XIV] Hybrids breeding true 249 



found in the course of contemporary experiments. It is to 

 be hoped that the original experiment may be repeated, for 

 Millardet's note is very brief, and we have no means of 

 judging whether the various possibihties of error were 

 excluded (e.g, heterozygosis of the mother-plant). 



Whatever the future may reveal as to the significance of 

 these ambiguous occurrences it is certain that. in all cases 

 where irregular results are observed we may have to reckon 

 with two possibilities: (i) actual parthenogenesis, or the 

 developm.ent of unfertilised ova without fertilisation ; (2) a 

 phenomenon tantamount — as regards heredity — to partheno- 

 genesis, occurring after fertilisation '^. In both events the 

 offspring are purely maternal. The latter exemplifies the 

 conception of Strasburger and Boveri, that fertilisation may 

 consist of two distinct processes, the stimulus to develop- 

 ment, and the union of characters in the zygote. 



B. Sexual Types. 



The literature of hybridisation and heredity abounds 

 with examples of hybrids or cross-breds which are said 

 to have bred true without splitting up, or as we should 

 now say, without segregation. Some of these, so far as 

 they refer to plants at least, are open to the criticism 

 made in the last section, that either actual parthenogenesis 

 or monolepsis may be occurring. But an even simpler and 

 more probable account is hardly ever excluded, for we have 

 rarely if ever any means of knowing that the plants studied 

 were in reality first crosses and not members of derivative 

 generations. Moreover even in cases such as the famous 

 series of Salix hybrids made by Wichura (304), where this 

 objection does not apply, we know almost nothing as to the 

 numbers of individuals raised. These remarks probably 

 apply to all the illustrations given by Focke (123) so far as 

 I have been able to ascertain. In Mutationstheorie, 11. p. 66, 



* On a former occasion (i8, p. 156) I suggested that in mosaic forms 

 the presence of the recessive character in patches might be thought of as 

 due to failure of fertiHsation in those areas. More accurate knowledge of 

 the facts has practically disposed of this suggestion. It was chiefly based 

 on evidence relating to the mosaic seeds of Peas, patched with green and 

 yellow in the cotyledons, but we may now feel fairly sure that the 

 abnormalities seen in their behaviour are due to disturbance caused by 

 external influences. 



