83 
Fretp OBSERVATIONS ON Rusts FoR THE GENERAL BOTANIST. 
By J. C. ARTHUR. 
(Abstract). 
The polymorphic character of many species of rusts, together with 
the discontinuous growth between the forms of most of such species, and 
the further fact that some species live upon unlike hosts at different stages 
of their life cycle, make the study of the rusts unusually attractive for 
those who enjoy a varied problem. A very important part of the field 
observation consists in later visits to the spot where a rust has been found 
in order to ascertain if it is followed by another form of the same species 
either on the same host or on a host of some other kind. In either case, 
but especially the latter, a suitable specimen of the rust may be taken to 
a locality where it does not occur and placed beside a healthy plant of 
the kind observed. It is then watched to see if the rust infects the healthy 
plant. If it does, the correctness of the inference from the first observation, 
that the two forms of rust found to succeed each other in the same locality 
belonged to the same species, is established. But if the healthy plant does 
not become infected, either the two forms found in the original locality 
belong to distinct species and only incidentally happen to follow one 
another, or else the sowing of the rust in the new locality was not well 
done. In either case further trials and observations are required. There 
are many variations to the inferential and deductive reasoning required 
to solve these problems, and to come to a conclusion repeated observations 
may be required extending over many months or even years. 
