84 THE PERIODICAL CICADA. 
enter the soil. Unfortunately no examination was made for a number 
of years, but in 1892 the writer made excavations under a large num- 
ber of trees and found the larvae in their third stage, having passed 
their second molt three years from the egg. They were present in 
enormous numbers, so that a single spadeful of earth would often turn 
up a half-dozen or more larva?. An examination in tide in 1893 showed 
the larva? to be still in the third stage. ISTo examination was made 
thereafter until April, 1897, when the larva? were found in the fourth 
stage, some of the specimens having recently assumed this stage, but 
most of them probably a year back, judging from their size. The 
abundance of material in this experiment gives greater promise of 
successfully following the brood to the adult stage. 1 
While none of these broods have been followed through an entire 
cycle, the records are sufficiently complete to demonstrate conclusively 
enough the long underground life, if it required any proof in addition 
to the chronological records of appearances. A valuable outcome of the 
experiments has been that they have afforded the means of studying 
the different stages of growth represented in the underground life of 
the Cicada, which had never before been investigated. The following- 
history of the larval and pupal development is based for the most part 
on information and material secured in the experiments just outlined. 
HISTORY OF THE LARVAL AND PUPAL STAGES. 
A careful study of the material collected in the course of the experi- 
ments described in the last section demonstrates the interesting fact 
that this species, in spite of its very long period of growth, presents 
the same number of adolescent stages as is found in insects which go 
through their entire development within a single year or even of the 
more rapidly multiplying species, which have many annual generations. 
But six distinct stages are found, four of which belong to the larval 
condition and two to the pupal. In other words, the larval and pupal 
changes in the periodical Cicada are normal and are not increased by 
its long preparatory existence. 
It has been inferred hitherto, and notably by Professor Riley, that 
owing to the continual use of the claws in burrowing, this species found 
it necessary to shed its skin and undergo a molting once or twice 
a year, and instead of the normal number of changes or molts there 
were probably from twenty five to thirty. An examination of types of 
the different larval stages which Professor Eiley had provisionally 
separated demonstrates that the differences on which these supposed 
stages were based are either individual and exceptional or due to the 
difference of age within the same stage, and that as far as structure 
and size of the hard parts of the larva and pupa are concerned the 
normal number of stages only are represented in this species. 
XXII in 1885 and Brood VIII in 1889 are given iu Appendix B. 
