828 



SCIENCE 



[isr. S. Vol. XLI. No. 1066 



posal was first made in tbe correspondence col- 

 umns of a Syracuse paper, the Post-Standard, 

 and has since been taken up by local people 

 until there is considerable possibility of its 

 ultimate success. A committee of representa- 

 tive business men has been appointed to 

 further the project. 



The matter is here brought to the attention 

 of readers of Science in order to ask that any 

 who can help may be moved to use their influ- 

 ence with state officials or any others who 

 might render help. East Green Lake and the 

 surrounding region represent a larger and 

 wilder tract of land. The lake itself is of 

 equal geological interest and from the stand- 

 point of the hart's tongue fern, is of greater 

 interest than the west lake region because the 

 best specimens in the country grow near the 

 east lake. The west lake preserve includes 

 only seventy acres, as this was all that was of 

 special value as park. In the east lake re- 

 gion it is proposed to acquire two square miles 

 to include not only the lake plunge-basin itself 

 but also a marl-bottomed lake nearby and ad- 

 ditional acres of beautiful woodland. 



The preservation of the best stations for 

 the hart's tongue fern, probably the rarest and 

 most interesting fern in North America, is 

 sufficient reason for urging the acquirement 

 of this proposed new park. Besides this species 

 there are other ferns to the number of about 

 forty, making it probably the best fern pre- 

 serve in the country. The need of prompt 

 action is indicated by the fact that a lime 

 development company now holds an option on 

 the tract desired. 



E. C. Benedict 



SCIENTIFIC BOOKS 

 Theory of Measurements. By James S. Stev- 

 ens, Professor of Physics in the University 

 of Maine. New Tork, D. Van Nostrand 

 Company. 1915. Pp. vii .-f 81. Price $1.25 

 net. 



This little book is intended to fill the gap 

 between the fragmentary treatment of the 

 errors and adjustment of measurements, found 

 in most laboratory manuals, and the detailed 

 discussions given in formal treatises on the 



theory and precision of measurements. In 

 eight short chapters, the author deals with 

 the following topics : Measurements and 

 Errors; Probability, including a discussion of 

 the probability curve and integral; Adjust- 

 ment of Observations; Precision of Measure- 

 ments ; Propagation of Errors, applied to direct 

 and inverse problems; Plotting; Negligibil- 

 ity, including rules for the use of significant 

 figures; Empirical Formulae and Constants. 



The methods and notation adopted are sim- 

 ilar to those employed in Merriman's " Method 

 of Least Squares " and Holman's " Precision 

 of Measurements." The treatment is neces- 

 sarily abridged to comply with the limits set 

 by the scope of the book but the usual formu- 

 lae and methods are developed in sufficient 

 fullness for their practical application by the 

 intelligent student. Possibly owing to his de- 

 sire to save space, the author gives very little 

 discussion or explanation of the fundamental 

 principles and assumptions underlying his 

 mathematical derivations. Consequently the 

 true significance of his results is not always 

 apparent and the conditions essential for 

 their correct application are apt to be over- 

 looked. Eor example, the deductions from 

 the law of accidental errors do not apply to a 

 series of observations affected by systematic 

 errors but the author has nowhere pointed out 

 the necessity of considering such errors in 

 connection with the discussion of precision. 



Definitions and problems are frequently so 

 briefiy and inadequately stated that the un- 

 aided student is apt to misinterpret their 

 meaning. The following quotation is a fair 

 example : " Measurements are usually classi- 

 fied as follows: 1. Direct — when, for ex- 

 ample, a distance is measured with a tape line. 

 2. Indirect — when the density of a cylinder is 

 determined by measurements of its length, 

 diameter and mass. 3. Conditioned — when 

 the third angle of a triangle is restricted by 

 the values of the other two angles. Measure- 

 ments not so conditioned are called indepen- 

 dent." 



However, with the aid of a competent 

 teacher, the student should be able to make 

 profitable use of the book in connection with 



