QUANTITATIVE METHOD AND PRIMORDIA 31 



The importance of the ingenious statistical method initiated 

 by QUETELET was not realized for many years. His work 

 has been scorned and ridiculed by certain biologists. 



WALLACE has done perhaps more than any other scholar 

 to call attention to the method of QUETELET, by giving in 

 his widely distributed book on Darwinism ^ a series of diagrams 

 illustrating the variation of certain properties in several animal 

 species. 



QUETELET' S results have been confirmed and extended by 

 FRANCIS GALTON;^ who ultimately introduced the statistical 

 method into biological science. 



Among the numerous statistical researches which have 

 already been published, the experiments of JOHANSEN on the 

 pure lines (see § 19) deserve a special mention, not only because 

 the notion of pure lines is important in itself, but because this 

 author has called attention to the importance of the extreme 

 •values and also to the fact that two or several distinct curves 

 may produce by their association one regular variation curve, in 

 which it is impossible to discover (without experiment) or even 

 to suspect the heterogeneity of the material. 



In 1 901 a special review, Biometrika,^ devoted to biostatis- 

 tical researches, was founded by WELD ON, PEARSON and 

 DAVENPORT, and this part of biological science has been 

 called Biometry. A large amount of valuable material has 

 already been collected. The biometrical researches, however, 

 have been hitherto rather fragmentary. 



The biometricians have constructed numerous variation 

 curves and skilful calculations have been made, but the in- 

 vestigated properties have been taken a good deal at random, 

 from species which belong to very different groups of the zoo- 

 logical and botanical kingdoms. Although numerous examples 

 of correlation have been studied, there exist on the whole neither 

 morphological nor physiological relations between the pro- 

 perties which have been the object of measurement, and there 

 is no systematic relation between the species of which pro- 

 perties have been measured. Therefore descriptive natural 

 science has only in exceptional cases found any help from the 

 results of biometrical work, however important this work may 

 be in itself. 



There seem to be, moreover, two weak points in biometry : 

 FIRSTLY, the exaggerated importance ascribed to the mean 

 value. A mean value is not always the quantitative expression 

 of a fact, of a reality ; it is often a rather artificial result of 



1 ALFRED RUSSEL WALLACE, Darwinism. London, Macmillan, 1890. 



*F. GALTON, Natural Inheritance. London, Macmillan, 1889. 



* Biometrika, a Journal for the Statistical Study of Biological Problems, edited 

 in consultation with FRANCIS GALTON by W. F. R. WELDON, KARL 

 PEARSON, C. B. DAVENPORT, Cambridge University Press. 



