K —BOTANY 175 
manner in which this comes about has only been explained for some of 
the most serious diseases during the last few years. ‘To explain how a 
fungus epidemic disease arises it is necessary to know what is the source 
of the abundant inoculum and to understand the precise environmental 
conditions in which infection of the host and rapid spread therein can take 
place. Not so long ago a sudden outbreak of Puccinia graminis in the 
wheat crop was looked upon as being something akin to magic. This 
holds no longer, for researches in countries where this fungus is rife have 
explained almost completely how such epidemics arise. In countries 
with mild winters, like the southern United States and Australia, it is now 
known that P. graminis survives from season to season by means of uredo- 
spores, so that the intervention of the barberry bush is unnecessary ; the 
same is true under somewhat different conditions for Kenya Colony. In 
the important wheat belt of North America, where epidemics of P. graminis 
are still severe, there are two serious sources of infection of the crop, as 
pointed out by Stakman?1°: firstly, zecidiospores from naturalised bar- 
berry bushes which have been infected from the teleutospores that survive 
on the straw of the previous crop notwithstanding the severity of the 
winter ; and secondly, uredospores brought by winds from Mexico and 
Texas, for by using aeroplanes for trapping spores in the upper air it has 
been shewn that there is a drift of inoculum northwards. As might be 
expected from these considerations there is a lag in the time of development 
of epidemics of P. graminis as one proceeds northwards to the limit of 
wheat cultivation in Canada. One important outcome of these investiga- 
tions has been the demonstration that spores of parasitic fungi can be 
distributed by wind in a living condition over a much greater distance than 
was formerly thought possible. In the plains of northern India the 
problem of the annual recurrence of P. graminis is of a different nature. 
The barberry plays no part in the life-cycle, but the difficulty there is to 
account for the over-summering of the fungus owing to the very high 
temperatures in the plains, which kill the uredospores. Mehta, how- 
ever, working for several years under great difficulties, has solved the 
problem. He has shewn that at altitudes of about 4,000 feet along the 
flank of the Himalayas uredospores on volunteer wheat plants and stubble 
survive the moderate summer temperatures ; by this means the following 
wheat crop in the hills is infected during November and December, and 
the uredospore stage is maintained as the temperature does not fall suffi- 
ciently low to kill it. In this way a source of inoculum is provided in the 
hills, which is blown by the prevailing north-westerly winds to the wheat 
belt in the plains, where infection usually occurs at the end of January or 
early in February. Mehta has also demonstrated that similar foci of 
infection occur in the lower reaches of other mountainous regions in 
India. In Britain the barberry is necessary for the annual recurrence of 
P. graminis as the uredospores do not normally survive the vicissitudes of 
our winter climate. Unlike the position about a century or so ago, how- 
ever, this fungus is no longer a menace here, for nowadays it generally 
attacks the cereal crops too late in the season to do appreciable harm. 
10 Pyoc. Fifth Pacific Science Congress, 4, p. 3177 (1934). 
11 Indian Jour. Agric. Sci., 8, p. 939 (1933). 
