584 



SCIENCE 



[N. S. Vol. LIV. No. 1406. 



solubility is critically discussed. It is shown that: 

 (o) for a fixed solute in a series of different sol- 

 vents, increasing solubility and increasing com- 

 pound formation proceed in parallel; (b) for a 

 series of different solutes of high melting-point in 

 a fixed solvent, solubility and compound formation 

 also proceed in parallel at low temperatures. Salts 

 of a very weak base exhibit increasing hydrate 

 formation and increasing solubility in water as the 

 acid radical X diverges from OH; salts of a very 

 weak acid show the same behavior as R diverges 

 from H. The increase in the solubility of a diffi- 

 cultly soluble salt in water on addition of a second 

 salt containing a common ion, due to complex salt 

 formation, is dependent upon the diversity of the 

 variable radicals. The extension of these rules to 

 non-aqueous solutions and their importance in 

 analytical chemistry are noted. 



The complete analysis of an insoluble silicate 

 with a single fusion: F. P. Dtjnnincton. Fuse the 

 powdered silicate with six parts of lithium car- 

 bonate in a gold crucible. The melt is dissolved 

 in dilute acid, evaporated, heated and the siHoa 

 separated as usual. To the solution of chlorides 

 add ammonia, etc. To remove alumina, iron and 

 manganese, precipitate lime as oxalate; magnesia 

 by ammonium phosphate and then, with little cal- 

 cium chloride and ammonium carbonate remove all 

 excess of phosphoric oxide ; evaporate filtrate, vola- 

 tilize animonium salts. The residue is digested in 

 a mixture of absolute alcohol and ether, which 

 readily dissolves the lithium chloride; filter off the 

 potassium and sodium chlorides, weigh and sep- 

 arate them. 



Alisarine-iron lalces: A. W. Bull and J. E. 

 Adams. 



Adsorption of tannin hy gelatine: A. W. Bull 

 AND J. E. Adams. 



The theory of molecular-compound formation : V. 

 E. Kokatnue and H. W. Stiegler. This theory 

 is based on an observation that molecules in molec- 

 ular compounds invariably contain elements that 

 belong to 5, 6, 7, 8 groups of the periodic system. 

 Assumptions : ( 1 ) Molecules combine through unsat- 

 uration or through latent valences of elements, 

 especially non-metallic, belonging to aforesaid per- 

 iodic groups. (2) These elements exhibit their 

 highest capable valence and combine through these 

 by single or double bonds. But all their valences 

 may not be satisfied. (3) Active groups and con- 

 ditions of molecules may influence this latent va- 

 lency and give rise to chain-compounds and conse- 

 quent isomerism. 



The diffusion of hydrogen through metals: H. 

 Gr. Deming and B. C. Hendricks. Sheet metal of 

 0.15 mm. thickness was clamped between heavy 

 steel blocks in an electric furnace, the diffusion 

 area being circumscribed on the face of each block 

 by a pair of concentric circular knife-edges. The 

 channel between the knife-edges in the block on the 

 incoming side was connected to a vacuum-pump; 

 on the outgoing side to compressed nitrogen. The 

 diffusion was thus limited to a definite area of 

 metal or perfectly uniform temperature, even 

 though the blocks were never pressed against the 

 metal tight enough to make a gas-tight joint. 

 Aluminum is impervious to hydrogen up to its melt- 

 ing point. Quantitative data have been obtained 

 for copper, iron, and other metals. 



The adsorptive property of fullers earth : Stuart 

 J. Bates and Alfred Stamm. 



Charles L. Parsons, 



Secretary 



AMERICAN MATHEMATICAL SOCIETY 



The two hundred and seventeenth regular meet- 

 ing of the American Mathematical Society was 

 held at Columbia University, on Saturday, October 

 29, 1921, extending through the usual morning and 

 afternoon sessions. The attendance included forty 

 members of the society. Thirty new members were 

 elected. 



The following papers were read at this meeting: 



Total geodesic curvature : J. K. Whittemore. 

 On the composition of polynomials: J. F. ElTT. 

 Complete determination of polynomials whose in- 

 verses can he expressed in terms of radicals: J. F. 



ElTT. 



Concerning continuous curves in the plane: E. L. 



MOOKE. 



Concerning the relation of a continuous curve to 

 its complement in space of three dimensions: E. L. 

 Moore. 



An algeiraic solution of Einstein's cosmological 

 equations: Edward Kasner. 



On hiharmonic functions: T. H. Gronwall. , 



General formulation of a combinatory method 

 used by William Emerson and others: L. H. ElCE. 



A theorem on loci connected with cross-ratios: 

 J. L. "Walsh. 



A generalization of the notion of covariants : Ii. 



B. EOBINSON. 



Inductances of grounded circuits: G. A. Camp- 

 bell. 



E. G. D. ElCHARDSON, 



Secretary 



