A Multivariate Growth-form Analysis of Grassland and Forest Forbs in South-eastern Australia

Abstract

The growth-form composition of grazed and unburnt, grassy forest remnants and ungrazed, frequently burnt, anthropogenic native grasslands in Gippsland, Victoria were compared, using a multivariate, clustering analysis of the growth-form and life-form attributes of 53 forb species. Groups comprising (1) annual forbs, (2) clambering, repent and decumbent perennials, and (3) rosette perennials and rhizomic ground-cover forbs occurred in significantly more forest than grassland quadrats. One group, mostly containing tall erect geophytes with linear basal leaves, occurred in significantly more grassland than forest quadrats. Grassland quadrats contained significantly more tall forbs (> 20 cm) than forest quadrats, whilst forest quadrats contained significantly more forbs of short to medium height (< 20 cm). There was a significant, positive correlation between plant height and frequency of occurrence in grassland quadrats (r_s = 0.58, P < 0.001), and a significant, although weak, negative correlation in forest quadrats (r_s = -0.29, P < 0.05). Short forbs are likely to have been depleted from grassland sites owing to competition from the dominant tussock grass, Themeda triandra Forsskal. By contrast, ground cover in forest sites is of relatively low stature, biomass and cover, allowing short forbs to persist. The relative paucity of tall forbs from forest remnants is suspected to be at least partly due to intensive stock grazing in the past.