170 BRITISH SPECIES AND SUBSPECIES. OF POLYGALA, 
Schrift. d. Ges. Nat. Marburg, p.1. P. serpyllacea, Weihe, Bot. Zei 
vol. ii., p. 705. Stems elongated, slender, flexuose ; the leaves om 
and the lower ones often nearly or quite opposite ; ; racemes fewer- 
flowered and flowers rather smaller than in var. 1; central bract 
shorter than the flowering pedicel; wing-sepals broad ; veins but 
slightly anastomosing ; ; appendages of aril short, blunt. 
F eLec.. t.1; Reich. Ic. Flor. Germ. et. Helv.,t. eee nae < 
f. 1 (very eee) Engl. Bot., t. elxxxvii. (mach to 
regular and fo ['Tab. 189, fig. 2 
Differs from var. 1 shied f in its more wiry and anes cane the 
flowers are also somewhat smaller and fewer, and the of the 
wing sepals rather straighter, with not so many fine sbesiivsiaie 
branches ; but it is difficult to draw up any other distinctive cha- 
racters, and it seems to me very questionable whether it is entitled to 
the dignity of a subspecies. DP. depressa is widely distributed through- 
‘tish T 
typical form, Our illustration is from a French specimen authenti- 
cated by M. Cosson. 
Var. 3. rove Engl. Bot., 8rd ed., vol. i., p. 836; Bab. Man., 
th ed., p. Hook. Stud. Flor., "2nd ed., p. 48. P. ciliata, 
Lebel i in ae a Godr. Flor. France, ‘vol. ip: 195: ; Reich. Ic. Flor. 
Germ. et. Helv., vol. xviii., p. 90. Stone we and flexuose; wing- 
sepals, and often the bracts, pedicel and capsule ciliated. 
Fics. Reich. Ic. Flor. Germ. et Helv. xviii., t. mecexlvi., f. 5. 
Although this form has been erected into a species by some Conti- 
nental botanists, Iam very doubtful whether it ought to be retained 
even as a good variety. The ciliation of the wing- -sepals, as well as 
of the capsule, seems a mere accidental local character ; in habit and in 
every other character it is, as far as I can see, indistinguishable from 
depressa ; and even the ciliationis not constant. Dr. Boswell says, in a 
letter to Prof. Babington, ‘‘ Two-thirds of the plants picked at random 
are not ciliated, but quite indistinguishable in habit”; while, on the 
other hand, a specimen of depress 8 from the Orkneys in my herba- — 
i 
characters of the former ; the wing-sepals are always broad, as far as 1 
have seen. The only ieardal} habitat i in these islands for this variety 
is the Gogmagog Hills, Cambridgeshire ; but it is probably not very 
uncommon. On the Continent it is reported fi rom several widely dis- 
persed localities. Our illustration is from a sia from the English 
locality, pathone by Dr. Boswell. 
Var. < oxyptera, Engl. Bot., 3rd ed., vol. ii., A 36; Bab. Man., 
7th ed., p. 44; Hook. Stud.’ Flor., 2nd. ed., p. 48; Reich. Ic. 
