102 Scientific Intelligence. 



atom losing a single charge. In the other type of ionization we 

 suppose that the mercury atom is struck by a rapidly moving 

 atom and not by a corpuscle ; after the collision the mercury atom 

 starts off with a very considerable velocity, which at first is not 

 shared by the corpuscles inside it. The tendency of the cor- 

 puscles to leave the atom depends only upon the relative velocity 

 of the atom and the corpuscles inside it, so that the ionizing effect 

 produced by the collision is the same as if the atom were at rest, 

 and all the corpuscles were moving with the velocity acquired by 

 the atom in the collision. Thus if there were eight corpuscles in 

 the mercury atom connected with about the same firmness to the 

 atom, the result of the atom acquiring a high velocity in a col- 

 lision might be the detachment of the set of eight, leaving the 

 atom with a charge of 8 units of positive electricity. We see in 

 this way how the cathode particles might produce one type of 

 ionization resulting in singly charged atoms, while the atoms 

 forming the positive rays might produce another type of ioni- 

 zation resulting in multiply- charged atoms." — Phil. Mag., vol. 

 xxiv, October, 1912, p. 668. 



10. Handbuch der Spectroscopic ;* by H. Katser. Vol. VI. 

 Pp. vi, 1067, with 1 plate. Leipzig, 1912 (S. Hirzel).— This vol- 

 ume supplements the preceding one, and forms with the latter a 

 complete unit. All the reliable wave-lengths and associated data 

 pertaining to the elements not given in the fifth volume are col- 

 lected in the sixth. As before, the alphabetical sequence is based 

 upon the chemical symbols for the respective elements, so that 

 sodium (Na) comes first and zircon (Zr) last. The bibliographical 

 references have been made as complete as possible. For the first 

 elements in the list these references close about Easter, 1911, 

 whereas they approach July, 1912, for the last elements. The 

 practical value of this volume is greatly increased by the pres- 

 ence in it of a group of special tables beginning with page 887. 

 The first of these tables gives the wave-lengths (chiefly) of iron 

 lines as obtained by the interferometer method alongside of the 

 corresponding data as given by Rowland. The differences are 

 given in the last columns and are also plotted on a large folded 

 plate. With the aid of this graph it will be relatively easy to 

 reduce from the older system of wave-lengths to the new one, and 

 vice versa. However, since the data given by Rowland are 

 inadequate, the next table contains a very complete set of wave- 

 lengths of iron lines in the International System. o These wave- 

 lengths extend from 2212*7 A. U. to 8863*75 A. IT., and the 

 intensities, when known, are given both for the arc and for the 

 spark. The next table, likewise of inestimable value, deals with 

 the principal lines of all the elements, the interval having the limits 

 1854*04 A. IT. and 9*0850^. The last table in the volume gives 

 the wave-lengths of 2800 heads of emission band spectra together 

 with symbols showing whether each band is shaded off towards 

 the red or towards the violet. Furthermore, the presumable 



* See this Journal, vol. xxx, p. 349, 1910. 



