136 BLASTOGENIC VARIATIONS. 



ated like organs/ 7 or " homotypes." It is found, how- 

 ever, that the variability of these like organs in an in- 

 dividual is less than that of similar like organs in all the 

 members of a race (it being as a rule 80 to 90 per cent, 

 as great), and that therefore there is a considerable cor- 

 relation between them. The principle that like organs 

 are correlated, or that the undiiferentiated like organs 

 of individuals have a certain degree of resemblance, 

 Professor Pearson speaks of as Jiomotyposis. 



Professor Pearson and his collaborators have deter- 

 mined the degree of homotyposis in 22 distinct series, 

 and have determined, for instance, the numbers of leaf- 

 lets on the leaf of the Ash (26 leaves being taken from 

 each of 329 trees), the number of veins in the leaf of 

 the Spanish Chestnut (26 leaves from 204 trees), and in 

 that of the Beech (26 leaves from 100 trees), the 

 prickles on the leaf of the Holly (26 leaves from 299 

 trees), the stigmatic bands on the seed capsules of 

 poppies (10,435 capsules, taken from 1064 plants), the 

 sori on 8 to 12 fronds of each of 101 Hartstongue 

 ferns, etc. The mean correlation for all the 22 series 

 was .4570, or practically the same value as was obtained 

 for fraternal correlation. The extreme values ranged 

 from .6311 to .1733, but there are numerous causes 

 which will account, at least in part, for these wide de- 

 viations from the average. Supposing that any of the 

 organs measured had undergone a certain amount of 

 differentiation or splitting up in various directions (and 

 this, it must be remembered, is always possible, as there 

 is no real criterion as to whether any given organ is 

 really undiiferentiated, or differentiated), this would 

 generally result in a great reduction in the correlation; 



