Chemistry and Physics. 89 



to the surrounding systems ; this would tend to raise the tem- 

 perature of the system. He proposes to test the question whether 

 the temperature in the middle of a mass of a magnetic substance 

 like iron, whose surface is kept at a constant temperature, differs 

 from the temperature inside a mass of a non-magnetic substance 

 like brass, whose surface is kept at the same temperature. — Phil. 

 Mag., Dec, 1903, pp. 673-693. j. t. 



9. Laboratory Physics. A Student's Manual for Colleges and 

 Scientific Schools; by D. C. Miller. Pp. xv-f-403. New York, 

 1903 (Ginn & Co.). — This is a book to be highly commended. The 

 experiments given with the object of teaching manipulation are not 

 too trivial as it is too often the case, and those given to illustrate 

 or demonstrate principles are not too intricate or difficult for the 

 grade of students for which the book is written. The introductory 

 remarks on observations, errors, corrections, probable error, signifi- 

 cant figures, graphical methods, etc., are admirably done. The book 

 is kept within a reasonable compass by means of giving the theory 

 of the experiments for the most part in references. In order for 

 this method to work satisfactorily, an adequate quiz system must 

 accompany the laboratory work. This is in line with the author's 

 aim to make the laboratory an intimate and not a separate ; 



of the general instruction in physics. He advises that more than 

 half of the whole time spent on physics be given to lectures and 

 recitations. This idea of the unity of the subject is one which 

 the prevalence of the elective system in this country has tended 

 to obscure. The separation between experimental and theoretical 

 physics is intellectually vicious, though, unfortunately, often made 

 in our colleges. Hence it is doubly a pleasure to welcome a book 

 that aims to aid in the teaching in the laboratory of that part of 

 theoretical physics which can advantageously be so taught; and 

 not to teach "Experimental Physics" as if it had a separate 

 existence. 



The book closes with an adequate set of tables of physical con- 

 stants, etc., though why such a useless table of trigonometric func- 

 tions should be included is hard to see. l. p. w. 



10. Physical Laboratory Manual for Secondary Schools; by 

 S. E. Coleman. Pp. 234. New York, 1903 (American Book Co.). 

 — This appears to be an excellent example of this class of manual. 

 Seventy-seven exercises, of which about two thirds are quantita- 

 tive, are outlined. Of these some forty or fifty are suggested as 

 covering the usual ground of the laboratory courses of secondary 

 schools, thus leaving considerable margin of choice in experiments 

 to the teacher. The directions, as far as examined, seem to be 

 clear and concise. The attention paid to the computation of per- 

 centage errors in data and result is especially to be commended. 

 The main adverse criticism to be made on the book is one that 

 applies to the majority of its class, namely, that the teaching of 

 quantitative relations by means of equations is practically ignored. 

 Why is mathematical reasoning, one of the most fundamental 

 and important methods of reasoning known to physics, entirely 



