102 
saltation is clear, since transfers made from these plates to wheat shoots 
(standard conditions; see App., p. 180) gave normal curves. 
A striking temporary modification of conidial shape, due presumably 
to nutritive or osmotic conditions, was exhibited by H. Nos. 5, 23, and 36, 
in which a considerable number of the conidia were tri-pointed, owing to 
the central cell becoming inequilateral, then extending as is shown in Fig. 
4. The fungus had long been cultured on rich carbohydrate agar, and 
when transferred to corn-meal agar these peculiar tri-pointed conidia were 
no longer common. (In this connection, see also page 154.) 
SUMMARY CONCERNING ENVIRONMENTAL FACTORS WHICH 
INDUCE VARIATION 
It is obvious from the foregoing record that very slight changes in en- 
vironment may produce marked alterations in colony characters—growth- 
rate, color, aerial mycelium, clumping of mycelium, and even in the shape, 
septation, and size of the conidia. The quantity and concentration of 
nutrients, particularly of proteids and carbohydrates, have a very import- 
ant influence on conidia-production and colony-color. Steinberg (114) and 
Javillier (69) have shown that even the zinc in the glass of the culture- 
flask has marked effect on the culture-characters. Dastur (40) has noted 
the following effect of media on Gloeosporium: after long culture on agar 
the conidia-bearing capacity is lost, though return to favorable media leads 
to its recovery unless the stay on agar has been too prolonged, in which 
case the capacity to bear conidia is completely lost. Air-humitidy is es- 
pecially significant as regards aerial mycelium, and has also an appreciable 
influence on the size of the conidia. Light seems of less importance, though 
experiments were insufficient with this factor to be conclusive. Tempera- 
ture in the upper and lower ranges has a marked influence on many mor- 
phological characters. 
In the light of the foregoing experiments it is clear that comparisons of 
colony-characters on artificial media should be made (especially as regards 
points of minor difference} only on media known to be as nearly as possible 
of the same composition, and, when practicable, upon the same lot of medi- 
um and at the same time; thus only can truly parallel conditions be secured. 
Petri dishes with flat bottoms should be used, and in order to secure equal 
thickness of medium the agar should solidify when level. Straw, leaves, 
etc., in test-tubes, give different quantities of aerial mycelium and conidia 
of different size at different levels, both of which are due to humidity and 
