114 THE QUANTITATIVE METHOD IN BIOLOGY 



the description of the species, they may -be regarded without inconvenience 

 as simple. 



Several authors have tried to find in the dimensions of the shells (for instance, 

 the ratio breadth : length, etc.) characters of the species. This method has met 

 hitherto with little success, because the majority of the authors did not know 

 in which way variable properties ought to be measured (see Part VI. ).^ 



In § 80, p. 98, I have called attention to the importance of 

 the study of the shells. Since the notes and the material which 

 I have collected for years are not within my reach, I must con- 

 tent myself with the above very incomplete indications. 



FOURTH EXAMPLE : The retina of the vertebrate animals 

 is a triaxial system, the structure of which is complicated but 

 comparatively regular. I think that it is practically possible 

 to measure certain of its primordia.^ A comparative investi- 

 gation of the retina of several allied species would be here, as 

 for many other subjects, the best method. 



iSee HENRY EDWARD CRAMPTON (Professor of Zoology. Barnard 

 College, Columbia University), Studies on the Variation, Distribution and 

 Evolution of the Genus Partula (the species inhabiting Tahiti). 314 pages, 

 34 + 19 (very beautiful) plates and numerous diagrams and tables. (Carnegie 

 Institution of Washington . ( Publication No .228.4°. Issued 20 th J anuary 1 9 1 7 . ) 



In this interesting work (with which I became acquainted through the 

 kindness of Prof. S. J. HICKSON, Manchester University) four properties 

 (length and width of the shell, id. id. of the aperture) axe measured in numerous 

 specimens of a number of species and local races. 



2 Dimensions of the cones, rods, pigmentary cells, etc. 



