ROSA MOLLIS 59 
tomentosa form. Dr. St. Brody’s Painswick, Gloster, specimens 
elled R. pomifera are rightly placed here by Déséglise, and 
Mr. Marshall’s Tidenham Chase, W. Gloster, specimens belong 
here also. 
Ley quotes fifteen vice-counties for R. recondita, but from 
his description and from the specimens I have seen so labelled by 
him, some of these are probably forms of mollis, and not of poma- 
fera. There are specimens in herb. Bailey from W. Argyle and 
N.E. Kincardine, which are doubtless referable to this species. 
Rosa MOLLIS 
Smith, Eng. Bot. t. 2459 (1812). 
“Fruit globose, half as long as the segments of the calyx, 
bristly as well as the flower stalks. Prickles of the stems 
straight. Leaflets elliptical-ovate, downy on both sides.” 
Smith adds that he thought this different from R. vellosa Linn. 
[R. pomifera Herrm.?], being of lower growth, with less elongate 
more ovate leaflets. Petals deeper red. Ripe fruit much smaller. 
nate. He thought it not very uncommon in England and Wales. 
R. mollis has perhaps been more misunderstood in Britain 
i it 
in most herbarium specimens, which are seldom collected in ripe 
fruit, and even in that state the sepals of the Sherardi group are 
mostly still adherent, but there should be little difficulty in sepa- 
rating R. mollis from the tomentosa group by these organs. 4n 
R. mollis, and indeed in the whole pomifera group, the sepals are 
long and narrow, much rounded on the back, so as to be almost 
the colour of the ripening fruit, while in the tomentosa group they 
are much thinner, and show signs of disarticulation soon after the 
always obvious in this group. Keller, however, remarks on the 
difficulty in rating R. mollis from R. on the on 
hand, and from R. omissa on the other, observing that Crépin 
