July 31, 1896.] 



SCIENCE. 



147 



these systems — in two hours ! And this at the 

 beginning of the course ! This same course also 

 provides laboratory work on Apus and Hydroph- 

 ilus in addition to that on two other Arthro- 

 pods, but can not get time for any work on the 

 fish or the Echinodermata or the Coelenterata. 

 We can crowd ourselves enough to pass the 

 Tunicata, but we can hardly comprehend how 

 a man capable of beginning so promising a 

 work as the ' Lehrbuch der Zoologie ' can so 

 lightly pass over these magnificent groups. 



As to the figures we are bound to say that 

 in point of distinctness most of them are in- 

 ferior to those of well known standard works. 

 The shading is often poorly done, though they 

 are all neat and clean. Originality, however, 

 is hardly enough to justify a new figure, partic- 

 ularly if it is to appear in a text-book for some 

 good. Unless a figure is a decided improve- 

 ment in some respect or other we prefer to have 

 it remain the exclusive property of the author 

 and his students. Some of the figures are of 

 necessity so much like well known old ones 

 that they have nothing to specially recommend 

 them. 



It is undoubtedly a mistake to introduce into 

 an elementary course individual views of mat- 

 ters in dispute. The authors have not sinned 

 grossly in this respect ; still the ' Cladus ' and 

 ' Anhang ' ideas might have been kept out with- 

 out impairing the value of the book. We even 

 get the accounts of Apus and Hydrophilus as 

 Anhange. 



The hope of the authors that their ' Biichlein ' 

 will prove useful to others may be realized to 

 some extent in Germany ; but for the English 

 teacher and student, excepting the descriptions 

 and illustrations of animals not taken up so 

 fully in other books, there is nothing in it to 

 recommend it above any of the good books now 

 available. Henry F. Nachtrieb. 



SCIENTIFIC JOURNALS. 

 THE PHYSICAL REVIEW, JULY-AUGUST. 



On the Measurement of the Expansion of Metals 

 by the Interferential Method: By E. W. MoR- 

 LEY and Wm. A. Rogers. The first part of 

 this article by Prof. Morley is devoted to a de- 

 scription of the method used, which consists 



essentially in measuring the change in length, 

 during the heating or cooling of a test bar, by 

 observing the displacement of interference 

 bands. These bands result from the interfer- 

 ence of beams of monochromatic light reflected 

 from mirrors which are placed at the ends of 

 the test bar and a constant temperature bar re- 

 spectively. A later article will contain the 

 results of an actual trial of the method. As in 

 all cases where interference phenomena are 

 used, the method demands great care in all de- 

 tails and is capable of extreme accuracy. Prof. 

 Morley discusses the sources of error and the 

 means of avoiding them at some length and in 

 a very practical manner. A difficulty which 

 one would not be apt to think of arose when 

 the sodium flame was used as a source of light ; 

 for the number of bands to be counted was so 

 great (five or six thousand) that the slight dif- 

 ference in wave-length between the two sodium 

 lines caused a great decrease in the visibility of 

 the bands in certain regions. Thus when about 

 3,400 bands had been counted, the fringes were so 

 obscure that the next hundred could hardly be 

 observed at all. 



The Viscosity of Polarized Dielectrics : By A, 

 W. Duff. Professor Dufi" finds that the vis- 

 cosity of certain non-conducting liquids, such as 

 glycerine and castor oil, is increased when the 

 liquid is subjected to considerable electrostatic 

 stress. The increase observed was small, being 

 expressed in fractions of one per cent, , but the 

 fact of its existence seems to be definitely 

 proved. The change was measured by observ- 

 ing the rate of descent of small spheres which 

 were allowed to fall through the liquid between 

 two oppositely charged plates. If, as seems 

 probable. Prof. Dufi"'s results are confirmed, 

 we have here a new class of electrical pheno- 

 mena, doubtless intimately connected with the 

 electrostatic Kerr effect. 



Note on the Theorem of Clausius : By Edgar 

 Buckingham. Prof. Buckingham's object is to 

 give a demonstration of what we usually refer to 

 as the 'Second Law of Thermodynamics,' for 

 cases not considered in Clausius' original proof 

 and usually treated very briefly by his followers. 

 The cases considered are (1) systems whose state 

 is defined by only one coordinate besides the 



