NOTE 66 (page 226)
 
Endemic Genera of Ferns in Hawaii

Hillebrand gives two genera of ferns peculiar to Hawaii, one, Sadleria of Kaulfuss, “scarcely distinct from Blechnum,” and containing four species; the other, Schizostege, constituted by himself, and represented by a single species found in only one or two of the islands.

NOTE 67 (page 241)
 
On the Dispersal of Compositæ by Birds

The goldfinch’s habit of pecking at the heads of thistles, and pulling out the achenes in bundles, is well known. Gätke mentions two suggestive instances of birds feeding on the fruits of a Composite plant. According to this observer, the Scarlet Grosbeak (Pyrrhula erythrina), when it alights on Heligoland, always feeds on the achenes of Sonchus oleraceus, which it picks off the plant; whilst the Parrot Crossbill (Loxia sp.), feeds in Heligoland on burrs and thistles (Heligoland as an Ornithological Observatory, pp. 407, 409). See Note 91.

Haplostachys, Phyllostegia, and Stenogyne, all Labiate Genera.—Phyllostegia is not strictly peculiar to Hawaii, since out of the 17 species enumerated in the Index Kewensis, 15 are Hawaiian, 1 Tahitian, and 1 is accredited to Unalaska (one of the Aleutian Islands). The last locality appears to be an error. The species in question is P. microphylla, Benth.; and on looking up the original authority in Linnæa (vi. 570, 1831), I find the locality is thus given: “insula coralligena Romanzoffii,” which is either one of the atolls of the Paumotu Islands in about lat. 15° S. and long. 144° W., or a coral island of the Marshall Group, most probably the former.... I paid some attention to the suitability of the fruits of these three Labiate genera for dispersal by frugivorous birds, for which the fleshy nucules in the cases of Phyllostegia and Stenogyne apparently fit them. Out of the fruits of five species of Phyllostegia examined by me, the seed-coverings in three species, after the removal of the fleshy covering of the nucule, were too soft for the protection of the seed in a bird’s stomach. Hillebrand also observes (p. 347) that the nucules when dried are wrinkled, and absorb moisture easily, a quality which, if true of all the species, would make the distribution of the genus by birds impossible. However, in two species I found the seed-coverings somewhat harder. It would seem that since birds have largely ceased to disperse these plants, the soft-skinned nucules would in the absence of their selective agency more frequently characterise the genus. It is possible that the dry nucules of Haplostachys, which according to Hillebrand are not affected by drying, represent the original condition of those of Phyllostegia, and that the fleshy character has been acquired in this archipelago. It will be seen in the list on page 263, that Haplostachys is regarded by Gray as a section of Phyllostegia. The remarks under Phyllostegia, regarding the softness of the seed-coverings beneath the fleshy coat of the nucule, also apply to Stenogyne; and Hillebrand, in contrasting its fleshy nucules with the dry nucules of Haplostachys, implies that they absorb water, which, I may remark, would render them quite unfit for dispersal by frugivorous birds.

Touchardia (Urticaceæ).—According to Hillebrand, the solitary species is by no means common in the group now. In 1897 I found it growing abundantly some miles up the Waipio gorge, Hawaii.

Cheirodendron (Araliaceæ).—C. Gaudichaudii, the well-known “Olapa” tree, is common in the forests of all the Hawaiian Islands between 2,000 and 5,000 feet; but I noticed it occasionally at greater elevations, as on the south-east slopes of Mauna Kea, where it extends to 7,000 feet. As described on page 343, the “Olapa” often grows in close contact with the Lehua (Metrosideros polymorpha), the two trunks appearing as one. The drupes would attract frugivorous birds and the pyrenes are well adapted for this mode of dispersal. Mr. Perkins states that the drupes are much sought after by the various species of Phæornis, a genus of birds peculiar to Hawaii.

Deterioration of Fruits for Purposes of Dispersal.—Among fruits or endemic genera that have evidently deteriorated in the Hawaiian Group as far as fitness for dispersal is concerned, may be mentioned, in addition to those of Phyllostegia and Stenogyne above noticed, those of the Araliaceous genera, Pterotropia and Triplasandra, and the Amarantaceous Nototrichium. The pyrenes of the first two genera on account of their thin covering, and the seed of the last-named genus on account of its thin testa, seem ill-fitted now for transport in a bird’s stomach, yet we cannot doubt that their ancestors originally arrived in this fashion. The same principle is also illustrated by some Hawaiian non-endemic genera of later eras that possess peculiar species, such, for instance, as in the case of Elæocarpus discussed in Chapter XXVI.

NOTE 69 (page 366)
 
On the Germination of Cuscuta

My observations were made on the Hawaiian endemic species (C. sandwichiana) and on a Fijian introduced species. Germination occurs readily in fresh water, the floating seedling growing rapidly. When the germinating seed is placed on wet soil in the shade, the seedling grows at the rate of ³⁄₄ inch (19 mm.) a day. The store of nutriment contained in the swollen radicular end will support the seedling for a couple of days, and if it has not then found a host it withers and dies. At first lying prone the seedling then lifts its upper end into the air, and it was almost pathetic to notice it moving round and round, endeavouring vainly to find some object near. The seedlings make no effort to strike into the soil, and when they are allowed to attach themselves to a plant they ascend rapidly, growing at the upper end and dying at the lower end.

NOTE 70 (pages 477, 480-1)
 
On Beach-Temperature

My data are rather scanty; but, judging from observations made in Hawaii, in South America, and in the south of England, the following scale would probably be true of typical beaches where the sand is found relatively cool and moist at a depth of four or five inches. This moisture seems to arise entirely from subsoil drainage seaward. When a beach fronts an arid, rainless region, few if any plants grow on it; the sand is loose, hot, and dry at the depth indicated; and the temperature of the surface half-inch rises to between 130° and 140° F., whilst four inches down it is 95° to 100°. Salt-marshes situated behind a beach even in a desert-region change its thermal behaviour, and it would then be more like a beach skirting a vegetated sea-border in the same latitude. The method of observation was as follows:—An unmounted thermometer of the size of a clinical thermometer, but graduated higher, was placed horizontally in the sand half an inch below the surface and a reading taken. It was then pushed vertically into the sand until the bulb was four inches deep and another reading taken. Provided that the sand is moist beneath, the colour does not seem to make much difference, except perhaps in very dark sands, none of which were tested.