THE MUTATIONS OF LYCOPERSICUM. 159 



certainly have been incomplete as to the whole crop and various as to 

 the kinds of hybrids produced. Even if it were credible that the 

 first case of complete aggregate mutation was due to fortuitous cross 

 fertilization from some unknown source, it would still be too much to 

 believe that exactly the same hybridizing process should have been 

 repeated in the same manner in a following year. It may be added 

 that there is now much reason to doubt that hybridization, although 

 always imminent among tomatoes, has ever been so effective an agent 

 in producing improved varieties of either plants or fruit as has been 

 generally believed. Indeed,' saltatory mutation and racial variation 

 have doubtless produced many of the results among plants that have 

 been attributed to hybridization; although the latter has produced 

 many wonderful results. 



At the close of this narrative of experimental observations it is 

 well to call special attention to the assumed fact that the mutative 

 process which produced the new plant form that has been described 

 was essentially separate from the accompanying process of fruit varia- 

 tion, although the two processes were intimately associated in both 

 their origin and development. The plant mutation was from L. 

 esculentum to L. solanopsis; the fruit variation was from the Acme to 

 the Washington variety. The new fruit variety which accompanied the 

 new plant form is of fine quality and therefore of horticultural value; 

 but the origination of any fruit variety is, from a naturalist's point of 

 view, of far less importance than the origination of a species. Plant 

 mutation produces species which are real entities. Fruit variation is 

 limited to changes in the pericarp; and the most improved and herit- 

 able fruit variety thus produced may, by degeneration, become dis- 

 associated from the plant entity without any impairment of that 

 specific condition. 



There are two extraordinary features of the foregoing narrative 

 of my observations. One of them relates to the sudden and complete 

 mutation of every plant of a crop of twenty-four Acme tomato plants 

 to another specific plant form bearing a new variety of fruit. The 

 other relates to a subsequent exact duplication of that mutation in all 

 its details as to both plant and fruit, in a crop of thirty plants, also 

 of the Acme variety. It is apparent that both cases were initiated and 

 consummated in the plants while they were growing in my garden 

 because the germ cells which gave origin to the mutated plants were all 

 formed there; and the mutated plants were there grown to maturity. 

 Another fact, important in this connection, although stated in a pre- 

 vious paragraph, is that this new specific form had been previously 

 produced by gardeners, who had given to it the name of potato-leafed 

 tomato. That is, this one and the same species, L. solanopsis, has 

 arisen suddenly and independently from L. esculentum at not less than 

 three different times, each in a different locality. 



