ON VARIETIES, RACES, SUB-SPECIES, AND SPECIES. 85 



by lorluring a Helianthus anniius, by twisting-, cutting, layer- 

 ing, binding, and making circular incisions on it, made it yield 

 seeds which produced plants with variegated leaves : now this 

 result is conformable to what we know of the influence of seeds 

 which have, from some cause or other, lost some of their 

 qualities ; the plants they yield are weak, and their leaves are 

 often variegated. 



o. A circular incision in many cases favours the production of 

 fruit. M. Sageret having tried it on a Capucine rose, which 

 rarely bears any fruit, at least in Paris, obtained a tolerable 

 quantity, some of which had seeds ; one of these seeds yielded 

 a dwarf ro3e tree with apetalous flowers. A wild Quince, 

 treated the same way by the same gentleman, was so modified 

 that those flowers which were above the incision gave eoAable 

 fruit. 



We will now proceed to the application of the foregoing re- 

 marks to the sowing of fruit trees for the purpose of propagating 

 them, or of improving and getting new varieties from them. 



It cannot be doubted, after the experiments of M. Sageret and 

 others, that a great number of the varieties of our fruit trees can 

 be raised froui seed : we may mention as examples Doyennes, 

 Saint Germains, Eeinettes ; some sorts of Peaches, especially 

 that called Teton de Venus ; some sorts of Apricots ; most of 

 the varieties of Cherry — the Quetsche, Perdrigon blanc, Reine 

 Claude, St. Catherine, Red Damsons, &c., among the Plums ; but 

 not to depart from our definition of species, we add that these 

 varieties can only be thus propagated under certain circum- 

 stances ; we must therefore expect to find situations where these 

 varieties cannot be raised from seed, and also, remembering what 

 we have before stated with respect to the different individuals 

 yielded by seeds of one and the same plant, under similar circum- 

 stances, that every seed of a fruit tree will not reproduce the 

 one from which it was taken. 



When ice loant to ohtain from seed varieties with any pecu- 

 liarity, loe must take the seed from that plant ivhich possesses the 

 same peculiarity in the greatest degree, supposing always this to 

 be possible. For example, if we want early varieties of a fruit 

 tree, we should sow the seetls of that tree the fruit of which is 

 ripe first ; and as far as we can in the same conditions in which 

 this tree itself grows. 



This rule is generally true ; we therefore think that M. 

 Sageret was more correct than M. Van Mons, who, without 

 being wholly ignorant of it, said what we cannot exactly under- 

 stand, " I prefer the seed of a fruit less good, but oftener re- 

 newed, to that of a fruit less often renewed." According to 

 our notions, for a given number of sowings, we ought to follow 



