F£b. s, 1917 
Sorosporella uvella 
193 
employed, in both of which growth is exasperatingly slow. The outer 
cells of the spore masses, however, when plated in such media, give rise* 
to a mycelium a portion of which is represented on Plate 66, AT. No 
spores have as yet developed in such cultures, although they are now 
several weeks old. 
No attempts have been made to study the development of the second¬ 
ary spores or conidia in culture media, but their germination in water 
has been observed (PI. 66, M, O). 
From the studies thus far conducted it seems probable that the large, 
spherical, thick-walled cells function as resting spores, which in this 
fungus are closely attached to one another in masses. Such a condi¬ 
tion is, as noted by Sorokin, very similar to that found in certain smuts, 
such as Urocystis and Tubercinia. 
The early stages of germination of the spore balls of such smuts and 
of the spore masses of Sorosporella are also similar. The later devel¬ 
opment of the germinating spore masses of Sorosporella, however, more 
closely resembles that of. the verticillate Hyphomycetes (PI. 66, 
H , K , L). The branched sporophores ending in bottle-shaped tips and 
the abjunction of the typical conidia which adhere to one another in 
groups as they are cut off suggest this relationship. 
While the spore masses, as mentioned above, bear a certain super¬ 
ficial resemblance to the spore balls of Ustilaginales, they nevertheless 
more closely resemble that class of propagative bodies known as bulbils, 
which are known to occur in the life histories of certain Ascomycetes 
and Basidiomycetes. 
Hot son (3) and others have shown that bulbils are often directly asso¬ 
ciated with hyphomycetous growths. In some cases, as in Papulospora 
aspergilliformis as figured by Hotson, the bulbils upon germination give 
rise to typical hyphomycetous forms. In other cases the same mycelium 
produces bulbils and conidia at different places. 
What the early stages in the development of Sorosporella are in the 
insect is not known. The presence of papillae on the spores suggests 
that vegetative development is carried on by a budding process, but, 
as mentioned above, this point has yet to be proved. It is known, 
however, that finally the entire larval body is filled with the above- 
described large spore aggregations, which will undoubtedly be incor¬ 
porated in the soil when the larvae disintegrate. In a moist chamber 
such spore aggregations give rise to the secondary spores mentioned 
above, and it may be assumed that the same process takes place in the 
soil. In such a position the secondary spores will readily be able to 
infect fresh cutworms, which habitually burrow in the earth. 
Thus far no experiments have been conducted to test the parasitism of 
Sorosporella uvella , but it is hoped that these experiments, as well as a 
more complete study of the life history of the organism, will be accom- * 
plished soon. 
