ERYTHRyE^ IN THE ISLE OF WIGHT. 329 



description, and the only one I have seen of Roemer's;is that in 

 Lloyd's ' Flore de I'Ouest de la France,' ed. iii. ' E. Centaiuiuni, 

 Pers. . . .ft. E. caiiitata, Roem. — Plaute courte, ramassee a 

 corymbe toujours compacte ; fenil. rad. nombreuses, grandes, 

 arrondies, obtuses, a 5-7 nervures. Eochers, pelouses sablonneuses 

 graviers de la region maritime, surtoiit de la cote du nord.' This 

 description applies to the dwarf states of E. Centaurium, but not to 

 our plant. I do not know whether Eoemer's E. capitata is the same 

 as that of Willdenow, the following description of which has been 

 copied out from Roemer and Sclmltes' ' Syst. Veg.' : — " Eri/thra^a 

 capitata, Willd. — Foliis ellipticis lanceolatis obovatisque 3- et 5- 

 nervibus sessilibus, floribus capitatis bracteatis, Herb. Willd. MSS. 

 Caulis 2 poll, simplex, basi dense tectus foliis obovatis lanceolatis, 

 ovalibusque obtusis apice subnudus tetragonns, capitulo multifloro, 

 foliis binis ternisve cincto. Bractes lineares acutae, floribus inter- 

 mixt^e. Calycis laciniae acut«, longitudine tubi. Corolla Cm- 

 taurii sed lacini^ acutiores, angustiores. Caulis sgepe in superiore 

 parte, immo ad ipsum cax3itulum, ramum unum alterumve emittit 

 nudiusculum capitulo terminatum ut planta prolifera videatur. 

 Willd. In Markbrandenbergensi." R. & S., ' Syst. Veg.,' vol. iv. 

 p. 786 (1819). It applies fairly to our plant, except in the characters 

 '' BractaB acutse " and "Corolla Centaurii sed laciniae acutiores." 

 I am not, however, disposed to place great value on the degree of 

 obtuseness of the corolla segments, for they are variable in 

 E. Centaurium. They are usually obtuse, but I have seen them even 

 acute. Willdenow does not allude to the place of insertion of the 

 stamens. 



Erythema tenuiflora. Link. — I am not aware that there has 

 been any notice of ErythrcEa tenuiflora, Link., as a British plant. 

 I gathered specimens lately in the Isle of Wight, on the west bank 

 of the Medina, between West Cowes and Newport, which I believe 

 to be Link's plant. The Isle of Wight specimens approach E. 

 jmlchella in the flowers, but are very different in habit from that 

 species, the stem being as stout as that of E. Centaurium , simple, 

 dichotomously branched above the middle, the branches allfastigiate, 

 the flowers axillary and terminal, subsessile, hardly corymbose, but 

 forming numerous few-flowered clusters. The radical leaves are few 

 and do not form a rosette. I have not with me sufficient materials 

 to give an opinion of the distinctness of the plant as a species, but 

 Mr. James Lloyd now accords it that rank, having formerly united 

 it to E. pulchella as " var. scojjaria.'" I only noticed it in one spot, 

 where there might have been from twenty to thii'ty plants. The 

 idea then occurred to me that it might be a hybrid between 

 E. Centaurium and E. pulchella, both of which are very common 

 througliout the Island. 



2u 



