86 ON VARIETIES, RACES, SUB-SPECIES, AXD SPECIES. 



the advice of M. Sageret and M. Puvis, and sow the seeds of the 

 best fruit. 



We say then in fact, that the best results are obtained by- 

 taking the seeds of the best fruit for the purposes of sowing, by 

 collecting the seeds yielded by the first sowing, and by sowing 

 them, and so on, in the most favourable circumstances. 



M. Sageret makes an important observation on the influence 

 of the parents on the goodness of the fruit of their descendants ; 

 it is that the seeds of a bad melon of a good variety have yielded 

 melons of excellent quality. 



It is important to ascertain whether an annular incision which 

 produced such curious modifications in the hands of M. Sageret, 

 would produce the same effects on the fruits of the individuals 

 yielded by the seeds of the modified fruit. 



b. Modifications which a plant can undergo in the se- 

 cond period of its existence. 



If the influence of external agents on the production of seeds, 

 and consequently on such of their qualities as depend on their 

 organization, is undoubted, the influence of the same agents on 

 the germination of the seed and the development of the individual 

 which arises tlierefrom is much more manifest in this period of 

 the life of the plant tlian in the first, for the double reason that 

 we observe its effects at every moment, and that we can compare 

 them in the individuals sprung from seeds that are identical, but 

 placed in circumstances which are not so. 



To study methodically the effects of external agents, viz. heat, 

 light, electricity, atmosphere, soil, and water, on plants, we must 

 consider them with respect to places and times. 



A. Influence of external agents in the same place to modify 

 seeds that are identical ; (a) in one and the same time ; (b) in 

 different times. 



(a) External agents may act at one and the same time. 



For example, identical seeds may be modified in a place where 

 they have been sown, because there may be veins of earth dif- 

 ferent from the rest of the soil, and because the water may be 

 luiequally distributed over the soil ; in such cases each individual 

 \v\\\ find itself exposed to different external conditions. 



Since seeds, having the same origin, may not be exactly 

 identical and yet be so nearly so as to deceive an experimenter, 

 it follows that if, out of a certain number of seeds that are sown, 

 one or more plants are obtained that differ from the others, we 

 can never be certain whether this difference should be attributed 

 to external or to internal influences. To avoid this source of 

 error a^ much as is, in the nature of things, possible, seeds 



