A XEW EUROPEAN serapias 243 



flowers, never brightly coloured as in lingua. Though colour is of 

 little value as a specitic character, the uniformity of the sombre 

 flowers of gregaria is in marked contrast with the wide range of colours 

 in lingua. (5) The dark petals, almost orbicular at base, passing 

 abruptly into the long acumen, not tapering gradually as in (B). 

 (6) The different shape, size, and colouring of the lip. (7) The 

 marked structural difference of the base of the lip, with its two 

 parallel calli. 



The following characters of lingua, taken from Grenicr et Godron, 

 Fl. France (G. G.) ; Bouy, Fl. France (U.) ; and Camus, Mon. Orch. 

 Europe (C.) will facilitate comparison. The main points of dif- 

 ference from S. gregaria are in italics : — 



Tubers 2. Stem 1-4 dm. not spotted at base (It.), pale green 

 (C). Leaves lanceolate, acuti (It.). Flowers 2-6 in a lengthened 

 spike (C). Bracts lanceolate acute, almost half as short as flowers 

 (G. & G.). Sepals oval lanceolate, pale violet (C). Petals lanceo- 

 late, insensibly tapering into an acumen, with 5 strong nerves at 

 the base (G. & Gr.), pale violet (C). Lip about twice as long as 

 helmet, directed forwards, with one single blackish callus at the 

 base (B.), pale violet, reddish, sometimes rose or yellow, always 

 paler in the middle (C), usually glabrous, or offeiiig a few rare 

 slender hairs (B.); middobe broadly oval acute, glabrous or very 

 finely pubescent (G. & G.). 



S. gregaria is abundant in the Maurettes, a low range of schistose 

 hills at the back of Hyeres. I carefully considered the possibility of 

 its being a hybrid. The only other species of Serapias growing there 

 are cordigera and longipetala, but, as both of these have large 

 flowers, a cross between them could hardly produce gregaria, which 

 has the smallest flowers of any Serapias except parvijtora, and is not 

 in any way intermediate between the two. I at first thought it 

 might be lingua Xjjarvijlora, but not a single specimen of either of 

 these two species could be found in the Maurettes. How far gregaria 

 extends into the geologically similar Maure mountains, I cannot say, 

 my only visit having been to Bormes, when I found lingua but 

 not gregaria. In 1908 I stayed at Bormes, and was much puzzled 

 by a two-flowered Serapias, which I could not reconcile with the 

 descriptions of lingua, and now believe to have been gregaria. 



Before the war the late Mr. F. Baine, of Hyeres, a most accurate 

 field-botanist, showed me, on the isthmus of Giens, Serapias occultata 

 var. anomala Albert. On my remarking that it looked more like 

 lingua than occultata, he said that in his opinion it had nothing 

 to do with the latter. In April 1920 I found it, much reduced in 

 numbers, in the same station. It agreed admirably with Bouy's 

 description of S. lingua (except as stated below), but I could find no 

 character in which it was nearer to parviflora than to lingua. In 

 Apiil 1921 I compared it carefully with living specimens of lingua 

 from Bonnes, when it became evident that they were conspecific, 

 The plant should be re-named as follows : — 



Serapias lingua var. anomala (n. comb.). Tubers 3. Stem 

 10-15 dm. Flowers smaller. Lip smaller, sidedobes more rhomboid, 

 middobe shorter, callus with yellow base and a shining blackishr 

 red ridge on each side of furrow, 



