| 
SECTIONAL TRANSACTIONS.—B. 359 
Hudson’s lactone rule, both tetramethyl y-gluconolactone (I) and tetramethyl 
y-mannonolactone (II) would be dextro-rotatory. This is true for (I) in 
all solvents, but the rotation of (II) may be strongly positive or negative, 
according tothe solvent used. The rotatory power of (II) is expressible by the 
_ formula «, = k,/(A* — 0°06) — k2/(M* — 002) and is dependent upon absorp- 
tion bands at wave lengths given by 4? = 0-06 and \2 = 0-02, and associated 
respectively with the dissymmetric double bond of the carbonyl group and 
with the single linkages inthe molecule. In the far ultra-violet the lactone 
rule is always obeyed, but wide departures from the rule are encountered in 
the visible region in those solvents for which kz > ky. 
The amide rule, which applies to many «-hydroxy- and «-methoxy- 
aliphatic acid amides, depends on the induced dissymmetry of the CO group 
and the configuration of the second carbon atom determines the rotation. 
Failure to follow the rule cannot be explained in terms similar to those used 
in the case of (II). In 4 : 6-dimethyl mannonamide the sign of the induced 
term is opposite to that which would be expected. ‘The result is attributable 
to the vicinal effect of the cis hydroxy groups which are responsible also for 
the failure of «- and 8-mannose to follow the iso-rotation rules. 
Dr. S. R. Carter.—The determination of molecular weights of carbohydrate 
derivatives by osmotic pressure measurements (11.45). 
This investigation (with B. R. Record) deals mainly with the methylated 
and acetylated derivatives of carbohydrates of molecular weight between 
1,500 and 50,000 in solutions of organic solvents. -According to the theory 
of van’t Hoff a 1 per cent. solution of a substance of molecular weight 
20,000 should produce an osmotic pressure of ca. 10 cm. water pressure. 
The liquid may be a colloid dispersion rather than a molecular dispersion, 
but we have assumed that the osmotic pressure P = RTn/N where n = 
number of particles in unit volume and N = Avogadro’s Constant. ‘The 
solutions were usually made up in chloroform and the semipermeable 
membrane was a disc of ‘ Viscacelle’ adjusted for porosity by immersion 
in alcohol-water mixtures. 'The osmometer consisted of two glass bells, 
accurately ground and clamped together, the membrane being enclosed 
between the ground joints. One bell contained the solution (5 c.c.) and 
Was connected to an air chamber and a water or mercury manometer, whilst 
the other bell contained pure solvent. The pressure was adjusted by trial 
until no movement of the solvent occurred as judged by the position of the 
Meniscus in a capillary connected with the solvent bell. ‘The thermostat 
was kept at 20°:00° + o0-’o1° C. The concentration of the solution was 
determined by weighing the evaporated residue from 1 c.c. on the 
-micro-balance. The observed pressure P is related to the concentration 
by the equation P = cRT + kc”. The values of P/c are graphed against 
¢ and extrapolated to zero concentration, when the last term vanishes 
and P =cRT. The molecular weights so determined agree with the values 
obtained by the chemical methods, 
Mr. E. G. Cox.—Crystallographic evidence on the form of the pyranose 
ring (12.5). 
Owing to the complexity of the problem, no complete structure deter- 
Mination by X-ray methods has so far been carried out in the case of 
‘crystalline carbohydrates. Much preliminary work has been done during 
the past few years, sufficient in fact to establish certain results not directly 
accessible by purely chemical methods. The most important of these is 
N2 
