but with fewer data, in Figure 8, lines of equal specific gravity (2.8, 2.7, 2.6) 

 meet at 100 percent porosity and 1 .03 g/cm^, the unit weight of sea water; at zero 

 porosity specific gravity equals unit weight. 



Discussion — Ratcliffe (1960, p. 1538) reports quite a different relationship. 

 His wet density-water content (in percent wet weight) line (Ratcliffe, Fig. 4) inter- 

 sects zero water content (zero porosity) at about 2.1 g/cm^, and at 1 .03 g/cnrW a water 

 content of about 82 percent is indicated, which corresponds to a porosity of only 91 

 or 92 percent depending on whether 2.1 or 2.35 g/cm^is used for the particle density. 

 Ratcliffe's data indicate abnormally low water contents, suggesting that the sediment 

 tested was desiccated. Alternatively, a different test procedure may account for 

 different values. The likelihood of this second possibility is increased with reference 

 to wet unit weights and water contents of selected samples presented by von Herzen 

 and Maxwell (1959, p. 1562), whose values fall to the left (low water content) of the 

 2.6 particle specific gravity straight line in Figure 8. 



E. WATER CONTENT 



introduction — Water content, w, used herein is the ratio, in percent, of the 

 weight of water in a given sediment mass, W w , to the weight of the ovendry solid 

 particles, W s (ASCE). It is determined by weighing a representative portion of the 

 sample, ovendrying at 110°C overnight, cooling in a desiccator, and rewelghing. 



Lambe (1950, p. 494-495; 1951, p. 10-12) demonstrated that variations of tem- 

 perature in different locations in non-heat distributing "constant-temperature" ovens 

 may exceed 100°C; and that variations in the amount of water driven off at any given 

 temperature are greatest in fine-grained sediments having a high colloid content, 

 investigators would do well to follow the example of Correns (1937, p. 38) and specify 

 the percentage deviation from 105° or 110°C whenever possible. Temperature varia- 

 tions of 0° to -41° from 110°C were found in the BUDOCKS oven by C. M. Yeomans 

 (1961, written communication), corresponding to a precision of water content measure- 

 ment of about one percentage point. Although reproducibility may be less than indi- 

 cated by the number of significant figures given in this report, these figures are retained 

 following engineering convention. 



Corrections have not been made for salt content; values of water content conse- 

 quently are slightly low. 



In addition to water content (also called moisture content), previously defined as 



W 

 w = ttt x 100, (7) 



s 



22 



