122 THE JOURNAL OF BOTANY 



clearly exceed the aclaclium). Heads 1 to 6, rather large for this 

 subsection, campanulate. Outer phyllaries short, blunt, linear or 

 oblong ; inner remarkably porrect in bud, narrow-linear, very long 

 and acute ; all decidedly senescent, cano-floccose, clothed with 

 short, medium- sized, and long gland-tipped bristles, and with many 

 long, white, simple hairs. Ligules lemon-yellow, not golden, 

 about twice as long as the inner phyllaries, very pilose at the tips. 

 Styles blackish. Ripe achenes black. Pappus white, rather 

 shining. 



Exsiccata.—E. S. Marshall, Nos. 3284-5 ; also 3602-9, inclusive. 



I dedicate this species to my old friend W. A. Shoolbred, in 

 whose company the whole series was collected. It is very 

 striking when alive ; but some of the characters are obscured in 

 the process of drying. Both W. A. S. and Rev. E. F. Linton 

 have suggested H. sanguineum Ley for it ; but Rev. Augustine Ley 

 remarked, on the original gatherings : — " Not sanguineum, but sil- 

 vaticum var. tricolor " — an identification which cannot be upheld. 

 Later, Mr. Linton thought that it might be "a local species"; 

 his present view is that " this species will have to be placed 

 near H. sanguineum Ley," and I fully concur. Superficially, the 

 resemblance between herbarium examples of both is very striking; 

 but a closer comparison renders their union impossible. In 

 H. Shoolbredii the leaves are extraordinarily vivid green, not dull, 

 with longer ciliation ; the heads are still larger, but not so broad 

 in proportion, and the very porrect phyllaries (in bud) are a good 

 distinctive mark. The stalked head-glands are far more nume- 

 rous, stouter, and more unequal ; while the simple hairs are long, 

 white (not dusky), and considerably more abundant. Instead of 

 being slightly ciliate, the ligules are extremely so ; and the styles 

 are a good deal darker. Several minor divergences might be 

 added ; but this is hardly needed. 



We first found this hawkweed in 1908, near Inchnadamph, in 

 three stations, and used to speak of it as " false hyparcticum," 

 from a superficial resemblance in their habit (the lower branch of 

 a specimen in my herbarium, including the peduncles and heads, 

 rather exceeds 1-5 decimetres!). Up to 1911 we regarded it as a 

 purely limestone species ; but in that year it was again obtained 

 near Dalwhinnie in several spots, both on gneiss and on granite, 

 so it is probably of wide distribution in the Highlands. It has, 

 so far, been observed as follows, chiefly on fairly dry rocks, but 

 sometimes by alpine streams ; the range of elevation varying from 

 about 800 feet, in the north, to 2200 feet or more, further south: — 



V.-c. 88, Mid Perth. — Hillside, Sgairneach Mor, facing and 

 west of the Sow of Athole. 



V.-c. 96, East Inverness. — Creag Dhubh, south of Loch 

 Ericht ; Coire Chuirn (adjoining Atholl Forest, v.-c. 89, East 

 Perth) ; and Allt Beal an Sporain (halfway between Dalwhinnie 

 and Dalnaspidal). 



V.-c. 108, West Sutherland. — Near the Traligill River and the 

 Allt-na-Glaice-Moire: also on a hill overlooking Loch Maol-a- 

 Choire (" Gillaroo Loch "), near Inchnadamph. 



