There is no evidence to suggest that tree roots have the faculty of searching out food deposits in the same way as they will grow toward a continuing source of moisture. Even if they had such a faculty, Mr. Stout argues, it would not be good for them to exercise it, because they would then concentrate root growth at the points of feeding, which are only temporary. This is exactly what takes place when you do happen to hit a root with a food deposit. The dense feeder ganglia that form, at the expense of root growth elsewhere, give the tree an abnormal root pattern, vulnerable to drought.

Moreover, Mr. Stout claims, sod does not restrict the air and moisture of tree roots. On the contrary, sod breathes better than baked bare ground and it slows the evaporation of moisture from beneath it. As for competition for nourishment between surface growths and tree roots, Mr. Stout believes the latter can more than hold their own. This has been shown by experiments in dense woodland where the absence of grass and weeds from the forest floor might have been supposed to be caused by shading. Ten-foot squares between groups of trees were trenched around to a depth well below the tree roots, which were all cut off as encountered. The trees' heads were left untouched, their shade unbroken. Within a year, each square filled up with surface vegetation, flourishing in the forest gloom wherever it had no tree roots to contend with.

Experts to the contrary notwithstanding, Professor Stout believes that broadcasting dry fertilizer to trees is surer, sounder, more economical practice than punching it down. He concedes that where tree food is broadcasted, spike-rolling the sod might be wise to speed the fertilizer's movement downward, and that when the needle or punch-bar methods are used, their efficacy can be improved by doubling the points in the usual pattern—i.e., putting them only one foot apart instead of two—and halving their depth, to nine to twelve inches.

Mr. Stout believes that most tree feeding is more arbitrary and haphazard than it might be. He recommends that before any feeding is done in their grounds, home owners dig some test holes to find out just how their soils lie and their tree roots run, then serve them accordingly.