20 PRODUCTION OF NEW VARIETIES. 



stigma unfertilized. To prevent the possibilities of impreg- 

 nation by bees or insects, or the wind conveying pollen from 

 other sources, the prepared blossom-bunch is inclosed in a 

 tight, oil-silk case, and pollen supplied at the proper time 

 from whatever variety it is proposed to cross, or hybridize 

 with. When the berries swell, and commence growing, it is 

 an indication that the process has been successful; and the 

 oil-silk covering may then be removed, the bunch carefully 

 labelled, and the seeds from these berries, when planted, are 

 expected to produce crosses or hybrids having characteristics of 

 both parents. 



" I have also tested the accuracy of my experiments in va- 

 rious ways. In one instance I prepared a bunch, as if for 

 crossing, by removing all the stamens, and inclosed it in the 

 usual manner, but applied no pollen. Upon removing the 

 covering some days after, every berry but one had blasted, 

 and fell off at a touch. This one berry, being from some 

 cause later than the rest, was just in condition to receive 

 pollen, which I supplied from the Chasselas Musque, and pro- 

 duced a grape, from which I have a seedling that may prove 

 valuable. Other bunches, prepared at the same time, upon 

 the same vine, and supplied with pollen at the proper time, 

 were all fertilized, and produced full and perfect bunches. 

 The Logan and Taylor's Bullitt both set their fruit unevenly 

 and imperfectly, and produce usually small, straggling, and 

 unhandsome bunches. When fertilized in the manner above 

 stated they have produced handsome and compact bunches, 

 the only ones of that character upon the vines. 



" Seedlings almost uniformly indicate their parentage by 

 their foliage. That of hybrids with the foreign vines is usu- 

 ally deeply lobed ; often having much more the form of the 

 foreign than the native leaf, although grown from the seed 

 of the native parent. Some have foliage intermediate or re- 

 sembling both in some degree. Also, in the crosses between 

 natives, some resemble one parent and some the other. 

 Others again seem a mixture of both." 



An easier process is to plant them in close contact, so that 

 the fruiting branches may intermingle. Out of a large num- 

 ber of seedlings thus obtained, there is a chance of a fair por- 

 tion of them being crosses. It was in this way that Dr. Kirt- 



