166 THE JOURNAL OF BOTANY 
than in a. genuinum, it is yet very much Seal than the calyx, 
and the long styles of H. linartifolium are equally evident. The 
Jersey plant commonly referred to H. li eavseiliurn exhibits all of 
these characters, which appear to be quite constant, and Borrer’s 
Cape Cornwall specimen differs only in its less revolute leaves—a 
difference very apparent in the Eng. Bot. r. 
Bailey possesses a French form with identical foliage collected at 
Le Mans. The intermediate individuals mentioned by Syme in 
Eng. Bot. ed. 3, ii. p. 156 (1864), as occurring in Jersey, were 
Bay, with barren capsules and probably of this origin, may be seen 
in Herb. C. Bailey. 
From these considerations, and as the varieties obtusisepalum 
and length of styles and capsules, and differs only in its peck 
broader foliage, glabrous or more ee ia calyx and r 
veined petals. The exsiccata, Bourgea d’Espagne, 1863, 
No. 237, and Bornmiiller, Pl. Exs. “one 1900. No. ape sent 
out as H. linariifolium, bear a strong resemblance to H. australe. 
It should be observed that H. linariifolium a. genwinum R. "k EF, 
the form growing with us in South Devon, appears to differ from 
a rites 8 
diagnosis, ‘Corolla parva, calyce parum longior.” I have seen no 
specimen of H. linariifolium with so small a corolla as this, but 
Vahl’s plant may be identical with H. linariifoliwm B. parviflorwm 
ilk. nge, Fl. Hisp. iii. p. 594, which is ase recorded 
from a locality i in the North of Spain. 
Reverting to H. humifusum, Rouy & Foucaud’s a. genuinum, as 
described and as illustrated by Magnier, Fl. Sel. 2684, seems 
essentially identical with the plant of the Linnean Herbarium, 
ough Linneus’s specimen, being gathered late, does not show 
well the lakee ea entire and very pee sepals, which are 
always most marked in the earlier flow In the case of the | 
variety decumbens, the eee hens ‘extended gnipens s 
definition to include Neyraut’s variety radicans, and to cover 
plants of similar habit with i glandular-serrate or even races 
sepals, provided the three larger ones are subacute. I believe 
