THE ANALYSIS OF VIRGIN SOILS. 351 



so far as these could be observed; the continued attacks of 

 mildew and plant lice preventing full maturity being attained. 



The restricted volume of soil occupied by the roots in the undiluted 

 adobe soil, together with the very abundant development of root-hairs, 

 is very striking. A marked change in these respects is manifest in the 

 first dilution, and increasingly so as dilution increases ; the paucity of 

 root-hairs is very marked in the last (greatest) dilution, in which, as 

 the photograph of the plants shows, the development was decidedly 

 behind that in the pot containing dilution i : 4. The latter in fact 

 showed the best development not only in this case, but in two other 

 series of tests conducted at the same and subsequent times ; and strangely 

 enough, also in the pulverulent, " sandy loam " soil of the southern 

 California substation tract. In the latter series, which for lack of space 

 cannot be figured here, the main difference was that in the undiluted 

 soil the roots filled the entire soil mass, instead of remaining near the 

 surface, as in the pure adobe. It is possible that the latter was too wet 

 when given the full half of its water-capacity, although, as the figures 

 show, the water was slowly introduced from below by means of glass 

 tubes, ending within a shield to prevent puddling. 



Limitation of Root Action. These results, representing five 

 soils of different percentage-composition and physical char- 

 acter, but identical chemical composition and ratios between 

 the several ingredients, and similarly acted upon by the atmos- 

 pheric agencies in the past, illustrate strikingly the impossi- 

 bility of judging correctly of a soil's productiveness from per- 

 centages of chemical ingredients alone. It is clear that the 

 physical characters of the land as well as its depth, must be 

 essentially taken into account. But there is obviously a cer- 

 tain limit beyond which greater perviousness and root-penetra- 

 tion cannot make up for deficiency in the absolute amounts of 

 plant-food within possible reach of the plant ; for in the case of 

 excessive dilution these are rendered partially inaccessible 

 within the time-limits of a season's growth. 



It is hardly necessary to say that these experiments require 

 repetition with the aid of the experience acquired in these first 

 trials, not only in the laboratory but also in the field. It will 

 be especially interesting to compare with the results obtained 

 in these strongly calcareous soils, the effects of dilution in such 



