November 4, 1922] 



NA TURE 



601 



Our Bookshelf. 



How to Measure in Education. By Prof. W. A. McCall. 

 Pp. xiii + 416. (New York: The Macmillan Co.; 

 London : Macmillan and Co., Ltd., 1922.) 155-. net. 



In the work under notice an attempt is made to show 

 that everything in education must submit to statistical 

 measurement or be condemned as worthless. Fourteen 

 theses in praise of measurement are blazoned in large 

 type at the outset. One of them states, " To the extent 

 that any goal of education is intangible it is worthless " ; 

 as this is given not only the dignity of capitals but also 

 the embellishment of inverted commas it presumably 

 conveys some meaning to the author. Education in 

 this book means a few of the elements of instruction, 

 such as reading, writing, and the mechanical parts of 

 arithmetic and composition. An elaborate analysis 

 is given of how to diagnose defects in reading, and 

 ignoring the analysis, we are told that " there are more 

 failures due to failure of interest than the world 

 dreams of." The schoolmaster who has important 

 tasks in education other than those of teaching mechani- 

 cal elements would greatly value some help in measuring 

 the interest of his pupils, but will ask in vain. The 

 importance of carefully framed instructions in giving 

 tests is rightly stressed, but we are told that such 

 " instructions should equalize interest." To accom- 

 plish this the pupil must, apparently, be told how 

 important it is to do well in a test. If he is refractory, 

 or keen on other things in school besides tests, the 

 advice may not be effective. Masters, however, are 

 in a worse plight, for the tests are used not only to 

 measure the pupils, but also the teaching and the 

 teachers ; and that form of measurement is said to be 

 of most service " which does not require a previous 

 acquaintance with the pupils." 



The most valuable part of the book is that concerned 

 with " scaling the test," as it shows the large amount 

 of statistical treatment necessary to prepare a suitable 

 test of skill. A useful suggestion is made for fixing 

 a single common unit of measurement for all mental 

 scales for elementary schools, namely, some function 

 of the variability of pupils of twelve years. The 

 standard deviation of pupils of sixteen years is also 

 suggested as a unit for measuring older scholars. 



The final section of the book deals with tabular, 

 graphic, and statistical devices. Each part has a 

 useful students' bibliography, but it is strange to note 

 the omission, in a work of this nature, of all reference 

 to the writings of Udny Yule, from whom the student 

 of statistical methods will derive more real help than 

 from any of the authors quoted. 



Fruit Farming : Practical and Scientific for Commercial 

 Fruit Growers and Others. By C. H. Hooper. Second 

 edition, Revised and Extended. Pp. xxiii + 212. 

 (London : The Lockwood Press, 1921.) 6s. net. 



Within the limits of two hundred pages Mr. Hooper 

 has aimed at the production of a text-book of fruit 

 farming under English conditions. After a brief 

 introductory section on the training of the prospective 

 grower, he deals in succession with the capital required, 

 the selection of suitable land, the law in relation to 

 fruit farms and market gardens, the laying-out of 



NO. 2766, VOL. I io] 



plantations and orchards, and the cultural details and 

 costings of the more important hardy fruits of this 

 country. The later chapters are concerned largely 

 with the more scientific aspects of the subject, such 

 matters as soils, manuring, insect and fungoid pests, 

 and spraying and other forms of disease-control in turn 

 receiving attention. Also included are brief histories 

 of many of the well-known varieties of apples, pears, 

 plums, and cherries. The numerous interesting and 

 economically important problems relating to pollination 

 and fertilisation and the setting of fruit are briefly 

 considered and the author is able here to provide data 

 from his own investigations. The volume is completed 

 by several sections dealing with special points of a 

 purely commercial character, as, for example, the 

 marketing of fruit and book-keeping. 



Many chapters have been contributed by specialists 

 and present in an abbreviated form the results of recent 

 research. A large part of the volume has in fact been 

 re-written since the first edition and brought up-to- 

 date. With these alterations and some useful additions 

 the second edition justifies its appearance. 



In a work of so small a compass, which seeks to stand 

 as a text-book of its subject, the selection of matter for 

 inclusion calls for careful discrimination. It is open 

 to some criticism in this respect and also for occasional 

 repetition. Printers' errors are not infrequent in 

 certain chapters. Much of the information given on 

 economic points is not readily accessible elsewhere, and 

 it is in this direction probably that the volume will 

 prove to be of most service to its readers. 



First Lessons in Practical Biology. By E. W. Shann. 

 Pp. xv + 256. (London: G. Bell and Sons, Ltd., 

 1922.) 5s. 



Mr. Shann's endeavour to provide a course of biology 

 suitable for lower fifth forms, and within the means of 

 the average school, is not entirely satisfying. Rightly 

 he relies on plants for the experimental work, and on 

 both plants and animals, employing them in alternate 

 chapters, for the observational. But experiments on 

 plant physiology are not reached until chapters 16 and 

 17 ; and by that time the preceding lessons have 

 incidentally given the very information which the 

 experiments should surely be intended to enable the 

 pupils to discover for themselves. There are good 

 chapters on variation and heredity, soils, insect pests, 

 and other topics of general biological interest ; but 

 neither with plants nor with animals does the author 

 make the best use of his material as a means of educa- 

 tion and of training the powers of observation and 

 reasoning. If he disapproves of the heuristic method, 

 he should at any rate indicate the evidence on which 

 conclusions as to homologies are based, and not be 

 content with mere statements. 



Apart from general considerations, sentences are in 

 several instances faultily composed ; and there is a 

 large number of actual errors : e.g. " false " fruits are 

 wrongly defined ; rose-hips will germinate without 

 passing through the digestive tract of an animal ; rose 

 stamens are peri-, and not epigynous ; Urlica urens 

 has not a creeping stem ; the biramous appendage is 

 not the primitive form of crustacean appendage ; the 

 telson is not a segment ; the abdomen of Blowfly does 

 not exhibit respiratory movements ; the embryo in a 



