90 



SCIENCE 



[N. S. Vol. XL VI. No. 1178 



tary laboratory work in physics presented by 

 average students must have been impressed 

 frequently by the writer's lack of familiarity 

 with ordinary methods of computation and by 

 his inability to draw rational conclusions re- 

 grading the accuracy and significance of his 

 results. Unfortunately, the instruction in these 

 matters presented by many widely used labora- 

 tory manuals is very inadequate and frequently 

 misleading. We all admit that the primary 

 object of elementary laboratory work is to put 

 the student in personal touch with the facts 

 and principles of physical science. But every 

 experienced teacher knows that this object is 

 not attainable without more or less formal 

 instruction in the methods of reduction and 

 interpretation of observations. Moreover, the 

 student is seriously handicapped by the long- 

 hand arithmetical processes taught in second- 

 ary schools when greater precision and facility 

 can be attained by the shortened methods of 

 computation adopted by every competent phys- 

 icist. 



A number of books designed to fill this gap 

 by a detailed discussion of methods of com- 

 putation and the theory of errors have ap- 

 peared during the past few years. Dr. Tuttle's 

 " Theory of Measurements " belongs in this 

 group and it meets the needs of students in 

 elementary physics more adequately than any 

 other text that has come to the reviewer's 

 attention. For the most part, concrete ex- 

 amples are developed to illustrate general prin- 

 ciples and the discussions are so clear and well 

 stated that the student can hardly fail to grasp 

 their significance. The treatment presupposes 

 no training in mathematics beyond that 

 usually required for admission to college. In 

 fact capable high-school pupils should find 

 little difficulty in following the discussions. 



The most important topics treated in the 

 first one hundred pages of the book are as 

 follows: fundamental ideas, abridged methods 

 of multiplication and division, units and meas- 

 urements, angles and circular functions, ac- 

 curacy and the correct use of significant fig- 

 ures, logarithms, computations involving small 

 magnitudes, and the use of the slide rule. The 

 reviewer would be inclined to place more 



emphasis on the importance of systematic 

 orderliness in computation and exact specifi- 

 cation of units in writing numerical results. 

 But on the whole the treatment is very good 

 and guards against most of the common 

 errors of inexperienced computers. 



About seventy pages are devoted to a very 

 illuminating discussion of the methods of 

 graphical representation and reduction of ob- 

 servations, including a brief treatment of in- 

 terpolation and extrapolation. The possibil- 

 ity of emphasizing the significance of the 

 plotted data by a suitable choice of scales is 

 illustrated by numerical examples and the 

 advantages of so choosing the variables that 

 the graph will be linear are pointed out. . The 

 uses of logarithmic and semi-logarithmic 

 papers are also illustrated. 



The remaining portion of the book deals 

 with errors of observation and measurement, 

 statistical methods, the determination of the 

 best representative value from a series of dis- 

 cordant observations, the estimation of the 

 precision of direct and indirect measirrements, 

 and simple applications of the method of least 

 squares. The formulse of the theory of errors 

 are not derived mathematically but their sig- 

 nificance and use are very clearly explained 

 and illustrated by numerical examples. 



The book is neatly printed and substantially 

 bound. It should find a place in every phys- 

 ical laboratory devoted to the- instruction of 

 students. 



A. DEFOREST Palmer 



SPECIAL ARTICLES 



LITHOLOGIC EVIDENCE OF CLIMATIC 

 PULSATIONS 



The geologic evidences of changes of cli- 

 mate, as is well known, are numerous and 

 incontrovertible, particularly as regards ex- 

 tremes of temperature and their accompany- 

 ing variations of flora and fauna. The cli- 

 matic changes which have produced the most 

 widespread changes in life forms, as well as 

 physiographic features, have been the ones 

 most clearly recognized and easily studied. 

 These changes are known to have been pulsa- 

 tory or periodic, but with periods or cycles 



