Cli 



PHYSICS, PROGRESS OF, IN 1902. 



549 



ay finds that the absorption band of the green- 

 absorbing ink should be a good deal narrower 

 than the region covered by the green sensation. 

 Thus the tints of the inks will not be truly com- 

 plementary to the three-color sensations. 



Spectroscopy. E. L. Nichols (Physical Review, 

 August and September, 1901) has compared the 

 visible radiation from incandescent carbon-rods 

 in vacuo with the radiation from an acetylene 

 flame, by means of a spectrophotometer. Since 

 the radiation in both cases is from incandescent 

 carbon, it was anticipated that similarity would 

 be found in the two spectra. This anticipation, 

 however, was far from being realized. The dis- 

 tribution of energy in the spectrum of the carbon 

 rod, instead of approaching that of the acetylene 

 flame, as the temperature of the rod is increased, 

 assumes an entirely unexpected character. Even 

 at low temperatures the change in the spectrum is 

 not simple, and after passing 1,100 the energy 

 in the yellow, which from the beginning increases 

 at a relatively more rapid rate than either the 

 red or the blue, becomes unexpectedly great. G. 

 Laubenthal (Annalen der Physik, April 8) has 

 measured in various ways the bands in the ab- 

 sorption spectra of the lakes formed by alkan- 

 nin and certain metallic salt solutions. In each 

 of the two groups studied the bands shift toward 

 the red with increasing atomic weights, in such 

 wise that the ratio between the wave-lengths of 

 the two bands of each spectrum is constant for 

 each group of metals. If the wave-lengths of the 

 absorption bands are plotted against atomic 

 weights, the curves have almost identical shapes 

 with those given in Ramage's paper on the emis- 

 sion spectra of the same groups of metals (Lon- 

 don Royal Society, May 12). In both there is a 

 break between sodium and potassium, and prob- 

 ably a similar break in the beryllium to calcium 



molecule will have, generally, a rotational vibra- 

 tion about an axis, which will emit . electro- 

 magnetic waves. Under certain circumstances 

 this may be converted into continuous rotation, 

 the atom or molecule thus becoming a perma- 

 nent magnet. 



Refraction. B. V. Hill (ibid., November, 1901) 

 has further investigated the property of dilute 

 solutions of gelatin and gum-arabic by which 

 when strained they become doubly refracting. 

 The dilute jellies were placed in thin brass tubes 

 closed by glass plates at the ends, and were 

 strained by squeezing the tubes between clamps 

 so that their cross-section was elliptical. The 

 double refraction first increases with the compres- 

 sion, then remains stationary, and finally dimin- 

 ishes. The solutions thus behave like solids that 

 can sustain only a small amount of strain. G. 

 Kucera and C. Forch (Physikalische Zeitschrift, 

 Jan. 1), having in mind the fact that the dielec- 

 tric constant of a liquid decreases with the tem- 

 perature, but in a different manner for different 

 substances, have studied the temperature varia- 

 tion of the refractive index, which should be 

 connected with the dielectric constant by Max- 

 well's law. The liquid is enclosed in a prism 

 whose refracting edge is horizontal, and is then 

 placed in a freezing mixture, where its refractive 

 index is measured by total reflection. The results 

 show that the refractive index, n, may be cal- 

 culated from formulae of the form n = a -f- bt + 

 ct 2 , where for alcohols a varies between 1.34 and 

 1.42 and 6 from 0.00046 to 0.000596. For carbon 

 disulfid, n =1.64362 0,000733 ;t + 0,000009^. P. 

 Zeeman (Archives N6erlandaises, 4, 1901) has 

 determined the optical constants of platinum up 

 to 200 to decide certain questions in dispute. 

 Kundt found that the index of refraction of plati- 

 num increases by about 27 per cent, for a tern- 



curve. This is interesting as correlating absorp- perature rise of 100, while Crude found for plati 



tion spectra with emission spectra, and also the 

 densities and melting-points, as having their ori- 

 gin in the same fundamental cause. J. Jeans 

 (Philosophical Magazine, November, 1901) ad- 

 vances the following hypotheses concerning the 

 structure of molecules and atoms to explain the 

 phenomena observed in the spectroscope. If each 



num, silver, and gold only very small varia- 

 tions. The author finds very small effects. F. 

 Pockels (Annalen der Physik, April 8) has experi- 

 mented on the effect of deformation on the opti- 

 cal properties of glass, and arrives at the follow- 

 ing conclusions: (1) Double refraction due to 

 thrust is negative for most glasses, but positive 



vibrating element of a gas were capable of rotation for flint glasses containing much lead. It should 

 about an axis through its center, and if that axis thus be possible to make a flint glass having no 

 were itself rotating, there would be in the spectrum double refraction for light of a certain wave- 

 generally only a bright line and a luminous band, length. Dispersion due to mechanical thrust is 

 If there is to be a line spectrum only, it is found considerable only in the heaviest flint glasses, 

 that the secondary vibration must be small com- (2) Change of index, absolute or relative, due to 

 pared with the primary. From considerations of hydrostatic compression increases with the dei 

 the law of equal partition of energy Jeans con- sity and index, but does not agree with any for- 

 cludes that the atom, and not the molecule, is mulse which have been put forward, 

 the vibrator, except in monatomic gases. He in- pure temperature coefficient of the index (U 

 fers that in all gases giving line spectra the vi- result obtained by subtracting the portw 

 brating parts are dissociated atoms, approxi- coefficient due to change of density > 

 mately spherical, which would explain why pure and in / c ff a ^f. s _ ra ? ldl y_^ lt |j ul _ ; J^J 

 line spectra are given only by elements. The 

 atom is supposed by Jeans to consist of a great 

 number of point electric charges arranged in 

 spherical shells of alternate signs. The outer 



shell is assumed to be always negative, to ac- May 20) finds that the well-known fluorescence 



count for the greater importance of negative ions, of the diamond in violet light is ini 



as shown by Zeeman and others. Incidentally lated to its brilliancy in artificial light, especially 



the calculations lead to an estimate of the size candle-light. The most brilliant stones are those 



of an atom, namelv, the radius of an atom of which are most clearly fluorescent in vio 



atomic weight n should be, at least as regards they exhibit a clear blue very lumm 



order of magnitude n-|10 J centimeter. The mole- rescence, while less brilliant diamonds 



lead. (4) Dispersion may either increase or de- 

 crease as the pure temperature coefficient in- 

 creases. 

 Fluorescence. Chaumet (Comptes Rendus, 



sume a violet color. A brilliant yellow diamond 

 when placed in violet light for n few minutes 

 shone vivid red, and its color changed to dull 



be liberated by chemical change, and that the brown, but it recovered color and brilliancy in 

 force exerted by a molecule at external points twenty-four hours. The action of viole 

 falls off very rapidly with the distance. Each distinguishes the rubies of Siam, which exni 



cule. w T hich is a combination of atoms, is defined 

 by Jeans as a system of ions of which the total 

 charge is zero. It follows that no free ions can 



