THE AMERICAN NATURALIST 



[Vol. XLII 



There are as many species as the Infinite Being orig- 

 inally produced different forms; and these forms, fol- 

 lowing the laws of reproduction imposed upon them, 

 have produced more, but always similar to themselves. 

 Therefore, there are as many species as there are dif- 

 ferent forms or structure met with to-day. 



From Adanson, Families des I Mantes, 1763. 



The moderns define a species of plant as a collection 

 of several individuals which resemble each other per- 

 fectly, yet not in everything, but in the essential parts 

 and qualities, without, however, giving attention to the 

 differences caused in these individuals either by sex or 

 accidental varieties. 



According to Linna?us (Phil, bot., p. 99) "the species 

 of plants are natural and constant, as their propagation 

 either by seeds or cuttings is only a continuation of the 

 same species. Individuals die, but the species does not. ' ' 



But we wish to make a distinction between reproduc- 

 tion by seed and that by shoots, offsets, conns, cuttings, 

 suckers or by grafting. These last simply continue the 

 individual from which they are taken and consequently 

 are opposed to the production of new species in plants; 

 whereas seeds are the source of a prodigious number of 

 varieties, sometimes so changed that they may pass for 

 new species. He cites, among other examples : 



"In 1715 Marchant found in his garden a new species 

 of Mercurialis and the following year it came from self- 

 sown seed ; again, four resembled the parent and two were 

 so different that he made another species of Mercurialis. 

 These two new plants were cultivated and continued to 

 grow T each year." 



It is well known that without foreign fecundation in 

 plants that reproduce by seed, similar changes are induced 

 either by reciprocal fecundation of two different indi- 

 viduals or owing to cultivation, the soil, the climate, 

 dryness or moisture, light or shade, etc. These changes 

 are more or less prompt, more or less durable, disap- 



