Because the root system is striving hard to live, and because it is usually the stronger, it may force the top to accept certain of its characteristics.  Occasionally, it may assume some qualities of the original top.  Such cooperation is necessary if either is to survive.  First of all, the grafted scions must accept the vital quality of climatic hardiness, a powerful factor developed through ages spent in a certain climate.  To hasten the acclimatization of a tender variety, I cut scionwood from such unions early in the winter, storing it until spring.  When these scions are grafted on new root systems, I find that they are much more readily accepted than the first grafts were.  The following season, I allow the grafts of this later union to go through their first winter of exposure.  Early each spring I continue to cut scions from the most recent unions and graft them to new root systems, so hastening and setting the factor of hardiness through frequent asexual propagation.

Because my observations of the effects of scion on root and vice versa, have not extended over a sufficient period of time, I think it is possible that the changes I have seen may be only transient.  In any case, I do know that the phenomenon occurs, for I have seen many examples of it.

One instance in which the stock was apparently affecting the scions, occurred in the case of several varieties of black walnuts which had been grafted on wild butternut stock over a period of sixteen years.  The walnut top flourished but tended to outgrow the butternut, so that the caliber of the walnut was greater than that of the stock a few inches below the graft union.  I also noticed that, although the graft began to bear about as early as black walnuts do when they are grafted on their own species, the nuts did not mature at all during the first few years of bearing.  In 1938, after a favorable season, I found mature nuts on one variety, the Thomas.  These nuts varied in size more than they do when grafted on black walnut.  The most surprising thing about them, though, was that they did not have the characteristic black walnut flavor.  When properly dried and cured, they could have passed as an entirely different nut since they tasted like neither the black walnut, the butternut nor the Persian walnut.

The overgrowth of the Ohio black walnut, grafted on butternut, was even more apparent than that of the Thomas.  These nuts were, as I have said, immature the first few years they appeared and they, too, lacked the usual black walnut flavor.  In their case, however, the most striking change was in the shape and structure of their shells which were elongated like butternuts, with corrugations typical of those found on butternuts and nearly as deep and sharp. (See Illustration in Chapter 1, Page 5.)