Chemistry and Physics. 393 



graphed by Cochin. For this purpose he used a spectrograph 

 with quartz prisms and quartz and Iceland spar lenses, the 

 gelatino-bromide film being curved so that all the lines were in 

 focus. The water bands at wave-length 309 were obtained in all 

 cases, but there were no lines or bands of higher refrangibility 

 than this observed in any case, even when the exposure was quite 

 long. In the case of the hydrogen flame the results are practically 

 the same, though the spectra are somewhat more luminous. In 

 these spectra, the line A413 of lithium is shown, and the double 

 line A330-3 and A330-2 of sodium, obtained in the electric arc 

 spectrum by Kayser and Runge. Rubidium, in addition to the 

 visible line A420, gives the two doublets at A359*l-358-7 and 

 A335-1-334-8, observed by Kayser and Runge ; especially if the 

 hydrogen flame is employed. Caesium gives the lines A388, 

 A361-5 and A347*75, in addition to the visible blue lines A459'7- 

 450 ; also corresponding apparently to lines seen by Kayser and 

 Runge. ~No ultra-violet lines were observed in the spectra of 

 calcium, strontium or barium, though the visible lines were pho- 

 tographed. Thallium gave lines at A353 and A378 in the ultra- 

 violet; thus resembling the alkali-metals. — C. i?., cxvi, 1055; 

 J. Chem. Soc, lxiv, II, 402, September, 1893. g. f. b. 



3. On the Spectrum given by Nickel-carbonyl. — According to 

 Liveixg and Dewae, nickel-carbonyl, when mixed with hydrogen 

 and burned in oxygen at a porcelain jet, exhibits a spectrum which 

 consists of two parts: (1) the spectrum of the main body of the 

 green flame ; and (2) that of the base of the flame. The spectrum 

 of the main body consists of a series of shaded bands, brightest 

 in the green but extending on the red side beyond the red lithium 

 line and on the violet side well into the blue. These bands have 

 their sharp edges on the more refrangible side. Besides the 

 bands a few visible lines appear, only one of which A5476 is a 

 known line of nickel. In the ultra-violet part, however, a great 

 number of nickel lines were photographed. In the spectrum of 

 the base of the flame there are two principal groups of lines, one 

 in the orange and red, the other in the citron and yellow. None 

 of them appear to be nickel lines. — Proc. Roy. Soc, lii, 117; J. 

 Chem. Soc, lxiv, II, 404, Sept. 1893. G. f. b. 



4. On the Optical Properties of Liquid Oxygen. — By means of 

 a specially constructed apparatus Olszewski and Witkowski 

 have succeeded in determining the angle of total reflection in liquid 

 oxygen ; and from this they have deduced the index of refraction 

 of this substance. As thus found, the relative refractive index 

 of liquid oxygen appears to be 1*2232 and the absolute index 

 1*2235 ; this latter value agreeing closely with 1*2236, that found 

 by Liveing and Dewar. The absorptive power of a column of 

 liquid oxygen one millimeter in thickness, between A577 and A570, 

 i. e., in the most intense portion of the yellowish-green absorption 

 band, was found to be between 84 and 89 per cent ; while for 

 the red absorption band between A630 and A638, the mean value 

 of the absorption obtained was 88 per cent. — Chem. Centr., i, 595 ; 

 J. Chem. Soc, lxiv, II, 353, August, 1893. g. f. b. 



