No. 381.] THE CONCEPTION OF SPECIES. 689 
researches of Eriksson and Henning in Sweden, whose work is 
certainly a model of careful investigation. I take it for granted 
that most of my hearers are already acquainted with the char- 
acter of the work in question, and we need stop to consider 
only those points which bear upon the subjects we are discuss- 
‘ing. Of the three common rusts which affect grains, Puccinia 
graminis, P. rubigo-vera, and P. coronata, the æcidia are to be 
found, respectively, in Æcidium berberidis, Æc. asperifoliz, and 
“ic. rhamni, according to the previously accepted view in 
regard to those species. Judging by the morphological charac- 
ters of the teleutospores and the uredospores alone, these three 
species occur on a large number of different grasses. In mak- 
ing inoculations to ascertain the facts in regard to the æcidia 
of the species, Eriksson and Henning found that what was sup- 
posed to be P. graminis growing on Phleum pratense and Fes- 
tuca elatior had no ecidia, and they described this form under 
the name of P. phlei-pratensis. Puccinia coronata is separated 
into two species, P. coronifera and P. coronata, the former hav- 
ing its æcidium on Rhamnus catharticus, the latter with zecidia 
on Rhamnus frangula, with perhaps two other forms to be 
separated from the old P. coronata. Puccinia rubigo-vera is 
separated into three species: P. glumarum, P. dispersa, and 
P. simplex —the distinctions based largely on the presence 
or absence of the ecidium, although there are also certain 
differences in the habit and color of the other stages. The 
three original species are split up into seven species, besides 
two uncertain forms, characterized in the main by physiological 
characters. Furthermore, of P. graminis, six specialized forms 
are enumerated, characterized by differences in the inoculating 
capacity of the uredo or teleutospores on different hosts. The 
other species also have their specialized forms, the total num- 
ber being, I believe, twenty-eight. We may consider the spe- 
cialized forms to be races, and, in that case certainly, we shall 
have to agree with Eriksson and Henning in considering their 
seven species as species rather than races. The important 
point is to know whether the differences observed are tempo- 
rary and accidental or permanent. It is too much to ask for 
the confirmation of the results of these two experimenters just 
