384 THE GRAMMAR OF SCIENCE 



We now reach the conception of race type as distinct 

 from individual type.^ The mode for beech leaves in 

 general is 1 6, and the average 1 6. i i . The mode for 

 poppies in general is lo, and the average 9,84. We can 

 now look upon individuality as the divergence in any 

 special case of individual type from racial type, or of 

 individual from racial average. The possibility of in- 

 dividuality existing within the race depends upon all the 

 racial frequency not being concentrated under one class, 

 i.e. upon the racial mode not being universal. Here we 

 have the numerical conception of individuality consisting 

 in everybody not being alike, i.e. everybody not in the mode 

 or fashion. Here in variety, the various deviations from 

 the racial type, we find the material for selection. We 

 have next to see how this variation is to be measured. 



S 6. — Variation : Continuo7is and Abnormal 



Taking the poppy capsules, we notice a gradual decrease 

 in the frequency of the stigmatic bands from the mode out- 

 wards. The numbers 9 and 1 1 are more frequent than 8 

 and I 2, and these more frequent than 7 and i 3, while series 

 of few bands like 5 and 6, or of many like i 5 and 1 6, are very 

 infrequent indeed. The entire range is from 5 to 1 6. Within 

 this range lies the variation. But to take the range itself as 

 a measure of variation is a very rough estimate. It gives us 

 no idea of how the frequency is distributed within that range." 

 For example, we should have had the same variation, taking 

 this to be measured by range, if there had been 20 poppy 

 capsules in the 2268 with 5, and 60 with 16 bands, instead 



1 We have deduced our conception of the individual type by taking the 

 average character for a number of like organs in the individual ; but we have 

 practically the same conception if we deal with organs of which the individual 

 has not a multiplicity. The individual type is then what the average becomes 

 as we pass from many to few, or ultimately to a single one. Yet even in this 

 latter case the single organ will generally be found to be really a complex of 

 like elements, be they cells or what not ; and its individuality or type must 

 be looked upon as ultimately the product of the individuality of these like 

 elements. This, it seems to me, is not only the basis of likeness in heredity, 

 but, further, of the individuality in the growth of the individual. 



2 See on this point the essay on "Variation in Man and Woman" in my 

 The Chances of Death and other Studies in Evolution, vol. i. p. 275. 



