SELECTION AND HYBRIDIZATION AMONG GRAPES. 165 



types of Labrusca yet found, along with Clinton (an accidental wild hybrid 

 of Labrusca X Vulpina, found in New York) and the best of the Taylor 

 family (accidentally originating in Kentucky as a hybrid of Labrusca X 

 Vulpina}, and we would include some of the very best of the / '. bicolor of 

 Ohio (such as Kohr and others found by Dr. G. L. Tinker of New Philadel- 

 phia, Ohio, who is most intelligently working to produce a family of Bicolor 

 hybrids), and some of the large-berried, large-clustered V. Lincccumii of 

 Southwest Missouri, Arkansas, the Indian Territory and North Texas (such 

 as Jaeger's 43 and my Lucky, which have proven perfectly hardy in Massa- 

 chusetts, Ohio and New York, and very resistant to rot and mildew), and in- 

 termingle them by twos and threes with each other and the hardiest, healthiest, 

 large-clustered, large-berried Vinifcra varieties. 



Thus, several distinct families should be started, and from time to time 

 various selections from these different families of hybrids should be chosen, 

 according to their peculiar fitness for some special purpose or location, and 

 combined to form secondary families, that would be decided improvements 

 upon their progenitors. Such a large basis to use for development would 

 serve for many generations, and the limits of evolution of colors, seasons, 

 flavors, etc., could never be reached. Clusters requiring two men with a staff 

 to carry them, as in old Canaan, might be produced, with berries on them as 

 large as Kelsey plums! Who knows? If nature, by its haphazard selection, 

 aided by the wild animals, produced the cocoanut, the banana, the bread 

 fruit and the marvelous durion fruit of the tropics, what may not nature and 

 man together produce, with scientific knowledge, keenly sharpened observa- 

 tion and skilled technique on the part of man? 



As an example of cluster lengthening in two or three generations from 

 varieties, none of which had half so long clusters, those of the Captain grape 

 have come, and, withal, much better qualities. As an example of refining 

 and compounding flavors by complex hybridization, I mention but one among 

 a large number produced, of which this is an exact natural size cluster, repro- 

 duced somewhat nearly in color. It is named Wapanuka, and contains in 

 combination at least three species, and probably four, partly by accidental 

 and partly by careful hand hybridization. 



This variety was produced by me by pollenizing my Rommel with Bril- 

 liant pollen. Rommel was produced by me by pollenizing Elvira with Camp- 

 bell's Triumph. Elvira was produced by Jacob Rommel of Missouri from 

 seed of Taylor, probably accidentally crossed by some Labrusca, unknown, or 

 may have been a pure seedling of Taylor. Taylor was an accidental hybrid of 

 some unknown Labrusca with some unknown Vulpina. Its blood is proven by 

 its characters being of these two species, without doubt. Triumph was pro- 

 duced by the lamented George W. Campbell by pollenizing Concord with 

 Muscat of Alexandria (Vinifera). Concord, as you all know, is a seedling 

 of second generation by E. W. Bull from some wild Labrusca ("Fox Grape"), 

 found in the woods by some boys near where Mr. Bull lived in Massachusetts, 

 and given to him to taste. Brilliant was produced by me by pollenizing 

 Roger's No. 9 (Lindley) with Delaware. Lindley was produced by E. S. 

 Rogers, the pioneer hybridizer of grapes in America, by pollenizing some 

 large Labrusca variety, found by him in the woods of Massachusetts, with 



