the Cretaceous-Tertiary Problem. 227 



the evidence is more conjectural and the results pro- 

 visional. If correctly and judicially used they can even 

 then provide very precise correlation data; but like all 

 other instruments of precision they require expert knowl- 

 edge and handling. A scalpel is much better for accurate 

 dissecting work than an ax; but not in the hands of a 

 lumberjack. Doctor Cross quite naturally despises the 

 ^^poor little mammals. '' But they have their uses none 

 the less, to those who know how to use them. 



American Museum of Natural History, 

 New York City. 



SCIENTIFIC INTELLIGENCE 



I. Chemistry and Physics. 



1. The Estimation of Sodium Hydrosulphite. — James Hol- 

 LiNGSWORTH Smith, who uses for this salt the name ''sodium 

 hyposulphite," which, although not incorrect, is somewhat con- 

 fusing on account of the extensive commercial use of this old 

 name for sodium thiosulphate, has found that the usual methods 

 for determining its purity are unsatisfactory because they require 

 either extreme care to exclude air or unusual reagents. He has 

 therefore modified the gravimetric method of Seyewetz and Bloch 

 which consists in treating the dry salt with an ammoniacal solu- 

 tion of silver chloride and collecting and weighing in a Gooch 

 crucible the metallic silver formed according to the equation 

 Na^S.O, + 2AgCl + 4NH4OH = 2(NHJ2S03 + 2NaCl + 2H2O 

 -|- 2Ag. This method has the advantage over the use of iodine, 

 permanganate or ferric sulphate as oxidizing agents in the fact 

 that sulphite or thiosulphate as impurities do not affect the 

 result, but it has the disadvantage of being inaccurate on account 

 of the fact that any insoluble impurity in the salt is weighed with 

 the metallic silver. Smith has therefore modified the method by 

 dissolving the metallic silver, after washing it with an ammonia- 

 cal solution of ammonium nitrate, in nitric acid and determining 

 the silver volumetrically by means of a thiocyanate solution. 

 Instead of silver chloride he uses silver nitrate in ammoniacal 

 solution for the original precipitation. The method appears to 

 be a very good one. — Jour. Amer. Chem. Sac, 43, 1307. 



H. L. w. 



2. The Calculations of Analytical Chemistry ; by Edmund H. 

 Miller. 8vo, pp. 201. New York, 1921 (The Macmillan Com- 



