620 



PHYSICS, PROGRESS OF, IN 1893. 



film of the metal being incased in nitrogen to 

 avoid oxidation. He found that as the tem- 

 perature increased the transparency also in- 

 creased, so that since electrical resistance also 

 increases with temperature, transparency and re- 

 sistance increase together, as is required by the 

 electromagnetic theory of light, which thus re- 

 ceives additional confirmation. 



Color. Capt. W. de W. Abney (London Royal 

 Society, June 1), in a comparison of natural 

 colors with pigments, finds that the blue of the 

 sky is about wave length 4,800, and that of 

 artists' cobalt about 4,812, so that painters in 

 this case have chosen the proper pigment with 

 remarkable exactitude. 



Effect of Light on Pigments. W. S. Simpson, 

 acting on the report of a commission which in 

 1888 reported that water colors fade only when 

 in contact with air, has devised an air-tight 

 frame in which such pictures may be kept fresh 

 indefinitely. An attached manometer enables a 

 leak to be detected at once. 



Actinometry. H. Rigollot (" Journal de Phy- 

 sique," November, 1892) has constructed a chem- 

 ical actinometer of two plates of oxidized copper 

 in water containing one part in a thousand of a 

 haloid salt of sodium. One plate b exposed to 

 light, and the resulting photoelectric currrent is 

 measured. On investigating the effects on this 

 instrument of different parts of the spectrum, 

 the inventor finds that it agrees with the eye in 

 being most sensitive to red, not to violet, as are 

 the salts used in photography. 



Polarization. Rubens and Du Bois (Berlin 

 Physical Society, June 2) find that heat rays of 

 long wave length are polarized by a grating of 

 metallic wire, just as eleetro-magentic radiation 

 is. Such a grating will not polarize light, but 

 the result shows that this is only because of the 

 short wave length of the luminous radiation. 

 The results show also that in the ultra-red 

 waves, as in electro-magnetic waves, the vi- 

 brations are at right angles to the plane of po- 

 larization. The measurements were made with 

 the bolometer. Emil Hirsch (Wiedemann's 

 ' Annalen," No. 3), in experiments on the circular 

 ferro magnetic polarization of transparent films 

 of iron, nickel, and cobalt, finds that Kundt's 

 constant (the ratio of the rotation of polariza- 

 tion to the increase of magnetization potential 

 from one side to the other) is independent of the 

 temperature. 



Diffraction. Gouy (Paris Academy of Sci- 

 ence, Nov. 6) has succeeded in obtaining the 

 image of an opaque object by diffracted light 

 alone by intercepting the geometrical rays by a 

 screen near the focus of the object glass. The 

 diffracted rays form an outline of the object 

 traced in a thin bright double line separated by 

 a black interval due to the interference of the 

 rays within the shadow with those without it. 



Photometry. Charles Henry (Paris Academy 

 of Science. Jan. 16), in further tests of his phos- 

 phorescent zinc-sulphide photometer (see "An- 

 nual Cyclopedia," 1892. p. 640), finds that the 

 sulphide emits at any given moment light whose 

 intensity is independent of the intensity and 

 duration of the primary illumination, and of the 

 thickness of the layer, and not varying with the 

 sample examined, thus fulfilling all requisites of 

 a good photometric standard. Prof. A. M. May- 



er, of Hoboken (" Philosophical Magazine," Au- 

 gust), uses a rotating disk with translucent sec- 

 tors for the photometric comparison of two 

 lights of different tints on opposite sides of the 

 disk. The result is to produce a nearly white 

 light, which is made to appear quite white by 

 the use of contrast colors. Lights of different 

 tints, as lamplight and daylight, can thus be 

 compared with good results. To the same end 

 Prof. 0. N. Rood (" American Journal of Sci- 

 ence," September) has devised a photometric 

 method, independent of color, that depends on 

 the shock experienced by the retina when one 

 surface is withdrawn and replaced by another. 

 The two surfaces whose luminosities are to be 

 compared are arranged as halves of a disk. 

 When this is rotated a flickering sensation is 

 perceived unless the luminosities are the same, 

 and this sensation is entirely independent of 

 color. A. P. Trotter (" Philosophical Magazine," 

 July) has devised an improved photometer, using 

 the principle of two screens set at such an angle, 

 or in such a way, that each receives light from 

 one and only one of the sources to be compared, 

 while the observer can see both at once. He 

 employs a plain screen and one in which there 

 are numerous slots, through which the first can 

 be seen. Slots are preferred to holes on account 

 of their symmetry. The best material lias been 

 found to be zinc. S. P. Thompson (ibid.) dis- 

 cusses what he calls the vibration principle in 

 photometry namely, that a more exact measure- 

 ment can be made when the intensity of one of 

 the compared lights is made alternately greater 

 and less than that of the other. To this end (1) 

 one of the screens may be vibrated, (2) one of 

 the lights may be vibrated, or (3) one of the 

 lights may be caused to vary in intensity. The 

 author puts the principle in practice by mount- 

 ing the paraffin block of a Jolly photometer on 

 a spring. 



Spectroscope/. J. R. Rydberg ("Philosophical 

 Magazine," March) finds that the focal curve 

 that passes through the center of curvature of a 

 Rowland's grating is a circle, but has not the 

 radius of curvature at the apex of the grating 

 as a diameter. It is possible to make the proper 

 adjustments, but the girder on which the tele- 

 scope moves must be a diameter of the true focal 

 circle. The fact discovered may result from the 

 noncoincidence of the point of symmetry of the 

 grating with the apex of the mirror, or from a 

 systematic variation in the distances between 

 the lines, due to an irregularity in the screw 

 of the dividing engine. Prof. Rowland (ibid., 

 May) has himself investigated irregularities or 

 " ghosts " in his gratings. The effect of small 

 errors of ruling, he finds, is to produce a diffused 

 light around the spectral lines. This is sub- 

 tracted from the light of the primary line, and 

 its comparative amount varies as the square of 

 the relative error of the ruling and the square 

 of the order of the spectrum. Thus, a periodic 

 error diminishes the intensity of a line, and sur- 

 rounds it with a symmetrical system of lines 

 called " ghosts," which are found most trouble- 

 some in spectrum photography. A periodic dis- 

 placement of one millionth of an inch in the 

 ruling produces visible ghosts, and sometimes 

 the primary line is replaced by them altogether, 

 but from a formula a table of the ghosts of 



