62 
thirty growing plants, Mr. Marten found a single root aphis dead on the 
inner surface of the leaf. 
Transfer of Root Aphis to the Leaf—This spontaneous appear- 
ance and breeding of the winged root aphis on the leaves of corn was 
strongly suggestive of a permanent and regular migration, and of a 
possible transformation there into the ordinary Aphis maidis in late 
June or early July. This hypothesis was tested by a series of experi- 
ments made in 1890 and 1891, the most conclusive of which was begun 
June 22 of the latter year. Corn growing in a trench in the insectary 
was stocked with root lice from the field, several “adult females” being 
confined to the leaf in a bag of Swiss muslin.* From this beginning, 
the successive generations were bred, specimens of each generation being 
transferred to a new leaf, until September 16 (at which time the sexual 
generation appeared) without the slightest approximation in any gen- 
eration to the characters of the aérial Aphis maidis. Several similar 
experiments made with sorghum plants instead of corn were even less 
successful, transferred root lice, though living and breeding for a short 
time, all dying, nevertheless, in from ten to sixteen days without more 
definite result. From these experiments we may conclude with some 
confidence that the corn leaf aphis does not originate in a migration of 
wingless lice from the roots.f 
Another class of experiments, less precise but less artificial, points 
to a similar conclusion concerning the winged root louse also. In these 
experiments large rectangular frames covered with cheese cloth were 
used to enclose tightly entire hills of corn in the field, as well as spe- 
cially planted hills in a plot of ground adjoining my insectary. In the 
field experiments these screens were made without doors, the hills being 
plentifully stocked with corn root lice in the beginning, and with their 
attendant ants, before the screens were set. In these cases the corn was 
left to itself until a sufficient time had elapsed, when the cover was re- 
moved and the plant searched for the corn leaf louse. In the small gar- 
den plot a screen was built large enough to enclose four hills of corn at 
once. At first the mistake was made of using too small a frame, and 
our results in 1889 were vitiated by the discovery that winged Aphis 
maidis from the fields would sometimes settle on the cheese cloth cover 
where the corn leaves pressed against it, and there extrude their young, 
which were small enough to pass through the meshes of the cloth. There- 
after the enclosure was made so large that no part of the plant could 
touch it, even when full grown, access to the plants being given by 
means of a closely fitting door. The chief experiments of this series 
were the following: Six corn hills were covered in the field June 28, 
1889, selected because already stocked with the corn root aphis and the 
small brown ant Lasius niger alienus attending it. July 15 one hill was 
opened up and found unchanged, except that the ants had burrowed un- 
der ground and come to the surface far outside the space enclosed, thus 
* On account of a deficiency in the notes of this experiment, it is now impossi- 
ble to say whether any winged females were included in this lot. 
+ Our notes of breeding-cage work and observation in the field show, never- 
theless, an occasional appearance of the wingless root form above ground on corn, 
in one case on the stalk, a few inches above the surface, and in another upon the 
leaves at a much greater distance, where young were being produced. 
