68 
of frequent cold rains which had greatly retarded the work of 
the ants and the development of the corn root aphis. A few of 
the hills already burrowed by the ants were found with the grasa 
louse only (Schizoneura panicola ), and one with a single mealy 
bug ( Uactylopius ), while twenty contained no insect whatever 
ex< ept the little brown ant, which was in full possession. May 
28 two long rows of a field likewise on sod were searched care¬ 
fully. Many lulls were burrowed by ants, but only two contained 
the corn root louse, and in these were winged adults—one speci¬ 
men in one hill and two in the other, the latter with a single 
young one but just born. Our records are full of notes of ob- 
sei vations made during the latter part of May in several years 
according to which ants were just beginning to mine corn hilla 
on sod, no lice of any kind yet occurring in their burrows. 
Occasionally, in fact, an ant has been seen to seize a winged 
louse in the field, commonly by the wings themselves, and^to 
cai ry it down out of sight. In one instance such a winged adult 
v as found in a corn hill with its wings gnawed away near the 
body as if to prevent its escape from the earth; an operation 
to which ants are well known to subject their winged captives 
of the migrant generation. Confirmation of this practice is fur¬ 
nished by an observation made September 22, 1884. A winged 
coin loot aphis placed near an ant in the field was seized by the 
latter and carried about four feet to a newly opened formicarv, 
and was then laid down while the ant went inside. An ant 
coming out of the nest afterwards carried the aphis into the 
earth. 
I ne ed hai dly sav that the relations above described between 
the corn root aphis and these ants continue without cessation 
throughout the year, the succeeding generations being quite as 
useful to the ants as those whose history I have thought it 
worth while to follow in detail. In order to determine more pre¬ 
cisely the value of the services performed by their guardians, I 
arranged in several years a series of experiments designed to 
show to what extent the plant lice could help themselves if left 
unattended. Owing to the waywardness of the ants, which in 
most cases refused to content themselves in confinement, but 
?QQn° f theSe ex P erime ! lts came to a successful issue. April 13, 
188J, corn root aphis eggs were placed in the earth among 
smart weed roots to test the ability of the young lice hatching 
to find the roots for themselves. A check experiment was started 
at the same time with eggs placed in artificial cavities beside 
smartweed roots. April 25 no insects could be fonnd on the 
plants of the first experiment, while the cavities made in the 
second experiment contained young lice upon the roots in fine 
condition. 
