256 PHYSICS. 



the temperature of the iii)per plate. By the help of a pulley aud cord, 

 the two plates thus prepared -were placed on a thick and flat plate 

 of ice and immediately covered with a double cylinder filled with ice. 

 The lower plate takes at once the temperature of the ice and remains so; 

 in one or two minutes the circuit of the galvanometer is closed and the 

 I)osition of the needle read every 20 seconds. The value of Ic, in absolute 

 C. G. S. unit-s, was found for carbon disulphide, 0.0250 ; for benzene, 

 0.0200 ; for water, 0.0745 ; copper sulj^hate, 0.0710 ; for zinc sulphate, 

 0.0691 to 0.0711 ; salt, 0.0692 ; alcohol, 0.0292 ; ether, 0.0243 ; chloroform, 

 0.0220 ; oil lemon, 0.0210 ; olive oil, 0.0235 ; glycerin, 0.0402. These results 

 in general agree with those obtained by Lnndquist. — {J. Phi/s., ix, 169, 

 May, 1880.) 



CrovaJias x^roposed a jiurely optical method for meavsuring high tem- 

 peratures, founded uj^on the principle that two incandescent bodies, 

 having the same radiating power, are at the same temperature when 

 the spectra which they give are identical in extent. The instrument 

 which he has devised for these measurements he calls a spectropyrom- 

 eter. It is composed of a direct-vision spectroscope, having a slit in 

 the eye-piece to limit the field, and having the observing telescope mov- 

 able about an axis perpendicular to the length of the spectrum, its play 

 being limited to wave-length 676, and on the other side to wave-length 523. 

 Over half the slit is a total-reflection prism, and in front of the slit is a 

 Nicol prism in the center of a graduated circle, a second Mcol being 

 jilaced either between this and the lamp, or preferably just behind the 

 slit of the collimator. To measure the temperature of a body in optical 

 degrees the telescope is first directed to the fixed point in the red, the two 

 Nicols are put at 0°, and the intensities of the two spectra made equal, 

 either by moving the moderator-lamp used as the standard farther off, or 

 by rotating a Nicol placed between the body and the reflecting prism. 

 The telescoiie is then moved to wave-length 523 in the green, and two 

 green images are obtained of unequal intensities. If the temperature of 

 the body is less than that of the lamp (below 2000° C.) the greens are 

 made equal by rotating the IS^icol. If a be the angle of rotation the in- 

 tensity is equal to 1000 cos ^a if the temperature of the lamp is 1000 optical 

 degrees. To calibrate the instrument, the temperature of the lamp flame ?jfi 

 was comijared with that of an air-thermometer with a porcelain bulb 

 heated in a furnace until the spectra of the two were identical throughout. 

 In this way Crova fiuds in optical degrees the following values for the i . 

 sources examined: Platinum strip heated to redness in a gas flame, 524; ;] 

 the same heated to a white heat in a gas blow-pipe flame, 810; moderator- ! 

 lamp fed with colza oil, 1000 ; stearin candle, 1162 ; illuminating gas (Ar-^ ; 

 gand burner), 1373; oxyhydrogen light (oxygen and coal gas on lime), ).' 

 1806 ; electric light (60 Bunsen cells), 3060 ; solar light, 4049. Thq author r||| 

 hopes to compare in this way the intensities of the simple radiations of 

 incandescent vapors, provided that the spectrum consists of more than < j 

 a single line. — [Ann. Chim. Phys., V, xix, 472, April, 1880.) ; 



