1866,] 



Mr. Abel on Gun-cotton, 



105 



of weight varying between 81*8 and 82'5 upon 100 parts of cotton, if 

 submitted for 24-48 hours to treatment with a very considerable excess 

 (about 50 parts to 1 of cotton) of the acid mixture. Similar results may also 

 be obtained by repeatedly treating the same sample of cotton for compara- 

 tively brief periods with fresh quantities of acid, provided this treatment 

 be not too greatly prolonged. Lower results (somewhat above or below 

 78 upon 100 parts of cotton) are obtained if the cotton be submitted to 

 treatment with a large excess of acid for only brief or for very protracted 

 periods, or if it be left for about 24 hours in contact with a comparatively 

 limited proportion of acid (10 or 15 to 1 of cotton). The increase of 

 weight which 100 parts of pure cellulose should sustain by complete con- 

 version into a substance of the formula Gg N3 O^^, is 83*3 ; if converted 

 completely into a substance of the composition Q^^ O^g, 5 Og, it 

 should sustain an increase in weight of 77 '7^. 



There is strong evidence that the differences between the highest results 

 furnished by carefully purified cotton-wool, and the number 83*3, are to 

 be principally ascribed to the small proportions of foreign matter still ex- 

 isting in the fibre at the time of its conversion. 



The maximum increase of weight sustained by cotton of ordinary quality, 

 such as is used in gun-cotton-manufacture, is, as might have been antici- 

 pated, below the result obtained, under similar conditions, with cotton of 

 finer quality and more thoroughly purified. The highest numbers ob- 

 tained by treatment of such cotton, in small quantities, with a considerable 

 excess of acid, were somewhat below 181, from 100 of cotton. The in- 

 crease of weight which this quality of cotton sustains is, however, more 

 generally about 78 per cent. 



Experiments are quoted which show that the attainment of lower results 

 with cotton of ordinary quality is ascribable to the existence of higher 

 proportions of foreign matters in the cotton under treatment. 



Some quantitative manufacturing experiments yielded results consider- 

 ably below those obtained with some of the same cotton in laboratory 

 operations (171 and 1/6 of gun-cotton having been produced from 100 of 

 cotton). The causes of these differences are investigated and explained. 



The identity in their characters, and close resemblance in composition, of 

 the most perfect results of laboratory experiments, and of the purified 

 products of manufacture, the close approximation frequently exhibited by 

 the weight of the former to the theoretical demands of the formula 

 Gg N3 O^^ (which may be expressed as 



and the satisfactory manner in which the unavoidable production of some- 

 what lower results in the manufacturing operations admits of practical 

 demonstration, appear to afford conclusive evidence of the correctness of 



K 2 



