268 THE INDIAN JOURNAL OF VETERINARY SCIENCE AND ANIMAL HUSBANDRY [ I, IV.

        but are more than that noticed in the earlier experiment. The
        average loss for the 4 pits is 8.15 per cent., the individual figures being
        4.2 per cent., 7.8 per cent., 12.7 per cent. and 7.9 per cent. respective-
        ly in the various pits in serial order. Rapid filling has caused the
        largest loss (Pit 3). Watering also seems to have inhibited this loss
        to some extent as is seen from the lower loss in No. 1 than in No. 2.
        Hence the loss in No. 3 would have been more serious if no watering
        had been done.

    (3) Changes in the Total Carbohydrates.—The losses of total carbohydrates
         (i.e., nitrogen free extractives+crude fibre) were found to be 10.4 per
        cent., 10.3 per cent., 24.1 per cent. and 15.3 per cent. respectively, in
        pits 1 to 4. The losses are fairly uniform, and as has already been
        pointed out, there is an increased loss due to rapid filling.

                                E. Changes in the Inorganic Constituents.

The changes in ash (Tables XII to XV) are very important. While No. 1 pit
shows a loss of 13.3 per cent., No. 2 denotes a loss of only 2.9 per cent., the former
must evidently be due to the enhanced leaching action of the added water. It
seems, apparently, surprising that pits 3 and 4 should show gains. But the
reason is not far to seek. It must be due to the self same leaching action on the
large amount of decomposing surface material. This seems to be the only possible
explanation for the remarkably large gains of 25.6 per cent. and 13.2 per cent.
respectively (Table XV) in the bags in pit 4. In pit 3 the upper bag shows a gain
of 0.9 per cent. while the lower one has gained to the extent of 20.1 per cent. It
may be noted that watering and rapid filling have acted as a fillip to the leaching
action of the drainage juice. Thus while the lower bags in the slowly filled pits
show gains of 0.0 per cent. and 1.9 per cent. only, the corresponding upper ones
show losses of 26.6 per cent. and 7.8 per cent. respectively. The higher loss has
taken place in the watered pit.

It is rather interesting that variations in silica are not insignificant when com-
pared to the soluble ash. The above figures, for the total ash variations due to
leaching, are also repeated for the soluble ash and silica. The movement of silica
down the pit emphasises, more than ever, the washing down of various constituents
by the expressed juice which travels down the silo. This considerable movement
of silica has also been observed in other experiments at Bangalore. This point is
of considerable practical significance in that a very great increase of insoluble ash
must always be expected in the lowest layers of silage.

A detailed analysis of the components of the soluble ash and the changes
undergone by them as the result of ensilage is given in Tables XXI to XXIV. As