190 
AMERICAN JOURNAL OF BOTANY 
[Vol. 9, 
when, shortly after inoculation, the host is placed under conditions where 
it cannot manufacture carbohydrates, as by keeping it in the dark or in 
light from which the red-orange end of the spectrum is filtered off or in air 
deprived of carbon dioxide, the development of the rust is inhibited. 
Fromme (1913, p. 516) found that placing oat plants recently inoculated 
with Puccinia coronifera in the dark for a period of several days increased 
the length of the incubation period of the rust by a corresponding interval. 
Fromme interprets the observation to indicate that the fungus is dependent 
for its nutrition on some intermediate product of photosynthesis. 
Mains (1917, p. 191) confirms Fromme's observation that the develop- 
ment of the crown rust of oats is retarded in the absence of light, and adds 
that if the infected plant is left in the dark too long (which would greatly 
impair the vitality and vigor of the host tissue) the rust is killed. Mains 
also independently repeated Ward's observation that growing the host 
plants in an atmosphere free from carbon dioxide inhibits the development 
of the rust. Similar experiments with Puccinia Sorghi on seedling plants of 
Zea Mays, however, failed to arrest the development of the rust. Further 
experiments showed that if the host leaf is supplied with carbohydrates, 
either from the reserve stores of the endosperm or by being floated under 
aseptic conditions on a sugar solution, then the rust develops successfully, 
even if, because of the absence of light or of carbon dioxide, the host tissue 
cannot manufacture its own carbohydrates. Mains is therefore prompted 
to qualify Fromme's inference that the rust is dependent upon intermediate 
products of photosynthesis into the statement that the rust is dependent 
upon transitory carbohydrates. 
It does not necessarily follow from an observation that the development 
of the rust is inhibited upon a plant starved of an essential nutrient — 
whether carbon, or nitrogen, or potassium, etc. — that the rust fungus is 
dependent for its nutrition upon compounds of that substance. As long 
as the host plant is at all alive, or, even if it is dead, before disintegration 
of its substance has set in, it contains carbon, nitrogen, etc., compounds, 
and we cannot say that the rust could not develop because of the absence of 
such compounds. Such observations are best interpreted on the basis of 
the physiological condition of the host when it is starved of an CvSsential 
nutrient substance. We can not say that a host plant starved of an essential 
nutrient is a host plant deficient in that particular class of substances; but 
we can say that a host plant starved of an essential nutrient is a host plant 
that is not assimilating, that is not growing, a plant in which anabolic proc- 
esses are at a standstill and katabolic processes predominate. And we 
are justified in inferring from the observed behavior of rust fungi on host 
plants starved of essential nutrients that a plant which is not assimilating, 
which is not growing, in which anabolic processes are at a standstill and 
katabolic processes predominate, does not make a congenial host for the 
rust fungus. The suggestion that the rust is dependent for nutrition upon 
