, Eaymond C. and 

 Cole, Leon J., "A Biological Survey of the Waters 

 of Woods Hole and Vicinity," Part I., See. 1, 

 Physical and Zoological, Bull. U. S. Bureau of 

 Fisheries, Vol. 31, Part I., 53 (for 1911), 1913. 



