6 TRANSPIRATION IN A DESERT PERENNIAL. 
The work is being carried on at the Desert Laboratory at Tucson, Arizona, 
and is made possible through the kindness of Dr. D. T. MacDougal, who has 
placed the facilities of the laboratory at the writer’s disposal. 
Parkinsonia microphylla is a leguminous tree with minute deciduous 
leaves. Plate IC shows the general appearance of the adult tree, while No. 
4 of plate IB gives an idea of a seedling which has grown under natural con- 
ditions for a year. No. 5 of the same figure is a seedling, about eight years 
old, which was killed in transplanting. All adult trees in natural conditions 
carry from one to eight dead limbs and have from 5 to 10 per cent of the 
medium-sized branches dead; about 30 per cent of the twigs are dead-for a 
distance of 2 to 5 em. from the ends, and after the first dry season succeeding 
growth practically all twigs are dead for a distance of 0.2 to 0.5 cm. from the 
ends. (See also Shreve, F., loc. cit.) The above is merely an estimate from 
general observation and is not based on statistics. 
The leaves appear in the late winter after the rains, are usually shed 
during the arid fore-summer, reappear within a few days after the summer 
rains begin, and persist for varying lengths of time, according to the amount 
of rain falling during the early autumn. During the autumn of 1910 the 
leaves had all fallen from the trees on Tumamoc Hill before the end of 
September, but in 1912 many leaves persisted until killed by frost in January 
of 1913. The rain records taken at the Desert Laboratory show a much 
drier autumn for 1910 than for 1912. During both these years a tree near 
the laboratory, where it probably received water from an artificial source, 
retained its leaves until January. 
Observation of the trees on Tumamoc Hill showed the following general 
response to progressive desiccation of air and soil: First, the leaflets drop, 
leaving the rachises still attached; then if no rain comes the rachises are 
shed, and after a month or so the ends of the twigs die. All leaves do not 
necessarily fall at once, for sometimes as many as half of the leaflets may 
remain for a month after the others have fallen. These observations were 
made throughout 1910 and the summer, autumn, and winter of 1912. At 
the same time, observations were made on two plants which probably 
received an abnormal supply of water. The first was a tree situated near 
a south door of the laboratory, where its roots probably received a reserve 
supply of water from beneath thelaboratory porch. This tree hashad no new 
dead twigs or limbs for the last three years at least, and during those years 
has retained its leaves from four to six weeks after the other trees were 
bare. The second was a plant about 3 feet high; whose main trunk had 
been cut off in the spring of 1909. Its root was thus unusually large in 
proportion to its top, and it was otherwise under unnatural conditions, since 
it was shaded by the laboratory during a large part of the day and the soil 
about its roots was covered with a mulch of crushed stone. This plant 
has lost no twigs or branches since the cutting took place and has leaves 
throughout the year, except in mid-winter. 
During the above period many trees were observed both on Tumamoe 
Hill and on the foot-hills about Tucson for evidence of the place where dying 
