75 2 
Journal of Agricultural Research voi. xxvn, No. i© 
epidermis. The shading in this figure is simply to indicate cytoplasmic 
contents. The open spaces shown in this section are clearly due to 
shrinkage in fixation. If the aecidium were open and the degenerated 
spores cleared away, the spore just below X would be in a position to 
be shot out with considerable force. 
Every spore is at first interlocked with its neighbors through the 
plugs indenting their walls and because the spore walls are rough or 
finely warted. At first the system only gradually moves out of the state 
of equilibrium. New adjustments are continually taking place until one 
spore is forced into the critical position necessary before it can be set 
free. 
The correctness of the above explanation of the mechanism by means 
of which the aecidiospores of G. myricatum are ejected from the sori 
with such velocity can be demonstrated by the use of tennis balls and 
marbles. Confine the balls under pressure from all sides, except from 
Fig. 3. —G . myricatum. A: Section of three adjacent spores showing pore plugs as they lie in the spaces 
which will become the germ pores. Note that the wall of the adjacent spore is indented in each case, 
and that the plug is covered on the outer and inner sides by very thin membranes. B: Section of a young 
spore parallel to the surface of the leaf. Two pore plugs and the edge of a third show in the section; 
one plug in an adjacent spore. C: Section of young spore with two large plugs. The indentations are 
due to side pressures and not to shrinkage. D, E show shrinkage of spore contents away 
from the wall. Material preserved in alcohol several years before imbedding. F: Surface view of pore 
plugs, showing that these bodies do not have a homogeneous structure. G, H: Position of pore plugs 
in dry mounts from herbarium specimens. I: Germ spore as it appears after dislodging the pore plugs 
above, with marbles thrust in between them where they touch. By 
gradually increasing the pressure from the sides the balls become more 
and more compressed but remain in position. Now if pressure is applied 
to one of the balls from below it gradually moves up until the resultant 
of forces is such as to overcome the friction between the balls and be¬ 
tween them and the marbles, and the ball will be liberated and suddenly 
shot upward. If the convexed bottom of a well chosen tin pan is pressed 
inward it will remain concave until set in motion by a slight pressure, 
or by warming the pan. The bottom moves out slowly at first until the 
critical position is reached, when it will suddenly complete the return 
with much noise. By placing two such pans together, bottom to bot¬ 
tom, one will be thrown upward violently, especially if the two bottoms 
in contact suddenly become convexed at the same time. The conidium 
of Empusa grylli is thrown off violently on the same principle. The two 
elastic membranes between the spore and its support suddenly become 
convexed. In this case the little columella may in reality function the 
same as the pore plug does in the Myrica rust, namely, as a protuberance 
against which the elastic membrane of the spore can react all the more 
