182 BLASTOGENIC VARIATIONS. 



Such union may have been the starting point of con- 

 siderably increased variation, which was never lost, 

 even through innumerable subsequent asexual genera- 

 tions. Thus Professor Pearson has shown that if the 

 ancestors of individuals be selected so as to be abso- 

 lutely similar in character for an indefinite number of 

 generations back, such individuals will still have a 

 variability of upwards of 89 per cent, of that of the 

 original race. Though produced sexually, these indi- 

 viduals are in reality comparable to asexually repro- 

 duced forms, as by hypothesis no new characters were 

 introduced by any of their ancestors. 



Whether the difference between racial and individual 

 variability is as small as Pearson maintains, or not, de- 

 pends solely on what is meant by the word " race." If 

 " species," in the generally accepted sense, is meant, 

 then the view is certainly incorrect. If, however, a group 

 of individuals is implied, all of which have been exposed 

 during several generations to practically identical con- 

 ditions of environment, then the view must be admitted. 

 It is of little practical value, however, as may be real- 

 ised by quoting certain data which Pearson has himself 

 brought forward in another connection.* Thus the 

 variability in the number of stigmatic bands in a sample 

 of the wild poppy, Papaver Rh&as, collected in a corn 

 field at the foot of the Chiltern Hills, was found to be 

 1.473, that in two individual poppy plants being, on an 

 average, 1.166, or 20.9 per cent. less. Another sample 

 was collected in some fields at the top of the Chilterns, 

 and in this case the variability was 1.770, or 20.2 per 

 cent, greater than in the other sample. Moreover, the 



* " Grammar of Science." p. 387. 



