It is a well-known fact that if bark-shavings be allowed to accumulate in a heap for any but a short period, a fermentative and heating action is set up. The heat developed in these piles of shavings is so considerable that it is impossible to keep the hand in a heap for more than a second or two. Should this be allowed to persist, as would happen in the case of a breakdown of engine or machines, it usually results in the final crepe rubber becoming tacky when approaching dryness. To avoid this heating effect it is necessary to have spare jars or proper tanks in which the shavings may be soaked in water. In this condition bark-shavings may be kept for many days. For the same reason (_i.e._, the heating effect and consequent tackiness) the custom followed on some estates of allowing coolies to keep bark-shavings in their "lines" until they have accumulated a fair quantity cannot be commended, quite apart from the possibility of actual loss by theft, which is thus rendered easy. It will be clear that where the trees are scrapped efficiently before tapping, the amount of rubber to be obtained from the treatment of pure dry shavings would be almost nil, and would scarcely repay the cost of collection and working. In actual practice, however, it is not possible to guarantee that the shavings are free from some scrap-rubber. Shavings brought in by Tamils and Javanese carry only a small amount of rubber, whereas where Chinese tappers are employed the yield of rubber may be as high as 35 to 40 per cent. upon the total weight of the material treated. Few estates now are not equipped with "scrap-washers"--machines specially designed for removing the bark from the rubber--and if they function efficiently the resulting crepe should be free from bark-particles. COLLECTION OF EARTH-SCRAP.--This, the lowest grade of rubber, is found at the base of the tree. Theoretically, if proper precautions are observed, the amount should be comparatively small, but in actual practice it may be very appreciable. The usual contributory causes are: (_a_) Failure to replace cups beneath the spouts of trees which continue to drip latex after collection. (_b_) Collection of latex at too early a stage. (_c_) Failure on the part of the tapper to ensure the flow of latex, by means of the spout, into the cup. (_d_) Flowing of latex over the edge of the cut before it reaches the vertical channel. (_e_) "Wash-cuts" on wet days, when the volume of rainwater down the tree is sufficient to wash the latex out of the cup.