The paper is the first of several to present the results of a study, over the last 12 years, of alpine vegetation of the Bogong High Plains (north-eastern Victoria, 5400–6200 ft elevation). Attention is focused on the grassland communities and the response of soils and vegetation to the exclusion of cattle grazing. The first paper outlines the problem, briefly describes the major grasslands, and discusses the climatic and biotic factors concerned.

A 10-year fencing experiment in the eroded grassland, type C, of the Bogong High Plains has established the value of the point analysis method for the quantitative study of continuing change in vegetation. The results show that the exclusion of cattle from this widespread type of subalpine grassland results in a gradual improvement in the amount of vegetative cover, especially in that provided by the herb Celmisia. Over the 10-year period there have been striking soil changes in the fenced plot, resulting in an increase in organic matter and in field capacity, and a significant decrease in bulk density. These findings enable one to make recommendations on catchment management in this area. They have also led to the hypothesis that the eroded grassland has been derived from a herb-field similar to that still existing in the higher parts of the area.

An initial survey of the diversity of early lignotuber development in Eucalyptus and an analytical study of the anatomy of young lignotubers and the seedling stem are presented. Studies of the early stages of the morphological development of the lignotuber in 13 species, representative of five taxonomic groups, resulted in the recognition of four modes of lignotuber initiation. The importance to lignotuber formation of the presence of a suite of accessory buds, adaxial to the axillary bud, is emphasized but lignotuber initiation is not in all cases associated with these buds. Lignotuber buds are derived by branching from existing buds, ultimately from the accessory buds of the node. Following its initiation, the possibilities of later morphological development of the lignotuber are discussed. Lignotuber growth may dominate over stem growth and the lignotubers at a node may then fuse laterally to encircle the stem. Stem growth, on the other hand, may dominate over lignotuber growth and the lignotuber then appears to regress. The consequences for the growth habit of the plant of these alternative pathways of development are outlined. The wood of young lignotubers (and that of the swollen hypocotyl) is shown to be different in composition and in the sizes of its elements from that of seedling stem wood; these differences owe their origin to differences in the nature and performance of the cambia of the lignotuber and stem. In lateral fusion of the lignotubers at a node, and their upward and downwards extension over the stem, e.g. over the hypocotyl, stem cambial initials are either progressively lost or, more likely, converted to lignotuber-type initials. The possibility of the reverse process occumng in stem dominance is discussed.

Most (94 %) of the known eucalypts have been examined for the presence of oil glands in vegetative tissues other than the mesophyll and the primary cortex of the stem. In certain species oil glands occur in the pith and, in a few cases, in the midrib as well. In others (series Corymbosae) oil ducts occur in the pith and in the midrib. In others again (Macrantherae Normales) glands are formed in the phloem of both root and shoot. As a general rule, species which have glands or ducts in the pith do not have glands in the phloem, and vice versa. The only exceptions are provided by nine species of the "eudesmioid complex" (defined below). Of eucalypt species, 58 % have no glands in either the pith or the phloem. It is believed that glands never develop in the primary cortex of the root but always in that of the shoot. The occurrence of glands in the pith or the phloem is constant in many species but others show great variability in the numbers of glands present. Possible explanations for the observed variability are suggested and further lines of investigation are indicated. The age of the tree at which glands first appear in the bark varies widely from one species to another and less widely within species; glands appear precociously in root bark. The age of the secondary phloem itself at which it produces glands also shows wide variations between species. In the Corymbosae the full development of the oil duct system is closely correlated with the ontogenetic juvenile-adult foliage change. The taxonomic implications of the presence or absence of glands in the pith and phloem are discussed. The characters have proved useful for field and herbarium identification. The presence or absence of pith glands is particularly valuable in dealing with Western Australian species.

Rows of pectic strands, each 0.3-0.4�m in diameter, are shown to connect palisade mesophyll cells in regular ladder-like configurations ('pectic scala'). These structures are illustrated in some species of eucalypts, but probably occur in other kinds of plants. Less regular wall filaments can be observed in the intercellular spaces between other types of cells. They are particularly numerous in the parenchyma of species of ferns. These filaments and the pectic scala are readily observable by scanning electron microscopy, but can also be seen in conventional preparations for the light microscope. The structure, formation, chemical composition and possible function of these and other kinds of cell wall protuberances, described in the literature, are discussed.