BRITISH FOKMS OF JASIONE MONTANA 213 



and Lange, and later Kony, were obviously right in separating 

 it from J. montana. There is excellent material of this plant in 

 Herb. C. Bailey from the Gironde district (Magnier, No. 2519, &c), 

 from which its features can be well seen. As stated by Duby and 

 subsequent authors, it produces a number of long, prostrate stems 

 from the rootstock, each terminating in a single flower-head and 

 remaining permanently unbranched. Its stem-leaves are numerous, 

 very small (5-8 mm. long), thick and recurved, and extend further 

 towards the flower-head than in any form of J. montana. The 

 umbels, which are quite small (10-15 mm. broad) and composed 

 of shortly pedicelled flowers, are surrounded by very small, obtuse 

 bracts, much shorter than the heads. The calyces are more or less 

 pubescent, with broad and densely long-hirsute teeth, which impart, a 

 hairy aspect to the flower-heads ; and this condition persists in fruit. 

 This plant seems to be of more than biennial duration. 



I am not aware whether any Portsmouth specimen exists of the 

 var. maritime/, mentioned by Macreight (Manual, 1. c), but his 

 description, as far as it goes, embodies some of the salient features 

 (including the hirsute calyx) of the var. niaritima of Duby and 

 De Candolle ; and his plant might have been rightly named, albeit 

 Townsend, in his Flora of Hampshire, refers it to var. littoralis Fr: 



Touching the Holyhead plant identified with var. niaritima Duby. 

 (Journ. Bot. p. 321 "(1917)), Mr. Salmon has kindly sent me his: 

 original specimens for examination, and these prove identical with 

 myllfracombe form. They are in the flowering stage, and show the 

 calyx-tube thickly clothed with the characteristic minute hairs often 

 seen in J. montana. In some cases a few of these minute hairs are 

 scattered on the calyx-teeth, and it is no doubt owing to the presence 

 of these hairs that this usually erect and large-headed plant was 

 associated with Duby's prostrate sand-dune variety. 



It now remains to determine our handsome coast form, which has 

 been named var. major, var. niaritima, and even var. littoralis. If 

 has already been shown that it is distinct from var. major Mert. & 

 Koch, and a similar tall habit, with narrow leaves, long-pedicelled 

 flowers, and very long, bristle-like calyx teeth, serves to separate 

 var. ecJiinata Willk. & Lange from it. It differs essentially, not only 

 from var. maritima Duby, but from var. littoralis Fr. and var. 

 sahularia Cout., by its more erect habit, and its large_ leaves, bracts 

 and flower-heads. Evidently it has no connection with the dwarf, 

 varieties gracilis Lange, Borcei Bouy, and Timbali Rouy. The 

 remaining form with which it invites comparison is Brebisson's 

 var: niaritima, which seems identical with var. nana Gren. ■& Godiv 

 This is said to-be a low, hairy plant (3-10 cm.), with short leaves 

 and large, pale heads, and is recorded from maritime rocks in Western 

 Normandy. I have seen no authentic examples of this plant, which 

 may possibly resemble the llf racombe form but for its dwarf growth ; 

 and unfortunately its description is very incomplete. But it is 

 strange that Rouv should reduce it to a sub-variety nana of var. 

 littoralis Fr., however dwarf its growth, if it is a form of our robust 

 coast plant. There is no Continental material that matches our form 

 in Herb. Mus. Brit., Herb. K'ew or Herb. C. Bailey, while these 



