198 THE JOURNAL OF BOTANY 



represented here). At present I am strongly inclined to believe 

 that we have at least three or four good species besides those 

 recognized in our text-books ; but the sjTionymy is very difficult, 

 and the task of distinguishing the segregates is complicated by 

 the fact that they certainly hybridize with much readiness in 

 cultivation, and probably also in nature, where different species 

 occur together. 



Saxifraga eiegans Mackay. An authentic specimen in the 

 Edinburgh Herbarium is clearly S. umhrosa L. var. jjunctata 

 (Haworth) D. Don x S. Geum L. var. serrata Syme. I gathered 

 very similar specimens on Brandon, at 2500 ft., and have found 

 various forms of the same hybrid on this mountain, Connor Hill, 

 &c. The small form of S. Geum var. dentata, often called ' eiegans,' 

 is quite different. 



S. hirsuta L. The plant of the Linnean Herbarium, which 

 appears to be the normal Pyrenean form, has crenate leaves, with 

 a very broad cartilaginous border ; it is glabrous but for some 

 scattered hairs on the petioles. In Ireland the type seems to be 

 extremely rare ; but there is a specimen in Smith's herbarium, 

 collected at the Gap of Dunloe in 1801 by Mackay, which plainly 

 comes under it. The leaves are oval, mostly with a shallow 

 cordate base, and have a broad cartilaginous border; lamina 

 glabrous ; petiole long, rather slender, very haiiy (the name 

 hirsuta seems to have been given in contrast to S. umhrosa, and 

 is not ver^' appropriate). The prevaihng Irish form of S. hirsuta 

 clearly deserves varietal distinction, and may be described as 

 follows : — " Yar. acutidens, n. var. Foliorum lamina dentibus 

 acutis serrata (neque crenata), basi subti-uncata, rarius leviter 

 cordata, margine cartilagineo angustissimo, in pagina superiore 

 ssepius pilis crispuHs validis albidis munita." This is parallel to 

 S. Geum var. serrata Syme ; which is by far the most frequent 

 Irish form of .S. Geum, the crenate-leaved type being quite rare. 



S. ? hirsuta Jj.YSiV.acutideiis x umhrosa yq.y. punctata. Connor 

 Hill, in a small form ; no doubt it is of fairly frequent occurrence, 

 as these plants cross very freely. Some leaves are broader than 

 long ; this points to a hirsuta rather than a Geum origin. 



S. Geum var. serrata x umhrosa var. serratifolia (Mackay) 

 D. Don, n. hybr. (Kef. No. 3611). Verj' scarce on Connor Hill. 

 The inflorescence closely resembles that of S. serratifolia (which 

 appears to me to deserve at least subspecific rank, and keeps 

 thoroughly constant after eight years' cultivation) ; petiole some- 

 what flattened, with a slightly raised, membranous, long-ciliate 

 margin, usually rather dilated at the top ; lamina coriaceous, 

 markedly cordate-truncate at the base, deeply and sharply serrate- 

 dentate, with a narrow cartilaginous margin, ovate to oblong in 

 outline, nearly parallel-sided, narrowed above to the acute or 

 rounded tip, punctulate on both surfaces, greyish beneath, mostly 

 with some scattered, stiff, whitish hairs, especially above, but 

 sometimes glabrous. This is a handsome, striking plant, which I 

 had not seen before ; it should also be found in the Gap of Dunloe 

 and other places where the parents are associated. 



