10 RELATIONSHIPS OF THE COCCACE^E 



either class. The problem for the systematist is: What 

 weight shall be given to these variations? Shall a 

 specific name be used for every discernible degree of 

 variation? Or shall the members of a great group be 

 included in one indistinguishable mass? The first 

 course leads to a hopeless confusion of detail; the second 

 conceals real differences in a cloud of vague generaliza- 

 tions. The conflict between the two views has been 

 the notable feature of much systematic bacteriology, as 

 in the endless discussions as to the "Vielheit" or the 

 "Einheit" of the streptococci. 



There has recently been placed in the hands of the 

 biologist a new instrument of research which promises 

 at last to solve these vexatious problems. This is the 

 statistical method, first suggested for the study of human 

 characteristics by Quetelet (1846); specifically applied 

 to the biological problems of variation and heredity by 

 Galton (1889); and extended and developed in detail 

 by Pearson and his pupils. The most important papers 

 on this subject may be found in the files of the Philo- 

 sophical Transactions of the Royal Society of London and 

 in Biometrika. Admirable brief summaries have been 

 prepared by Pearson (1900) and Bigelow (1904) and 

 Vernon (1903). 



From the systematic standpoint there are two impor- 

 tant generalizations from biometry, the statistical 

 method of studying biology. In the first place, normal 

 fluctuating variations, when measured in a considerable 



