When a solution which has been treated in this way is precipitated with lead acetate and kept neutral to litmus, the free phosphate and hexosephosphate are thrown down and the co-enzyme remains in solution. The filtrate can be freed from lead by means of sulphuretted hydrogen and neutralised, and then forms a solution of co-enzyme free from phosphate and hexosephosphate but still containing combined phosphorus. More complete purification than this has not yet been accomplished. Occasionally the precipitate of lead salts retains some of the co-enzyme, apparently by adsorption, but usually the greater part remains in the solution (Harden and Young). The co-enzyme is partially removed from yeast-juice by means of a colloidal solution of ferric hydroxide (Resenscheck). A precipitate is thus obtained which contains phosphorus and resembles boiled yeast-juice in its regenerative action on yeast-juice rendered inactive by fermentation. It has not, however, so far been found possible to isolate any definite compound from this precipitate. There are also indications that when yeast-juice, either fresh or boiled, is electrolysed, the co-enzyme tends to accumulate at the cathode [Resenscheck, 1908, 1, 2]. Buchner and Klatte [1908] made use of yeast-juice rendered free from co-enzyme by incubation with sugar solution to examine the nature of the agent by which the co-enzyme is destroyed. This agent is certainly an enzyme, since boiled yeast-juice can be preserved with unimpaired powers for a considerable length of time, and suspicion fell naturally, in the first instance, on the endotryptase of the yeast cell. Direct experiment showed, however, that yeast-juice, which, when fresh, rapidly destroyed the co-enzyme of boiled juice, lost this power on preservation, but retained its proteoclastic properties without diminution, so that the tryptic enzyme could not be the one concerned. The direct action of commercial trypsin on boiled yeast-juice also yielded [p068] a negative result, although this cannot strictly be regarded as an indication of the effect of the specific proteoclastic enzymes of yeast-juice. On the other hand, it was found that when boiled juice was treated for some time with an emulsion containing the lipase of castor oil seeds, the co-enzyme was completely destroyed. This is a result of great importance, inasmuch as it probably indicates that the co-enzyme is chemically allied to the class of substances hydrolysable by lipase, i.e. to the fats and other esters. Further, observations by Buchner and Haehn [1909] have shown that digestion with potassium carbonate solution containing 2·5 grams per 100 c.c. also brings about the destruction of the co-enzyme, and that this is also slowly accomplished by the repeated boiling of the juice. The co-enzyme is also destroyed both by acid and alkaline hydrolysis, and when the solution is evaporated to dryness and the residue ignited.