December 4, 1896.] 



SCIENCE. 



845 



to the adjacent dark space a sudden tem- 

 perature change amounting to four or five de- 

 grees was observed in almost all cases. 



It can hardly be doubted that Mr. Wood's 

 results will prove of great assistance in the de- 

 velopment of the theory and explanation of 

 Geissler discharge phenomena. 



The Specific Heats of the Metals: By F. A. 

 Waterman. Dr. Waterman's article contains, 

 first, a critical discussion of the methods hereto- 

 fore used in the determination of specific heat, 

 which is accompanied by a table giving the values 

 obtained by various experimenters. This table 

 appears to have been prepared with much care 

 and will be found extremely useful. The paper 

 also contains the description of a new form of ca- 

 lorimeter devised by Dr. Waterman and used by 

 Mm in determining the specific heats of bismuth, 

 tin, aluminum, copper, gold and zinc. A de- 

 scription of the apparatus is beyond the limits 

 of this abstract, but the instrument seems to 

 be capable of a high degree of accuracy, and the 

 results are thoroughly concordant. 



Dr. Waterman has used especial care in ob- 

 taining pure .specimens of the metals studied, 

 and in this way avoids what is perhaps the 

 commonest source of error in previous deter- 

 minations. 



The Viscosity of Mercury Vapor : By. A. A. 

 NoYES and H. M. Goodwin. Determinations 

 of the viscosity of mercury vapor, hydrogen and 

 carbon dioxide were undertaken by the writers, 

 with the object of finding whether the viscosity 

 of a gas can be made to furnish a criterion for 

 distinguishing between monatomic and poly- 

 atomic molecules. As the authors very justly 

 state, the recent discussions in connection with 

 the atomic weight of argon have thrown some 

 discredit on the indications of the specific heat 

 ratio in this connection. 



The kinetic gas theory shows that a simple 

 relation exists between the viscosity of a gas 

 and the cross section of the gaseous molecule. 

 If the space occupied by a molecule is widely 

 different from that occupied by an atom, we 

 should therefore expect the viscosity of a mona- 

 tomic gas such as Hg to differ widely from that 

 of a diatomic gas like H. Such differences were 

 however not found. Messrs. Noyes and Good- 

 win conclude that the space occupied by a mole- 



cule is of the same order of magnitude as that oc- 

 cupied by an atom, and that the viscosity can 

 therefore not be used to determine the molecular 

 complexity. 



An Example in Thermometry : By A. S. Cole 

 and E. L. Duegan. The piece of work de- 

 scribed in this paper was undertaken at the 

 suggestion of Prof. Wm. A. Rogers, and has 

 to do with the calibration of a mercury the- 

 mometer. The object of the paper is to give a 

 description of the methods used in such calibra- 

 tions, with sufficient detail to enable one to be- 

 come thoroughly familiar with the process. The 

 thermometer studied was one of those em- 

 ployed by Profs. Morley and Eogers in their 

 work on the coefficient of expansion of Jessop's 

 steel. 



A Study of the Apparent Capacity of Condensers 

 for Short-charge Periods : By H. V. Cabpenter. 

 The ' soaking in ' of a charge in a condenser 

 is a phenomenon whose existence has long been 

 known, and for which various explanations 

 have been suggested. As is well known, the 

 phenomenon leads to troublesome complications 

 when it is desired to measure the capacity of a 

 condenser, for the amount of charge taken up 

 by a condenser will depend upon the duration 

 of charging. Mr. Carpenter has undertaken to 

 determine experimentally the variation in the 

 apparent capacity due to this cause, as the 

 period of charge is increased from a few thou- 

 sandths of a second up to about half a second. 

 Data are given for a mica condenser, a conden- 

 ser made of paraffined paper, and one furnished 

 by the Stanley Electric Manufacturing Com- 

 pany. The charging circuit was made as nearly 

 as possible non-inductive, so that the effect 

 studied could be separated from the similar 

 effect which might be produced by self-induc- 

 tion. Eesults are shown in form of curves giv- 

 ing the variation in the charge as a function of 

 the time of charge. Observations were made at 

 various voltages. It appears that the rapidity 

 of absorption varies greatly with different con- 

 densers. It seems also to depend somewhat 

 upon the potential to which the condenser is 

 charged. The rate of absorption is found not 

 to be proportianal to the potential difference. 

 Mr. Carpenter's curves indicate great uniformity 

 in the behavior of a given condenser, and while 



