July 1, 1898.] 



SCIENCE. 



25 



steel cyliudei's referred to on p. 381 are quite 

 accurately tuned by Koenig by a special method. 

 In general, it may be said that one of the most 

 ■expensive ways of getting pieces of apparatus 

 is to have them constructed by ' any carpenter;' 

 e.g., the time required to explain the construc- 

 tion of the Wheatstone stereoscope to an ordin- 

 ary carpenter, the inevitable use of unseasoned 

 wood, and the high wages demanded by the 

 American workman, make the result ruinously 

 expensive. Successful apparatus can be fur- 

 nished at reasonable terms only by a conscien- 

 tious workman under the supervision of the 

 scientist. Can we not hope that Clark Univer- 

 sity will again add to its reputation by estab- 

 lishing a special mechanic who can make the 

 material for this course under Professor San- 

 ford's personal supervision ? This will aid in 

 the introduction of elementary laboratory work 

 throughout American institutions. 



In conclusion, it would be hard to overesti- 

 mate the labor, care and skill that show them- 

 selves in every line of Professor Sanford's 

 book ; as an elementary laboratory course it is 

 not only a pioneer — it is at the same time a 

 brilliant success. It is to be hoped that this 

 Part I. will be followed by a Part II., which 

 shall serve as a second-year course of a quanti- 

 tative character ; the subject of Time — which 

 has been, I believe, promised for this part — 

 lends itself readily and elegantly to this method 

 ■of treatment. 



As an episode in the history of science this 

 book marks the introduction into psychological 

 work of the elementary qualitative laboratory 

 method which has, for example, been so suc- 

 ■cessful in chemistry ; it also bears some resem- 

 blance to elementary courses in physics given 

 in some high, normal and grammar schools. 

 It is, of course, not intended that such labo- 

 ratory work should form the whole of the psy- 

 chological instruction ; a more general treatise 

 would probably be read at the same time, such 

 as Ladd's Outlines of Descriptive Psychology, or 

 Titchener's Primer of Psychology, or, possibly, 

 my own New Psychology, with the omission of 

 the couple of ditHcult chapters on statistics and 

 color. This elementary qualitative work should 

 be followed by most carefully planned exercises 

 in elementary psychological measurements ; at 



Yale such a set of thirty exercises has been de- 

 signed to teach such elementary concepts and 

 methods as 'average,' probable error,' 'func- 

 tion,' 'plotting,' determination of constant 

 errors, compensation of progressive errors, in 

 addition to the usual psychological concepts and 

 observations in touch, hearing, sight, time, etc. 

 This course, in turn, should be followed by ad- 

 vanced work in psychological measurements 

 analogous to that in astronomy, geodesy, etc.; 

 such a course includes a discussion of proba- 

 bilities, statistics, least squares, etc., and their 

 application to psychological work. With the 

 completion of Professor Sanford's book and the 

 appearance of more advanced laboratory man- 

 uals we may hope to find the methods of in- 

 struction as well systematized in psychology as 

 in physics or chemistry. 



B. W. SCKIPTUEE. 



SCIENTIFIC JOURNALS . 

 The Journal of Geology for April-May, 1898 

 (Vol. VI., No. 3), contains the following papers : 

 ' Chemical and Mineral Relationships in Igne- 

 ous Rocks,' by Joseph P. Iddings. Professor 

 Iddings continues the interesting discussion of 

 the chemistry of igneous rocks by means of 

 plotted curves, which was begun in a previous 

 number. He first establishes the formulas and 

 molecular ratios of the chief rock-making min- 

 erals. Next from a series of diagrams which 

 are plotted by using silica-percentages as ab- 

 scissas and the ratio of the molecular ratios of 

 alkalies to silica as ordinates for a great num- 

 ber of rock analyses, illustrations and curves of 

 extreme mineralogical composition are drawn. 

 They serve very neatly to localize and group 

 within limits many rock analyses of more com- 

 plex relationships and cast much light on the 

 minerals that must result in the crystallization 

 of maqueas whose composition is known. 'The 

 Weathered Zone (Yarmouth) between the II- 

 linoian and Kansan Till Sheets,' by Frank Lev- 

 erett. This weathered zone is most pronounced 

 and best recognized in the region between 

 Davenport, Iowa, and Quincy, 111. Its char- 

 acter is illustrated by various well-sections. 

 ' The Peorian Soil and the Weathered Zone 

 (Toronto Formation),' by Frank Leverett. A 

 bed of muck and weathered soil, for which the 



