108 NUT GROWING 



ing at the distal end of a branch to be grafted we 

 usually observe that some one part of the circum- 

 ference of the branch is less rounded than the rest 

 and the flatter part is to be chosen for the scion site. 

 One may pick out in the same way a comparatively 

 flat surface for a proximal graft. 



When the scion has been inserted in a cleft graft 

 it is bound in place with raffia or with twine in the 

 usual way. Raffia is always preferable because it 

 does not cut into the bark of the growing stock and 

 it is usually burst asunder by the enlarging of the 

 stock. If twine is used for binding material it will 

 cut in and choke off growth unless the grafter keeps 

 close watch and loosens the binding himself instead 

 of allowing the tree to do it for him. 



For bark slot grafting, on the other hand, very 

 strong twine is preferable as a rule excepting with 

 smaller limbs. Large limbs use up too much raffia 

 and the strands are too short. The tongue of bark 

 and the hard bark of these larger limbs will not allow 

 twine to cut in deeply enough to do much choking. 

 When the bark is very thick and hard no binding 

 material o-f any sort is required for the bark slot 

 graft because the tongue pried free with a lever will 

 go back with such pressure upon the introduced 

 scion that the latter is held snugly enough in place. 

 For all of the bark slot grafting which requires 

 binding material I have introduced the Spanish wind- 

 lass method of fixation of the scion. The Spanish 

 windlass which is used in surgery for controlling 

 hemorrhage seemed to me to be applicable for fasten- 



