Febeuaky 18, 1910] 



SCIENCE 



273 



wider limit than their heats of combustion, being 

 comparable to the variations in the boiling point. 

 In general, the isomers with higher boiling points 

 possess lower compressibility, and the greater the 

 density the less the compressibility; but there are 

 interesting minor variations in these relationships 

 which deserve further investigation. Ortho- and 

 metasylene and ethyl benzene also were investi- 

 gated. The authors received important pecuniary 

 assistance from the Carnegie Institution of 

 Washington. 



Electrochemical Investigation of Liquid Amalgams 

 of Thallium, Indium, Tin, Zinc, Cadmium, Lead, 

 Copper and Lithium: T. W. Eichaeds, J. H. 

 Wilson and R. N. Gaebod-Thomas. 

 The investigation was a continuation of the re- 

 search concerning amalgams of zinc and cadmium, 

 described in a recent paper by Richards and 

 Forbes. The object was to extend the study to 

 elements possessing other valences and to study 

 more accurately the phenomena investigated. The 

 electromotive forces (and their temperature-co- 

 efficients) of various cells containing amalgams 

 of the eight metals named in the title were meas- 

 ured with many precautions against experimental 

 errors. Tliallium and indium were found to be- 

 have in the same manner as cadmium, but in a 

 much more exaggerated degree. Tin and lead 

 were found to behave in the same manner as zinc, 

 but likewise in a more exaggerated degree. It 

 was shown that the greater part of these devia- 

 tions from the concentration law may be explained 

 by the heat of dilution of the amalgam, according 

 to the equation of Cady. The temperature co- 

 efficient of a cell of this type was shown to corre- 

 spond closely with the requirement of this equa- 

 tion. The difficulties of the actual measurement 

 of thermochemical data involving amalgams were 

 emphasized, and many errors in the work of 

 previous investigators were discovered. It was 

 shown that the deviations from the simple con- 

 centration law in every case decreased as the dilu- 

 tion increased, so that upon reaching a concentra- 

 tion of 0.01 gram-atom per liter all the amal- 

 gams investigated behaved practically as ideal so- 

 lutions. 



Further Investigation concerning the Atomic 

 Weights of Silver, Lithium and Chlorine: Theo- 



DOEE W. RiCHABDS and HOBAET HUED WiLLAED. 



This investigation consisted in a careful study 

 of three ratios, namely, LiCl/AgCl, LiCl/Ag and 

 LiClOj/LiCl. By means of the latter two ratios 

 the ratio of O^/Ag was calculated, and new values 



were obtained in an entirely original way for the 

 atomic weight of silver, lithium and chlorine. In 

 the process of this work new methods of purify- 

 ing lithium salts better than any preceding were 

 devised. The lithium chloride was fused in such 

 a way as to show perfect neutrality to the most 

 sensitive indicators, and was weighed in a strictly 

 anhydrous condition. The preparation of per- 

 chloric acid also was subjected to rigid scrutiny, 

 and this substance was made in a state of un- 

 usual purity. A new precise method was devised 

 for converting lithium chloride into lithium per- 

 chlorate, and its sources of error were carefully 

 examined. 



Tlie atomic weight of lithium was found to be 

 very nearly 6.94 (much less than Stas's value) 

 and that of silver 107.871, if oxygen is taken as 

 16.000. 



On the Velocities of Certain Reactions between 

 Metals and Dissolved Halogens: Ralph G. van 

 Name and Geaham Edgae. 

 Under like conditions the metals mercury, 

 cadmium, zinc, copper and silver were foimd to 

 dissolve in an aqueous iodine solution containing 

 a large excess of potassium iodide at practically 

 the same rate. In bromine mercury dissolves 

 somewhat faster, in euprie bromide much slower 

 than in iodine. The so-called diffusion theory of 

 reaction veloeitj', of Noyes, Whitney and Nernst, 

 seems to give a satisfactory explanation of the 

 results obtained, as regards both the observed 

 agreements and the change in the velocities with 

 the conditions. 



The Estimation of Radium Emanation and of 

 Radium in Common Materials : Meele Randall. 

 A definite quantity of radium emanation, ob- 

 tained from a definite volume of a solution of the 

 mineral uraninite, was introduced into electro- 

 scopes of the various types now in use for de- 

 termining radium emanation. The values for the 

 ionization current due to the emanation associ- 

 ated with one gram of uranium varied from 2.36 

 X 10"" amperes for a Boltwood type to 4.80 X 

 10"" for a Schmidt type. Thus it is incorrect to 

 assume, as many European investigators have 

 done, that data, expressed in amperes or C.G.S. 

 units, obtained with one instrument, are directly 

 comparable with those obtained with another. 

 Some forms of apparatiis for separating the 

 emanation from solution removed a greater per- 

 centage than others. With all types the per- 

 centage loss was greater when the total amount 

 of emanation present was small. Accurate de- 



