Apeil 2, 1920] 



SCIENCE 



349 



eral principles of physics and give the student 

 a first-hand contact with the notions discussed 

 in text-books, lectures and recitations. The 

 emphasis is on the underlying ideas and the 

 discussion of methods and accuracy of meas- 

 urement is purely incidental. Books of this 

 type are eminently suitable for students in 

 elementary physics. On the other hand, the 

 purpose of the manual may be to develop the 

 theory and practise of physical measurements 

 and to describe the construction and operation 

 of standard measuring instruments. Such 

 manuals are essential to the advanced student 

 in physics and, if sufficiently comprehensive, 

 they are useful to the student in chemistry or 

 biology. 



Professor Ferry's work belongs to the second 

 category although a few of the experiments de- 

 scribed would not be out of place in a manual 

 of the first type. It is a thorough revision and 

 rearrangement of an earlier book on " Prac- 

 tical Physics," by Ervin S. Ferry and Arthur 

 T. Jones, to which chapters on sound, optics, 

 electricity and magnetism have been added. 

 The scope and method of the work are ade- 

 quately indicated by the following quotations 

 from the 'author's preface : " Only those ex- 

 perimental methods have been included that 

 are strictly scientific and that can be depended 

 upon to give good results in the hands of the 

 average student. Although several pieces of 

 apparatus, experimental methods and deriva- 

 tions of formulffi that possess some novelty 

 appear, our fixed purpose has been to use the 

 standard forms except in cases w'here an ex- 

 tended trial in large classes has demonstrated 

 the superiority of the proposed innovation." 

 " It has been assumed that the experiment is 

 rare that should be performed before the stu- 

 dent understands the theory involved and the 

 derivation of the formula required. Conse- 

 quently the theory of each experiment is given 

 in detail and the required formula developed 

 at length. The more important sources of 

 error are pointed out, and means are indicated 

 by which these errors may be minimized or ac- 

 counted for." 



Several of the methods of measurement de- 

 scribed involve the use of instruments of spe- 



cial design not likely to be found outside of 

 the author's laboratory but the greater part of 

 them can be carried out with the apparatus 

 that should be found in any well-equipped lab- 

 oratory. The theory and manipulation of the 

 more important modern instruments of pre- 

 cision are comprehensively treated and any 

 student who has occasion to use such instru- 

 ments will find these sections of the work very 

 useful. The work is well adapted for use 

 as a text in second- or third-year laboratory 

 courses in physics. It should also find wide 

 use as a reference book, in any laboratory 

 where physical instruments and methods are 

 occasionally used. 



A. deF. p. 



SPECIAL ARTICLES 



NOTICE OF A RECENT CONTRIBUTION TO 

 STATISTICAL METHODS 



Progress in science is measured, among 

 other things, by the extent to which the qual- 

 itative treatment of problems is supplemented 

 by a more rigorous quantitative treatment. 

 The introduction of quantitative methods into 

 the biological sciences, however, is beset with 

 unusual difficulties. The highly complex and 

 variable nature of the subject matter generally 

 demands the empirical procedure of the statis- 

 tician rather than the deductive one of the 

 mathematician, and this is true of many prob- 

 lems of physical science as well, for example, 

 those of meteorology. One of the main diffi- 

 culties to be overcome arises from the simul- 

 taneous variation in the magnitudes of the 

 many variables concerned. Especially is this 

 true in " field " investigations where artificial 

 control over the variable is impossible; as, 

 for example, in marine ecology. In order to 

 meet this difficulty the authors have prepared 

 a paper entitled : " The functional relation of 

 one variable to each of a nmnber of correlated 

 variables determined by a method of suc- 

 cessive approximation to group averages." 

 The introduction is written by Wm. E. Hitter 

 under the title : " A step forward in the 

 methodology of natural science." 



1 Proc. Amer. Acad. Arts. Soi., Vol. 55, Dee., 

 1919, pp. 89-133. 



