29

C—l 3

As the sand-grass dunes become more stable they afford sufficient shelter for I he seedlings of shrubs to establish themselves on the lee side. Thus, as already shown, shrubs occur to some extent on the sand-giass dunes, becoming more abundant with increase of distance from the shore, until finally they dominate, and a shrub association occupies the ground. Besides capturing sand-grass dunes, the shrubs under discussion arc also dune-builders, their form favouring the accumulation of sand. Especially is this the case with the prostrate, spreading, much-branched Coprosma and Pimelea, which function as veritable sand-traps (see Photo No. '22). the sand accumulating in their interior, but eventually, if the drift continues, burying them altogether. The various species of Cassinia, much taller and more erect plants than the two preceding, function somewhat differently. Their closely branching portion is at some feet above the .sand-surface. The sand is at first held, but not strongly, by the basal stems, and as the drift continues it mounts up to ihe above twiggy portion, where it accumulates more rapidly, so that the shrub may be altogether buried, or a few twigs alone project above the sand, which, unlike those of the Coprosma or Pimelea. cannot lengthen to any noticeable extent. Even in that case the plant is not necessarily doomed, for the. loosely held sand at its base is frequently blown away, the stiff naked stems being again exposed. and the shrub none the worse for its burial. On the other hand, sand held in the network of branches of the Coprosma or Pimelea cannot again be removed, except under very special circumstances. Cassinias at all stages of burial and disinterment are a common feature of the shrub dunes. To a minor degree the sand-shrubs function as sand-binders, since their upper branches some limes put forth roots, thus enabling the plant to grow upwards with the drift, while the cord-like old stems are buried deep in the dunes. But, generally speaking, all the indigenous shrubs of the New Zealand dunes function as SB,ndi-holders rather than as s&ni-binders, such as are certain species of Europe or North America, whose upper steins root freely and grow rapidly—various willows (Salix), and dogwoods (Cornus). The association is grey or yellow in colour, according to the dominance of I he Cassinia or Coprosma, but generally both colours are in evidence. Certain other indigenous plants are common in this association. More or less Scirpus frondosus, or Spinifex hirsutus if near the sea and according to the locality, will be present, especially on the lee slopes, the accumulating of sand there being favourable to their development. Tussocks of the pale green stems of Scirpus nodosus and shining green mats of Calysleijia Soldanella will be sometimes abundant. The grass Calamagrostis Billardieri will bo dotted about. Where the wind is not especially strong I'hormium lena.r and Arundo conspicua may enter in. On the Southland dunes where the sand is coarse the silvery mats of Raoulia australis, so frequently a denizen of stony river-beds, will appear early. There will be usually more or less introduced plants which are able to tolerate a dry station if there bo sufficient shelter, Trifolium arvense, Hypochoeris radieata, Sonchus asper, ami Bromus hordeaeeus being especially common. As for the stability of these Cassinia-Coprosma dunes, all depends upon their position—though if stable they, belong more properly to the next class —and the degree of covering. Where the shrubs quite cover the sand—a not infrequent occurrence —and if. in addition, the dune is on the lee side of a well-fixed sand-plain, it is quite stable, and would remain so for years were there no advance of sand, or did no animals or fire disturb its surface. This stability is important, insomuch as it shows that under certain conditions a dune exposed to wind-tearing action may be naturally covered with shrubs and rendered stable without any previous preparation, except such shelter as is afforded by sand-grasses. (b.) Heath Dunes. —The heath dunes are the third stage towards the evolution of the fixed sandhill, with its loam-covered surface. They occur at a greater distance from the sea than the sand-shrub dunes, or even quite close to the shore where the wind strikes with less power. They are closely allied to shrub dunes, and differ only in being more stable and containing generally more species. Very often 'one or other of the species of Cassinia is the dominant plant. In many localities, however, the manuka (Leptospermum scopariutn) gives the character to the association. The appearance of this shrub is an interesting plant-geographical phenomenon, since it seems at first thought that, the dune-vegetation being correlated with the wind-velocity, manuka must be. to a smaller degree wind-tolerating than are the dune-plants proper. But it is nothing of the kind; on the contrary, it can tolerate more wind than almost any other indigenous shrub, as I have shown elsewhere (Cockayne, 92). On a dune-area it is clear that the presence of the manuka is altogether dependent on the strength of the sand-drift and on nothing else, not even on excess of sea-spray. Thus, in selecting shelter-plants for dune-afforestation purposes, tolerance of drifting sand is a matter of prime importance, without which drought- or salt-resisting power are as nothing. Besides the above shrubs, other species may make up the association. The great tussocks of the toeloe (Arundo conspicua) and the New Zealand flax (Phormium tenax) may be present, and in some localities dominant. These two can tolerate a certain amount of sand-burial. The toetoe, for instance, grows upwards as it is buried, and thus ascends a slowly advancing lee slope, finally possibly capturing it and leading to its fixing (see Photos Nos. 33 and 34). If such lee slopes were not interfered with, and beyond the influence of cattle and burning, many which are now advancing, menacing fertile land, would be naturally fastened. Even under the present adverse conditions stability is in some places being established. In Southland, so far as IJhave observed, true heath dune is absent, but Phormium and toetoe (Arundo conspicua) are common on many hills. There in some places the sand-grass dune gives place, on the immediate ridge behind, to fixed dune, which is perhaps grassed or was originally covered with sou! hern rata forest (Mclrosideros lucida), now replaced by a scrub of species of Coprosma, &c. Heath dune seems wanting in Stewart Island, except on the ancient inland dunes. In north-western and north-eastern Auckland more species enter into the heath than on the dunes of the central botanical province. Styphelia fasciculata, flattened close to the ground, is very common. The wdiite tea-tree (Leptospermum ericoides), of exactly the same habit as the Styphelia, is frequent.