140 



SCIENCE. 



[N. S. Vol. XII. No. 291. 



measurement of a velocity does not require the 

 determination of the change in the direction 

 of motion ; the discussion of Avogadro's law 

 (p. 171) contains a deduction of the Maxwell- 

 Boltzmann theorem which is certainly illogical, 

 the deduction being based on the constant rela- 

 tion between the temperature and the kinetic 

 energy of the molecules of all gases which was 

 established by that theorem ; electricity is not 

 energy (p. 673), although its manifestation re- 

 quires the expenditure of energy ; electromotive 

 force is not equivalent to difference of potential 

 (p. 674), the former term including cases which 

 cannot be described in terms of the latter ; the 

 formula for the velocity of electric waves is 

 given incorrectly on p. 858, and the mistake is 

 repeated on p. 861, where Maxwell's relation 

 between the index of refraction and the specific 

 inductive capacity is deduced from it by a series 

 of algebraic errors. 



One other matter needs to be noticed more 

 particularly. In the section on the Liquefac- 

 tion of Gases (p. 286), after giving an account 

 of the method of Wroblewski, so efficiently 

 employed by Olszewski, the author describes 

 Dewar's method, attributing its operation to 

 the principle that when a gas expands against 

 pressure it does work and hence becomes cooled. 

 This principle was the one employed by Cail- 

 letet and by Pictet in their successful attempts 

 to liquefy gases. In their experiments the 

 liquid product was obtained in the tube in 

 which the gas was compressed, the gas emitted 

 when the stopcock was opened acting as a pis- 

 ton pushed out by the pressure of the gas left 

 in the tube, and the cooling effect was, at least 

 partially, due to the work done by this remain- 

 ing gas and was experienced by it. When we 

 examine the description of the Dewar method 

 it appears that the expansion is so gradual that 

 it cannot be considered even approximately 

 adiabatic and that the gas which is cooled is that 

 which has passed out of the chamber in which 

 it is compressed. A comparison of this de- 

 scription with that of the Linde method (p. 320), 

 shows that the methods are alike in every es- 

 sential particular, including the important fea- 

 ture of 'the regenerative process,' and that 

 the principle which applies to both of them is 

 that which is so well explained by the author 



on page 318. Surely it cannot be contended 

 that different principles apply in the two cases 

 because in the Dewar method the gas to be 

 cooled is contained in a vessel in which the 

 pressure gradually falls, while in the Linde 

 method the supply of gas is renewed by a 

 pump so that the pressure is kept approxi- 

 mately constant. In view of the claims made 

 by Linde (Wied. Ann. 57, p. 332), which have 

 never been successfully controverted, such an 

 account of the Dewar method should never 

 have been given, or if given it should have 

 been accompanied with some adequate justifi- 

 cation for it. It is incumbent on the writer of 

 a text-book to be unusually careful in making 

 statements on disputed points, and particularly 

 on questions of priority, since his opinions are 

 naturally adopted by his readers as those of an 

 impartial umpire. 



The book is well printed, its diagrams and 

 illustrations are excellent, and it contains much 

 new matter, and old matter put in a new way. 

 It deserves to take a high place among the text- 

 books of physics. 



W. F. Magie. 



Princeton University. 



ornithology. 

 In ' The Birds of Rhode Island ' by Howe and 

 Sturtevant, we have a very acceptable addition 

 to the excellent lists already published of the 

 birds of several of the States. Lists of this 

 character are useful in bringing together the 

 scattered notes pertaining to a given region, 

 thereby saving the reader the time and trouble 

 of hunting through many volumes. The au- 

 thors have arranged their book in two parts : 

 The first reviews the former publications on the 

 birds of Rhode Island as well as the State col- 

 lections, gives some details on migration, and a 

 full account of the historic ' Cormorant Rock' ; 

 the second part includes an annotated list of 

 three hundred and three species, and a bib- 

 liography of one hundred and eighty-five titles. 

 Of the three hundred and three birds accredited 

 to the State, two hundred and ninety are based 

 on positive records, three have been exter- 

 minated through the agency of man, and ten 

 are placed in a hypothetical list as the evidence 

 of their occurrence is not absolutely conclusive. 



