966 



SCIENCE 



[N. S. Vol. XXXIII. No. 860 



lowing this with a statement of the principle 

 of association, Mast continues : 



Every step in the development of the theory is 

 supported by numerous experimental facts and all 

 seems to fit with what is known concerning the 

 reactions of organisms. Reactions, according to 

 tlie theory, are as stated above, primarily due to 

 physiological states. External agents ordinarily 

 produce reactions through the effect they have on 

 these states. By the application of this idea all 

 the different phenomena connected with reactions 

 to light as summarized at the beginning of this 

 chapter can be accounted for. 



All this sounds rather naive as compared 

 with the critical exposition of the preceding 

 part of the vohime. It would indeed be com- 

 forting to be able to repose with such a spirit of 

 confidence and contentment in a general phi- 

 losophy of behavior, but it is perhaps pertinent 

 to enquire if the author has not been deceived 

 with the delusive appearance of explanation 

 where no real explanation has been given. It 

 is an obvious truism that external factors 

 cause reactions by altering internal processes; 

 else they would not be reactions at all. It is 

 equally obvious that if changes of behavior 

 occur where external conditions do not vary 

 they must be due to the fact that internal con- 

 ditions do vary; or, in other words, if the 

 cause of a change is not outside of the organ- 

 ism it must perforce be inside of the organism. 

 Phenomena may thus be " accounted for " on 

 the basis of varying internal states, but as it 

 is admitted that in most cases we are entirely 

 ignorant of what these states are we are about 

 as much enlightened as we are by the cele- 

 brated explanation of the sleep-producing ef- 

 fect of opium by attributing it to a dormitive 

 principle. 



There is a useful bibliography in which the 

 efPort is made to include all the important 

 works on reactions to light in both animals 

 and plants, but several noteworthy contribu- 

 tions are not included. Notwithstanding 

 minor defects, the work of Mast will prove of 

 great value to students of the effect of light on 

 the behavior of organisms, and the author is 

 to be congratulated on having made so sub- 

 stantial a contribution to the subject. 



S. J. Holmes 



A Laboratory Manual of Physical Geography. 



By E. S. Tarr and O. D. von Engeln. 



New York, The Macmillan Company. 1910. 



Tarr and von Engeln's " Laboratory Manual 

 of Physical Geography " is the most practical 

 and best organized manual that has yet ap- 

 peared. Prepared as an exercise and note- 

 book, with detachable leaves, and containing 

 within its covers a large part of the necessary 

 equipment for work, except for topographic 

 maps, minerals and certain simple pieces of 

 physical apparatus, it is a hand-book and 

 guide available for both the expert and inex- 

 pert teacher. Although primarily devoted to 

 the study of physical geography, much em- 

 phasis is madg of life relations and therefore 

 the book not only meets a present condition, 

 but makes possible a development of the 

 phase of geography which is beginning to be 

 emphasized by the better secondary school 

 teachers. 



Of the seventy-three exercises in the book, 

 nine are devoted to the earth as a globe, seven 

 to excursion^, eight to minerals and rocks, five 

 to map study, twenty-six to the physiography 

 of the lands, two to the ocean, fourteen to the 

 atmosphere and one each to life zones and 

 magnetism. 



Thus the special emphasis is good, though 

 it is questionable whether mineral and rock 

 study deserves to remain in physical geog- 

 raphy and whether the ocean is not given too 

 little emphasis. The larger life relations to- 

 the ocean, apart from the phenomena associ- 

 ated with ocean currents and tides, are so im- 

 portant and fascinating that it seems unrea- 

 sonable to omit them, while space is, given to 

 minerals and to the physical phenomena of 

 condensation. Laboratory work should be de- 

 voted to topics that are not capable of being 

 learned more easily and more effectively 

 through demonstration, and certainly conden- 

 sation can not be included in this class. 



The order of treatment under the land 

 forms is original in that types of plains are 

 studied in relation to drainage and not follow- 

 ing drainage, and thus a better unity is pre- 

 served. Mountains are grouped according t» 



