The combined pipette-sieve method of grain-size analysis resulted in the use of 13 size fractions in this study. These fractions are the following (grain size in millimeters): >1.0, 1.0-0.63, 0.63-0.40, 0.40-0.315, 0.315-0.200, 0.200-0.160, 0.160-0.100, 0.100-0.063, 0.063-0.050, 0.050-0.010, 0.010-0.005, 0.005-0.001, and <0.001. The State USSR Standard (GOST 12536-79, 1998) provides a basis for this method.
To calculate grain-size fractions content of dry sediment, weight percent of water in the sediment sample was first determined.
For grain-size analysis, a weighed sediment sample (20-35 g) was added to distilled water, mixed, and held for 12-15 hr. Then, according to Lapina (1974), the sample was dispersed with ultrasound (frequency 22 kHz) for 3-5 min. Thereupon the specimen was ground (by rubber pestle, without force), poured into a glass beaker, topped up to the definite level by water (with the addition of an anticoagulant [sodium phyrophosphate, 50 mL of a saturated solution per 4 L of treated suspension]), mixed again, and held at room temperature (18º-22ºC). Three subsamples were taken, one after another, from a tabulated depth (7-10 cm) below the surface of suspension, at tabulated time intervals of "minute" (17-19 min), "hour" (71-78 min), and "diurnal" (21-23 hr) for subsamples representing the grain-size fraction <0.01, <0.005, and <0.001 mm, respectively. A 25- to 30-mL pipette was used for taking these subsamples. Each subsample was transferred to a porcelain bowl, dried, and weighed. Using the data obtained and taking into account the water content of the sample (see above), the mass of three clay grain-size fractions of the sediment were calculated. Thereupon the suspension (situated in the glass beaker) was completely freed of clay particles by repeated settling and decantation (through a siphon); the duration of settling and level of decantation are tabulated in the method prescription. These operations were repeated until the layer to be decanted became transparent. The remainder of the specimen (grain-size fractions > 0.01 mm) was transferred into a porcelain bowl, dried, weighed, and sieved manually by means of a bank of woven sieves, for a period of 20 min. Each obtained fraction was weighed, and the mass of the remaining grain-size fractions was calculated.
The results of grain-size analysis of all 375 samples from Sites 994, 995, and 997 are presented in Table 1, Table 2, Table 3, respectively. Seventeen samples were analyzed a second time. These results (provided in Table 4) show that the accuracy of determination of the isolated grain-size fractions, particularly the fine ones, is rather low; however, the discrepancy becomes essentially less important when considering fractions jointly (Table 5).