EPILOBIUM DURLEI IN ENGLAND ? 297 
as covering the base of the stems. All our species die down to the 
e new flowering-stems are produced from stolons or 
rosettes which have outlived the winter. The withered leaves 
is not limite f usual in 
analogous states of H. Lamyi and EF. obscurum which I have found 
in its company. This oct disappeared in my garden; an 
the presence of such a feature in an accidental state of one species 
does not destroy its sed as a constant character, in another species. 
Mr. Clar self are thus fully agreed on one pains, 
namely, in Sotiitlering ‘that his Hants form belongs to FE. 
tanum, and may be traced into the type by Sidanaiblg gradations. 
Our difference amounts to thi cae asserts “that H. Durie 
Gay, is an English plant’’; I say, with great respect, but without 
a shadow of doubt, Ac it has not yet ae found in Britain 
Mr. N. E. Bro is well wn an able and accurate 
draughtsman, and I have no fault to find with ‘the figures given ; 
but I must say that the shabby little specimen of EF. Duriai here 
depicted does not give an adequate notion of the species, as repre- 
sented in the Kew set. Still, the lower half of the plant is quite 
unlike any example of FE. montanum that I have yet come across. 
M. Barbey’s fine plate, on the contrary, hits off the average FE. 
Durizi remarkably well. Is not Mr. Clarke mistaken in stating 
J ; 
and I should suppose a specimen of this gathering to have been 
To avoid further tediousness, I pass on to the best book- 
charaotees by which to differentiate FE. sagen (including the 
. minor aprica) from EF. Duriei ; remarking, ever, that British 
—to say nothing of foreign——willow-herbs oanat be mast 
without patient and systematic study of their various forms, in the 
field as well as in print and in he rbaria 
ist uri@i (as in FE. collinum) the buds are obtuse; in F. 
montanum they are always apiculate or mucronate. 
2. The he doce di of H. Duriai are sly stouter, and the flowers 
larger, in proportion, than in normal montanum; they are also 
described” Shae “ saturate violacei,”’ instead of rose- -coloured, 
pink or wliite. 
8. The stem is always simple in E. Duriai; in E. montanwm it 
is pereey branche 
: an uriai ‘the general cut of the leaves recalls /. lan- 
sistant rather + an E. montanum, and their texture seems to be 
narrowed into a rather long petiole, whereas in E. rp oat they 
are sessile or shortly gies with a broad, rounded bas 
5. The stolons of F. @i are ‘‘ entiérement stellen aux 
stolons de l’F. alsinefolivom,” This, we my thinking by far the 
most important distinctive character, is exceedingly well illustrated 
by Barbey, and can be fairly well seen in several of the Kew 
