A PEAE-TREE PUZZLE. 
57 
the wall. The more upright of the two^ on reaching the top, 
bends over and hangs down some distance on the other side. 
At the end of March, or early in April, 1889, the branch 
had been cut through, and the stump grafted with another 
variety of pear ; but the grafts did not take.” The severed 
branch should have been removed at the time, but was for- 
gotten and left in situ ; and during the thirteen years which 
have elapsed it has been in vigorous health, blossoming 
freely each year, and bearing proportionately more fruit than 
any other part of the tree. 
A year later (early in 1890), it was seen to be showing plenty 
of flower-buds ; and then, for the first time it was examined; 
but no adventitious roots to the wall were found. Not until 
the following year was the actual cause of the continued 
vitality detected by the owner’s personal examination of the 
branch where it crossed the top of the wall. 
A twig about as thick as a cedar-pencil (J-inch) was then 
found coming from the top of the main stem of the tree, and 
this lay along the upper part of the wall, passing under the 
severed branch, which was about If inches in diameter. The 
distal portion of the twig was about the thickness of a straw, 
hardly half the size of the proximal part. The large branch 
and this twig had growm together, and the small “ foster- 
mother ” had successfully provided for all the needs of its 
big protege from end to end. The union of the two must, I 
think, have taken place one, or possibly two, years before 
the grafting of the stump was attempted ; and it was a very 
curious coincidence that the only large branch on the tree 
capable of surviving by reason of its vital union with another 
should have been left in situ quite unintentionally after 
severance from the trunk. 
In 1895 the “ feeder ” twig was about f inch thick on the 
near side, and f inch on the far side of the large bough, 
which had here increased to about 2 inches in diameter. 
