82 
Journal of Agricultural Research 
Vol. XXI, No. I 
METHODS OF EXPERIMENT 
The methods of preparing and examining the material have been 
adapted to the needs of the work here reported. As the work of making 
several hundred determinations progressed, a routine procedure was 
followed which stabilized, if it did not remove, the sources of error. 
The chief object of this work has been to study comparative concentra¬ 
tion of sap as related to the growth rate. The results obtained, though 
not free from error, are comparable with each other and give a satisfac¬ 
tory picture of an important physiological relation. 
The plant material was collected in the field and packed at once in a 
tight container. For small samples, the container was a quart fruit jar 
with a screw top. For large samples, especially constructed steel cylin¬ 
ders, 6 inches in diameter by 14 inch tall, were used. The open end of 
the cylinder was provided with an accurately fitting metal cap and rubber 
gasket. The cap could be drawn down tightly on the rubber gasket by 
means of three bolts, and the cylinder was thus effectively sealed. 
The samples were always collected between 9 and 10 o’clock in the 
forenoon, in order to avoid the possibility of diurnal fluctuations in con¬ 
centration. As soon as filled, the containers were brought to the labora¬ 
tory and packed in an ice-salt mixture for the preliminary freezing which 
killed the protoplasm and increased the permeability of the material. 
The supply of ice and salt was renewed as occasion required. The con¬ 
tainers remained in this mixture for 18 to 24 hours, to insure complete 
freezing of the material. When the containers were removed they were 
washed to remove all adhering brine and wiped dry. The material was 
removed and ground in a small hand mill and then pressed in a strong 
screw press. Not more than 20 minutes were required to grind a sample 
and express the sap. The mill and press were heavily tinned, and there 
was no apparent corrosive action of the plant juices on any part of them 
after continued use. 
The expressed sap was received in small bottles which were closed with 
rubber stoppers and placed on ice, in case the sap was not to be imme- 
diately used. Samples which had stood for more than two or three hours 
were discarded. In spite of these precautions, there was some oxida¬ 
tion of the plant saps and probable change in concentration; but the 
error, if any, due to such changes was shared to approximately the same 
extent by all samples and is not believed to vitiate the results. 
The freezing point of the sap was determined in the usual freezing 
apparatus with the use of a Beckmann thermometer. At least two deter¬ 
minations were made upon each sample, and the average of closely agree¬ 
ing duplicates was taken as the freezing point. The osmotic pressures 
were calculated by the method given by Harris and Gortner (4). 
The rate of growth of the trees was obtained by measuring the length 
of certain selected shoots at intervals of seven days. Each shoot bore a 
