52 E. J. BUTLER. 
23. Onion—In one tube the growth resembled that on plan- 
tain, but the surface growth on the liquid below was brick-red in 
colour, and no sclerotia were formed. In another tube the growth 
was moist and pink as on normal potato and there were no coremial 
strands. 
24. Sugarcane pith—Growth very feeble and scarcely visible 
‘after a week. It ceased entirely soon after. A few microconidia 
only were formed. 
25. Tapioca.—After 15 days growth slight and confined to the 
point of inoculation. Colour pale salmon pink. A few micro- 
conidia. 
26. Nutrient agar—tIn streak cultures the growth appears 
slowly, being visible on the third day, and forming a raised distinct 
band by the 6th day. The band is strongly convex, limited closely 
to the neighbourhood of the streak, and the rest of the surface of 
the agar long remains free from the fungus. There 1s usually little 
aerial growth, the streak being not unlike a bacterial growth. In 
the centre it is from salmon pink to brick-red, and around this is a 
zone of scanty superficial growth distinctly filamentous under a 
lens. The raised part of the streak is uniformly moist and shiny 
as soon as spore-formation becomes considerable. After about 2 
weeks portions of the streak become greyish black and this may 
gradually extend to the whole growth. The spores are very 
variable even in young cultures, those figured on plate IV, fig. 4, 
being sketched from a single culture 3 days old. In old tubes 
macroconidia occur, but they are never numerous on this medium. 
Chlamydospores form on agar, as on most of the other media, after 
some months. 
27. Glucose agar.—The addition of glucose does not appear 
to affect the character of the growth to any extent. It is rather 
more vigorous, and the black colour may develop more quickly. 
In seeking for diagnostic characters in the above series of cul- 
tures, most weight naturally attaches to such characters as appear 
