376 THE JOURNAL OF BOTANY | 
cells with an oblique sigmoid curve, usually having the walls con- 
siderably thickened ; at times this is carried to such an extent that 
they are properly described as vermicular. In the variety under 
notice the transition rarely takes place; the upper cells differing 
little from those of the uppermost part of the expanded base, only 
showing a tendency to become rounded at the angles and hence 
rhomboid-elliptical, and at the same time somewhat oblique in their 
direction. In this, however, there is some variation, as plants occur 
that must be referred to the variety, but with the areolation almost 
as in the type. As is frequent in this species, the hair-points are 
often very short or wanting in many of the leaves. 
I sent a specimen of the Welsh plant to Prof. Barker, who 
nm 
S 
@ 
te 
oO 
= 
~y 
wm 
2 
leje} 
° 
Oo mw 
5 
ea 
< 
= 
~ 
s 
= 
~ 
~ 
= 
~” 
eS 
myself at Broadford, from the wall where Barbula nitida grew. 
Perhaps the one moss is different from the other. I cannot now 
examine them. There is, I see, another variety or form in my 
There is a specimen of this var. in Wilson’s herbarium consisting 
of three tufts labelled ‘ Campylopus longipilus var.; Glen Phee, 1868, 
ergusson. No. 2, W. 11/68 
plant. 
__C. adustus De Not. is of a different habit, very short, but with 
wider, shorter leaves, and, according to Limpricht, the stereid cells 
of the nerve are absent; so that it cannot belong to C. atrovirens, 
to which it has been referred by Husnot and other writers. 
abit, pale colour, and leaf-characters are so marked in the 
plant here described, and the localities in which it has been gathered so 
numerous and widespread, that I have felt no hesitation in describing 
it as a variety; at the same time I am quite aware that transition 
