SECTIONAL TRANSACTIONS.— B. 319 



The configuration of the tetramino metalhc complex — whether tetra- 

 hedral or uniplanar — is still in doubt, however, since the available experi- 

 mental evidence remains indecisive. 



The salient points in this evidence are briefly reviewed, and their signifi- 

 cance discussed. 



Dr. S. SuGDEN. — Planar configuration of diamagnetic nickel complexes. 



The wave-mechanics valency theory of Pauling gives a correlation 

 between {a) the space distribution of valency links ; {b) the azimuthal 

 quantum numbers of the electron levels taking part in the links , and 

 (c) in some cases the magnetic moment of the atoms. 



The case of nickel is of special interest, since not only 5 and />, electrons 

 (with azimuthal quantum numbers i and 2) are concerned , but also d electrons 

 (with azimuthal quantum number 3). If one (f electron level is concerned, 

 then Pauling finds that four strong links should be formed in one plane. 

 At the same time the nickel should change from a paramagnetic ion to a 

 diamagnetic complex. 



Co-ordination complexes of nickel are found to be sharply divided into 

 two groups as regards their magnetic behaviour. One group is paramagnetic, 

 the other diamagnetic. Of the latter the double cyanide K2Ni(CN)4 and 

 the dimethylglyoxime are the best-known compounds. 



By synthesising unsymmetrical glyoximes it has been found that the 

 nickel derivatives occur in two forms, which are interconvertible. These 

 appear to be the cis-trans isomerides required by Pauling's theory of a 

 planar configuration. 



Mr. J. D. Bernal. — Crystal structure and stereochemistry. 



Friday, September 2. 



Discussion on The constitution of polysaccharides , with special reference to 

 fibres (Prof. W. N. Haworth, F.R.S. ; Prof. Dr. L. Zechmeister ; 

 Prof. Dr. H. Staudinger ; Dr. E. L. Hirst ; Prof. Dr. H. Mark ; 

 Mr. W. T. Astbury) :— 



Prof. W. N. Haworth, F.R.S. — Introduction. 



In a brief historical survey of the development of the constitution of 

 polysaccharides, it is shown that the occurrence of cellobiose as preformed 

 units in cellulose is established by chenaical methods, and that the structure 

 of cellulose rests ultimately on the constitution previously assigned to this 

 biose. The mutual linking through position i : 4 of p-glucopyranose units 

 in chain fashion is thus the fundamental principle of modern cellulose 

 structures. Recent work by Haworth and Machemer has indicated that a 

 methylated cellulose can be utilised to gain valuable information (i) as 

 to the character of the chain, whether open or closed, and (2) as to the 

 approximate chain length in this very representative derivative. Tetra- 

 methyl glucose, in yield of o-6 per cent., is obtained by hydrolysis of this 

 specimen, so that the minimum mean value for the chain length of cellulose is 

 given as 200 glucose units. Similar experiments with intermediate hydrolysis 

 products of cellulose have been carried out and their molecular size 

 determined. 



