78 Scientific Intelligence. 



the points corresponding to the thin end of the wedge and grad- 

 ually l :nles off toward the thick region of the absorbing soreen, 



It is evident that the relative intensities of two lines can be cal- 

 culated from the lengths of their images, provided the lines are 

 so close together that errors due to the variation of the sensibil- 

 ity of the plate to different wave lengths are negligible. A 

 broadened line gives a wedge-shaped impression on the plate, the 

 apex corresponding to the maximum of intensity in the radia- 

 tion, and, from the shape of these wedges, the intensities at 

 different distances from the maximum can be deduced. It is thus 

 clear that the method consists in picking out points of equal 

 density at different wave-lengths and determining the thicknesses 

 of the wedge to which they correspond. Points of equal photo- 

 graphic density must indicate equal illumination since they are 

 exposed for the same time and subjected to the same chenVical 

 treatment. The neutral-tinted wedge, of course, does not vary 

 in absorbing power for neighboring wave-lengths. Accordingly 

 the method is unaffected by the eccentricities of the photographic 

 emulsion, and it is only necessary to assume that there is one par- 

 ticular density which can be recognized at different points. 



Enlargements of spectrograms of the first three series lines pro- 

 duced by the passage of condensed sparks through hydrogen at 

 atmospheric pressure are reproduced in the paper. Ha consists 

 of a strong maximum falling off rather rapidly and apparently 

 regularly. The intensity decreases much less rapidly for Up and 

 there is a distinct minimum at the center of the line, which 

 appears to be a close doublet. Hy has a bright axis with very wide, 

 nebulous wings. The distribution of intensity in each of these 

 lines is in complete accord with that deduced from Stark's data 

 on the electric resolution and polarization of the lines. — Proc. 

 Roy. Sac, vol. xcii (A), p. 322. h. s. u. 



8. The Single-Line Radiation of Magnesium. — The earlier 

 work of McLennan and Henderson showed that the vapors of 

 each of the elements cadmium, mercury, and zinc, can be caused 

 to emit a single spectral line by bombarding the particles with 

 electrons possessing a certain amount of kinetic energy. For 

 cadmium and zinc the range of kinetic energy corresponds to 

 potential differences lying between 4 volts and 13*6 volts. This 

 investigation has been recently extended by McLennan to mag- 

 nesium. The vapor of this metal was found to emit the line of 

 wave-length 2852*22 A, and no other line, when the kinetic 

 energy was included within definite limits. The numerical values 

 of the limits have not yet been determined, but the range 

 agrees, in part at least, with the interval for cadmium and zinc. 

 Absorption lines homologous to those already observed for cad- 

 mium, mercury, and zinc were photographed in the case of mag- 

 nesium. The wave-Jengths of the two absorption lines ara 

 recorded as 2852-22 A and 2073-36 A. When sufficiently magni- 

 fied the less refrangible band is seen to consist of two very close 

 narrow bands similar to the mercury doublet at A2536 - 72. On the 



