PHYSICS. 469 



J. 8cL, September, 1884, III, xxviti, 1G3; Fhil Mac., October, 1884, V, 

 XVIII, 289.) 



Abney and Festing, by photographiug the iufia red solar spectrum 

 under different atmospheric conditions, have shown that when the air 

 is nearly saturated with moisture this part of the spectrum is covered 

 by a dark band extending to wave-length 8330. In a v^ry dry time 

 the band extends only from wave-length 9800 to wave-length 9420. At 

 a great altitude, with a cold northeast wind, this band disapjiears 

 almost entirely, but can be reproduced by placing a suitable thickness 

 of water before the slit. These i^hotographs serve as indicators of at- 

 mospheric moisture. {Proc. 'Boy. Soc, XXXV, 80 ; J. Phys., May, 1884, 

 II, III, 219.) 



Nichols has studied the character of thclight reflected from j^igments 

 by means of a spectrophotometer, in which one-half of the slit is cov- 

 ered by a right-angled prism, the other half by a Nicol prism, while the 

 eye end of the observing telescope is furnished with a second Nicol and an 

 adjustable diaphragm . By these means two spectra, one above the other, 

 are produced, the lower one polarized in a vertical i^lane. In order to 

 compare the spectrum of the light reflected by any object with the 

 spectrum of daylight, the object is placed beneath the reflecting i^rism, 

 and illuminated either by direct sunlight or diffused daylight. Sky- 

 light reflected through the Mcol gives the polarized spectrum, and by 

 rotating the Nicol in the eye-piece it may be given any intensity re- 

 quired. Whatever be the character of the light reflected by the object 

 to be studied, it is always possible to find a position of this Nicol for 

 which any region of the spectrum under observation and the corre- 

 sponding wave-length of the jiolarized spectrum are equally bright. 

 Knowing the angle between the principal sections of the Nicols, the in- 

 tensity of this region, in terms of the intensity of the corresponding 

 wave-length in the spectrum of daylight, can be calculated. The au- 

 thor examined in this way red lead, chrome yellow, chrome green (chro- 

 mic oxide), and artificial ultramarine, and gives the results iu tabulated 

 form and in the form of curves. {Am. J. ScL, November, 1884, III, 

 XXVIII, 342.) 



Eosenstiehl has made a valuable contribution to the science of color 

 in a little book published by the Society Industrielle of Eouen, intended 

 to show the imperfection of methods founded on the study of coloring 

 matters in distinction froija those which recognize color simply as a prop- 

 erty of matter, and, in the physiological sense, as simply an affection 

 of the organ of sight. Attention is specially directed to the study of 

 color by means of color sensations ; and it is shown that it is to the 

 analysis and synthesis of the retinal impressions that we are to look for 

 exact views on the relationships of the colors. For the investigation 

 of colors the author uses concentric disks rapi illy rotated. These disks 

 are divided into sectors of different magnitudes, variously colored. 

 For the white disk barium sulphate is used, applied to a. suitable sur- 



