Page:The Botany of the Antarctic Voyage.djvu/45

Campbell's Islands.] them both numerically and in the amount of space they occupy. In Antarctic America they are represented by a very few Stellat&aelig;, which group is here entirely absent. As no other order exhibits so remarkable an excess, they probably balance the strangely disproportionate want of Composit&aelig;, which appear to have almost as few representatives in proportion to the mass of exogenous vegetation as any other island. Comparing the dicotyledonous vegetation of the Falkland Islands with that of Lord Auckland's, it will be seen, that in the former the Composit&aelig; are to the other Dicot. as 1:2.8, and that Rubiace&aelig; (Galium) are to Compos. as 1:21: but in the latter group, Compos. are to the other Dicot. only as 1:4.5, and Rubiace&aelig; to Composit&aelig; as 1:1.6! If in each we add these two Nat. Orders together, it will be found, that in the Falklands the proportion which the sum of Rubiace&aelig; and Composit&aelig; bear to other Dicotyledonous plants, is as 1:2.7, and in Lord Auckland's group as 1:2.3: proving, that as far as these two remote localities are comparable, Rubiace&aelig; only balance in the latter the want of what is generally, in all climates, the preponderating natural order. This is one only of many equally singular proofs, which a little patient investigation may deduce, that a harmony exists and may be traced in the vegetation of remote climates, whose Floras are otherwise totally dissimilar.

XVI. Fig. 1, a ripe berry, ''nat. size; fig. 2, transverse section of do., showing the nucules; fig. 3, nucules removed; fig. 4, transverse section of the latter, showing the seed; fig. 5, lateral, and fig. 6, front view of a seed; fig. 7, vertical section of do.:—all magnified''.

B. Flowering portion from Tasmanian specimens, ''nat. size; fig. 1, a male flower; fig. 2, a female flower:—both magnified.'' 

 1. depressa, Banks in G&aelig;rtu. i. t. 26. ''et Icon. ined. Plant. Nov. Zel. in Mus. Brit.'' t. 22. ''Forst. Prodr.'' n. 501. ''Smith, Icon. ined.'' t. 28. ''Carmichael in Linn. Trans.'' vol. xii. p. 505. ''Gaudich. Flor. des lies Malouines in Ann. Sc. Nat.'' vol. v. p. 104. Gaud, in Freycinet, Voy. p. 135. ''D'Urville, Flor. Ins. Mal. in Annal. Soc. Linn. Paris,'' vol. iv. p. 612. ''Pet. Thouars, Flor. Trist. d'Acun.'' p. 42. t. 10. DeC. Prodr. vol. iv. p. 451. ''A. Cunn. Flor. Nov. Zel.'' l.c. p. 208.

. Lord Auckland's group; creeping amongst moss in the woods, where its bright red berries give it a pretty appearance.

My specimens are unfortunately not in flower; they however entirely resemble the figures of N. depressa above quoted, and agree with numerous Falkland Island and other southern specimens of that plant with which I have compared it. In Mr. Cunningham's 'Flora of New Zealand,' its precise habitat is omitted; but it is inserted in a MS. copy of that 'Flora' which formed part of my library at sea. There he mentions the "Falls of the Keri-Keri river" as the only locality in which he gathered it. In botanizing over that spot repeatedly in September and October 1841, in company with Mr. Colenso, we often met with Cunningham's plant, both there and afterwards in other moist places near cataracts; it is however entirely different from the true N. depressa, being much smaller in all its parts, with narrower and more acuminated leaves. The berries of the Auckland Island specimens are very much vertically depressed, and their structure is entirely that of the genus Coprosma.  



Capitulum sub-12-florum; floribus exterioribus 8–10, f&oelig;mineis, 2 serialibus; interioribus abortu masculis; omnibus ut videtur tubulosis. Involucrum octophyllum, subbiseriale, squamis inter se sub&aelig;qualibus oblongo-lanceolatis obtusis trinerviis, nervis latiusculis pellucidis transversim septatis. Receptaculum nudum, minutum, convexiusculum. Corolla tubulosa, basi globosa, medio cylindracea et constricta, ore obliquo