95 
In order to test the stability of the species Klebahn has since 
1892 conducted a series of cultures in which the gecidiospores 
from one species of host plant have been year by year sown 
on other host plants known to harbour the parasite. P. 
smilacearum-digraphidis has been the species selected which 
as will be seen from the table has its zcidiospores on five plants 
viz.: Maianthemum bifolium, Polygonatum multiflorum, P. verti- 
cillatum, Convallaria majalis and Paris quadrifolia. ‘The zcidio- 
spores from Polygonatum only were employed, these were placed 
on Phalaris and the resulting P. teleutospores next year applied 
to the above named plants: from the resulting ecidia only those 
on Polygonatum were again used, and so on. In 1895 the 
acidia were less abundant and only poorly developed on Paris. 
In 1897 none were formed on Paris at all, and it has failed to 
develope the parasite since, at the same time the fungus was not 
2 ee developed on Maianthemum, and in the following year 
onvallaria was not so freely infected ; while all through the series 
the Polygonatum plants produced the ecidia most freely and 
most perfectly. ‘This looks as if there were a tendency for the 
species to become specialised on one particular host, although 
the results obtained were not absolute. 
Klebahn found that the common ecidium on Orchis latifolia 
ps connected with a Puccinia on Phalaris (P. orchidearum- 
phataridis K.) Originally Rostrup led us to believe that the 
orchid zcidium was connected with the Puccinia on Mbolinia 
ae but from several experiments I have made at King’s 
ynn it was apparent that our zcidium on Orchis latifolia had 
some other life history. Accordingly, I determined to repeat 
Klebahn’s cultures upon Phalaris with it. A suppl f th 
Phataris having been obtained f ai ee 
ene a ae oe Barton Broad from a place 
€ other host plants 
ve ee were this year poriieen a ee ee 
or plants viz.: Orchis latifoli 
ursinum and Convallaria ae oie eee - ae 
a he were observed on the orchid ‘oliewet in : ae Hts 
c 
ae Py ee a 0 was observable on the other pnapeee a 
Convallaria which by die ate rekon 
mises (The eae: were so evident as to be un- 
hardly look like at first = ee ee ee 
of the parasitic m pa ge ee Uo oT Ee 
a chat cara - ae of a uredine, for although there is 
of fhe eee A € chlorophyll, yet there is no thickening 
paaiihe ae ee case when these mycelia are present, in 
iter Scheci up to the light it will be seen th 
rophyll colour has been @ischarecd h oe 
gone and but little else left e fee Moe ee 
xcept the epidermis of the upper and 
