208 



THE AMEBIC AN NATURALIST [Vol. XLVII 



stant fasciated race 12 with which he worked is the cocks- 

 comb and his experimental researches on this plant led 

 him to conclude that "complete atavists," or normal 

 plants carrying the fasciated character in a latent state, 

 are very rare, and even under repeated selection are to 

 be obtained in very small numbers. Further, the normal 

 plants thus obtained do not breed true, but revert very 

 soon to the abnormal condition. While investigations 

 have not been made showing definitely that many of De 

 Vries's fasciations were not heritable, but were simply 

 somatic modifications, enough evidence is at hand from 

 numerous sources to justify at least the expression of a 

 strong doubt of their heritable character. According to 

 the observations of Knox, fasciated stems in Oenotheras 

 are not germinal in origin, but traceable directly, in 

 most eases, to insect injuries. Observations by Molliard 

 on Raphanus and Picris support this conclusion, while 

 Godron was unable to secure fasciated individuals from 

 the seeds of a Picris plant thus affected. The fact that 

 fasciation appears in every generation of (Enothera 

 plants in varying percentages, in certain cultures, espe- 

 cially those of a biennial nature, is best explainable on a 

 re-infection basis. Spiral torsion races such as Dipsams 

 sylvestris torsus in De Vries's cultures behaved, from 

 a genetic standpoint, in the same manner as his fasciated 

 races. Races of Dipsacus species are rich in torsions in 

 Holland and Denmark, but, according to Johannsen, 13 

 the seeds of torsus strains when grown in England pro- 

 duced normal progeny. This would indicate an environ- 

 mental rather than a germinal basis as a causal factor. 



In all of De Vries's experimental cultures of fasciated 

 races (with the exception of Celosia) only a certain per 

 cent, (averaging in most races 50 per cent, or less) of 

 the individuals in each generation possessed the abnor- 

 mality, and he was never able to breed a constant and 



u Possibly Geranium molle fasciatum may be an exception in which more 

 than one unit factor is responsible for the anomaly. Otherwise it should have 



13 Johannsen, W., public lecture IV, Boston, 1911. 



