652 LECTURE XXIII. 



But it does not always happen that an individual produced 

 sexually presents the characters of its parents ; it is some- 

 times the case that it possesses new characters. This is true, 

 though it is much less common, even of individuals produced 

 vegetatively. This development of new characters constitutes 

 what is known as Variation ; when it occurs in an individual 

 developed from a seed, it is termed seed-variation, when in 

 one produced vegetatively bud-variation. Bud-variation is 

 most commonly manifested in this way, that a branch of the 

 plant produces leaves which differ in form or colour varie- 

 gated leaves, for example, from those of the other branches ; 

 or a flower which is abnormal in form or colour ; or a peculiar 

 fruit, as when a branch of a Peach-tree bears a nectarine. 

 By these means new forms, termed varieties, are produced. 

 The varietal characters are not, however, reproducible with 

 certainty by means of sexual reproduction, for there is in the 

 offspring of varieties a tendency to assume the specific form, a 

 tendency which is known as Reversion. The hereditary cha- 

 racters are, namely, of different values. There are, first of 

 all, those which are characteristic of the class to which the 

 plant belongs : then those which are characteristic of the 

 Natural Order, of the genus, of the species, and finally those 

 of the variety. Of these, those which are characteristic of 

 the wider groups are the most constantly transmitted ; and 

 even the specific characters are transmitted almost equally 

 well by sexual as by asexual reproduction. But this is not 

 the case with regard to the varietal characters. As Darwin 

 says, when a new peculiarity appears, we can never predict 

 with certainty that it will be transmitted by sexual reproduc- 

 tion; but if both parents present the same peculiarity, the 

 probability is great that it will be transmitted to at least 

 some of their progeny. The varietal peculiarities can only 

 be reproduced with any certainty by vegetative reproduction, 

 and it is on this account that propagation by means of 

 cuttings, grafts, etc., is so much resorted to in horticulture. 



The discussion of the causes of variation will be deferred 

 for the present, but it may be pointed out that variability is 

 promoted by cultivation and by crossing. In fact there is 



