The trouble becomes acute sometimes where wooden tanks are employed. Unless the tank and the partitions are thoroughly and regularly cleansed, the wood may become coated with a bacterial slime, which is capable of causing what may be termed "fermentation" of the latex layers in contact. The tank should be thoroughly cleaned occasionally with a weak (5 per cent.) solution of sodium bisulphite. The partitions should be scrubbed and placed in the sun twice or three times a week. (2) Allowing latex to stand too long before treatment. This point needs no further expansion. (3) The use of a latex of too high a rubber content. Such latices are difficult to handle in order to secure uniform mixture with the coagulant. (4) The use of too concentrated a solution of coagulant. In conjunction with (3) there may be a rapid and irregular coagulation, giving rise not only to decomposition in parts (and subsequent formation of gas), but also to the formation of true "air-bubbles" by inclusion of air during stirring. (5) The use of insufficient coagulant. Coagulation is slow and incomplete. (6) Defective straining and skimming. Small flocculated particles of rubber may pass, or be rubbed through, the strainer. If allowed to remain, they act as local points of danger. (7) The proximity of the coagulating latex to some source of heat, or exposure to sunlight. (8) Any delay of drying in the preliminary stages, either before or after the rubber enters the smoke-house. BLISTERS.--This description aptly fits the case in which sheet-rubber in the smoke-house exhibits large bubbles of gas which distend the surface of the rubber. When subjected to pressure, small "balloons" are formed, which burst with a perceptible report. It was formerly the belief that this defect was occasioned solely by an abnormally high temperature. That such is not the case can be shown by the experience of estates which have had only the rubber of a particular day or short period affected under normal factory conditions. At the same time it is not disputed that the heat of the smoke-house exerts an influence (causing expansion and distension), but it is advanced that the gases had begun to form before the rubber entered the house. The view held is that decomposition had supervened or was taking place--probably from one or more of the causes enumerated in the preceding paragraphs. The heat of the smoke-house only serves to exaggerate the effect. It is acknowledged that the degree of decomposition must be initially greater than in the ordinary incidence of "bubbles." Beyond this point we are not in a position to put forward any definite supposition as to the apparently haphazard occurrence of the phenomenon.