90 MASSACHUSETTS HORTICULTURAL SOCIETY. 



hand by applying a definition. One must show that his new type 

 — if it is a plant — has botanical characters as well marked as 

 similar accepted species have, and these characters must show, 

 as a whole, a general tendency towards permanency when the 

 plant is normally propagated by seeds. He must measure his 

 type by the rule of accepted botanical practice. If the same plant 

 were found wild, so that all prejudice might be removed, would 

 the botanist unhesitatingly describe it as a new species? If yes, 

 then we should say that a new species had been created under the 

 hand of man ; and this rule I wish now to apply to a very few 

 familiar plants. In doing so, I do not wish to be understood as 

 saying that I consider it advisable to describe these plants 

 as species under the existing methods of botanical description and 

 nomenclature, for, merely as a matter of convenience and perspi- 

 cuity, I do not ; but I wish to show that they really are, in every 

 essential character, just as much species as very many other 

 universally accepted species are. 



The evolution of forms which any botanist would at once 

 designate as species, were he ignorant of their origin, is well 

 illustrated in the tomato. Dunal, the accepted authority upon the 

 genus Lycopersicum, admits ten unqualified species into his 

 account in De Candolle's Prodromus. Two of these, L. pyriforme 

 and L. cerasiforme, are generally regarded as mere forms of the 

 -common garden species, L. esculentum, both because they are 

 very like the common tomato in botanical characters and because 

 we know, as a matter of history and experiment, that all three of 

 these reputed species are modifications of one type. Omitting 

 these two species, then, there remain eight to which we cannot 

 attach any such suspicion as a knowledge of their origin. These 

 are what botanists call good species. These species agree in 

 having a weak and spreading habit of growth, much like the 

 common tomato. The features by which they differ amongst 

 themselves, that is, the specific characters, are founded chiefly 

 upon the manner of division of the leaves, the shape of the leaf- 

 lets, the character of the flower cluster, and the relative hairiness 

 or smoothness of the parts. If one applies these same tests in 

 the same degree, to the two modern offshoots of the tomato — the 

 Upright and the Mikado types — he will find that these offshoots 

 differ as much or even more from each other and from their own 

 ■common parent, than auy one of the wild species differs from any 



