February 1, 1895.] 



SCIENCE. 



131 



cietj- and in the Athenseum Club, where it 

 delighted Sir AVm. Thomson, now Lord 

 Kelvin, and led to the extraordinaiy lecture 

 On Recent DUcureries in Mechanical Conver- 

 sion of Motion, delivered by Sylvester before 

 the Royal Institution on January 23, 1874. 

 This in turn led to Kempe's reniaikablo de- 

 velopment of the subject, and to Hart's dis- 

 covery of a five-bar linkage which does the 

 same work as PeaucelUer's of seven. 



Henceforth Poaucellier's Cell and Hart's 

 Contraparallelograni will take their jilace in 

 our text-books of geometrj-, and straight 

 lines can be drawn without begging the 

 question by assuming first a straight edge 

 or ruler as does Euclid. 



Thus Kempe's charming book, ' Hoic to 

 Draw a Straight Line,' is a direct outcome 

 of Tchebychev's sketch for Sylvester. As 

 might perhaps have been expected, the im- 

 mortal Lobachcvsky found in his compatriot 

 a devoted admirer. Not only was Tcheby- 

 chev an active member of the committee of 

 the Lobachevsky fund, but he took the 

 deepest interest in all connected with the 

 spread of the profound ideas typified in the 

 non-Euclidean geometry. Knowing this, 

 Vasiliev in his last letter asked that a copj' 

 of my translation of his address on Loba- 

 chevsky be forwarded to the great man. 

 His active participation in scientific assem- 

 blies is also worthy of note ; for example, at 

 the ' Congivs de V association fran^aise pour 

 Tavancement des sciences, a Lyon,' he read 

 two interesting papers, Sur le-i valeurs limites 

 des inti'yrales. and Sur les quadratures, after- 

 wards published in lAoitville''^ Journal. 



George Bruce Halsted. 



UXIVERSITV OF TkXAS. 



SCIEXTIFIC LITERATURE. 

 Les Oscillations Electriques. H. Poixcare. 



(COSCLUDED. ) 



Propagation of Electrical 0'<eillation.'< Through 

 Air. — The velocity of propagation of electi-o- 

 magnetic induction through dielectrics of- 



fered the fii-st experimental test of supe- 

 riority of the Faraday-Maxwell theory over 

 the older theories. According to these that 

 velocity' should be infinite ; according to the 

 Faraday-^Iaxwcll view of elcc^tromagnetic 

 phenomena it should be the same as that of 

 light. Poincare reviews carefully all the 

 experimental evidences bearing upon this 

 point. Hertz's experiments in Carlsruhe 

 are first discussed and his early failures in 

 arriving at a satisfactory result are pointed 

 out. Two methods emploj'cd in these 

 measurements liy Hertz at Carlsruhe and at 

 Bonn are described briefly. One of these 

 consisted in measuring by means of a reso- 

 nator the difference of phase between two 

 waves sent forth by the same oscillator, one 

 wave along a conducting wire and the other 

 through the dielectric in the vicinity of the 

 wire. The other method consisted in meas- 

 uring what Hertz considered the wave 

 length of stationary electric waves in air 

 formed by the interference between the di- 

 rect waves sent forth bj' an oscillator and 

 the waves reflected by a large flat mirror 

 consisting of a metal sheet 2 meters wide 

 and 4 meters high. In all these experiments 

 the velocity of propagation along the wire 

 seemed to come out considerably dift'crent 

 from and generally less than that in air. 

 But the methods were open to several critic- 

 isms. In the first place, the hall in which 

 these experiments were caiTied out was too 

 small for the wave lengths employed ; sec- 

 ondly, the influence of the waves reflec'ted 

 from the walls was entirely neglected ; 

 thirdly, the dimensions of the reflecting 

 mirror were not large enough in comparison 

 to the wave length to prevent errors of ob- 

 servation due to the misleading influence of 

 diffraction phenomena. All these objections 

 were in a measure overcome in the earliest 

 experiments of Sarasin and de la Rive (C. 

 R. t. ex. p. 72). In these experiments the 

 methods of Hertz were employed, but they 

 were performed in a large hall, with a large 



