The ecology of two rare Chamelaucium species (Myrtaceae) from Southwestern Australia

Studies were carried out into various aspects of the ecology of Chamelaucium erythrochlorum m.s. and C. roycei m.s. (Myrtaceae), two declared rare and endangered shrubs restricted to the south west of Western Australia. Initially a soil and vegetation survey was a carried out at the site of all known populations of C erythrochlorum and C. roycei and at some nearby environmentally similar areas where the species did not occur. In addition, aspects of the water relations, phenology and reproductive biology of both species was investigated. Measurements of various floral characters and leaf length were compared for a number of the populations of C. erythrochlorum. Chamelaucium erythrochlorum was found on a range of soils from the infertile gravelly soils at Dardanup Block near Bunbury to fertile red-brown alluvial loams at the Yoongarillup site near Busselton. The soil of the C. roycei sites in the Tutunup-Ruabon area near Busselton varied from moderately acid, brown sandy loams to red-brown loams over a sheet laterite hardpan at a shallow depth. Classification of the C. erythrochlorum quadrats produced four community groups, determined primarily by understorey species composition. These groups were partly based on geographic factors and partly on topographic position. All quadrats were in the Eucalyptus marginata - E. calophylla or E. marginata - E. haemotoxylon forest structural type. Classification of the C. roycei sites and other heathland sites, produced three vegetation groups, with two of the groups containing C. roycei. All of the vegetation groups are structurally mid-dense to dense heathland 0.5 to 1.2 metres high with occasional emergents to 3 metres. Studies into the water relations of C. erythrochlorum, Bossieaea omata and Hypocalymma robustum at Dardanup Block showed that pre-dawn xylem pressure potentials (XPP) were not significantly different between the three species within the C. erythrochlorum "band" in either summer or autumn. C. erythrochlorum was apparently not restricted to this "band" because of increased moisture availability to the adult plants. Mid-morning XPP only was measured at the C. roycei site (Tutunup), where C. roycei had the second lowest XPP of the six species measured although it was not significantly different from any of them. Some of the heathland species, but not C. roycei, showed no fall in late-moming XPP between December and March which indicates that they were probably accessing abundant water below the laterite hardpan. Shoot growth of C. erythrochlorum in 1992/93 at Sabina River started in mid July which was about two weeks earlier than at Dardanup. Flowering was also more advanced at Sabina River (August compared to December) and the flowering season lasted considerably longer (until May) than at Dardanup Block where it finished in February. More plants in the long-unburnt area at Sabina River produced flowers than in the recently burnt area, and they also produced more flowers per plant. In contrast to those at Sabina River very few of the C. erythrochlorum plants monitored at Dardanup Block flowered over the three seasons of the study. Many of the flowers of C. erythrochlorum at Sabina River were damaged by ant grazing in February. The damage in most cases was caused by the insect chewing a hole in the ovary or eating part of the stamen or petals. Compared to Sabina River there was very little insect grazing of C. erythrochlorum at Dardanup Block. All adult plants censused at Sabina River survived over the two and a half years of the study. However 15% of the monitored plants at Dardanup Block had died by the end of October 1993. Most of the plant deaths occurred in late autumn or early winter, primarily from the effects of the summer/autumn drought. Shoot growth in C. roycei started in June and extended through to late January. Flower formation, which was prolific, tended to coincide with shoot growth and plants in the wetter area had the longest flowering season. By February 1992 almost three quarters of the adult C. roycei at Tutunup had died, apparently from drought. All of the seedlings in the burnt area at Tutunup died in mid-summer while only 17.5% of the seedlings on the railway embankment did so probably because those on the embankment were more mature. A leaf and floral morphometric study showed that the populations of C. erythrochlorum could be split into those with larger flowers with longer styles and those with smaller flowers with shorter styles. The flowers at Dardanup Block are intermediate between the two groups. The C. erythrochlorum populations can also be split into two groups in regard to leaf length. A study of ovule and pollen numbers in both rare species showed that the most common number of ovules/flower for C. roycei is 8 (62% of flowers) and that for C. erythrochlorum is 9 (37%) though 10 and 8 ovules/flower were also quite common. In C. erythrochlorum from Sabina River the lower ring of anthers had 20 to 30% more pollen grains than the upper anthers. There was a large difference in the proportion of apparently inviable pollen between the large-flowered and small-flowered types. The total number of pollen grains per ovule (pollen/ovule ratio) for C. erythrochlorum was about 610 for the large-flowered and small-flowered types assuming 9 ovules/flower which indicates that this species is facultatively xenogamous. However the high proportion of inviable pollen in some plants, and the fact that their ovaries develop only a single ovule, will have a large bearing on the reproductive success of individual plants. The lower anthers in C. roycei had more pollen grains than those in the upper whorl, though the difference between the two anther positions was less than for C. erythrochlorum. The proportion of apparently inviable grains in C. roycei was much lower than in C. erythrochlorum. The pollen/ovule ratio in C. roycei from Tutunup is about 540, assuming 8 ovules/flower which indicates facultative xenogamy. However, retention of self pollen on the stigma would seem to promote autogamy, or at least prevent cross pollen from reaching the stigma