164 



HORTICULTURAL SOCIETY OF NEW YORK. 



greater root power, with nearly all of the good points of the Berlandieri, with 

 larger berries, highly colored juice, easy growth from cuttings and desirable 

 flavors and wine properties, not found in either Berlandieri or Vinifera, native 

 of the same region with Berlandieri, I would also surely use V. Champini, V. 

 rupestris, and perhaps V . candicans for special wine purposes. If, on account 

 of the non-resistance of Vinifera to Phylloxera, the hybrids might not be 

 sufficiently resistant thereto, I would graft them upon Berlandieri, Champini, 

 Candicans and Rupestris of that region, and continue to introduce into them 

 the finer hybrid strains of native blood. 



So I would do for other regions with their native species in combinations 

 with the best selections of Vinifera for such regions. The Vinifera we cannot 

 afford to omit, as it is the embodiment, in fine qualities, of thousands of 

 years of improvement. We can fortify its weakness with native blood, and 

 besides greatly enrich its qualities for every use. Yet we have produced 

 hybrids without any Vinifera blood of very good qualities for market, table 

 and wine, and could, so I confidently predict, erect a splendid viticulture, if 

 Vinifera were all destroyed. 



It is quite clear, from all our experience, that little can be gained by 

 combining more than three or four species for any particular region, and 

 those species should be, as far as possible, the best selected natives of the 

 region to be supplied. This applies only to those parts of the United States 

 east of the Rocky Mountains, for no other regions of the earth have any 

 native species, except the Vinifera, suitable to yield any very good results in 

 hybridization. 



For other parts of the world some combination of disease-resisting Amer- 

 ican varieties with Vinifera promise the best results, as in France many direct 

 producers, especially for wine, have been originated by combining V. rupestris 

 chiefly, and V. Lincecumii with Vinifera, The French taste greatly dislikes 

 the "foxy" and earthy flavors of the Librusca and its low percentage of sugar, 

 hence eschews it for new-variety-making purposes. 



The primary specific selections should always be, as nearly as possible, 

 from similar climatic and soil conditions to those to be served. 



If we should start with only one variety, or original selection, from each 

 of two or three species of any particular region, and continue indefinitely to 

 use these through a long course of selection and recrossing, eventually we 

 would tend too much to in and in breeding to permanently maintain en- 

 durance. Such seems to have been somewhat the character of development 

 of many Vinifera varieties, while others appear to have come down purely of 

 one species by a long course of pure selection ; and I believe this too close in 

 and in breeding has had much to do with the feebleness of many varieties 

 of that class of grapes. Hence, for each distinct set of climatic and soil 

 conditions we should have recourse to several different varieties of each 

 species used, collected from widely separated localities of that region, if 

 possible. 



For example : If we were endeavoring to produce the most desirable of 

 market and table grapec for New England, the Atlantic Slope and the Great 

 Lakes region, where the flavor of Labrusca is relished, we would begin by se- 

 lecting Concord, Ives, Perkins, Wyoming, Columbian and others of the best 



