D. Keilin 
87 
2-5/* in width (Fig. II, 13, 14). These elongated forms correspond to asci and 
are at first transparent having finely granulated protoplasm. As develop¬ 
ment proceeds, the protoplasm near one end of the ascus begins to show a 
small triangular refractive body which gradually elongates until it occupies 
almost the whole length of ascus. The refractive body (Fig. II, sp. 14 , 15 ) 
gradually becomes clearly defined and finally develops into a long needle- 
shaped unicellular spore with one end sharply pointed, the other end truncated. 
The space between the spore and walls of the ascus is filled with transparent 
fluid, and the end of the ascus facing the pointed portion of the spore is 
thickened. In some cases, after the death of the larva, when the asci escape 
into the fluid surrounding the insect, the ascus walls become deformed so that 
the thickened end is bent to one side of the spore (Fig. II, 16). The asci and 
spores vary in size, the asci measure 30//, to 40//, and the spores 24//, to 35//, 
in length; the truncated end of the spore usually measures T8//, across. 
I have failed to observe the liberation of the spores from the asci and the 
germination of spores. In only one larva was the alimentary canal found to 
contain several free spores (Fig. II, 17 ), but unfortunately the larva was 
damaged during examination whereby further observation was precluded. 
As the spores of M. unicuspidata have only one end pointed, whilst 
M. bicuspidata (Metsch.) has spores with both ends pointed, it is probable 
that the first named species has a poorer chance of perforating the alimentary 
canal of its host and this may account for the smaller proportion of infected 
hosts as compared to what has been observed with M. bicuspidata and Daphnia. 
It is worthy of note that other Dipterous larvae (those of Rhyphus fene- 
stralis Scop., Mycetobia pallipes Meig., Aulacogaster rufitarsis Mcq., Phaonia 
cincta Zett. and a few Eristalines, Drosophilids and Dolichopodids) living under 
the same conditions as Dasyhelea obscura, were not found to be infected with 
Monosporella unicuspidata. 
The genus Monosporella , hitherto known as Monospora Metschnikoff, is 
often placed by systematists near to the genus Nematospora Peglion (Fig. III). 
The latter, which contains but one species N. coryli, discovered and named 
by Peglion (1901), is a parasite of the hazel-nut in Italy. It is a budding, 
yeast-like fungus with elongated cells; the ascus is sausage-shaped 65/X-70//, 
long by 6//,-8/x broad, and contains 8 spores in two longitudinally disposed 
bundles of 4, separated by an interval midway along the length of the ascus. 
These spores are elongate spindle-shaped, with a long flagellum at one end 
and measure 38//,-40/z without the flagellum which is about 35//, to 40//, long. 
Before germination, the spore loses its flagellum and broadens. Peglion 
succeeded in cultivating N. coryli , finding that it grew well on sterilized sugar- 
beet or meat-broth gelatin and badly in fluid media, where it formed only a 
mycelium. 
The systematic position of the genera Monosporella and Nematospora is 
not yet clearly defined. Metschnikoff and Zopf placed Monosporella among 
