SPECIES OF RANUNCULUS. 67 





Stem-leaves pinnatipartite, often three-cleft, nearly or quite sessile, 

 the divisions linear or lanceolate ; uppermost leaves simple, lanceolate or 

 linear entire. Sepals horizontally spreading. Eeceptacle more or less 

 villose at the base of the spike. Beak of achaenia hooked. In these 

 four points there is no constant or marked difference between the pre- 

 sent plant and the two following. They are here added therefore once 

 for all. 



In habit, size of parts and stature, this species resembles much more 

 the Canarian than any form or state of the Madeiran plant. This is 

 perhaps conformable to the greater dryness of the Cretan and Canarian, 

 compared with the Madeiran climate, however disaccordant with the 

 geographic station of the plants. 



In the Appendix to the second edition of c Primitiae Faunae et Florae 

 Mad.,' I was induced by the observations of my late friend Dr. Charles 

 Lemann, made in 1844, to refer the Madeiran R. grandifolim to this 

 species. But after recent careful re-examination of the very same ma- 

 terials in the Banksian and Hookerian Herbaria, on which he formed 

 his opinion, with later additions of importance, I am constrained to 

 draw a different conclusion. R. Creticus, L., differs from R. grandi- 

 folim in its altogether smaller size, humbler stature, weaker, more 

 slender habit, more copious, softer, spreading, cottony pubescence, 

 smaller, reniform, softer, coarsely and remotely inciso-crenate leaves, 

 with few, short, broad, bluntish or rounded teeth or lobes, smaller, 

 fewer, not corymbose flowers, larger spikes and achaenia (the latter 

 often hispid), and tufted, not palmate root, with slender instead of 

 fleshy, thick divisions : approaching in fact nearer to R. cortuses/olius, 

 Willd., though certainly distinct from that species by its copious, soft, 

 villose pubescence, simple, not bulbous, cottony or silky, not substri- 

 gose or setose hairs, light green, unspotted leaves, and remarkably 

 short, thick, globose-ovate or oval spikes, with larger achaenia. 



R. Creticus, L., should be placed between R. cortusafolius, Willd., 

 and R. grandifolius : both of these being more nearly allied to it thau 

 to each other. 



R. CORTUS/EFOLIUS, Willd. 



R. sparsim parciusve substrigoso-pubescens, pilis subsetosis brevibus 

 rariusculis, caulinis patentibus superne erccto-pat< utibus, fohorum 

 pnesertim bulbosis; fol. submembranaccis sa'pius nigro-macuJ.M.-s 



