114 THE JOURNAL OF BOTANY 
E. tees Jordan; though that is described as having veined 
I 
petals failed to observe in the fresh gee Ni Plants 
often irae for the poneeicy —Epwarp S. Marswatu. ‘Cannot get 
this named.’’—Ep. 
sabadeoie se te L. Llyn-y-fan-fach, Carmarthenshire, v.-c. 44. 
Very small plants, but quite characteristic and in good flower, with 
fruit well ei on the more forward specimens, 21st July, 1902. 
Apparently a new record for South Wales province of H. C. Watson. 
H. . RIDDELSDELL. 
Viola Riviniana var. nemorosa Neum., W., & Murb.  Plentiful 
in Comber Wood, near Houghton, West Sussex, v.-c. 18, 8rd May, 
902. Flowers large and handsome; spur sla scarcely fur- 
rowed. Unfortunately, the colour is very fugitiv : 1 have never 
yet cg apael in Seen Droverving it,—Epwarp 8. MarsHaLL. 
‘* Rightly n W. 
Cer ties weishekas L., forma. This plant grows rather 
plentifully on the greensand formation at Bow Brickhill, Bucking- 
hamshire, June, 1902. It has a very different facies from the 
ordinary plant of = Midlands, but I have been unable to match it 
with any named form.—G. Craripce Druce. 
Sagina ‘aide Boiss., glandular form. Birkenhead Docks, 
Cheshire, 31st July, 1902. This has been referred to three different 
species by as many well- known botanists. I cannot agree that it 
belongs to any of our British forms. Mr. Williams says: ‘I have 
ae Pet that it is S. Reutert. You will notice that the valves 
wo are oar meat scone at the tip. In S. apetal 
as indigenous.” ennett, to whom also I submitted specimens, 
sent me 8. "Reuter fo or comparison with my plants, and I could see 
no difference, except in the relative length of the internodes and the 
denser glandular pedicels of my plant. It is new to county 58.— 
J. A. WHELDoN 
Rubus holerythros Focke, variety. Moorland ditches and planta- 
tion sides on Mitcheldean Meend, West Gloucester ; altitude about 
830 ft.; 20th August, 1902. This plant was aaa n in situ by Rev. 
. M. Rogers in 1896. On taking these specimens to him in the 
present year, Mr. Rogers made the we pees rehaniie on them: 
‘On the whole, this plait is nearest to R. holerythros Focke; but 
receding towards R. latifolius Bab,, in the shape of the leaves. 
