January, 1922 GROWTH OF TREE SEEDLINGS AND WHEAT ' 67 



rock itself was a granite, its decomposed particles underlying the humus and 

 forming the only soil in the crevices. Almost none of these particles were 

 included in the humus used. The forest which created the humus was of 

 mixed hardwoods characteristic of moist rocky ravines in southern New York, 

 Connecticut, and Pennsylvania (9). It contained white oak, Quercus alba, 

 chestnut oak, Q. prinus (on the rockier places), red oak, Q. rubra, much 

 sugar maple, Acer saccharum, black birch, Betula lenta, a little beech, Fagus 

 Americana, a little shagbark hickory, Hicoria ovata, some white ash, Fraxinus 

 Americana, and a scattering of dead chestnut, Castanea dentata, which had 

 probably contributed its quota to the leaf mold. The reproduction was fairly 

 abundant of chestnut oak, red oak, sugar maple, and white ash. Among the 

 shrubs of the undergrowth was some laurel, Kalmia latifolia, viburnum, and 

 a little dogwood, Cornus florida. Higher up the slope and about 50 to 75 

 yards away was a group of hemlock. 



The wilting coefficient of this humus, on the basis of air-dry weight, was 

 43.5 percent. The acidity, determined by titration of an aqueous extract 

 with .05 normal sodium hydroxide in the manner described by Coville (i) 

 in his blueberry study, was .00114 normal at the start of the experiment, and 

 was reduced by exposure from June 8 to September 12 to .00075 normal. 

 By the Truog test (10), applied before exposure, it was between "medium" 

 and " strong " acidity, nearly " strong " ; after exposure this test revealed 

 practically no change. Wherry's colorimetric method of determining the 

 hydrogen and hydroxyl-ion concentration (11) had not been made available 

 at the time these tests were carried out. 



The intermediate soil was a mixture of the sand and humus above de- 

 scribed in the proportion of 80 percent sand and 20 percent humus by volume. 

 Owing to the lightness of the humus it made only 4 percent by weight and 

 did not give the sand an appearance of a high humus content; in fact, it is 

 probably somewhat lower than the surface layers under a typical forest of 

 pitch pine. The addition of this amount of humus raised the wilting coeffi- 

 cient from .85 percent to 2.4 percent. The acidity became .001 normal before 

 exposure and was of " medium " acidity by the Truog test. 



In order to eliminate the effects of greater moisture-holding capacity, and 

 to find the influence of humus on the nutrient properties of the soil, all soils 

 were kept continually moist. 



In the alkalinity tests a set of three soils, similar to the three above 

 described except for the addition of lime, was used. The sand received 

 2 percent of air-slaked burnt lime by volume and the humus received 4 per- 

 cent. The sand and humus received 2 percent of lime for the sand and 

 4 percent for the humus. It has been commonly supposed that burnt lime in 

 the soil quickly absorbs carbon dioxide and assumes the neutral carbonate 

 form. In a study of the comparative effects of burnt lime (CaO) and 

 calcium carbonate (CaCOs) on the soil reaction Hoagland and Christie (6) 



