Table 4 presents fractions of sunk, suspended, and floating material after a single separation. The residual grain represents grain that was hand sorted from either the suspended or floating material. The DM in the efflu­ent was obtained by mass balance after other components had been dried and separated.

Table 4 Grain and Stover Proportions After One Water Separation and Various Pretreatments Pretreatment Silage used (%) MC (%) Sunk grain DM after one Sunk Suspended stover stover separation (%) Floating Residual stover grain DM in effluent Grain concentration in sunk material (%) 1. Untreated silage 100 64.6 21.3 c 8.4 a 7.1 b 44.1 ad 1.5 b 17.6 a 71.8 d 2. Partially dried to lose 100 54.8 22.9 c 9.5 a 5.2 c 45.4 c 0.5 c 16.5 ab 70.6 d 10% units of MC 3. Partially dried to lose 100 45.3 25.8 b 2.2 c 2.1 d 54.6 b 1.0 bc 14.3 bc 92.3 b 20% units of MC 4. Oven-dried 100 0.0 23.8 bc 0.1 c 0.6 d 62.3 a 2.5 a 10.8 d 99.4 a 5. Sieved and fresh 50.3 64.6 37.0 a 9.8 a 9.5 a 29.5 e 0.6 c 13.6 cd 79.1 c 6. Sieved, partially dried 50.3 55.5 34.6 a 4.7 b 5.1 c 41.1 d 1.7 ab 12.8 cd 88.0 b to lose 10% units of MC SEM 1.4 1.0 0.8 1.8 0.4 1.3 2.7 LSD 3.1 2.2 1.7 3.9 0.9 2.8 5.8 a Experiment 3 with silage from Manthe Farm: 64.6% MC, unprocressed and 10.3mm MPL; average of three replications. Values with the same superscript letter in a given column indicate no significant difference (p < 0.05). SEM, standard error of means; LSD, least significant difference.

Sieving increased the proportion of grain collected (37 vs 21%, with­out drying), and improved the grain concentration in the sunk material (79 vs 72%). However, the total amount of grain obtained in the sunk material was lower after sieving (18.6 vs 21.3%) because about half the silage remained in the other sieves that were not used for the water sepa­ration. The smaller-size sieves would contain a significant amount of broken grain.

The proportion of sunk stover decreased significantly with partial drying (20 percentage units of moisture loss) and with complete drying in the oven. The grain concentration was enhanced as high as 99.4% for bone-dry material. The proportion of suspended stover was highest for sieved and fresh pretreatment. This material was of relatively uniform length (between 9 and 18 mm), so grain could sink rapidly in the absence of long stover pieces that tended to float and hinder the descent of smaller grain. The proportion of floating material was significantly higher for dry material. Residual grain was highest also in dry material. The proportion of DM in the effluent decreased as the silage was dried. The reduction in soluble and fine particles after drying might be owing to a loss of volatile organic acids.

In experiment 3, effluent contained an average of 17.6% of fresh silage DM after one separation. In experiment 2, in which eight successive water separations occurred, the effluent water contained between 20.7 and 26.0% of the fresh silage DM. A large proportion of the soluble and fine particles mixed rapidly in the effluent water. These values suggest a range between 68 and 85% for the ratio between DM in the effluent after one separation and DM in the effluent after eight separations. A ratio of 70% was assumed in Figs. 1 and 2 to illustrate initial DM in effluent and is within the experimental range of 68 to 85%.