1886. ] BOTANICAL GAZETTE. 235 
tively clear notion of the species as distinguishable by their gross 
and microscopic appearances, for, until that was the case, any ac- 
count of cultures made would be quite unintelligible. Nor could 
we hastily assume that those of our Gymnosporangia, which ap- 
pear to be very much like European species, must have the same 
Roestelia form as in Europe. Practical experiment by cultures 
is the only method of settling the question effectually. If the 
anatomical characters are the same, and if the sowings of the 
spores are followed by the same Roestelia in both cases, then our 
form and the European belong to the same species, otherwise not. 
In a paper published in 1880, I attempted to take the first 
step by collating what had been written about our species and 
supplementing it by observations ona large amount of living 
and herbarium material, so as to be able to form an estimate of 
the comparative distribution of both our Gymnosporangia and 
Roestelize, and, as far as our knowledge then allowed, to learn 
something of the probabilities of the connection between different 
orms; for we must naturally assume that, if there is any natural 
connection between Gymnosporangia and Roestelie, the con- 
nected forms will be found growing near together rather than 
botanists who, on sowing Gymmosporangium spores 
tained only spermogonia have at once assumed that they belonged 
several cases, the same host is known to support several different 
forms of Roestelia. Although my cultures were not conclusive 
at all, taken in connection with what was known of the compar- 
ative natural distribution of the forms in question, they afforded, 
in a measure, some guide to the i in which we might ex- 
t more definite information in the future. — : 
ray ibe 1880 my cultures have been continued at interes 
without, however, the production of ecidia, and in the meanwhile 
