418 



SCIENCE. 



[N. S. Vol. XXIII. No. 585. 



made by the electromotive force method 

 recently described by Hulett and Minchin.' 

 One millimeter deflection of the gal- 

 vanometer corresponded approximately to 

 .0005 volt. The results are given in the 

 following table: 



From the above a deflection of 1 mm. 

 corresponds to the presence of zinc in the 

 ratio of 1 : 1,500,000. In numbers 5 and 3 

 the degree of zinc impurity vi^as known 

 only approximately, and since there was 

 also present some lead and tin little weight 

 should be given these results. In numbers 

 a, b and c the ratio of zinc to mercury was 

 defiinitely Imown. The distillate from 

 this zinc amalgam condensed in three sepa- 

 rate condensing chambers showed prac- 

 tically no trace of zinc. The mercury 

 against which the above was balanced in the 

 test cell was carefully and repeatedly puri- 

 fied by the 'wet' method. The degree of 

 purity indicated above was altogether un- 

 locked for, since the physical conditions 

 in the still-temperature, vacuum, etc., are 

 such as favor the vaporization of zinc too, 

 and hence we should naturally expect zinc 

 present to a more or less extent iu the dis- 

 tillate. The result seemingly points to a 

 suppressing action exerted by the electric 

 forces upon the zinc ions. This phase of 

 the phenomena is the subject of further 

 inquiry. 



Difference in the Coefficient of Discharge 

 of Steam through a Single Orifice and 

 through a Number of Orifices near Each 



' Phys. Rev., Vol. XXI., December, 1905. 



Otlter: D. S. Jacobus, Stevens Institute 

 of Technology. (To be published in the 

 Transactions of the American Society of 

 Mechanical Engineers.) Section D. 



Note on the Distribution of Energy in 

 Fluorescence Spectra: Edward L. Nich- 

 ols, Cornell University. 

 The fluorescence spectra of solids and 

 liquids are, so far as known, confined to 

 the visible wave length. Observers in this 

 field of optics have until very recently 

 contented themselves with a description of 

 the appearance of the fluorescence band or 

 bands and an indication of its approximate 

 limits towards the red and violet. Pro- 

 fessor Merritt and the present writer have, 

 however, succeeded in making spectropho- 

 tometric measurements of the fluorescence 

 of numerous substances and have published 

 curves in which the distribution of inten- 

 sities of fluorescence spectra are expressed 

 in terms of the intensities of the corre- 

 sponding wave lengths in the spectrum of 

 the acetylene flame.^ 



By means of measurements of this source 

 of light made by G. W. Stewart and inde- 

 pendently by W. W. Coblentz, using a 

 mirror spectrometer with rock salt prism 

 and a radiometer, a curve showing the dis- 

 tribution of energy in the visible spectrum 

 may be plotted. The writei-'' has piiblished 

 in a recent paper a curve based upon these 

 data which gives the distribution of energy 

 in the acetylene flame and has checked the 

 values thus obtained by means of spectro- 

 photometric comparisons between the Hef- 

 ner and the acetylene flames and Ang- 

 strom's curve for the distribution of en- 

 ergy in the spectrum of the Hefner flame. 

 This curve makes it possible to convert 

 the spectrophotometric curves for the flu- 

 orescence of any given substance into 

 curves of the distribution of energy in its 



' Nichols and Merritt, Physical Review. 

 " Nichols, Physical Revieio, Vol. 21, p. 147. 



