| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Ecosystem Management, University of New England, Armidale, NSW, 2351, Australia
Received for publication March 16, 2006. Accepted for publication October 16, 2006.
ABSTRACT
Plant breeding systems are seldom studied across the breadth of a species' range. For many systems, this precludes an informed assessment of the evolutionary biology of a species, particularly of the factors that shape fecundity. Grevillea rhizomatosa is a threatened species of shrub known only from a 7 x 8 km area and c. 2000 plants in northern New South Wales, Australia. The species reproduces asexually from rhizomatous suckers, and fruit are only produced in a few populations. Over two flowering seasons, we investigated the extent of sexual reproduction and the mechanisms of infertility in five populations that span the range of the species. Seed were produced in three of the five populations. The breeding system varied among populations from obligate outcrossing to facultative outcrossing to fully sterile. Fruit to flower ratios were below 0.13 in the fertile populations but within the range found in other species of Grevillea. Pollinator limitation was not linked to infertility. Stigmatic opening and receptivity were functional in all populations. Interpopulation crosses using fertile pollen failed to recover fertility in an infertile population. A breakdown in female and male fertility mechanisms was found including a fault in the mechanical release of pollen from anthers, <10% viable pollen, and a post-pollen-deposition event that prevents fertile pollen from effecting seed-set. In the infertile populations pollen was not released from anthers, resulting in flowers projecting barren pollen-presenters. Sexually reproducing populations are threatened by the incursion of asexual forms that may be favored by frequent disturbance from wild-fires.
Key Words: Grevillea • infertility mechanisms • pollen augmentation • pollen viability • Proteaceae • restricted species • variable breeding systems
For some species, an absence of sexual reproduction is no handicap to dispersal—some of the world's most invasive species are sterile, yet clonal division of ramets and an effective dispersal mechanism contribute to their wide distribution (Salvinia molesta, Cilliers, 1991
; bamboo, Lazarides et al., 1997
). However, in the absence of an effective dispersal agent, infertility may correlate with rarity (Pilgrim et al., 2004
), and there are many threatened species with restricted ranges that never produce seed (Grevillea infecunda, Kimpton et al., 2002
) or are only partially fertile (Grevillea althoferorum, Burne et al., 2003
) with asexual reproduction also occurring in many cases (Haloragodendron lucasii, Sydes and Peakall, 1998
; Grevillea rhizomatosa, Caddy and Gross, 2006
). Prolonged clonal growth and lack of sexual recruitment have been found to affect within- and between-population genetic structure and the capability for sexual reproduction (Honnay et al., 2006
). The consequences of reduced sexual breeding for the viability of populations include the loss of the capacity to evolve broad ecological tolerances (Andersson, 1994
), and it is of particular concern where a species is threatened with extinction.
Absolute infertility in populations can be caused by pollinator limitation (Thelymitra epipactoides, Cropper, 1989
; Banksia goodii, Lamont et al., 1993
) but is more often due to a variety of aberrations with the fertilization process that may include a lack of suitable pollen donors (Elaeocarpus williamsianus, Rossetto et al., 2004), infertility induced by polyploidy (Spartina xtownsendii, Thompson, 1991
), prezygotic events such as the production of inviable pollen (Grevillea infecunda, Kimpton et al., 2002; G. althoferorum, Burne, et al. 2003
), disruptions to pollen tube growth (Decodon verticillatus, Eckert et al., 1999
), gynoecial malformations (Ulmus minor, Lopez-Almansa et al., 2003
) and postzygotic events such as seed abortion (Coptis teeta, Pandit and Babu, 2003
), or a combination of events (Bambusa vulgaris, Koshy and Jee, 2001
) that can be exacerbated by reduced genetic diversity (Decodon verticillatus, Dorken and Eckert, 2001
). Evaluating the mechanics of infertility and the relative importance of the corrupted processes in preventing or retarding seed set help unravel the evolutionary pathways to sterility. In addition, this information may assist with practical interventions (e.g., pollen augmentation) for the recovery of threatened species particularly where a sterile form may be invading a fertile population.
Grevillea rhizomatosa Olde & Marriot (Proteaceae) is a restricted and threatened species from northern NSW, Australia where shrubs spread asexually through suckers and less often through seed production. Caddy and Gross (2006)
determined that fruit production was completely absent in two of five study populations where only asexual individuals occurred. In the three fertile populations, sexual and asexual individuals co-occurred. Seed set was low (fruit : flower ratios < 13%), but seed were 100% viable (Caddy and Gross, 2006
). Our aim was to determine why some populations are infertile. We set out to determine the type of breeding system in operation in G. rhizomatosa so that limitations to seed set could be assessed. Within the breeding system, we investigated several important components of reproductive function including (1) the mechanics of anthesis, (2) pollen viability, (3) stigma receptivity, (4) the degree of self-compatibility, and (5) floral visitation. We also assessed interpopulation fertility in an attempt to restore fertility in an infertile population.
MATERIALS AND METHODS
Study species and study sites
Grevillea rhizomatosa Olde & Marriott (Proteaceae) is a shrub restricted to Gibraltar Range and Washpool National Parks, northern NSW, Australia (fig. 1 in Caddy and Gross, 2006
). The species is closely related to G. montana and G. arenaria but is unlikely to be a recent hybrid between them because these related species do not co-occur with G. rhizomatosa (Makinson, 2000
). Instead, these three related species are likely to have differentiated as a result of vicariant evolution. The Gibraltar plateau is a high altitude (980 m a.s.l.) granitic area of undulating to steep topography dominated by extensive rock outcrops and granite boulder fields. The area receives c. 1430 mm of rainfall and annual temperatures range from –3.3–35.6°C (Caddy, 2004
). The species is found in an area 8 x 7 km (Caddy and Gross, 2006
), and flowers mostly from August to November with sporadic flowering at other times of the year. Five populations that span the range of the species were chosen for detailed study. Populations were coded north to south as Wash (Washpool National Park, c. 225 plants), Mhut (Mulligan's Hut, c. 250 plants), Swamp (McClimonts Swamp, 165 plants), Dand (Dandahra Trail, c. 250 plants), and Cas (Murrumbooee Cascades, 41 plants). In all populations there were some groups of individuals that appeared connected (i.e., plants were growing in a line from a larger "mother" plant) or were actually connected (determined by excavating soil), <10 cm or up to several meters from the genet plant. The extent of clonality in populations is unknown but is currently being investigated with molecular techniques. Caddy and Gross (2006)
determined that Wash, Dand, and Cas were fertile, and Mhut and Swamp were infertile populations.
Floral structure
Inflorescences in G. rhizomatosa are comprised of loose clusters of paired flowers although through abscission only one flower may be present. Flowers are hermaphroditic, zygomorphic with a perianth of four petaloid tepals each with one fused anther. Nectaries are located as fused hypogynous glands at the base of the perianth. The ovary has two ovules. In common with most of the Proteaceae, G. rhizomatosa has a secondary pollen presentation system, in which pollen from the four anther lobes is deposited onto the specialized subapical region (pollen presenter) of each style before the flower opens. During anthesis, the pollen presenter, with its load of pollen, erupts and projects from the flower as a result of the elongation of the style (see Johnson and Briggs, 1975
, p. 125). This self pollen needs to be dislodged before incoming pollen can be deposited on the stigma. The pollen presenter in G. rhizomatosa is disc-like in shape with central circular depressions from which patches of stigmatic papillae protrude (Caddy, 2004
). In G. rhizomatosa the pollen from each anther can be distinguished on a pollen presenter as four hemispherical pollen mounds, as each pollen mound corresponds with the pollen load from one anther. It is thought that this form of pollen presentation in the Proteaceae may influence the onset of stigmatic receptivity, separation of male and female floral phases (protandry), pollen removal by pollinators, and the amount of outbreeding (Vaughton and Ramsey, 1991
; Howell et al., 1993
; Vaughton, 1996
; Goldingay and Carthew, 1998
).
Anthesis mechanism
Initial comparisons between Mhut and other populations showed differences in the amount of pollen adhering to the stigmatic surface following anthesis. To quantify the presentation of pollen loads, we bagged 50 mature buds using nylon organza in Wash, Dand, and Mhut. After anthesis, all flowers were examined, and the success of anther dehiscence onto the pollen presenter was recorded. A score, from 0 to 4 was given to each flower, which represented the number of anthers (0, 1, 2, 3, or 4) that had deposited a visible amount of pollen onto the stigmatic surface.
Breeding system
To determine whether G. rhizomatosa is self-fertile, a series of hand-pollination experiments were conducted in the field at Wash, Dand, Mhut, and Cas over two flowering seasons (1999–2000 and 2000–2001, Dand and Cas; and 2000–2001, Mhut and Wash). The Swamp population was not used for breeding experiments because of time restrictions, but plants in this population were extensively searched for fruit production. To prepare flowers, individual buds approaching anthesis were covered with bags of nylon organza (300 µm mesh) and sealed at both ends with wire ties to exclude floral visitors before pollination treatment. This produced virgin flowers (sensu Gross and Mackay, 1998
). All treated flowers were labelled with a jeweller's tag. Fruit set was recorded for each experimental flower 6–10 weeks after the treatment was applied. Four to 10 separate plants were used in each treatment (see Table 1 for sample sizes).
|
Self-pollination treatments
Some species are self-compatible but not automatically so because a vector is needed to transfer self pollen to the same flower's stigma (Gross, 1993
). We tested for self-compatibility in flowers of G. rhizomatosa by rubbing self pollen from the same flower into the stigmatic pore of virgin flowers (as described earlier) with a fingertip. Self-pollination was repeated 1–3 days after the initial pollination in case the stigma receptivity was delayed. Fingertips were cleaned with 100% ethanol between treatments to prevent accidental cross-pollination among flowers. Self-pollination treatments were not performed at Mhut because pollen was not found on any pollen presenters there.
Cross-pollination treatments
The level of fruit set resulting from hand cross-pollination and open pollination can be used to understand levels of pollination shortages in populations. The clonal nature of the species meant that to avoid near-neighbor mating, cross-pollination treatments used pollen from a different population. Donor flowers were placed in a labelled vial, and cross-pollination was effected by wiping the self pollen from the pollen presenters of treatment inflorescences with a cotton wool bud. The pollen presenter from donors was used to dab pollen onto the cleaned treatment presenters. Cross-pollination was repeated 1, 2, or 3 days after the initial pollination using donor pollen (N = 2–3 donor plants) from the same source.
Open treatments, measuring natural levels of fruit set
Inflorescences exposed to floral visitors provided an indication of natural fruit set and a standard to compare against other treatments. Mature inflorescences were tagged, and the number of flowers per inflorescence recorded and monitored for fruit set. In addition, all five populations were searched extensively in both seasons for new and old fruit on untagged plants.
Pollen viability
To determine the levels of viable pollen in populations, one mature bud from 8–10 discrete plants was collected from each of the five populations. All pollen was removed from each flower by scraping pollen from the four anthers onto a microscope slide. Pollen was stained in a 0.5% solution of 2,3,5 triphenyl tetrazolium chloride (TTC) in 12% sucrose, using the method of Cook and Stanley (1960)
. A cover slip was placed over the sample and sealed with clear nail varnish. Slides were left in the dark for a minimum of 3 days to let the TTC infiltrate pollen exine before examination under a stereomicroscope. Both viable and nonviable pollen grains were scored using hand counters. Pollen grains stain red in the presence of reductases, indicating enzyme activity. Pollen killed in formalin propionic acid in 70% ethanol (FPA) was used as a control against which viable pollen could be compared. Thus, red pollen grains were recorded as viable because nonviable pollen does not take up the stain (Smith and Gross, 2002
).
Stigma receptivity
Receptivity in stigmas was examined using two techniques; a physical examination of stigmatic opening using scanning electron microscopy (SEM) and a chemical assay for the peroxidase enzyme, the presence of which indicates receptivity (Galen and Plowright, 1987
). Three mature buds were collected from 10 plants at Mhut and another sample of two buds per 10 plants was collected from the fertile populations of Wash, Dand, and Cas and pooled. In the laboratory, peduncles of buds were placed in a vase of sugared water to promote longevity. Time 0 h began when the pollen presenter first projected past the tepals (anthesis). One of two treatment conditions was then applied to each pollen presenter. In the first treatment, pollen was left on the stigma and then wiped off with cotton wool at the designated time (0, 6, 12, 18, 24, 36, 48, or 72 h). In this manner, the impact of delayed pollen removal on stigma opening and receptivity could be examined. Removing pollen in this manner mostly removed all pollen although occasionally a few grains became lodged in the stigmatic pore, but we did not attempt to remove them. In the second treatment, pollen on all flowers was removed at time 0 h, and then pollen presenters were harvested sequentially at the designated time as described. Receptivity was then examined using SEM and peroxidase tests.
Scanning electron microscopy
To determine when the stigma opens during anthesis, stigmatic pores of different ages prepared using the two techniques described were examined using SEM. At the designated maturation times of 0, 6, 12, 18, 24, 36, 48, and 72 h from anthesis, pollen presenters were harvested and immediately placed in labelled Eppendorf tubes containing the fixative, formalin propionic acid in 70% ethanol (90 : 5 : 5 FPA). Samples were removed from FPA, placed into 70% ethanol, then washed for 10 min in an ethanol sequence, from 70% through 80%, 90%, 95%, and twice in 100% concentration. Grevillea rhizomatosa stigmas collapsed when exposed to air; consequently, critical point drying in liquid carbon dioxide was used to dehydrate stigmas. Stigmas were affixed to double-sided tape, placed on stubs, and splutter coated with 50-nm gold particles at 2.4 kV accelerating voltage. Scanning electron micrographs were taken using a JEOL JSM-5800LV at 20 kV.
Peroxidase activity
Stigmatic receptivity was also determined by testing stigmas for the production of peroxidase (Galen and Plowright, 1987
). Stigmas from flowers prepared with the two techniques described, at 0, 24, and 48 h after floral opening were harvested, and under a dissecting microscope, any remaining pollen was removed. Stigmas were then plunged into 6% w/v hydrogen peroxide; the presence of bubbling on the stigma surface indicates that the stigma is producing the enzyme peroxidase and is receptive. Stigmas were scored qualitatively as receptive (bubbling detected) or nonreceptive (no bubbling).
Floral visitation
Poor fecundity in plants can be due to pollinator limitation (see Gross et al., 2003
). Floral visitor activity was compared in three of the five study populations (Wash, Dand, Mhut) in the 2001 flowering season. Observations were made between 0630–0830 hours, although avian visitors observed at other times were also recorded. A total of 10.75 h was spent observing avian visitors. Floral visitors were viewed using binoculars from hide positions amid rocky outcrops at distances of 5–10 m from the plants. The following data were recorded during 0.5-h observation periods: bird identity, when possible the identity of each plant visited, the number of plants visited per bird, and the number of flowers visited per plant. Bird species were identified using Slater et al., (1986)
.
Statistical analyses
The statistical package STATGRAPHICS(Statgraphics Corp, Virginia, USA) was used to analyze data. The anthesis data were analyzed using a chi-squared test. Because variances could not be transformed for homogeneity, a nonparametric ANOVA (Kruskal–Wallis test) was used on the breeding system and floral visitation data. Parametric ANOVA was used for pollen viability data.
RESULTS
Anthesis mechanism
The amount of pollen released from anthers onto the pollen presenter of each style varied significantly among the three populations (
2 = 141.52; df = 2; P < 0.001, Fig. 1). Flowers at Mhut mostly did not release any pollen from anthers resulting in barren pollen presenters (Fig. 1). In contrast, all flowers in Wash opened with a fully laden pollen presenter bearing pollen from four anthers (Fig. 1). In Dand most flowers had fully laden pollen presenters with only 6% of sampled flowers presenting a low amount of pollen (Fig. 1).
|
|
|
|
|
|
Ecological factors, both biotic and abiotic, can reduce sexual reproduction, and as noted by Eckert (2001)
, these can affect seed production, germination, and recruitment in plants. In the present study, however, within flower factors such as pollen viability, the anthesis mechanism and female fertility factors diminished the capacity for sexual reproduction in two populations. Low fruit to flower ratios were detected in three fertile populations and experiments revealed that the cause of this was more likely to be ecological factors (resources for growth) and pollen quality than pollen quantity. Lowered fecundity in plants is often due to pollen limitation incurred by a shortage of pollinator visits (Copland and Whelan, 1989
; Martínez-Palle and Aronne, 2000
) although less often is an absolute failure in seed set caused by a total lack of pollinators (but see Lamont et al., 1993
). Presumably absolute pollinator limitation is rare because it would quickly lead to local extinctions when populations cannot be maintained by asexual reproduction. In populations of G. rhizomatosa, the eastern spinebill was the only avian species to visit flowers, but birds were just as active at Mhut, the infertile population, than at the two fertile populations where bird observations were conducted (Wash, Dand). Pollen limitation was only detected in the Wash population, despite the highest bird visitation among sites, and here this may have been due to a release from inbreeding depression because the outcross pollen was sourced from different populations. We thus rule out pollinator limitation as a major cause of low fecundity in G. rhizomatosa.
The structural characteristics of the stigmatic grooves in the Proteaceae may restrict pollen access and subsequent pollen tube growth (Dryandra formosa, Matthews et al., 1999
) but this would not appear to be a causal factor of poor fecundity in G. rhizomatosa. The stigmatic aperture was open at anthesis and remained open, increasing slightly, up to 72 h. Moreover, the presence of self pollen on the pollen presenter was not observed to delay opening of stigmatic grooves from 0 to 72 h in either the fertile or infertile populations. Peroxidase activity, an indicator of stigma receptivity, showed G. rhizomatosa flowers to be receptive from 0 to 72 h. Our results, one of the few studies that have combined SEM and chemical tests, suggest that stigma receptivity is triggered by anthesis and not by the removal of self pollen from the pollen presenter as has been suggested in other proteaceous systems (reviewed in Goldingay and Carthew, 1998
). The lack of protandry would assist self fertilization in G. rhizomatosa.
In this study sexual reproduction was absent from two of the five study populations, and the causes were found to occur at the flower level. Failure in the mechanical release of pollen from anthers was detected in two populations (Mhut, Swamp), and these populations failed to produce any viable seed. In addition, pollen viability was very low in the Mhut and Swamp populations because inviable pollen accounted for >90% of pollen in both locations (cf. with G. robusta, up to 95% viability, Kalinganire et al. [2000
] and c. 100% in G. beadleana, Smith and Gross [2002
]). The high pollen sterility in Mhut and Swamp is unlikely to be the ultimate cause of sterility in these populations because in one of these populations (Mhut), cross-pollination treatments using pollen from three fertile populations failed to effect seed set. Thus the Mhut population was found to be sterile in both male and female function. Investigations into pollen tube growth, and the construction of the transmitting tissue in the pistil in flowers from Mhut and Swamp plants may assist with understanding the ultimate cause of infertility in these populations (see Eckert et al., 1999
; Matthews et al., 1999
).
Overall though, low fruit to flower ratios were a feature of fertile populations (0.07–0.13). It is unlikely that this is due to poor floral visitation because birds were active in all sites (discussed earlier). Low fruit to flower ratios are common place in Grevillea (in both common and rare species, reviewed in Hermanutz et al., 1998
), and this has largely been attributed to a combination of interacting factors that includes resource limitation, pollen source, pollen limitation, and fruit predation (Hermanutz et al., 1998
). Within G. rhizomatosa, a variety of factors is implicated in the low fruit to flower ratios, including resource limitation at Dand and Cas (because fruit set could not be optimized with more pollen), pollen source and resources at Wash, and sterility factors at Mhut.
The morphology of some of the inviable pollen in the study populations was similar to that reported for G. banksii by Herscovitch and Martin (1989)
, in which pollen size and viability were correlated—aborted pollen differed from germinated grains in their smaller size, lack of cellular contents and abnormally thick endexine. In addition, we found large pollen grains without cellular content, and these only occurred in the infertile populations of Mhut and Swamp. Production of inviable pollen has been reported in other clonal species that appear to have lost the ability to reproduce sexually (Warburton et al., 2000
; Sharma, 2001
; Kimpton et al., 2002
). The amount of viable pollen produced per flower (2953.52 ± 1051.29) was low but within the range found for other species of Grevillea (753–15105, N = 5 species, Hermantutz et al., 1998; 1613 ± 68.64, N = 20 flowers G. beadleana, Smith, 1997
).
Most species in Grevillea are self-compatible (see table 1 in Smith and Gross, 2002
) with little intraspecies variation in breeding systems (Richardson et al., 2000, but see Hermanutz et al., 1998
). The breeding system of G. rhizomatosa, however, varied from self-compatible in the north of the species' distribution (Wash), through to self-incompatible in the center (Dand) and southern limits (Cas) of the species. The sterile populations (Swamp and Mhut) were located in the center of the species' range. Wash and Cas are the furthest apart populations but only c. 9.5 km separates them, and thus it would appear that there is unusual variation in breeding systems within a small area. However, it may be that breeding systems within species of Grevillea are more variable than reported, but because few studies (Richardson et al., 2000
; Smith and Gross, 2002
) incorporate multiple populations in their assessments of self-compatibility, we possibly have a skewed view of breeding systems variability in Grevillea. Breeding system variation within species has been recorded in other systems (Leavenworthia alabamica,Busch, 2005
) and may be associated with unreliable pollinator servicing (Blandfordia grandiflora, Ramsey et al., 1993
). We did detect pollen limitation at Wash, but a lack of pollinators seems an unlikely cause of this. Instead we suggest that the quality of the outcross pollen (sourced from different populations) in the Wash outcross experiments caused heterosis and elevated seed production compared with selfed treatments. It is unlikely that the elevated fruit production in the outcross treatment was a result of better growing conditions at Wash because natural levels of fruit set at Wash were not significantly different from the other fertile populations. The self-compatible population at Wash may have had different selection pressures over time compared with self-incompatible Dand and Cas such that the allele for self-incompatibility may have been lost from Wash. The Wash environment is more mesic than Dand or Cas (Caddy, 2004
) and may have experienced less fires than these two southern populations. Frequent fire events could lead to the loss of individuals and suppress sexual function. Hoffman (1999)
for example, found that frequent fires (every 1–3 years) in a neotropical savanna was associated with an increase in clonal growth in woody species and the importance of sexual reproduction was greatly reduced by frequent burning. Fire has the effect of suppressing flowering in resprouting plants of G. rhizomatosa—plants that resprouted after the 2002 fires have yet to flower (see Caddy and Gross, 2006
). Environmental suppression of sexual function may cause the evolution of genetic sterility because natural selection no longer retains the many traits involved in sex, i.e., the "use it or lose it" hypothesis (reviewed in Eckert, 2001
). Within Grevillea, Makinson (2000; R. Makinson, Royal Botanic Gardens, Sydney, unpublished data) estimates that about 14% of species have the capacity to reproduce asexually but that this is usually combined with sexual reproduction.
The regenerative capacity of clonal growth provides the ramet with increased longevity, but is this at the expense of sexual function? We do not know how long plants of G. rhizomatosa live, but Duncan (1992)
noted that a plant of Grevillia striata was still surviving after 150 years. We suspect that the individuals at Mhut (where seeds are not produced) are older than plants in the other study populations. Plants at Mhut are large (0.5–1.20 m tall x 0.5 x 1.40 m wide) and are nestled among granite boulders that afford protection from fire. Fire records indicate that the Mhut area was burnt in 1968 and possibly in 1988 (Caddy and Gross, 2006
), and although we do not know if G. rhizomatosa plants were killed during these fire events, we can hypothesize that because of the lack of seed production in this population and the species' resprouting ability (Caddy and Gross, 2006
; Croft et al., 2006
), that this isolated population of clonally reproducing plants is old. Clonal plants can have extraordinary life spans. Rossetto et al. (1999)
estimated a clonal mallee eucalypt to be over 6000 years old, much older than the usual age of single stemmed eucalypts (see also Tyson et al., 1998
). In a late successional habitat these old clones have the potential to occupy large areas that may reduce opportunities for seedling establishment. For example, Lomatia tasmanica (Proteaceae) is sterile, genetically uniform, and clonal, and the ramet spans c. 1.2 km (Lynch et al., 1998
). In addition, there appears to be no indication of seedling germination among well-established clones of the infrequent seeder G. kennedyana (Duncan, 1992
; Makinson, 2000
; von Richter et al., 2001
). Thus there is a potential threat to the persistence of sexual forms of G. rhizomatosa through the monopoly of sites by asexual forms that rapidly resprout and colonize bare ground after fire (Caddy and Gross, 2006
).
Loss of sexual function in plants is particularly intriguing when it accompanies clonal growth. Is the production of asexual clones a consequence of a decrease in vigor due to a build-up of deleterious mutations that are fixed in the absence of sex (mutational meltdown hypothesis), or is the asexual form selected over sexual forms (resource allocation theory, antagonistic pleiotropy hypothesis)? The loss of sexual traits may also be precipitated through genetic drift of selectively neutral mutations (Dorken et al., 2004; see Eckert, 2001
, for a review of hypotheses). The latter would seem unlikely if several discrete populations each have sterile forms with the same malfunction in sexual processes and if several genotypes exist in the asexual populations. In this study we have detected interpopulation variation in breeding system components, including two sterile populations, and we predict that we will also find genotypic diversity among these populations of G. rhizomatosa. This work is underway.
Seed production provides an important mechanism for species to escape a changing environment or overcrowding. A reduction in sexual regeneration can therefore have adverse impacts on clonal species when the balance that exists between the two forms of reproduction is lost. Rossetto et al. (2004)
investigated the occurrence and distribution of clonality in an endangered rainforest tree Elaeocarpus williamsianus (Elaeocarpaceae) and found only eight remaining genets and a high incidence of sterile fruit, concluding that habitat fragmentation had removed an existing balance between vegetative and sexual reproduction. The sexual forms of G. rhizomatosa are thus very valuable in the conservation of this restricted species.
Long-term monitoring of the fertile populations may be required to ensure that a balance is maintained in the proportion of sexual and asexual plants. If the interval between fires is too short (possibly <10 years), then this will prevent plants from reaching sexual maturity and developing a seed bank. Simultaneously, it is likely that the asexual form, with its suckering ability will colonize areas more rapidly than the infrequently produced seeds.
Increasingly, studies of threatened plants are being conducted on clonal species (Sydes and Peakall, 1998
; Rossetto et al., 2004), and the results are presenting many challenges for managers of these threatened species. A broader question for reproductive ecologists and managers of threatened species is the possibility that unnatural frequent fires or other disturbance events in the landscape are causing selection of clonal forms at the expense of sexually reproducing individuals.
FOOTNOTES
1 The project was supported by funds to C.L.G from UNE and NSW National Parks and Wildlife Service, Glen Innes District, and H. Caddy undertook much of the field work during his Honours year. P. Croft is thanked for providing additional support. The authors appreciate field assistance from A. Coventry, B. Tailor, and D. Mackay; W. Sheather for help with propagating cuttings of Grevillea rhizomatosa; P. Garlick and P. Littlefield (UNE) for helping with scanning electron microscopy. The Drirector of the New England herbarium is also thanked for access to specimens and records. B. Makinson (RBG-Sydney) is thanked for again generously sharing his knowledge of Grevillea. M. Ramsey, M. Rossetto, P. Nelson, G. Vaughton, and S. Simpson are thanked for valuable discussions. 
2 Author for correspondence (e-mail: cgross{at}une.edu.au
)
102 4 Current address: Namoi Catchment Management Authority, Walgett NSW 2832, Australia.
101 The project was supported by funds to C. L.G from UNE and NSW National Parks and Wildlife Service, Glen Innes District, and H. Caddy undertook much of the field work during his Honours year. P. Croft is thanked for providing additional support. The authors appreciate field assistance from A. Coventry, B. Tailor, and D. Mackay; W. Sheather for help with propagating cuttings of Grevillea rhizomatosa; P. Garlick and P. Littlefield (UNE) for helping with scanning electron microscopy. The Director of the New England herbarium is also thanked for access to specimens and records. B. Makinson (RBG-Sydney) is thanked for again generously sharing his knowledge of Grevillea. M. Ramsey, M. Rossetto, P. Nelson, G. Vaughton, and S. Simpson are thanked for valuable discussions.
LITERATURE CITED
Andersson S. W. B.. 1994. Reaction norm variation in a rare plant, Senecio integrifolius (Asteraceae). Heredity 73: 598-607.[ISI]
Burne H. M. Yates C. J. Ladd P. G.. 2003. Comparative population structure and reproductive biology of the critically endangered shrub Grevillea althoferorum and two closely related more common congeners. Biological Conservation 114: 53-65.[CrossRef][ISI]
Busch J. W.. 2005. The evolution of self-compatibility in geographically peripheral populations of Leavenworthia alabamica (Brassicaceae). American Journal of Botany 92: 1503-1512.
Caddy H. A. R.. 2004. Investigations of low fecundity in the vulnerable shrub Grevillea rhizomatosa Olde & Marriott (Proteaceae) Honours thesis, University of New England, Armidale, NSW, Australia.
Caddy H. A. R. Gross C. L.. 2006. Population structure and fecundity in the putative sterile shrub, Grevillea rhizomatosa Olde & Marriott (Proteaceae). Proceedings of the Linnean Society of New South Wales 127: 11-18.[ISI]
Cilliers C. J.. 1991. Biological control of water fern, Salvinia molesta, (Salviniaceae), in South Africa. Agriculture, Ecosystems and Environment 37: 219-224.[CrossRef]
Cook S. A. Stanley R. G.. 1960. Tetrazolium chloride as an indicator of pine pollen germinability. Silvae Genetica 9: 134-136.
Copland B. Whelan R.. 1989. Seasonal variation in flowering intensity and pollinator limitation and fruit-set in three co-occurring Banksia species. Journal of Ecology 77: 509-523.[CrossRef]
Croft P. Hofmeyer D. Hunter J. T.. 2006. Fire responses in four rare plant species at Gibraltar Range National Park, Northern Tablelands, NSW. Proceedings of the Linnean Society of New South Wales 127: 57-62.[ISI]
Cropper S. C. D. T. D.. 1989. Thelymitra epipactoides F. Muell. (Orchidaceae): the morphology, biology and conservation of an endangered species. Proceedings of the Royal Society of Victoria 101: 89-101.
Dorken M. E. Eckert C. G.. 2001. Severely reduced sexual reproduction in northern populations of a clonal plant, Decodon verticillatus (Lythraceae). Journal of Ecology 89: 339-350.[CrossRef]
Duncan A.. 1992. Aspects of the ecology of the rare Grevillea kennedyana (Proteaceae) in north-western New South Wales. Cunninghamia 2: 533-539.
Eckert C. G.. 2001. The loss of sex in clonal plants. Evolutionary Ecology 15: 501-520.[CrossRef][ISI]
Eckert C. G. Dorken M. E. Mitchell S. A.. 1999. Loss of sex in clonal populations of a flowering plant, Decodon verticillatus (Lythraceae). Evolution 53: 1079-1092.[CrossRef][ISI]
Galen C. Plowright R.. 1987. Testing the accuracy of using peroxidase activity to indicate stigma receptivity. Canadian Journal of Botany 65: 107-111.[ISI]
Goldingay R. L. Carthew S. M.. 1998. Breeding and mating systems of Australian Proteaceae. Australian Journal of Botany 46: 421-437.[CrossRef]
Gross C. L.. 1993. The breeding system and pollinators of Melastoma affine (Melastomataceae)—a pioneer shrub in tropical Australia. Biotropica 25: 468-474.[CrossRef][ISI]
Gross C. L. Bartier F. V. Mulligan D. R.. 2003. Floral structure, breeding system and fruit-set in the threatened sub-shrub Tetratheca juncea Smith (Tremandraceae). Annals of Botany 92: 771-777.
Gross C. L. Mackay D.. 1998. Honeybees reduce fitness in the pioneer shrub Melastoma affine (Melastomataceae). Biological Conservation 86: 169-178.[CrossRef][ISI]
Hermanutz L. Innes D. Denham A. J. Whelan R. J.. 1998. Very low fruit:flower ratios in Grevillea (Proteaceae) are independent of breeding system. Australian Journal of Botany 46: 465-478.[CrossRef]
Herscovitch C. J. Martin A. R. H.. 1989. Pollen–pistil interactions in Grevillea banksii. The pollen grain, stigma, transmitting tissue and in-vitro pollinations. Granna 28: 69-84.
Hoffmann W. A.. 1999. Fire and population dynamics of woody plants in a neotropical savanna: Matrix model projections. Ecology 80: 1354-1369.[CrossRef][ISI]
Honnay O. Jacquemyn H. Roldan-Ruiz I. Hermy M.. 2006. Consequences of prolonged clonal growth on local and regional genetic structure and fruiting success of the forest perennial Maianthemum bifolium. Oikos 112: 21-30.[CrossRef][ISI]
Howell G. J. Slater A. T. Knox R. B.. 1993. Secondary pollen presentation in angiosperms and its biological significance. Australian Journal of Botany 41: 417-438.[CrossRef][ISI]
Johnson L. A. S. Briggs B. G.. 1975. On the Proteaceae—the evolution and classification of a southern family. Botanical Journal of the Linnean Society 70: 83-182.
Kalinganire A. Harwood C. E. Slee M. U. Simons A. J.. 2000. Floral structure stigma receptivity and pollen viability in relation to protandry and self-incompatibility in silky oak (Grevillea robusta A.Cunn). Annals of Botany 86: 133-148.
Kimpton S. K. James E. A. Drinnan A. N.. 2002. Reproductive biology and genetic marker diversity in Grevillea infecunda (Proteaceae), a rare plant with no known seed production. Australian Systematic Botany 15: 485-492.[CrossRef][ISI]
Koshy K. C. Jee G.. 2001. Studies on the absence of seed set in Bambusa vulgaris. Current Science 81: 375-378.
Lamont B. Klinkhamer P. W. E. Klinkhamer P. Witkowski E.. 1993. Population fragmentation may reduce fertility to zero in Banksia goodii—a demonstration of the Allee effect. Oecologia 94: 446-450.[CrossRef][ISI]
Lazarides M. Cowley K. Hohnen P.. 1997. CSIRO Handbook of Australian weeds, 1st ed., 1–264 Commonwealth Scientific and Industrial Research Organisation Publishing, Collingwood, Victoria, Australia.
Lopez-Almansa J. C. Pannell J. R. Gil L.. 2003. Female sterility in Ulmus minor (Ulmaceae): a hypothesis invoking the cost of sex in a clonal plant. American Journal of Botany 90: 603-609.
Lynch A. J. J. Barnes R. W. Cambecedes J. Vaillancourt R. E.. 1998. Genetic evidence that Lomatia tasmanica (Proteaceae) is an ancient clone. Australian Journal of Botany 46: 25-33.
Makinson R. O.. 2000. Grevillea. Flora of Australia, vol. 17A 1-460 Australian Biological Resources Study, Australian Government Publishing Service Canberra, Australia.
Martínez-Palle E. Aronne G.. 2000. Pollination failure in Mediterranean Ruscus aculeatus L. Botanical Journal of the Linnean Society 134: 443-452.[CrossRef]
Matthews M. L. Gardner J. Sedgley M.. 1999. The proteaceous pistil: morphological and anatomical aspects of the pollen presenter and style of eight species across five genera. Annals of Botany 83: 385-399.
Pandit M. K. Babu C. R.. 2003. The effects of loss of sex in clonal populations of an endangered perennial Coptis teeta (Ranunculaceae). Botanical Journal of the Linnean Society 143: 47-54.[CrossRef]
Pilgrim E. S. Crawley M. J. Dolphin K.. 2004. Patterns of rarity in the native British flora. Biological Conservation 120: 161-170.[CrossRef][ISI]
Ramsey M. Prakash N. Cairns S.. 1993. Breeding systems of disjunct populations of Christmas Bells (Blandfordia grandiflora R.Br., Liliaceae): variation in self-fertility and an ovular mechanism regulating self-fertilisation. Australian Journal of Botany 41: 35-47.
Richardson M. B. G. Ayre D. J. Whelan R. J.. 2000. Pollinator behaviour, mate choice and the realised mating systems of Grevillea mucronulata and Grevillea sphacelata. Australian Journal of Botany 48: 357-366.[CrossRef]
Rossetto M. Gross C. L. Jones R. Hunter J.. 2004. The impact of clonality on an endangered tree (Elaeocarpus williamsianus) in a fragmented rainforest. Biological Conservation 117: 33-39.[CrossRef][ISI]
Rossetto M. Jezierski G. Hopper S. D. Dixon K. W.. 1999. Conservation genetics and clonality in two critically endangered eucalypts from the highly endemic south-western Australian flora. Biological Conservation 88: 321-331.[CrossRef][ISI]
Sharma I. K.. 2001. Understanding clonal diversity patterns through allozyme polymorphism in an endangered and geographically restricted Australian shrub, Zieria baeuerlenii and its implications for conservation. Biochemical Systematics and Ecology 29: 681-695.[CrossRef][ISI][Medline]
Slater P. Slater P. Slater R.. 1986. The Slater field guide to Australian birds, rev. ed Lansdowne, Sydney, Australia.
Smith J. A.. 1997. The reproductive ecology of the endangered plant Grevillea beadleana (Proteaceae) B.Sc. Honours thesis, University of New England, Armidale, NSW, Australia.
Smith J. A. Gross C. L.. 2002. The pollination ecology of Grevillea beadleana McGillivray, an endangered shrub from northern New South Wales, Australia. Annals of Botany 89: 97-108.
Sydes M. A. Peakall R.. 1998. Extensive clonality in the endangered shrub Haloragodendron lucasii (Haloragaceae) revealed by allozymes and RAPDSs. Molecular Ecology 7: 87-93.[CrossRef]
Thompson J. D.. 1991. The biology of an invasive plant. Bioscience 41: 393-401.[CrossRef][ISI]
Tyson M. Vaillancourt R. E. Reid J. B.. 1998. Determination of clone size and age in a mallee eucalypt using RAPDs. Australian Journal of Botany 46: 161-172.[CrossRef]
Vaughton G.. 1996. Pollination disruption by European honeybees in the Australian bird-pollinated shrub Grevillea barklyana (Proteaceae). Plant Systematics and Evolution 200: 89-100.[CrossRef][ISI]
Vaughton G. Ramsey M.. 1991. Floral biology and inefficient pollen removal in Banksia spinulosa var. neoanglica (Proteaceae). Australian Journal of Botany 39: 167-177.
von Richter L. Azzoparid A. Johnstone R. Offord C.. 2001. Seed germination in the rare shrub Grevillea kennedyana (Proteaceae). Cunninghamia 7: 205-212.
Warburton C. L. James E. A. Fripp Y. J. Trueman S. J. Wallace H. M.. 2000. Clonality and sexual reproductive failure in remnant populations of Santalum lanceolatum (Santalaceae). Biological Conservation 96: 45-54.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
