| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Department of Biology, Portland State University, Portland, Oregon 97207 USA; and UC Herbarium, Department of Integrative Biology, University of California, Berkeley, California 94720 USA
Received for publication September 26, 2000. Accepted for publication June 7, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: floral stages • fruit number • Iridaceae • Iris tenax • megagametogenesis • pollinator limitation • seed number
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
400 species) and its widespread occurrence in the Northern Hemisphere. Most accounts of Iris megagametogenesis are based on a study of I. fulva and I. hexagona var. giganticaerulea (Riley, 1942
). Other significant contributions include a report on megagametogenesis in I. tenax (Smith and Clarkson, 1956
) and a report on megagametogenesis and ovule development in I. kumaonensis and I. decora (Pande and Singh, 1981
). Abnormalities in megagametogenesis have been reported in >50% of the species observed: I. japonica (Yasui and Sawada, 1940a
), I. pseudacorus (Karagyozova, 1963
), I. kumaonensis, and I. decora (Pande and Singh, 1981
).
Smith and Clarkson (1956)
, in their study of megagametogenesis, reported several deviations from patterns previously reported for the genus. These deviations include the persistence of synergids and antipodals until fertilization and the migration of nucellar nuclei into the gametophyte following the disintegration of nucellar cells surrounding the embryo sac.
Piper (1924)
collected and described I. tenax var. gormanii from the only known population. He reported that sexual reproduction for I. tenax var. gormanii was low and estimated capsule production at 5%. No further reproductive work has been reported for this variety. There have been other reports of low reproductive success within Iris. Yasui and Sawada (1940a)
reported fruit production at 0.008% for the allotriploid I. japonica and ascribed this to developmental and other factors. Uno (1982a)
found that ovules of the diploid I. douglasiana had high mortality (70%) prior to maturation.
Several aspects of reproduction in Iris tenax were investigated during the present study. Stages of anthesis were identified as a preliminary step to pollination studies and were correlated with stages of ovule development. Ovule development was reexamined in I. tenax var. tenax and investigated in I. tenax var. gormanii to determine if abnormalities in early development might be responsible for the low fruit production previously reported for this subspecies. In addition, pollination studies were carried out to examine reproductive success in I. tenax. Ovule and fruit production of I. tenax var. gormanii was examined to quantify the low estimate of reproductive success previously reported.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Fieldwork was carried out at three sites in northwestern Oregon. Iris tenax var. tenax occurred at two sites, one 42 km east of Portland near the town of Corbett and one within the city of Portland. At the third site, along Scoggins Creek 33 km southwest of Portland, is the only known population of I. tenax var. gormanii. This population is subdivided into an upper, relatively flat area and a sloping area along a road cut. Voucher specimens are deposited at the Jepson Herbarium (JEPS) (collection numbers, 92-ph-28, 93-re-23, and 92-ph-29, respectively).
Flower and ovule development
Whole buds and flowers of I. tenax var. tenax and I. tenax var. gormanii were collected at eight stages: five stages during bud development and three stages during flowering. Line drawings were produced from computer-scanned images of buds and flowers.
Ovaries of I. tenax var. tenax and I. tenax var. gormanii were collected at nine stages: five stages during bud development, three stages during flowering, and one stage after pollination. Midsections of freshly collected ovaries were preserved in both formalin-propionic acid-alcohol (FPA) and 2% glutaraldehyde in 50 mmol/L sodium cacodylate buffer. Specimens from FPA were dehydrated in tertiary butyl alcohol, embedded in paraffin, and sectioned at 8–12 µm for light microscopy. Ovary material in glutaraldehyde was rinsed in sodium cacodylate buffer, postfixed in 1% buffered osmium tetroxide, dehydrated in ethanol, and embedded in LR White Resin, medium grade (Electron Microscopy Sciences, Fort Washington, Pennsylvania, USA). Resin embedded material was sectioned at 0.5–1 µm for light microscopy. Paraffin embedded material was stained with safranin-fast green, except for material viewed under fluorescence, which was stained with aniline blue. Resin embedded material was stained with toluidene blue.
Pollination studies
Fresh and 2-d-old pollen was tested for viability using an alcohol dehydrogenase test (Dafni, 1992
, protocol 5). Five flowers representing five individual genets were collected for analysis at each site. Immediately upon return to the laboratory, fresh pollen viability was determined using 200 pollen grains from one anther of each of the five flowers. Viability of 2-d-old pollen was determined by analyzing 200 pollen grains from a second anther of each flower after refrigeration at 4°C for 24–30 h. Time of receptivity of stigmas was also determined, based on a test for alcohol dehydrogenase (Dafni, 1992
, protocol 5).
For field pollination studies, two treatments were used: self-pollination and xenogamous outcrossing. Plants used in self-pollinations were prepared by removal of ripe anthers from terminal flowers that were not yet in female phase and by the placement of gelatin tubes (one-half of a gelatin capsule) over each of the three stigmatic lobes to exclude pollinators. Anthers and parent plants were tagged for identification. Anthers were stored in empty gelatin capsules at 4°C until the parent plant entered the female phase (usually the following afternoon). Pollen was then removed from stored anthers using toothpicks and placed on parent-plant stigmas. Gelatin tubes were replaced over the pollinated stigmas.
Outcrossing treatments followed the same procedures used for self-pollinations except that the pollen was removed from anthers of five to ten individuals of a population, mixed and placed on stigmas of different plants of the same population. The pollen mixture should have minimized any interactions between specific pollen donors and female plants.
Pollen tube growth within styles of I. tenax var. gormanii and I. tenax var. tenax was analyzed after each of the two pollination treatments. Whole styles were collected from plants at 12, 24, and 72 h posttreatment and preserved in 70% alcohol. Prior to study, styles were rinsed in distilled water, placed on a slide, stained with either 0.1% aniline blue (Dafni, 1992
, protocol 18) or 0.005% aniline blue (Jensen, 1962
) and observed using fluorescence microscopy.
Fruits resulting from each of the two pollination treatments were collected prior to the completion of fruit maturation at 13–17 d after pollination. In addition, all of the fruits from five clones at each of the two populations of I. tenax var. tenax and the two subpopulations of I. tenax var. gormanii were collected to represent an open-pollination treatment. Clumps of leaves were assumed to represent vegetative reproduction, although it is possible that clumps may have originated from more than one seed that germinated in close proximity. Collected fruits were identified as developed (enlarged and light green) or not developed (small and yellow-green to brown). Ovules within each of the three locules of developing fruits were identified as either developed or not developed. Ovules were considered developed only if they showed significant enlargement and did not display signs of deterioration, such as turning yellow or collapsing. Ovaries with insect damage were not included in the analysis. Parametric analysis of variance (ANOVA) statistics were performed to determine if taxon or treatment effects were significant.
Taxon self-compatibility indices were computed by dividing the average seed and fruit set after self-pollination by the seed and fruit set after cross-pollination (Becerra and Lloyd, 1992
). Indices used normally developing ovules to represent seed set because fruits were collected prior to seed maturation.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
Ovule development
Early development leads to the formation of the four megaspores within an anatropous ovule (Figs. 11–18). Ovules are arranged in two rows within each of three ovary locules. The ovary has axial placentation (Fig. 11). In general, one hypodermal cell enlarges and functions as an archesporium. The differentiation of the archesporium in I. tenax occurs early and was not observed in the present study. The youngest flower buds collected occurred at ground level among the new set of leaves.
|
In the second floral bud stage (Fig. 3), the outer integument of the ovule is well defined, having both elongated and expanded in thickness (Fig. 14). The first meiotic division of the megasporocyte has produced two cells that are easily identified, due to the many small vacuoles surrounding a centrally located nucleus (Fig. 14, at arrows). Concomitant with this stage, periclinal divisions in parietal cells add to the nonsporogenous layers that are typical of crassinucellate ovules (Fig. 15, at arrows). Development of parietal tissue is limited in this species; no more than two layers of parietal cells were observed.
By the third floral bud stage (Fig. 4), the ovule has become anatropous due to unequal growth primarily in the region below the attachment of the integuments (Fig. 16). The inner, two-layered integument continues to elongate (Fig. 17) and soon extends beyond the nucellus, defining the region that will become the micropyle (Figs. 16 and 17). Cells of the nucellar epidermis that flank the parietal tissue also divide periclinally (Fig. 17). A centrally located cell (or cells) of the nucellar epidermis (Fig. 17) also enlarges, but generally does not divide. Epidermal cells at the base of the funiculus are enlarged radially and are glandular in appearance, constituting an obturator (Fig. 18).
During this floral stage, megasporogenesis within the developing ovule has also been progressing. A second meiotic division has produced a T-shaped tetrad of megaspores (Fig. 17, at arrows). The micropylar megaspores soon degenerate (Fig. 17). In each ovule examined, the two megaspores closest to the micropyle were breaking down during this floral stage and the two chalazal cells were larger and showed no signs of degenerating. The most chalazal megaspore is large, with many small vacuoles (Fig. 17) and will function in megagametophyte development. Smith and Clarkson (1956)
reported that linear tetrads also occur in I. tenax, but they were not observed in the present study.
The development of the functional megaspore through meiotic divisions of the sporocyte and the degeneration of the three micropylar megaspores sets the stage for megagametogenesis (Figs. 19–28). In I. tenax, several changes take place in gross ovule morphology. The micropyle becomes well defined by final elongation of the inner integument and expansion of integumentary cells in the micropylar area. The outer integument undergoes further elongation and is barely subequal to the inner integument (Fig. 19). During this interval, the functional megaspore shows dramatic enlargement.
|
Development of a seven-celled megagametophyte occurs during the final floral bud stage (Fig. 6). The megagametophyte remains largely unchanged as the bracts diverge and the flower unfurls (Fig. 7). Figures 25–28 illustrate the seven-celled megagametophyte. The megagametophyte consists of three chalazal antipodals (Fig. 25), a secondary nucleus within the central cell (Fig. 26), and two synergids and an egg cell at the micropylar end (Figs. 27 and 28). At the chalazal pole the antipodals lie in a pouchlike extension of the embryo sac (Fig. 25). At this stage, the secondary nucleus, with two prominent nucleoli, lies slightly chalazal from median in the embryo sac (Fig. 26). The movement of the two polar nuclei to a median position and their fusion to form a secondary nucleus must occur rapidly in I. tenax because none of the embryo sacs observed had polar nuclei. The egg cell nucleus is chalazal to the two synergid nuclei (Figs. 27 and 28). The positions of nuclei in the egg apparatus are influenced by vacuolar location within each cell.
Final maturation of the megagametophyte is observed during the male phase of flowering (Fig. 8). No further changes were seen during the female phase of flowering (Fig. 9). The mature megagametophyte is shown in Figs. 29–32. The mature embryo sac has expanded considerably and is rounded in outline (Figs. 29 and 30). In Fig. 29, the nucellus has pulled away from the integuments showing that the nucellar epidermis is not crushed or used up during enlargement of the megagametophyte. At maturation, the synergids have degenerated and appear as two darkly stained bodies near the micropyle (Fig. 29, at arrows). The nucleus of the egg cell is visible above the degenerated synergids (Fig. 29). Also visible in this view is the nucleus of one antipodal at the chalazal pole. Near the chalazal end of the same ovule the secondary nucleus is large and prominent within the central cell (Fig. 30). The secondary nucleus now lies against the antipodals and has a single large nucleolus. The nuclei of the two most chalazal antipodals are visible as well as the outline of the third antipodal. The ovule in Fig. 31 shows the large egg cell at maturity. The egg nucleus (Fig. 31) is considerably smaller than the secondary nucleus (Fig. 30). Within the synergids, a filiform apparatus is apparent as a very darkly stained area with raylike projections (Fig. 32). The synergids in this ovule still have nuclei but are undergoing biochemical changes that are apparent by their dark red color when stained with safranin. This observable change in synergids is considered to reflect a degeneration of the cells.
|
Pollination studies
Pollen viability is high for both varieties. Viability of fresh pollen averaged 92% for I. tenax var. tenax and 93% for I. tenax var. gormanii. Viability of 2-d-old pollen averaged 90% for I. tenax var. tenax and 95% for I. tenax var. gormanii.
Figure 36 shows a typical ovary of I. tenax var. gormanii. Part of the ovary wall has been removed, exposing one of the three locules. Six ovules are developing normally, one ovule began to enlarge but is now degenerating (at arrow), and several ovules show little or no development. The percent of I. tenax var. tenax and I. tenax var. gormanii ovules with normal development at 17–20 d after pollination is similar for each of the two treatments and open pollination for both years (Table 1). When data is combined across all treatments and both taxa, there were significant differences (P < 0.02) between 1993 and 1994. None of the differences observed in ovule survival between treatments are significant (P > 0.05).
|
|
|
Pollen tube growth was similar after all treatments. Germinated and ungerminated pollen could be seen on stigma surfaces 12 h after pollination. By 24 h after pollination pollen tubes were observed within styles and ovaries (Fig. 38). Some pollen tubes could be seen entering ovules.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Ovule development
Megagametophyte development in I. tenax displayed the "Polygonum type" reported previously for other members of the genus (Riley, 1942
; Pande and Singh, 1981
). The ovule is crassinucellate and the mature ovule is anatropous and bitegmic.
The present study includes several new observations and clarifications for the genus Iris. Not reported previously in Iris is the presence of two regions within the developing ovule that have large, glandular-appearing cells. The inner integumentary cells lining the micropyle became radially elongate and glandular in appearance resembling an integumentary tapetum. An integumentary tapetum usually occurs when the embryo sac comes into direct contact with the inner integument because the nucellus breaks down early in embryogenesis. When an integumentary tapetum develops, the inner integumentary cells adjacent to the embryo sac become elongated radially and are thought to function in the transport of photosynthates to the developing embryo sac (Gasser and Robinson-Beers, 1993
). In I. tenax, only the cells lining the micropyle became radially elongate and glandular appearing, and the nucellus remained intact around the developing embryo sac. The glandular-appearing integumentary cells in I. tenax are similar in nature but do not occupy the position of an integumentary tapetum.
Adjacent to the micropyle along the base of the funiculus another region of cells elongated radially and became glandular in appearance. These cells represent a well-defined obturator. Preliminary studies indicate that the obturator in I. tenax may be similar to that in Ornithogalum caudatum in the Liliaceae. In Ornithogalum the obturator developed from cells at the base of the funiculus and along the placenta so that the obturator extended around the base of the funiculus and around the entire central column of ovary tissue (Tilton and Horner, 1980
). Obturators are considered to assist in directing pollen tubes toward ovules and into the micropyle (Maheshwari, 1950
). It is likely that a similar function can also be attributed to the integumentary cells lining the micropyle in I. tenax.
The occurrence of a parietal cell has been previously reported in Iris (Riley, 1942
; Smith and Clarkson, 1956
), but the nature of crassinucellate development of the ovule has not been described. In I. tenax, the parietal cell divided anticlinally and periclinally typically resulting in two layers of cells. Cells at the flanks of the nucellar epidermis also divided but did not contribute to the depth at which the developing megaspores were embedded. The embedded position of the functional megaspore was short-lived, however, because it soon elongated, crushing the cell layers separating it from the nucellar epidermis.
Finally, although the presence of a filiform apparatus has been reported in Iris (Smith and Clarkson, 1956
), studies have not documented when the filiform apparatus appears or how long it persists. The filiform apparatus is a complex of cell wall ingrowths that project into the synergids. During the present study, the filiform apparatus was found to be short-lived in I. tenax, appearing late in development during the male phase of flowering. The development of the filiform apparatus occurred just prior to or as the synergids began to degenerate, and it disappeared after pollination. The filiform apparatus is thought to function by directing the pollen tube into one of the synergids and possibly by nourishing the entering pollen tube (Jensen and Fisher, 1968
; Russell, 1992
).
Several earlier observations of megagametogenesis in I. tenax (Smith and Clarkson, 1956
) were not supported by this study. First, Smith and Clarkson concluded that the synergids persisted until fertilization and were the last of the seven embryo sac cells to disintegrate if fertilization did not occur. Smith and Clarkson reported that the first synergid to degenerate after fertilization was the one not receiving the pollen tube. In the majority of flowering plants, it is thought that one synergid begins to degenerate prior to fertilization and that the pollen tube enters this synergid (Jensen, 1974
). All of the mature embryo sacs observed in the present study had synergids that were both degenerating or had already degenerated. These synergids degenerated before pollination when the flower was functionally male.
Second, Smith and Clarkson (1956)
reported that membranes between the developing embryo sac and the nucellus disintegrated at about the two-nucleate stage and that nuclei migrated from the nucellus into the developing embryo. They reported that these nuclei soon degenerated. It is unclear to what exactly these authors are referring, as the cells of the embryo sac and nucellus have cell walls as well as membranes. None of the ovules observed in the present study showed disintegration of the embryo sac or migration of nucellar nuclei into the gametophyte. It is possible that Smith and Clarkson interpreted nuclei of the large cells surrounding the embryo sac (Figs. 21–23) as lying within the embryo sac. The nuclei of these large cells tend to be appressed to the embryo sac but are clearly separated from the embryo sac by cell walls. It is my interpretation that these large cells expand and define the space that the embryo sac will soon inhabit (Fig. 24).
Finally, Smith and Clarkson reported that fusion of the secondary nucleus and the sperm nucleus occurs in the chalazal end of the embryo sac but that the resulting primary endosperm nucleus then moves to a position near the egg cell and divides. All of the primary endosperm nuclei observed in my study divided while in the chalazal end of the embryo sac.
Pollination studies
Piper (1924)
reported that 5% of I. tenax var. gormanii produced fruit in 1922, but did not speculate on probable causes of the low fruit set observed. There are other reports of low reproductive success within Iris, although the findings reported in these studies are not directly applicable to the results found for I. tenax. In I. douglasiana, another member of Iris series Californicae, Uno (1982a)
reported >70% mortality of ovules prior to seed production. He found substantial insect predation of capsules and concluded that seed loss was due to predation. It is unknown if other factors also contributed to ovule mortality. Yasui and Sawada (1940b)
found that in the allotriploid I. japonica only 13% of embryo sacs developed normally and that only 63% of the pollen was viable. In addition, Yasui and Sawada (1940a)
observed that fruit production was low (0.008% of the ovaries developed into fruits), indicating that additional factors might be involved. It is probable that the reduced fertility observed is related to the triploid nature of this species. Pande and Singh (1981)
stated that ovule sterility was high in I. kumaonensis and I. decora and that most failures occurred by the four-nucleate stage of embryo sac development. They did not quantify ovule failure but stated that "only in a very few ovules organized mature embryo sacs could be observed." Karagyozova (1963)
noted abnormalities in megagametogenesis and embryology of I. pseudacorus including the occurrence of two embryos in some embryo sacs and the fusion of ovular and antipodal cells with the polar nuclei during ovule development.
My study supported Piper's estimate of low capsule production for I. tenax var. gormanii (Piper, 1924
). Seed set is estimated not to exceed 5% when the combined effects of ovary success and ovule success (22% x 22%) are considered. This estimate is likely to be considerably higher than final seed set because it is assumed that additional developmental failures would occur between 20 d postpollination and maturation of fruits and seeds. Sexual reproduction was considerably more successful in I. tenax var. tenax; seed set is estimated to be 26% when ovary and ovule development are both factored in (75% x 35%).
In this study, the success of ovules maturing into seeds for I. tenax var. gormanii did not deviate significantly from what was expected. The survival of ovules in I. tenax var. gormanii was only slightly less (22%) than in I. tenax var. tenax (35%). Other Iris species in series Californicae, e.g., I. innominata, I. munzii, and I. thompsonii, have ovule developmental success between 33% and 45% (C. Wilson, unpublished data). Studies of Iris species outside of series Californicae provide similar results: 51% of ovules developed into seeds for I. lacustris (Planisek, 1983
) and 41% for I. versicolor (Kron, Stewart, and Back, 1993
). The ovule success found in Iris was typical of reports of ovule success in other outcrossing perennials (Wiens, 1984
).
Sexual reproduction in I. tenax var. gormanii was limited by fruit success. Ovary success resulting from open pollination was only 22% for I. tenax var. gormanii compared with 75% for I. tenax var. tenax. A 70% depression of ovary success in I. tenax var. gormanii results from a comparison of open pollination in the two varieties. Ovary success in I. tenax var. tenax is similar to that of other plants. A survey of reproduction in 447 flowering plants found that in self-compatible hermaphroditic plants, 72.5% of ovaries developed into fruits (Sutherland, 1986
). Kron, Stewart, and Back (1993)
found 77% fruit development for I. versicolor.
My study suggests that pollinator limitation was responsible for a substantial portion of the ovary failure observed in I. tenax var. gormanii. With hand pollination of flowers, 56% of ovaries survived to 17–20 d postpollination as compared with 22% of ovaries from open-pollinated flowers. Although the percentage of normally developed ovaries of hand-pollinated flowers of I. tenax var. gormanii was still less than the 75% from open-pollinated flowers of I. tenax var. tenax, the difference in ovary success between the two varieties was reduced substantially with hand pollination of I. tenax var. gormanii. Others have reported lowered fruit (papers reviewed in Bierzychudek, 1981
; Karoly, 1992
; Burd, 1994
; Irwin, 2000
) and seed production (Snow, 1982
; Burd, 1994
; Irwin, 2000
) due to pollinator limitation.
Concerns have been raised about the experimental design of studies that test pollinator limitation. One concern is that unless all flowers on a plant are included in experimental treatments, the results from treatment and nontreatment plants are not comparable. Seed production for treatment plants is often based on individual flowers while seed production for nontreatment plants is usually based on whole plants (Bawa and Webb, 1984
; Zimmerman and Pyke, 1988
). This type of experimental deficiency may be especially relevant to clonal species such as Iris if the overall amount of available resources for a genet is fixed. Another possible problem related to comparisons of individual flowers with whole plants is the possible reduction of flower initiation due to the onset of maturation of seeds after intensive hand pollinations. Although this concern is relevant to plant species with many-flowered inflorescences, I. tenax typically has only one flower per inflorescence and only one inflorescence per rhizome branch. Another experimental design concern is that greater seed production in a year due to individuals receiving high pollen loads may limit seed production for the following year (Janzen et al., 1980
). Lastly, there is concern that single-year studies may not represent overall patterns due to an unusually favorable or unfavorable year.
The present study was designed to separate reproductive strategies from other limitations to reproduction by comparing reproductive success between sister taxa, as well as between pollination treatments. These sister taxa would be expected to share similar reproductive traits, just as they share geographical and ecological parameters. Both varieties are native to northwestern Oregon, where they are found in open Douglas fir forests. There is also no reason to suspect that availability of resources differs significantly between the three study sites. Pollination experiments were carried out over 2 yr, although the same individual plants were not necessarily used for both years.
Availability of suitable pollen may be limited (as opposed to availability of sufficient pollen), a concern that is less important with self-compatible plants. It is impossible to know what pollen was deposited on the stigmas of open-pollinated ovaries that failed to develop. However, the most abundant flower at the Scoggins Creek site during the study period was Iris, with relatively few other species also in bloom. The only conspicuous, concurrently blooming plant was the introduced shrub, Cytisus scoparius (scotch broom), and it occurs in only a small portion of the study site. The bloom period (early June) is later than most spring- and earlier than most summer-blooming species in northwestern Oregon.
The Scoggins Creek population of I. tenax var. gormanii may have pollinator species that are different (and less efficient) than those at the Corbett and Portland populations of I. tenax var. tenax, or low fruit set could be due to low numbers of pollinators at Scoggins Creek. Bumble bees were commonly seen visiting Iris flowers at both the Corbett and Portland populations but rarely seen at Scoggins Creek. Uno (1982b)
found nectar-collecting bees of three genera, Emphoropsis, Bombus, and Anthophora, were the most common pollinators of I. douglasiana and that Bombus visitors carried the purest pollen loads (94% of pollen was from Iris).
The difference between ovary success for hand-outcrossed vs. open pollinations in I. tenax var. gormanii cannot be explained by the effects of incompatible self-pollen or pollen from closely related plants because ovary success for the taxon was not significantly depressed by self-pollination treatments. This is confirmed by self-compatibility indices, where values of 0.96 (using ovule development) and 0.70 (using ovary development) indicate that I. tenax var. gormanii is self-compatible. Lloyd and Schoen (1992)
suggested that values at or above 0.75, although this number represents an arbitrary boundary, might be considered self-compatible. Given that Iris is a clonal species, a significant degree of self-incompatibility would limit sexual reproduction because pollen loads from insect vectors would typically consist of both outcross and self (geitonogamous, within-genet) pollen.
Although I. tenax var. gormanii was determined to be self-compatible, based upon ovule development, the relative success of ovaries resulting from outcross and self-pollination treatments was slightly below Lloyd and Schoen (1992)
suggested values. It is unknown whether this difference in ovary success was due to incomplete self-incompatibility or postzygotic inbreeding depression. The latter is suspected based on historical information suggesting that the one known population of this variety was flooded in the 1960s during reservoir construction, which may have subjected it to a bottleneck event. The variety is now known from a site 0.75 km from the original site; however, it is unknown whether this is a remnant of the original population or a population on a previously unknown second site.
|
FOOTNOTES |
|---|
2 Address for reprint requests (bwcw{at}odin.pdx.edu
).
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Becerra J. X. D. G. Lloyd 1992 Competition-dependent abscission of self-pollinated flowers of Phormium tenax (Agavaceae): a second action of self-incompatibility at the whole flower level?. Evolution 46: 458-469[CrossRef][ISI]
Bierzychudek P. 1981 Pollinator limitation of plant reproductive effort. American Naturalist 117: 838-840[CrossRef][ISI]
Burd M. 1994 Bateman's principle and plant reproduction: the role of pollen limitation in fruit and seed set. Botanical Review 60: 83-111[CrossRef][ISI]
Dafni A. 1992 Pollination ecology: a practical approach. Oxford University Press, New York, New York, USA
Gasser C. S. K. Robinson-Beers 1993 Pistil development. Plant Cell 5: 1231-1239
Irwin R. E. 2000 Morphological variation and female reproductive success in two sympatric Trillium species: evidence for phenotypic selection in Trillium erectum and Trillium grandiflorum (Liliaceae). American Journal of Botany 87: 205-214
Janzen D. H. P. De Vries D. E. Gladstone H. L. Higgins T. M. Lewisohn 1980 Self- and cross-pollination of Encyclia cordigera (Orchidaceae) in Santa Rosa National Park, Costa Rica. Biotropica 12: 72-74[CrossRef][ISI]
Jensen W. A. 1962 Botanical histochemistry. W. H. Freeman, San Francisco, California, USA
Jensen W. A. 1974 Reproduction in flowering plants. In A. W. Robards [ed.], Dynamic aspects of plant ultrastructure, 481–503. McGraw-Hill, London, UK
Jensen W. A. D. B. Fisher 1968 Cotton embryogenesis: the entrance and discharge of the pollen tube in the embryo sac. Planta 78: 158-183[ISI]
Karagyozova M. 1963 Embryological studies on Iris pseudacorus L. Bulgarische Akademie der Wissenschaften 11: 111-124
Karoly K. 1992 Pollinator limitation in the facultatively autogamous annual, Lupinus nanus (Leguminosae). American Journal of Botany 79: 49-56[CrossRef][ISI]
Kron P. S. C. Stewart A. Back 1993 Self-compatibility, autonomous self-pollination, and insect-mediated pollination in the clonal species Iris versicolor. Canadian Journal of Botany 71: 1503-1509
Lloyd D. G. D. J. Schoen 1992 Self- and cross-fertilization in plants. I. Functional dimensions. International Journal of Plant Sciences 153: 358-369[CrossRef]
Maheshwari P. 1950 An introduction to the embryology of angiosperms. McGraw-Hill, New York, New York, USA
Pande P. C. V. Singh 1981 A contribution to the embryology of the Iridaceae. Journal of the Indian Botanical Society 60: 160-167
Piper C. V. 1924 New flowering plants of the Pacific Coast. Proceedings of the Biological Society of Washington 37: 91-96
Planisek S. L. 1983 The breeding system, fecundity, and dispersal of Iris lacustris. Michigan Botanist 22: 93-102
Riley H. P. 1942 Development of the embryo sac of Iris fulva and I. hexagona var. giganticaerulea. Transactions of the American Microscopy Society 61: 328-335
Russell S. D. 1992 Double fertilization. International Review of Cytology 140: 357-388[CrossRef]
Smith F. H. Q. D. Clarkson 1956 Cytological studies of interspecific hybridization in Iris, subsection Californicae. American Journal of Botany 43: 582-588[CrossRef][ISI]
Snow A. A. 1982 Pollination intensity and potential seed set in Passiflora vitifolia. Oecologia 55: 231-237[CrossRef][ISI]
Sutherland S. 1986 Patterns of fruit-set: what controls fruit-flower ratios in plants?. Evolution 40: 117-128[CrossRef][ISI]
Tilton V. R. H. T. Horner Jr 1980 Stigma, style and obturator of Ornithogalum caudatum (Liliaceae) and their function in the reproductive process. American Journal of Botany 67: 1113-1131[CrossRef][ISI]
Uno G. E. 1982a The influence of pollinators on the breeding system of Iris douglasiana. American Midland Naturalist 108: 149-158[CrossRef][ISI]
Uno G. E. 1982b Comparative reproductive biology of hermaphroditic and male-sterile Iris douglasiana Herb. (Iridaceae). American Journal of Botany 69: 818-823[CrossRef][ISI]
Wiens D. 1984 Ovule survivorship, brood size, life history, breeding systems, and reproductive success in plants. Oecologia 64: 47-53[CrossRef][ISI]
Yasui K. N. Sawada 1940a On the capsule formation, a rare case with special reference to the sterility in Iris japonica Thunb. Botanical Magazine, Tokyo 54: 135-140
Yasui K. N. Sawada 1940b On the spore and embryo sac formation with special reference to the sterility of Iris japonica Thunb. Botanical Magazine, Tokyo 54: 96-102
Zimmerman M. G. H. Pyke 1988 Reproduction in Polemonium: assessing the factors limiting seed set. American Naturalist 131: 723-738[CrossRef][ISI]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
