CHAPTER. IV.— SPECIAL PHYSIOLOGY OF REPRODUCTION. 779 



It would seem natural to carry this rule further, applying it to 

 the case of asexual reproduction so as to lead up to the conclusion 

 that variation does not occur at all in plants produced asexually ; 

 (that variation is, in fact, necessarily associated with the sexual 

 Iprocess ; a conclusion which has met with a certain measure of 

 'acceptance. But it is a question whether this conclusion can be 

 fairly drawn when all the facts of reproduction in plants are duly 

 taken into account. Considering, first, the fact of asexual repro- 

 duction by means of gonidia, which is so common among the 

 Thallophytes, and which in some Fungi (e.g. Basidioraycetes and 

 many others), is the only known mode of reproduction, it is in- 

 conceivable, in the face of the vast number of families, genera, and 

 species of the plants under consideration, that variation has not 

 taken place among them, although they present no indications of 

 sexuality. Then there are the facts of hud-variation to be con- 

 sidered. Bud- variation is the term used to express the fact that 

 shoots, whether reproductive or vegetative, are sometimes de- 

 veloped, which differ in some striking manner from the other 

 normal shoots of the plant ; these varying shoots are known as 

 \" sports " : for instance, many varieties of Rose are known to have 

 "arisen in this way, such as the Moss-rose from the Provence rose 

 iRosa centifolia). The varieties arising in this way can, as a rule, 

 only be propagated vegetatively, by cuttings, grafts, etc., but oc- 

 casionally they come true from seed. Here again variation seems 

 to occur intlependently of sexual reproduction. 



With regard to the causes of variation, it would appear that the 

 capacity for variation, or variability, is increased (a) by sexual 

 reproduction, and (6) by external conditions, more particularly 

 high nutrition, as is shown by the fact that bud-variation, for 

 instance, is rare in wild plants but comparatively common in 

 cultivated plants. The character of the variation has also to be 

 considered. There is no doubt that variation is often the mani- 

 festation, not of a new character, but of an old ancestral character 

 which had become dormant : this is especially true of the varia- 

 tions of sexually-produced plants, but it is also true in many cases 

 of bud- variation : this kind of variation is termed reversion . But 

 there seem to be many cases of variation, and more particularly 

 of bud- variation, which cannot be attributed to reversion, and in 

 which the direction of variation does not appear to be traceable to 

 any deteimining external causes : such variations seem to be 

 altogether fortuitous. 



