85 
siderable profit and give promise of still more valuable harvests 
tocome. It may well serve us as a starting point in reviewing 
briefly the progress recently made. : 
Enksson’s primary object was to investigate the rusts of the 
cereal crops and determine whether anything could be done to 
check the enormous losses they cause year by year. There were 
some four species recognized by mycologists on the cereals 
wheat, barley, oats and rye, but Eriksson showed that in Sweden 
there were at least ten forms of rust to be reckoned with on 
these host plants. Taking, for instance, the single species 
Puccinia graminis, this was known on the various cereals and 
on a hundred of the native grasses and it was supposed that the 
fungus could pass indiscriminately from any one host to the 
other. By a long series of infection experiments Eriksson 
showed that this was by no means the case. If for instance the 
uredospores of P. graminis were taken from oat-plants they 
failed to reproduce the rust on rye, barley, or wheat; similarly 
if taken from meadow grass (Poa pratensis) they were incapable 
of infecting these same cereals. Thus this single species could 
be broken up into a series of races such as P. graminis secalis, 
P. graminis tritici, P. graminis avenae, P. graminis poae, etc., 
the infecting powers of which were confined to certain groups 
of plants. For the first time then we realized that the spreading 
of these parasites from wild to cultivated host-plants was not 
as widespread as we had believed, and it was also clearly proved 
that, with exceptions we need not describe here, the spreading 
from one cereal to another did not occur. These specialized 
forms of P. graminis have received various names. They are 
frequently described as “ biologic species,” for from a physiologi- 
cal point of view they are obviously distinct species, but syste- 
matists who can make use of external characteristics only for 
delimiting the species can hardly look upon them as such, as 
hephologically they are identical with one another. We may 
erefore compromise and call them “ biologic forms.” 
The distinctness of these biologic forms is further emphasized 
Bue i that the aecidiospores they give rise to on the bar- 
neve ee the same restricted powers of infection. If, for 
aa ae Be oes result from an infection with the form 
oe ae cy are incapable of reproducing the rust on barley 
and gh, eoeatches have proved the accuracy of Eriksson’s work 
cee i i a that this specialization of fungi to particular host- 
ee ee by no means confined to the rusts of the 
fee baa, fe i os Pepe of ey gens teacons 
as C laviceps E r Are ineae and in suc scomycetes 
eper Sai voascus and Erysiphe. The researches of 
> pa'mon and Marechal have made our knowledge of the 
