PUCCINIA GRAMINIS 43 
by its uredospores, (2) by a perennial mycelium, (3) by Eriks- 
son’s mycoplasm. 
The first possibility is entirely a matter of climate: it may 
take place in one country and not in another, or in the same 
country it may take place in one season and not in others. 
McAlpine and Cobb find viable uredospores all the year round 
in Australia, and Lagerheim says the same for Ecuador. But 
in northern climates it has been shown that the uredospores of 
P. graminis frequently lose all capacity for germination during 
the winter; this is proved true of Sweden, North Germany, 
North Dakota, etc., but in the United States, south of Ohio, 
Bolley found germinable uredospores all through the year. 
Similarly in Bohemia, uredospores of P. dispersa, P. glumarum 
and P. Loli can survive mild winters or in sheltered places 
(Baudys). Even though uredospores capable of germination 
may sometimes be found on wild grasses during the winter, it 
does not follow that those could start an epidemic next spring, 
owing to the specialisation which has been proved to exist, by 
which a form of P. graminis on one host is often incapable .of 
infecting another host. 
In regard to the second possibility, we find again two 
opposing views. De Bary and others have searched in vain for 
mycelium in the growing wheat plants, before infection becomes 
visible, but Pritchard (1911) found mycelium resembling that 
of P. graminis both in the pericarp of wheat grains and in 
various parts of wheat seedlings. He showed that large num- 
bers of wheat grains contained pustules of teleutospores, even 
visible in the neighbourhood of the hilum, but also hidden 
within the pericarp. He proved that the mycelium from the 
pericarp penetrates through the intercellular spaces, as well as 
through the cells, and “soon passes into the spaces between 
the leaf-sheaths where it grows rapidly and attacks the tissues 
at various points.” W.G. Smith figures teleutospores within 
the seed of Oat (Gard. Chron. 1885, xxiv, 245, f. 58) and ecidia 
in the pericarp and seed of Barberry (ibid. 1886, xxv, 309, 
f. 58). 
It is evident that, if this state of things prevailed on a 
large scale, nothing more would be required to explain the 
