338 Transactions. — Botany. 



B. Cauline leaves petiolate. 



7. G. saxosa, Forster. In Act. Holm., 1777, p. 183, t. 5. 

 Plate XXVIIc. 



Perennial ; stems prostrate or ascending, 3in.-5in. long, 

 sometimes inclined to be naked above. Leaves cartilaginous 

 or fleshy, radical rosulate, spathulate, fin.-l^in. long, nar- 

 rowed into long petiole ; cauline smaller, petiolate, sometimes 

 distant. Flowers terminal, solitary or in 3-5-flowered cymes. 

 Calyx about one-third as long as the corolla, or shorter; 

 segments fleshy, subulate, recurved at the tips ; corolla rotate, 

 segments rounded; ovary stipitate. Forster, Prodr., n. 132; 

 A. Rich., Fl. Nov. Zel., 202 ; Willd., Sp. PL, i., 1357 ; A. Cunm, 

 Precurs., n. 398; Grisebach, in DC, Prodr., ix., 99; Eaoul, 

 Choix, 44 ; Hook. 1'., Fl. N.Z., i., 178, et Handbook N.Z. FL, 

 190 (in part). G. saxosa, var. recurvata, T. Kirk, in Trans. 

 N.Z. Inst., xvii. (1884), 224. G. hookeri, J. B. Armstrong, 

 in Trans. N.Z. Inst., xiii., 340 (not of Grisebach). 



Hab. South Island : On maritime rocks, Dusky Bay ; 

 G. Forster, 1773 ; A. Menzies, 1791. Northern shore of 

 Foveaux Strait, Ruapuke, Dog Island, Green Island, and 

 other places in Foveaux Strait ; Earotonga Island, coast of 

 Stewart Island ; T. Kirk. Ascends the Bluff Hill to 800ft. ; 

 F. W. Hutton I Only found in situations exposed to the sea- 

 spray. 



Most closely related to G. cerina, Hook, f., the sepals of 

 which sometimes exhibit a slight tendency to become recurved. 

 It is singular that this plant should not have been observed 

 from the time it was collected by Menzies in 1791 until it was 

 collected by Professor F. W. Hutton on the Bluff Hill in 1873, 

 and it is still more remarkable that although so easily dis- 

 tinguished it has been confused with G. montana and other 

 species. Mr. N. E. Brown, of the Kew Herbarium, who has 

 kindly compared my specimens with the types in the Liver- 

 pool fasciculus of Forster' s plants recently presented to Kew, 

 assures me that it exactly corresponds : it is well known that 

 the fasciculi distributed by Forster are of uneven value, and it 

 may well be that in the set originally consulted by Sir Joseph 

 Hooker the present plant was represented by something else. 

 Mr. J. B. i\rmstrong, in his description of G. hookeri, does not 

 mention the recurved calyx segments, and has further con- 

 fused the plant with the alpine G. saxosa y of the Hand- 

 book — the G. corxjmbifera of this paper, a robust erect species 

 which bears no resemblance to Forster's plant. 

 • 



8. G. cerina, Hook. f. Fl. Antarctica, L, 55, t. 36. 



Perennial; stems few or many, lin.-15in. long, suberect, 

 decumbent or prostrate, slender or stout, leafy. Leaves 



