^58 SECTIONAL TRANSACTIONS.— B. 



The formation of this intermediate urethane affords striking evidence of the 

 stabilising effect of methyl groups in the sugar chain. It is also interesting 

 in showing the conversion of a carbohydrate into a derivative in which nitrogen 

 is present in a stable cyclic substituent. 



4. Dr. E. L. Hirst. — The Composiilon of Esparto Cellulose. 

 Esparto grass, after removal of waxes, lignins, &c., in the ordinary course 



of paper-making, gives a material which is homogeneous and is similar to cotton 

 cellulose in appearance, but differs markedly in that it gives on distillation 

 with 12 per cent, aqueous hydrochloric acid an amount of furfural corresponding 

 to the presence of i8 to 20 per cent, of a pentosan. 



Acetylation of this esparto cellulose cannot be effected so readily as in the 

 case of cotton cellulose, but by slight modification of Barnett's method, in 

 which sulphuryl chloride is used as a catalyst, almost quantitative yields of 

 acetates have been obtained v/ithout appreciable loss of the pentose residue as 

 shown by furfural estimations. This acetate mixture has been subjected to 

 the action of acid methyl alcohol in sealed tubes at 130° C, when it is found 

 that after prolonged digestion practically the whole of the material dissolves 

 and the solution then contains methylglucoside along with a proportion of a 

 methyl pentoside. The pentose has been identified as xylose, and confirmation 

 of this has been obtained by the isolation from esparto cellulose of a pentosan 

 which or. hydrolysis is converted into a reducing sugar identical with ordinary 

 xylose. In the course of quantitative experiments esparto cellulose has thus 

 been converted into methylglucoside and methylxyloside in such a manner that 

 90 per cent, of the whole material can be accounted for. On the assumption 

 that no other hexose or pentose is present, the analytical results indicate that 

 the overall yield of methylglucoside is 95 per cent, and that of methylxyloside 

 68.5 per cent, of tlie theoretical amount. The loss in yield, as indicated by 

 the results of control experiments, is due to the destruction of pentose owing 

 to furfural formation during the digestion in the sealed tubes. 



The evidence therefore points to esparto cellulose being, to the extent of 

 90 per cent, at least, a definite chemical substance composed of glucose residues 

 and xylose residues present together in the proportions of 80 per cent, and 

 20 per cent, respectively. 



5. Prof. Sir William H. Bragg, F.E.S.— T/te Crystalline Structure 



of Organic Compo^inds . 



6. Joint Discussion whh Section K on Photo-Synthesis. (See 



p. 395.) 



Friday, September 8. 



7. Discussion on Valency and Polarity in Organic Compounds. 



(a) Opener: Prof. E. Egbinson, F.E.S. 



(b) Dr. J. Kenner. — The Significance of Induced Polarity. 



It is pointed out that Lapworth's derivation of the principle of induced 

 alternate polarities (Trans. Chcin. Soc, 1922, 121, 416) is a further develop- 

 ment of the \iews of Werner and Fliirscheim, witli the aid of postulates which, 

 subject to apparently reasonable assumptions, conform to the thermodynamic 

 condition for the attainment of stable equilibrium. Fry's electronic theory of 

 valency is based on the fallacy that two atoms in direct combination are neces- 

 sarily of opposite polarities, and would seem to lead to mutually irreconcilable 

 conclusions when applied to the consideration of certain closely related com- 

 pounds. The observed results are explicable in terms of the view expressed 

 by Lapworth, and also by Kermack and Robinson [ibid., 427), that polarity is 

 consequent upon constraint of the molecule (resulting, in the cases referred to, 

 from formation of molecular compounds), and on the further assumption that 

 of two alternative reactions that one will predominate in which the free energy 

 gradient is the greater. The important reservation, therefore, seems necessary 

 that before the outcome of a reaction can lie predicted by the theory of alternate 



