370 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1926 
EXPERIMENTS WITH RHODODENDRON 
A soil mixture was made up as follows: rotted turf loam, 1 part, 
by bulk; well-rotted cow manure, 1 part; and sand, 1 part. A por- 
tion of this soil was treated with a 0.6 per cent (M/20) solution of 
magnesium sulphate in sufficient quantity to supersaturate the soil, 
and this was afterward leached repeatedly with distilled water until 
the excess of soluble salts was removed. Another portion was 
similarly treated with a solution (M/60) of crude aluminum sul- 
phate, and a third portion was leached with distilled water with- 
out other treatment. 
On March 30, 1921, in each of these three portions of soil were 
set six seedlings of Rhododendron catawbiense in six 2-inch glass 
pots, each provided with a drainage hole at the bottom. The pots 
were plunged in sand in a greenhouse maintained at a temperature 
of 55° to 70° F., and were given uniform treatment as to light and 
watering. Within a month the cultures were showing pronounced 
differences in behavior, and after three months, on June 29, when the 
photographs shown in Plate 1 were taken, these differences were 
conspicuous. The plants in the untreated garden soil had made no 
growth. In the soil treated with magnesium sulphate the plants 
had grown a little, the increase in diameter of the rosettes of leaves 
being about 30 per cent. In the soil treated with aluminum sulphate 
the stimulation of growth had been very great, the increase in the 
diameter of the rosettes being about 250 per cent. In fact, these 
plants were almost as large as plants grown in an ideal rhododendron 
soil consisting of peat and sand. On May 27, 1921, when the bene- 
ficial effect of the treatment with aluminum sulphate had become 
clearly apparent, two of the pots were given an additional application 
of 1 gram of aluminum sulphate in 5 grams of water. These two 
plants were kept for observation in comparison with the six plants 
treated with magnesium sulphate and the six untreated plants. 
The condition of the three lots on July 28, 1922, is shown in Plate 2. 
All the untreated plants were dead. Of the magnesium-sulphate 
plants, two were alive, but they had made little growth. The alumi- 
num-sulphate plants, however, had grown many times their original 
size. 
When the beneficial effect of aluminum sulphate began to show 
itself from the first experiment, it was determined to try further 
experiments, in which ordinary porous earthenware 2-inch pots were 
to be used in place of glass pots, and aluminum sulphate was to be 
applied to the plants after they had been potted. The soil was made 
up as before, equal parts of loam, manure, andsand. The mixture was 
neutral or slightly alkaline in reaction and proved so injurious to the 
healthy rhododendron seedlings set out in it that they all stagnated 
