80 Transactions of the Royal Society of South Africa. 
The present writer repeated Hesse's experiments many times, but never 
once was a growth of M. racemosus obtained. From Bernstein's account of 
the experiments conducted by himself and Hesse it would seem that the 
initial cultures were obtained by placing the infected flies on the slides 
bearing the culture medium. He definitely states (p. 28) that "it was 
impossible to sterilize the flies from which the cultures were obtained." 
Obviously a grave source of error was introduced in this way. In order to 
overcome this the writer sterilised the flies used in the experiments by 
soaking them for fifteen minutes in a 1 per cent, solution of corrosive 
sublimate, the specimens being afterwards well washed in sterile water. 
The soaking was not long enough to prevent conidia formation provided 
the specimens were so treated immediately after death, yet it served to kill 
any foreign spores or bacteria adhering to the exterior of the flies. 
Besides egg-yolk colostrum was also used as a culture medium, this 
being also a highly concentrated nutritive substance. The tubes containing 
the colostrum were sterilised in a slanting position, and in this manner 
excellent slopes of the coagulated milk were obtained. The infected flies, 
after sterilisation, were dropped into the tubes containing the egg-yolk and 
colostrum slopes, and next morning, after large numbers of conidia had 
been thrown off in each tube, the flies were removed. In no case was a 
growth of M. racemosus obtained. The conidia germinated freely and grew 
for two or three days, but no increase in bulk took place, and finally the 
germ tubes died and disintegrated. 
A large number of flies that had not been sterilised were dropped into 
the tubes, and in many cases profuse fungous growths were obtained. These 
were found to consist of Mucor, JRhizopus and Penicillium, spp., besides 
other saprophytic fungi that were not identified, but in no instance was a 
growth of M. racemosus found in the cultures. 
The Mucor spores obtained in the above cultures were mixed with a 
solution of sugar in water (as recommended by Hesse), and fed to fifty or 
more house-flies bred in the insectary and kept in a roomy cage. None of 
these flies died of Empusa, although in the house numbers of flies were 
dying of the disease at the time. 
Emptisa conglomerata, Sorokin. (Plate II, figs. 5 and 6.) 
Conidia ovoid, usually with a single large oil-globule, 20-25 /x x 25-40 ja, 
average length about 35 /x. Conidiophores simple. Secondary conidia like 
the primary, produced by direct budding. Resting spores not observed 
at Cedara, but according to Thaxter they are " azygospores, produced from 
spherical hyphal bodies, and borne on a neck-like process of variable length." 
Host. — Imago of Nephrotoma umhripennis, Alex. (Tipulid). 
Habitat. — South j^frica, U.S.A., and Europe. 
Only one specimen was found, the host clinging to a pine-needle by 
