THE UNDERGROUND LIFE OF THE CICADA. 83 
The earliest systematic attempts to follow the development of the 
Cicada were made in the field in Missouri by Professor Riley, and sub 
sequently continued under the hitters direction by Mr. J. G-. Barlow, 
an agent of the Division. They consisted in making diggings from 
year to year under trees which were known to have been thickly 
stocked with eggs. The first records approaching in any way to com- 
pleteness were obtained with the 13-year Brood XVIII, beginning with 
its appearance in 1881. Observations on this brood were continued by 
Mr. Barlow at Cadet. Mo., with a fair degree of regularity until duly. 
1891, when they unfortunately terminated. 
During the ten years over which these observations extended the 
insect had developed through all four larval stages and was ready to 
enter the first pupal stage. The first molt occurred after a period of 
from one year to eighteen months, the second molt after an additional 
period of two years, the third molt after an additional period of three 
years, and the fourth molt after an additional period of three or four 
years, leaving in this 13-year brood three or four years more for the 
pupal stages. 
A much more careful series of experiments were instituted in connec- 
tion with the 17-year Brood XXII, beginning with its last appearance 
in 1885. At the time that the eggs of the 13-year Brood VII were 
being distributed to various points in the Xorth in order to determine 
the effect of the temperature and climate (see p. 16). quantities of 
egg-laden twigs of the 17-year brood noted, collected in Virginia, were 
distributed under certain linden and oak trees on the grounds of 
the Department of Agriculture at Washington, D. C. Larva* came 
from these twigs in some numbers and went into the soil under the 
trees, but not in such abundance as could have been wished for the 
successful outcome of the experiment. This brood was followed in its 
underground life from 1885 to 1S96, at which time the specimens had 
become so rare that extensive digging resulted in the discovery of very 
few individuals, and further search was abandoned. With this brood 
the first molt occurred after one year, the second molt two years later. 
the third molt three or four years later, and the fourth molt after an 
additional three or four years, thus occupying upward of ten years 
with the four larval changes and bringing the insect into tin- last 
larval stage with some six or seven years lor the subsequent larval 
and pupal life. 
A much more promising experiment, because of more abundant mate- 
rial, was instituted on the Department grounds in L8S9with Brood V 1 1 1 
of the 17-year race, which will next appear in L906. The egg-infested 
twigs of this brood, obtained in North Carolina, Long Island, Kentucky, 
and Ohio, were distributed in enormous numbers under oak trees in the 
grounds of the Department of Agriculture and also under sycamore and 
willow trees. The eggs in most instances were hatching when received 
and were placed under trees in the very best condition for the larva to 
