368 BOTANICAL GAZETTE [DECEMBER 
reaction in the water drawn from the immediate surface of the roots 
and that taken from the peat near by. Hence it seems that all 
roots in a given area are subject to approximately the same soil 
reaction. Members of a given species, however, often showed con- 
siderable latitude in tolerance to soil reaction. The widest varia- 
tion found was that of Larix laricina. At Mineral Spring, Indiana, 
in a very old bog in which the peat has become casas 
decayed, the reaction about its rootlets was 10 alkaline; at 
side, Indiana, in a mature bog of fibrous peat, 300-1000 ae 
and in the comparatively young mat at Cedar Lake, 10 acid. No 
marked differences could be seen in subterranean systems of 
members of the same species growing in peat and mineral soils, or 
in various natural concentrations of H and OH ions. 
Other species which showed a narrower range of tolerance to 
reaction follow, with the extremes of reaction found in each. 
Water squeezed from Sphagnum had a specific acidity of 100 
to 1000; Aspidium Thelypteris, Scirpus validus, and Betula pumila 
all varied from ro alkaline to 10 acid; Sarracenia purpurea, Drosera 
rotundifolia, and Vaccinium macrocarpon varied from neutral to 
300 acid. The peat about the roots of Decodon verticillatus at the 
margin of the mat was approximately neutral, while the lake water 
into which this species was migrating was 30 alkaline. 
Experimentation 
In order to determine the factors involved in the horizontal 
placing of roots in bogs, the following experiment was carried 
out. Galvanized iron boxes, 10 cm. X15 cm. X55 cm., were made 
with the bottom and one side replaced by a diagonal pane of glass. 
This glass was covered on the outside by a piece of galvanized iron 
which could readily be removed, making it easy to make observa- 
tions of the roots, but at the same time keeping them protected 
from the light except during examination. In certain parts of this 
experiment, as indicated later, a fixed water table was maintained 
by keeping the boxes in pans of water of proper depth (figs. 7, 8)- 
All metal surfaces were given two coats of Acme asphalt varnish. 
Various germinating seeds were planted in these boxes. The most 
