154 THE JOURNAL OF BOTANY 



palmato-multiHdis sparsis : laciniis acutis, caule erecto," besides being 

 very vague, does not tally as to the stem-leaves with the carefully 

 drawn figure (of which a tracing, kindly furnished by Dr. Rendle 

 from the copy in the National Herbarium, lies before me), nor with 

 the full accovmt in Hortus EUhamensis, pp. 337-8. 1 strongly 

 suspect that he wrote from memory, oi' from im])erfect notes, and had 

 not the passage on which his species depends at hand for reference. 

 D. Don and Smith agree in reckoning S. (jronnlandica as a meie 

 synonym of aS'. cespitosa (the true Lapland plant). I can see 

 nothing in the figure, t. ccliii. f. 329, nor in what Dillenius wi'ote, 

 to separate the two, except that his drawing is decidedly more leafy 

 and stronger in the whole of the lower parts ; a diiierence which 

 might easily be due to an arctic climate. His roots from Greenland 

 Avere planted, and (as he says) "aliquousque gliverunt, sed postea 

 ])erierunt, ob aerem nostrum temperatiorem gelidarum regionum 

 plantis minus faventem " — just as usually happens with our Scottish 

 X cespitosa, when cultivated. He adds that s])eciraens of the same 

 thing are preserved " in Phytophylaceo Sherardino," sent on two 

 occasions bv different authors ; which accounts for the Linnean dis- 

 tribution : ''Habitat in Groenlandia, forte etiam in Pyrenaeis et 

 Helveticis alpibus." That these European plants were really con- 

 specific is prima facie unlikely ; by favour of Mr. G. C. Druce I have 

 examined one of them, which is a mere scrap, and hardly determinable, 

 but certainly not S. cespitosa, vera. I have only seen two British 

 examples which may be S. grcoilandica. The first Avas gathered by 

 Mr. Druce on Ben Lawers, and has been so named by Englei*. The 

 material is scanty, and rather far advanced ; it reminded me of very 

 dwarf S. Sternhergii, the leaves — at this stage — being glabrous ; but 

 I was not aware, when I saw them, how important this and the 

 Sherardian examples might be. The other is the cultivated plant 

 from Kew, in Herb. Smith, alread}^ mentioned. 



But this case is still further comiDlicated by the fact (as 

 Mr. Williams has informed me) that in Syst. Nat. ed. 12, p. 309, 

 Linnaeus quoted Flora Danica, t. Ixxi, as representing his S. groen- 

 landica. That figure, as w^as stated by Mo?nch, when publishing 

 S. rosacea, well depicts >S'. decipiens Ehrh., Exsiccata, No. 5 ! Conse- 

 quently, much of the "■ (jvoenlandica'" in our jniblic collections from 

 circumpolar countries is this ; and some good botanists claim that 

 ^.groinlandica^ being an older name, ought to displace H. rosacea {deci- 

 piens). My answer is threefold : — 1. Linnaeus did not know his own 

 species. 2. K nomen incert^imismxdiW^. 3. The Dillenian plate can- 

 not be S. rosacea ; and the description of the up])er root-leaves : " lajte 

 virentibus, crassiusculis " is quite different from the greyish villous 

 foliage of that species, while agreeing well enough with S. cespitosa. 



Haworth, whose judgement was much less sane, on the whole, 

 than either D. Don's or Smith's, and whose inadequate definitions 

 often make his species and varieties almost impossible to identify, 

 added to the muddle by referring the S. tridactylites groenlandica 

 of Dillenius to *' S. mnscoides Jacq." [Wulfen in Jacq. Misc. ii. 

 ]x VliV], which is a plant of the Alps, Pyrenees, <S:c., and not arctic; 

 alsd by placing -S'. granlandica L., as a different species, among his 



