Approved NSW and National Recovery Plan for the Grevillea beadleana
Threatened Species Unit, North East Branch
New South Wales Department of Environment and Conservation, 2004
ISBN: 174122 135 8
5. Life History
- 5.1 Flowering and fruiting phenology
- 5.2 Breeding system
- 5.3 Floral visitors
- 5.4 Seed dispersal, predation and dormancy
- 5.5 Germination
- 5.6 Response to fire and regeneration
Flowering mainly occurs in late spring or summer, but with good rains it can continue into early winter (C. Gross pers. obs.). The basal flowers of the inflorescence are three times more likely to set seed than the distal flowers (Gross & Smith unpublished data).
Grevillea beadleana is a self-compatible species. All pollination treatments at Guy Fawkes River National Park in 1995 yielded lower fruit set than at Binghi which may be the result of a resource deficiency (eg. lack of water) at the former site (Gross & Smith unpublished data).
Grevillea species are often abundant nectar producers with biotic pollen vectors being prominent as a result. A number of potential biotic pollinators have been recorded attending Grevillea beadleana flowers. At Binghi these include Yellow-tufted Honeyeater (Lichenostomus melanops), Spiny-cheeked Honeyeater (Acanthagenys rufogularis), White-naped Honeyeater (Melithriptus lunatus), Brown-headed Honeyeater (Melithriptus brevirostris ) and Fuscous Honeyeater (Lichenostomus fuscous). A small, unidentified native mammal was also briefly observed lapping at an inflorescence at this site (Smith pers. comm.). Crimson Rosellas (Platycercus elegans) were observed ripping flowers from inflorescences. At Guy Fawkes River National Park Eastern Spinebills (Acanthorhynchus tenuirostris ) have been the only birds observed at inflorescences of G. beadleana. European Honeybees (Apis mellifera) and some native bees have been observed foraging for nectar at both locations. Honeybees rob nectar from unopened flowers that may potentially render these flowers unattractive to native pollinators.
The seeds of Grevillea beadleana are about 9 mm long and 4 mm wide and are held in the pods for about four weeks once the pods have developed. Indications are that there is little predation of the seed while it is still within the pod. At Guy Fawkes River National Park the seeds on the ground appear to have been eaten by rodents and a trapping program may be needed to investigate which species are eating the seed and the level of this predation. Significant seed predation levels would affect the amount of time needed for seedbank replenishment after a fire.
Studies on seed predation of the closely-related, fire-sensitive and endangered shrub Grevillea caleyi found that replenishment of the soil seedbank is severely retarded by mammal seed predation, with seed loss levels reaching 90% at sites that were recently burnt (Auld & Denham 2001). Auld and Denham (2001) also argued that as seed production is lower in younger plants, seed accumulation in the seedbank would not occur for 46 years after a fire. With further depletion of the seedbank due to seed predation in the post-fire period, it was estimated that after a fire G. caleyi should be protected from the risk of a future fire well beyond three times the primary juvenile period in the species or local extinction will occur (Auld & Denham 2001). Thus a fire-free interval of around 15 years would be required to allow accumulation of seed in the soil. It is considered that such findings and implications are likely to apply to Grevillea beadleana and thus a fire-free interval of 15 years is considered to be an appropriate strategy to minimise the risk of ongoing depletion of the soil seedbank and the likelihood of local population extinction.
Research into Grevillea caleyi estimated a seed half-life of between 6 and 10 years and found that seeds exhibited a seasonal pattern of non-dormancy, possibly due to temperature or darkness explaining why non-dormant seeds do not germinate in the field (Auld & Bradstock 2000). It is likely that G. beadleana exhibits similar characteristics, although insufficient research has been undertaken to determine whether this is the case.
Heat has a positive effect upon the germination of a variety of species within the Proteaceae (Leigh et al. 1984; Edwards & Whelan 1995; Bradstock et al. 1996). This fire-triggered germination is a factor associated with innate dormancy, a characteristic of many Australian plant species (Edwards & Whelan 1995).
Durbin (1996) investigated the effects of heat on G. beadleana seeds collected from the seedbank at Binghi. Four samples of seed were heated for 10 minutes at 23oC (as a control), 50oC, 100oC and 165oC respectively then transferred to a growth cabinet and monitored for germination which occurred after about 20 days (Table 2).
Seeds do not need to be exposed to high temperatures for germination although rates may be enhanced by an appropriate heating event. At the end of the germination experiments ungerminated seeds were tested for viability (tetrazolium and indigocarmine) and many were shown to be still viable, ranging from 100% viability for the 23o control group to 50% for the 165o group (Table 2).
|Temperature0C||Percentage Germination||Sample Size||Percentage Viability|
The harvesting of seeds in the seedbank using a trowel and sieving to separate them from the soil may have inadvertently scarified them, thereby assisting the uptake of water and promoting germination at abnormal temperatures. Such scarified seeds may also respond differently to the heat treatments (e.g. by being killed by very hot temperatures). The effect of heat on germination needs to be investigated using seeds harvested passively from bushes, and replicated using a larger sample size to investigate the trends apparent in the data, ie. that 100oC is an optimal temperature to trigger germination and that 165oC may kill seeds.
Grevillea beadleana seeds were excavated from the leaf litter layer and soil beneath the litter to a depth of 4 cm. About 40 seeds per m2 were encountered in the soil layer, compared with about 9.5 seeds per m2 in the leaf litter layer. Intact seeds from both the litter and soil layers were found to be viable using tetrazolium and indigocarmine.
Grevillea beadleana plants are very often killed by fire, although coppicing does occur, but mainly from young plants (Dwyer unpublished data; Gross pers. comm.). G. beadleanas regeneration strategy is very similar to that of other obligate seeding species such as G. caleyi and G. barklyana. These species must replenish seed numbers before the next fire if populations are to be maintained, as the mature plant is usually killed by fire (Gill 1975; Auld et al. 1993; Edwards & Whelan 1995; Scott et al. 1995). In the closely related endangered shrub Grevillea caleyi, it was found that plants were killed by fire, which also promoted the breaking of seed dormancy in the soil seedbank (Regan et al. 2003). Regan et al. (2003) found that where seed predation levels remain high, the optimal fire management strategy is longer than when seed predation levels are reduced by 2030%.
In 1988 fire burnt 1.5 hectares of the 4.25 hectares containing the eastern population in Guy Fawkes River National Park and killed 321 plants out of 647 leaving 326 live plants. In 1989, 388 seedlings were present in the burnt area and 21 in the unburnt area (Dwyer unpublished data). In the same area in August 1995 after a very intense fire nine months earlier, 84 dead mature plants, 117 plants greater than 2 years old and 450 seedlings were recorded (Streat unpublished data). Thus in six years the population had been reduced from 735 to 567 as a result of two fires occurring in the area within those six years.
A survey in March 1996 of about 3.5 hectares of the eastern population in Guy Fawkes River National Park revealed two dead adults, 59 plants greater than two years old and 389 seedlings (Gross unpublished data). Drought stress appeared to be the cause of some of the plant deaths recorded by Gross et al. (1995). It would appear that less than 20% of the population might be capable of flowering and fruiting in the two to three years after a fire (Gross unpublished data).
In contrast to the fire prescription required at Guy Fawkes River National Park, a fire would be beneficial in some parts of the Binghi population where seedling recruitment is poor and adults appear to be senescent. Durbin's (1996) work reported that a seedbank of viable seeds exists in at least one of these senescent demes.
Frequent fires are also likely to lead to a narrowing of the genetic base if only a small proportion of the population is contributing seeds to the seedbank. This, in combination with the time required for juvenile plants to reach maturity and the likely increased seed predation levels after a fire, indicates that a fire frequency of around one in 15 years is considered likely to be optimal to allow the maturation and expansion of the population while avoiding dominance of senescent age classes.
The population in Chambigne Nature Reserve is so small that it is particularly vulnerable to fire and other stochastic events. It is not known how large the seedbank is at this site although the lack of soil on the rock ledge suggests that the seedbank would be limited to a few seeds in the rock crevices. It is considered that the risks to the very few plants at this site are too great to employ a control burn at this stage. Such a control or managed burn would at this stage be likely to kill existing plants, some of which may not have reached maturity and would also further deplete the existing seedbank. A control burn may, at some future stage, provide an appropriate management strategy to prevent unplanned fire entering the site, but the risk of burning the site during such a burn is currently considered unacceptable. Extinction of genotypes from this population due to wildfires and other events has been partially avoided by the successful propagation of material from four individuals at this site (see propagation section below).