Chemistry and Physics. 467 



10. Absorption of Gas in a Crookes Tube. — The anomalous 

 behavior of X-ray tubes has been the subject of many investiga- 

 tions, and it is generally supposed that gases are occluded by the 

 metallic terminals and by the glass tubes. In order to properly 

 exhaust a tube it is necessary to subject it for a long time to elec- 

 tric discharges and to external heat. Mr. R. S. Willow has 

 investigated this absorption of gases and concludes from experi- 

 ment with electrodeless tubes that the absorption is largely due 

 to the character of the glass employed. Jena glass absorbs 

 hydrogen to only a small amount : soda glass to a large amount. 

 The absorption is due to a chemical combination with the glass. 

 In general hydrogen is absorbed less than air or nitrogen. — Phil. 

 Mag., April 1901, pp. 503-517. j. t. 



1,1. Mechanical Movements of Wires produced by Electrical Dis- 

 charges which also make these Movements luminous. — When oscil- 

 lating discharges are sent over wires these become luminous under 

 certain conditions and apparently show stationary electric wavey. 

 Yarious investigators have described this phenomenon and have 

 measured wave-lengths, which they have considered true wave- 

 lengths due to the capacity and self-induction of the circuit. O. 

 Yiol, however, considers that these luminous effects are merely 

 mechanical vibrations due to the electric spark of a much lower 

 period than those which would arise from true electric oscilla- 

 tions, and that the nodes and ventral segments are rendered visi- 

 ble by a species of negative brush discharge along the vibrating 

 wires. — Ami. der PhysiJc., No. 4, 1901, pp. 7 34-761. j. t. 



1 2. The Ban d Spectrum of Oxides of A luminum and Nitrogen. 

 — Various authors have attributed the band spectrum observed 

 in the spectrum of aluminum between wave-lengths 5162*05 and 

 4470-63 to aluminum. G. Berndt finds, however, that this band 

 spectrum disappears when the spectrum of aluminum is obtained 

 in hydrogen gas, and becomes very strong when oxygen is sub- 

 stituted for hydrogen. He therefore attributes this band spec- 

 trum to oxides of aluminum. He also gives a very complete 

 table of the wave-lengths of the constituents of the bands of 

 nitrogen. — Ann. der Physih, No. 4, 1901, pp. 788-793. j. t. 



13. J^lectricite et Optique. La Lwnibre et les Theories Elec- 

 trodynamiques, par H. Poincare. Deuxieme- edition, revue et 

 completee par J. Blondin et E. Neculcea.. Pages x, 641. 

 Paris 1901 (Georges Carre et C. Naud).— The first half of this 

 remarkable book is nearly a repetition of the lectures given at 

 the Sorbonne in 1888 and 1890, except for the suppression of the 

 section relating to the experiments of Hertz, omitted because this 

 subject is treated in detail in the author's work on Electric Oscil- 

 lations, and for the addition of a little more than one hundred 

 pages which extends the general discussion of electromagnetic 

 theories of bodies at rest by including those of Ampere and of 

 Weber. The latter half of the book, also divided into two parts, 

 contains the lectures of 1899. These are devoted to a compari- 

 son of the different theories relative to the electrodynamics of 



Am. Jour. Sci.— Fourth Series, Vol. XI, No. 66.— June, 1901. 

 32 



