232 AMERICAN VINES. 



the contact can only, therefore, take place at two points, A 

 and B, and consequently the knitting is defective. There are 

 different means of bringing the sides of the scion parallel to 

 one another, and to bring them into juxtaposition for the whole 

 of their length with the generative layers of the stock. 



ist. By using scions larger than the stock, the contact 

 of the generative layers takes place, in this case, nearer the 

 top of the slit, the lower portion of the wedge slightly pro- 

 truding from the slit in the stock; but as the two tongues 

 are very thin and separated only by the pith, which has no 

 resistance, the binding brings them close together inside the 

 slit, and so makes them coincide exactly with the generative 

 layers of the stock. 



2nd. By making the lateral section as close as possible to 

 a node; at this point the dimensions of the canes used as 

 scions are larger, the edges of the section do not form an 

 exact parabola, and are closer for the greater part of their 

 length, at least, to the parallelism sought. 



3rd. By making two shoulders on the scion (Figs. 109 to 

 112), rendering the lateral section of the wedge almost 

 triangular in shape, and, therefore, its edges almost parallel. 

 By combining these three means, we may obtain an almost 

 perfect parallelism of the surfaces brought in contact, and, 

 consequently, a more perfect knitting. The shoulders may 

 be made with a very narrow-bladed knife; they are in this 

 case slightly rounded (Figs. 109 and no) ; or, better, by 

 means of special machines. Their object is not only to 

 facilitate the knitting and strengthen the joint, but also to 

 diminish the extent of the wound which always exists in an 

 English cleft-graft (Figs, in to 113). 



The cleft-graft, shouldered or not, never gives as perfect 

 a joint as the whip-tongue, especially during the first year. 

 The knitting only takes place along a longitudinal slip equal 

 to the thickness of the part of the wedge inserted into the 

 stock, therefore, the circulation of the sap can only take 

 place through that strip, and not through the whole circum- 

 ference (Fig. 113) ; on each side are large wounds or more or 

 less extended portions of dead wood, which only become 

 completely healed a few years after. This perhaps has not 

 very great importance for fine growing vines grafted by the 

 English cleft system are to be seen everywhere. And, again, 

 if theoretically the whip-tongue graft is perfect, practically 

 it rarely attains that degree of perfection. 



