28 DROUGHT EESISTANCE OF OLIVE IN SOUTHWESTERN STATES. 



inches, respectively, beyond which point they were not excavated 

 and were still one-eighth of an inch in diameter. 



The block of the Manzanillo variety at the north end of the 20 

 acres showed a different behavior from that of the main body of the 

 grove. The wood growth averaged much ranker, but the branches 

 were coarser and the tops more open to the sun. Far less adapta- 

 bility to conditions is evident. Dead trees, trees with dead tops but 

 with live sprouts from below, dead branches, and sun-scalded spots 

 on exposed places are to be seen on every side. In this block the tree 

 selected for study was 8 feet high, with 8 feet spread of top and a 

 trunk 5 inches in diameter. A still stronger root system was found 

 here, there being ten roots of from three-fourths of an inch to 1^ 

 inches in diameter springing from the burl at the following depths 

 below the surface: Two at 2 inches, one at 3 inches, one at 12 inches, 

 three at 14 inches, two at 16 inches, and one at 18 inches. It being 

 evidently impossible to keep track of all of these at one excavation 

 the surface roots were excavated by themselves. These comprised 

 three strong roots, issuing at 2 and 3 inches below the surface, and a 

 circle of short fine roots, which the writer called the shade roots from 

 the fact that they occupy the space immediately beneath the spread 

 to the top. 



Figure 7 shows these superficial roots represented in solid lines, 

 while the deeper roots are shown by dotted lines, but it should be 

 noticed that a number of the shallow roots figured came up to join 

 this class from roots of deep origin. These shallow roots taken 

 together may be regarded as a very important part of the equipment 

 of the tree. Their rootlets reach to quite near the surface, so that 

 they are prepared to gather moisture from a small rainfall. The shade 

 roots appear to collect also the water which is arrested by the top 

 and runs oft' to the ground from the trunk. It must be remembered 

 that the medium in which these trees grow is so nearly pure sand as 

 scarcely to be called a soil. There is a good deal of fine material in it, 

 but it does not bake or pack, and cracks never occur. It is very doubt- 

 ful whether any method of soil culture, dust mulch, or subsurface 

 packing would be of value here. All that such treatment is expected 

 to accomplish is already insured by the nature of this soil, and, 

 furthermore, any cultivation would only destroy these delicate root- 

 lets so nicely adapted to taking advantage of the lightest rainfall. 

 Referring to the deeper roots in figure 8, we notice first the large area 

 occupied by them as compared with the spread of the top. The roots 

 of this tree over one-eighth of an inch in diameter have a total length, 

 including the upper layer, of approximately 376 feet. The area of 

 the root spread, as compared with the spread of the top, would be a 

 little more than 7 to 1. 



192 



