Cockayne. — On Celmisia'lindsayi. 349 



From the above it may be seen that the Nuggets plant has 

 leaves longer considerably but narrower in proportion to the 

 Bonpland plant, which consequently gives them a different 

 shape. 



An examination of the literature on the subject, of plants in 

 my herbarium, and of notes which I took personally some years 

 ago at Nugget Point, shows that the Nuggets plants vary con- 

 siderablv in dimensions of leaf. For instance, my notes taken 

 on the 3rd October, 1902, give, as an average size of certain 

 leaves measured, 10 - 5 cm. by 1*9 cm., and a specimen collected 

 by me in the same locality some months later gives the following : 

 9 - 7 cm. by 2 3 cm.; 10 cm. by 22 cm.; 84 cm. by 19 cm.: 

 9-8 cm. by 2 -3 cm. ; 7-5 cm. by 1-7 cm. : 7 -2 cm. by 1-8 cm. 

 A Catlin's River specimen collected by D. Petrie measures 

 14-6 cm. by 25 cm. ; 112 cm. by 24 cm. ; 14-5 cm. by 2-3 cm. 

 A cultivated specimen sent me by T. Kirk of the Nuggets plant 

 measures 136 cm. by 17 cm. : 132 cm. by l - 9cm; 129 cm. 

 by 1 '9 cm. Hooker gives the dimensions of Lindsa)"'s speci- 

 mens as 6 - 4 cm. by 1*25 cm., and the plate in Lindsay's paper* 

 shows a comparatively small plant. But these figures on the 

 whole confirm the statement above, that the coastal plant has 

 longer but narrower leaves in proportion to their length than 

 the Bonpland plant. 



In my herbarium is also a flowering specimen of the alpine 

 plant collected in Matthews's garden by Petrie. This, so far 

 as the flower goes, exactly matches Nuggets specimens. Its 

 leaves measure 9 "3 cm. by 24 cm. ; 9'8 cm. by 2 - 35cm. The 

 scape of the alpine plant is of the characteristic flexuous form, 

 which Hooker called attention to as a peculiar characteristic 

 of C. lindsayi. Petrie (I.e., p. 558) remarks that the Lake 

 Harris specimens of Matthews " have a more robust habit than 

 the sea-coast form." 



In order to further test the differences between the two 

 plants. I cut a number of transverse sections of the leaves. 

 These, though agreeing in the main as to palisade, spongy paren- 

 chyma, &c, show one difference — viz, that the Bonpland plant 

 has invariably a two-layered epidermis, while the Nuggets plant 

 has one of one layer only, though the latter is occasionally two- 

 layered for a short distance. Thus, this anatomical distinction, 

 again, is one rather of degree than of actual difference. 



Such are the facts of the case with regard to the alpine and 

 coastal plants. There is undoubtedly a Celmisia on the Hum- 

 boldt Mountains, common also in many places, although Kirk, 

 Petrie, and myself, who have all been in the vicinity of its habitat, 

 have never collected it. Moreover, this alpine Bonpland form 



""Contributions to New Zealand Botany,'' 1868. 



