diluted acetone extract was neutral to test papers, yet the residue was acid in character. It contained combined nitric groups, fused below 200 deg. giving off acid vapours, and afterwards burning with a smoky flame. On adding lead acetate to the original clear solution, a well-marked precipitation was determined. The lead compounds thus isolated are characteristic. They have been obtained in various ways and analysed. The composition varies with the character of the solution in which the lead compound is formed. Thus in the opalescent or milky solutions in which a proportion of cellulose nitrate is held in solution or semi-solution by the acetone still present, the lead acetate causes a dense coagulation. The precipitates dried and analysed showed 16-20 p.ct. PbO and 11-9 p.ct. N. It is clear that the cellulose nitrates are associated in these precipitates with the lead salts of the acid compounds in question. When the latter are obtained from clear solutions, i.e. in absence of cellulose nitrates, they contain 60-63 p.ct. PbO and 3.5 p.ct. N (obtained as NO). In further confirmation of the conclusion from these results, viz. that the nitrocelluloses with no tendency to combine with PbO are associated with acid products or by-products of the ester reaction combining with the oxide, the lead reagent was allowed to react in the presence of 90 p.ct. acetone. Water was added, the disintegrated mass collected, washed with dilute acetic acid, and finally with water. Various estimations of the PbO fixed in this way have given numbers varying from 2 to 2.5 p.ct. Such products are perfectly stable. This particular effect of stabilisation appears, therefore, to depend upon the combination of certain acid products present in ordinary nitrocelluloses with metallic oxides. In order to further verify this conclusion, standard specimens of cellulose nitrates have been treated with a large number of metallic salts under varying conditions of action. It has been finally established (1) that the effects in question are more particularly determined by treatment with salts of lead and zinc, and (2) that the simplest method of treatment is that of boiling the cellulose nitrates with dilute aqueous solutions of salts of these metals, preferably the acetates. The following results may be cited, obtained by boiling a purified 'service' guncotton (sample C) with a 1 p.ct. solution of lead acetate and of zinc acetate respectively. After boiling 60 minutes the nitrates were washed free from the soluble metallic salts, dried and tested.