A GIRDLED TREE. 119 
an upright position. The tree was living and look- 
ing perfectly healthy. It was supported on three 
short columns, with the portion of the body that had 
been girdled, dead and removed, thus leaving an 
opening in the trunk of the tree between the col- 
umns. 
Remembering the successful experiment with the 
cherry tree, I now set to work; prepared the scions 
by beveling them on the same side at both ends, and 
with a budding knife made an opening both on the 
upper and lower edges of the bark. In each tree 
that was completely girdled, I inserted three scions, 
and where there was a small connection of bark left, 
I inserted one scion. I fastened them by a band, 

— “SSS 
PSSST 
Fig. 16, 
