WATER RELATIONS OF DESERT PLANTS. 85 
observations showed no important deviation from this rate. It is to be 
noted that the conditions differed from those of the desert plant in the 
lower temperature and the much lower intensity of the light. 
Experiment 4.—Eucalyptus globulus was used, being the shoot of a 
young plant grown from seed in the greenhouse and having a surface of 
352 square inches. The test was made at the same time of day (10 to 11 
a.m.), and under approximately the same conditions as experiment 3. 
Readings of 500 milligrams in 7.5 minutes, 9.5 minutes, 10.5 minutes, 10 
minutes, and ro minutes were made with an average rate of 0.79 to 1.11 
milligrams per second. 
The data furnished by the above tests afford a fair means of com- 
parison of the relations of moisture-loving and desert plants to water if 
due allowance is made for dissimilar conditions. It is to be seen that a 
given area of surface of Mentzelia at similar temperatures and in a light 
vastly more intense and in a drier atmosphere transpires water at a rate 
slightly less than the tomato and at a rate about a third to a half that of 
Eucalyptus. The exposure of the two last-named species to similar tem- 
peratures, insolation, and dryness of the air would doubtless show that 
the moisture-loving plants would take up and lose water at a much 
greater rate. 
The shrubby Artemzsza was found to use water at a rate per area about 
one-fourth that of the tomato under the dissimilar conditions offered. 
An increase of the temperature, insolation, and dryness of the air affect- 
ing the tomato would doubtless increase the ratio many times. 
Still another interesting suggestion arises from these results. Ment- 
zelta is an annual that carries on its growth only during the season of 
maximum humidity, while Artemisia is an example of the perennial 
shrubby plant which makes no reduction of its surfaces during the dry 
seasons. The latter therefore must be better protected against the danger 
of drought and actually uses only about half the amount of water per area 
of surface that is needed by Mentzelza, and it sends its roots to enormous 
depths to insure a constant supply to keep up a steady but slow rate of 
transpiration. 
The fleshy cacti are examples of wholly different types of shoots, in 
which the leaf-organs are reduced toa minimum. Dr. W. A. Cannon has 
found that a young sahuaro (Cereus giganteus) about 4 inches high and of 
cylindrical form transpired 33 milligrams of water per hour in an atmos- 
phere with a relative humidity of 32.5 to 35.5, while a bisnaga (Echino- 
cactus wislizent) of the same height, but globular form, and hence great 
surface, gave off but 9.6 milligrams per hour in an atmosphere with a 
relative humidity of 35 to 45 percent. Itis quite possible that the spe- 
cific rates correspond to the different types of root-systems with which 
these two forms are equipped. These fleshy forms, however, have such 
storage capacity that the rate of water loss does not have a direct con- 
