136 REVERSION AND ALLIED PHENOMENA 



downy, or when the white-flowering currant puts forth red 

 flowers. Such cases may be described as reversions to the 

 specific type, and they can be interpreted only in two ways. 

 Either we have to do with new variations which happen to hit 

 the old mark, or, as seems more probable, latent ancestral 

 characters have re-asserted themselves. 



It is a current belief that these " varieties " have a strong 

 tendency to " revert " to the parent species, but, according to 

 De Vries, this is, as regards pure varieties, not of hybrid origin, 

 and ordinarily propagated by seeds, a popular delusion. " In 

 the present state of our knowledge it is very difficult to decide 

 whether or not true reversion occurs in constant varieties. If 

 it does occur it surely does so very rarely, and only under unusual 

 circumstances, or in particular individuals " (1905, p. 155). It 

 must be noticed, however, that De Vries distinguishes true 

 reversion (due to a spontaneous germinal change) from false 

 reversion which is induced by hybridising. 



In illustration of the constancy of varieties he cites the wide- 

 spread rayless form of the wild camomile (Matricaria chamomilla 

 discoidea), which is so constant that many botanists have made 

 a species of it. De Vries raised in successive years between 

 1,000 and 2,000 seedlings, but observed no trace of reversion. 

 Similarly, the rayless " variety " of the common tansy ragwort 

 (Senecio jacobcea) is quite as stable as the radiate species. De 

 Vries also refers to the stability of white strawberries, green grapes, 

 white currants, crisped lettuce, crisped parsley, smooth spinach, 

 white flax, sugar-corn, and strawberries without runners. 



Seed-reversion very Rare. Excluding cases where it is 

 doubtful whether the variety has not a hybrid origin, and is 

 therefore liable to the peculiar phenomenon known as the 

 splitting up of hybrids, excluding also all cases of "sporting 

 varieties," where an apparent reversion might be a mere 

 coincidence in the crowd of variations, De Vries concludes 

 that " seed-reversions must be said to be extremely rare. . . . 



