114 
time. In old individuals the stroma frequently spreads over the sur¬ 
face of the branch, if only to a limited degree. • 
The structure of a fully-developed Cytispora is made clear by longi¬ 
tudinal and tranverse sections. The bark has been destroyed and 
broken through by a cone-shaped stroma which also sends hyphse into 
the surrounding cells of bark parenchyma and breaks them down. This 
stroma is composed of a dense mass of irregularly-interwoven, coarse, 
brownish hyphae. Within this are a greater or less number of cav¬ 
ities which are not surrounded in the young stages by specially-dif¬ 
ferentiated cell-layers, but in the older stages are inclosed by a brown 
or black wall; they are round at first, but finally irregular in form, 
possessing numerous folds and invariably lined with a hymeuium. 
The development of the stroma begius from below and grows strongly 
toward the surface and a little toward the sides. The hyphae at the 
apex which come in contact with the air swell up. * * .* Eveu be¬ 
fore the stroma has broken through the bark we can see in it dense 
knots of hyphae distinguished by their transparency, which stead¬ 
ily increases after they attain their full size, which is very indefinite. 
The growth ceases in the interior while it continues at the periphery. 
Consequently cavities arise into which spore-forming hyphae immedi¬ 
ately grow from all sides. The pycnidia retain the power of growth for 
some time, and, therefore, frequently become adjacent or grow against 
obstacles so that their circumference is often very irregularly wrinkled. 
By the growth of the stroma they come to the surface, where their wall 
disintegrates and the spores escape. The pore, a special arrangement 
for the perforation of the wall, which is often present, is lacking in this 
case. The wall of the pycnidium arises from the knot of hyphae and 
often can scarcely be distinguished from the stroma. In younger speci¬ 
mens it can usually be seen simply as a brown layer surrounding the 
cavity. In old specimens, however, and on pycnidia that have reached 
the surface, it often attains a considerable thickness and is dark 
brown or black. This is particularly the case with pycnidia, which 
together with the stroma extend over the surface of the branch. These 
protrude for a long distance out of the stroma; but the beginnings of 
the pycnidia occur only in the basal portion of the stroma. 
The basidia are simple and filiform and cut off spores in immense 
numbers. The spores are one celled, colorless, straight, and cylindrical. 
Their size is very variable; generally they are 3-4/* long and 1-2/* in 
diameter, but they have been observed 12/* long and 5/* in diameter. 
It may be added here that the Cytispora under consideration is ex¬ 
ceedingly variable. The size and shape of the stroma aud pycnidia 
and the number of the latter are very inconstant. In view of this case 
too much care can not be taken in systematic investigations of this 
kind of fungi. The greatest weight should be laid on the form of the 
basidia and spores, since these show relatively the greatest constancy. 
When the stroma of Cytispora has reached a certain size the forma- 
