282 
Journal of Agricultural Research voi. xxm, no. * 
aphids, it is readily seen that their position on the leaf after being killed 
is relatively insecure. They must be fastened in some way, and this 
was accomplished by using small drops of collodion as a cement to 
hold the feet in position. Since this material is soluble in absolute 
alcohol it was replaced with cellulose acetate when the insects were 
transferred to this reagent. The cellulose acetate, being soluble in 
chloroform, could not be used during fixing in Camoy’s fluid following 
killing. Sections lOfj, and 12/j in thickness were obtained and were 
stained with safranin and gentian violet. 
It was demonstrated that with Aphis maidis the beak is usually placed 
on the cuticle covering a stomate guard cell at the point where the cuticle 
is thinnest (PI. 1, A) and the setae thrust into the latter by pressure. 
During this process a copious secretion is excreted at the end of the setae 
from the salivary glands. The secretion is colored brilliant red by the 
safranin. This secretion continues to pour from the tip of the setae as 
the latter passes into the deeper tissues, and forms the sheath described 
for other aphids by Biisgen (3) in his work on honey dew. Typically, the 
setae of A. maidis pass through the sub-stomatal cavity, then through 
the mesophyl cells either intercellularly or intracellulary, continuing 
between two cells of the starch sheath and finally into the phloem of the 
vascular bundle. (Pi. i, B.) If the ligmified secondary thickening of 
the xylem elements are encountered, they get no farther in this direction. 
(PI. 1, C, D.) In one section (PI. 2, C) the preference of this insect for 
phloem is well illustrated, since an exploratory puncture leading to the 
xylem elements of the bundle is indicated by the empty sheath, while 
the setae which were partly withdrawn and again pushed forward are 
shown leading around the xylem and turning in the direction of the 
phloem. During the whole process the copious secretion from the insert 
pours into the practically uninjured and rapidly growing tissues of the 
leaf. The fact that it leads to the phloem cells, rich in substances of 
known nutritive value for microorganisms, seems especially significant. 
In all three sections of the series illustrated in Plate 2 the unusual amount 
of salivary secretion is clearly shown. It appears to the writer that this 
secretion is unquestionably the medium by which the infective principle 
of mosaic is carried into the plant. A more perfect mechanism for 
inoculation could scarcely be devised. There is no apparent wound 
reaction on the part of the plant; none could be seen in the phloem at any 
rate, and certainly none is apparent from the outside. Reaction to the 
virus itself does not cause great structural changes, as is well known. 
This phenomenon meets the requirements for successful inoculation in 
this disease in a peculiarly fitting way. It has been found that a definite, 
measurable quantity of “virus” is necessary. This would eliminate 
from consideration as an explanation the carrying into the plant of the 
scanty amount of virulent material adhering to the minute mouth parts. 
It is necessary to introduce the virus into rapidly growing tissues in the 
interior of the plant. Even in the much more easily communicated 
tobacco mosaic, a w^ound such as the crushing of trichomes seems neces¬ 
sary. Thus the anal secretion or honey dew falling on the normally 
unwounded surface leaves much to be desired as an explanation of the 
infection process. Aside, then, from the perfect adaptation of the setae 
to the requirements for successful inoculation, the phenomenon herein 
illustrated is the most likely explanation of the transfer of mosaic. 
