A survey of floral traits, breeding systems, floral visitors, and pollination systems of the angiosperms of the Juan Fernandez Islands (Chile).
The extraordinary faunas and floras of oceanic islands--that is, islands located over oceanic plates that have never been connected to continental land masses (Nunn, 1994)--with unique assemblages of species and high levels of endemism have historically been of interest to naturalists (e.g., Sloane, 1707-1725; Shelvocke, 1726; Linnaeus, 1747; Juan & Ulloa, 1813; Hooker, 1853). Biologists paid even more attention to them following the publications of Darwin's (1859) and Wallace's (1880) books that highlighted these fascinating small areas of the world as remarkable laboratories of evolution. Subsequently, a great deal of insightful work has been published on island biology, particularly in the last three decades (e.g., MacArthur & Wilson, 1967; Carlquist, 1974; Bramwell, 1979; Mueller-Dombois et al., 1981; Williamson, 1981; Bowman et al., 1983; Lawesson et al., 1990; Vitousek et al., 1995; Wagner & Funk, 1995; Keast & Miller, 1996; Grant, 1998; Stuessy & Ono, 1998; Whittaker, 1998). Although much on the biota, ecology, and evolution of islands has been published, much remains to be studied, especially in terms of the biology of the plants. There is increasing urgency for these studies (Carlquist, 1998; Raven, 1998), because insular species are being lost at a higher rate than are their continental relatives (Reid & Miller, 1989; Smith et al., 1993). This is not surprising, given the often smaller populations on islands and, for many island species, a relatively low competitive ability (Carlquist, 1974). Human disturbance of fragile island habitats is one of the principal causes. Indeed, anthropogenic alterations of oceanic islands have taken place on a greater scale than on most continental systems (Loope et al., 1988; MacDonald & Cooper, 1995; Mittermeier et al., 1999).
The biota of the oceanic islands often has a different combination of species than do equivalent mainland areas (Carlquist, 1965, 1974; Grant, 1998). Because islands are difficult to reach, and each continental species has different dispersal capability, it is inevitable that islands will possess a nonrepresentative sample of the species from the near continents. In addition, chance must play an important role in determining which species arrive, when, and in what numbers. Difficulties of establishment will further influence the composition of the island communities by favoring some types of colonists over others. Another basic obstacle is reproduction. For angiosperms in particular, hermaphroditic self-compatible (SC) species would be favored by enabling reproduction and establishment after long-distance dispersal (Baker, 1955, 1967; Stebbins, 1957). If pollen transfer is required--for example, when there is temporal or spatial separation of the sexes, or self-incompatibility (SI) or dioecy--then the presenc e of the pollinator fauna must also be a consideration in establishment. For the reasons given above regarding plant colonization, pollinator faunas on islands also are comparatively small, with many kinds of pollinators either absent or poorly represented (Ehrendorfer, 1979; Barrett, 1998). Thus, unless abiotic pollination, in particular wind pollination (Carlquist, 1974; Ehrendorfer, 1979; Anderson et al., 2000a, 2001a), can be a factor, the lack of pollinators can lead to failure of successful colonization.
In spite of its importance for conservation and restoration purposes, studies of the reproductive biology of island plants in general (Carlquist, 1974; Ehrendorfer, 1979; Godley, 1979; Lloyd, 1985; Barrett, 1998), and of their breeding systems in particular, are comparatively uncommon (e.g., Rick, 1966; Carlquist, 1974; Pandey, 1979; Carr et al., 1986; McMullen, 1987, 1990; Connor, 1988; Webb & Kelly, 1993; Anderson et al., 2001a). There are no detailed, comprehensive species-based reviews of these aspects on any oceanic island system. Technical and economic difficulties in reaching islands, in locating species in the commonly inaccessible, steep habitats (for islands of volcanic origin) where they grow, and in finding the plants in the right reproductive stage all seem to have contributed to this paucity of studies. Data on pollination-related floral morphological traits of the species may be used to postulate how pollination is accomplished in these isolated island floras. There has been some analysis of th e general characteristics of flowers in the context of plant reproduction. For instance, some authors have called attention to the high representation of small, white or green, simple-shaped flowers in the floras of oceanic islands and have suggested a correlation with the paucity of native pollinators (Wallace, 1895; Carlquist, 1974; Webb & Kelly, 1993).
We chose the Juan Fernandez Archipelago flora for a comprehensive study for a number of reasons. First, we had the opportunity to do much original fieldwork via several expeditions to the islands, mainly to Robinson Crusoe Island, and have published on the reproductive biology of elements of the flora (Bernardello et al., 1999, 2000; Anderson et al., 2000a, 2000b, 2001a, 2001b). Secondly, the angiosperm flora of the archipelago is, like the area of the islands (100.2 [km.sup.2]; Stuessy, 1995), comparatively small (156 taxa, 152 species; Marti-corena et al., 1998) and fairly well known. The flora also has a high level of endemism (about 63%). At 2.1 native species/[km.sup.2] and approximately. 1 endemic species/[km.sup.2], the autochthonous species density of the native flora and the density of endemics is higher on Robinson Crusoe Island than on any other oceanic island. Finally, this flora is among the most threatened in the world (Allen, 1984; Davis et al., 1995; Mittermeier et al., 1999). The native flora is characterized by low fire tolerance and poor adaptation to herbivore resistance (Skottsberg, 1953). The species are especially vulnerable to human-induced disturbance--that is, to the historical and continued foraging by feral goats and rabbits--and to losses of habitat due to aggressive introduced exotic weeds (Perry, 1984; Wester, 1991; Bourne et al., 1992; Stuessy et al., 1997; Jaksic, 1998). Thus, the flora is of manageable size, of particular interest, and highly vulnerable and threatened.
The Juan Fernandez Archipelago is one of the few regions where there were no permanent human settlements before the sixteenth century (Woodward, 1969; Wester, 1991). It consists of three islands, all of volcanic origin: 667 km W of continental Chile are Robinson Crusoe Island (= Masatierra) and Santa Clara Island; Alejandro Selkirk Island (= Masafuera) is 181 km farther west. Robinson Crusoe Island has been dated at approximately 4 million years old; Alejandro Selkirk Island, at 1-2 million years old; and Santa Clara Island, at 5.8 million years old (Stuessy et al., 1984).
In this article, features of the angiosperm flora of the archipelago are analyzed and statistically compared. Although in progress (Stuessy, in prep.) there is not yet available a comprehensive modern flora of the archipelago; thus, the collection of original data was extensive. Information was gathered primarily from our own fieldwork, and also from the existing literature. The published literature offered relatively little direct information on reproductive biology, but much on floral features. Data recorded include habit, plant sex, several flower features, such as size, shape, and color, and the hypothesized pollination system of the first colonizers. In addition, we summarized the available data on compatibility, presence and type of dichogamy, observed floral visitors, presence of floral rewards, and currently known pollination systems. Our goals were to provide a species-based review for as much of the flora as possible of the features relevant to reproduction and pollination, to identify associations and generalizations, to contribute to the understanding of the colonization and evolution of the angiosperms of the archipelago, and to promote use of these data for conservation, preservation, and restoration.
III. Methods and Materials
The recent catalogue of the vascular flora of the Juan Fernandez Islands by Marticorena et al. (1998) was used for the list of species on the archipelago and their habit. Information on sex and flower features are our own observations, mainly on Robinson Crusoe Island, supplemented by data extracted from Gay (1845-1854), Hemsley (1884), Johow (1896), Reiche (1896-1911), Skottsberg (1921, 1928, 1953), and Moore (1983). Data on breeding system, floral biology, presence of reward in the flowers, floral visitors, and pollination are largely personal observations, supplemented by the published data on visitors extracted from Johow (1896), Skottsberg (1928), Brooke (1987), Meza (1988), and Colwell (1989), and on breeding systems of some species from Ramanna and Hermsen (1981) and Moore (1983). The pollination biology of the hypothesized progenitors or colonizers was estimated after reviewing the literature on the closest continental relatives and publications on other southern Pacific oceanic islands (e.g., Carlqui st, 1974; Porter, 1983; Wagner et al., 1990).
Our observations and experiments were largely confined to natural populations on Robinson Crusoe Island (cf. Bernardello et al., 1999, 2000; Anderson et al., 2000a, 2001 a). In addition, many experimental studies, especially on compatibility, were conducted with native plants grown in the Corporacion Nacional Forestal (CONAF) gardens adjacent to San Juan Bautista, the only permanent settlement on the archipelago. Finally, a limited number of observations were made with material from Alejandro Selkirk Island.
To facilitate the comparisons, flower sizes were placed into five categories based on the flower area (lengthxwidth) as follows: very small = <9 [mm.sup.2], small = 10-25 [mm.sup.2], medium-sized = 30-110 [mm.sup.2], large = 120-400 [mm.sup.2], very large = >400 [mm.sup.2]. We recognized these groups by plotting the average data obtained for every species of the flora and identifying the emergent groups. Shape categories follow the classification of Faegri and van der Piji (1979)--but see our comment in section A.3 below. Color categories were defined by reducing all of the kaleidoscope to just six color categories: yellow, white, green, blue (including purple, lavender, and violet), brown (including straw colored), and red (including pink and orange); orange was considered separately only when comparing flower color with current pollination. When a species had flowers with two or three colors (either multicolored or with different nuances), we coded the most prominent color. All comparisons were made at the species level, except for the hypothesized pollination of colonizers, which we recorded at the genus level.
Statistical comparisons were made using the Statistical Analysis System (SAS for Windows version 8.00, 1999, Cary, NC). The hypothesis tested was that there was no association among the variables. The PROC FREQ was performed using the general association statistic ([Q.sub.CMH], Stokes et al., 1995), which produces both the Pearson chi-square statistic and the Mantel-Haenszel statistic as well as the general association coefficient. The analyses were applied first to the whole series of data in each category. Then, if the result was significant at p <0.01, the same analyses were applied to pairwise combinations of variables to identify significant associations.
All relevant data from our studies and from the literature are summarized in Table I. An analysis of each feature, and pairwise comparisons, where appropriate, follow.
A. HABIT AND REPRODUCTIVE FEATURES
About 85% of the flora comprises perennials: 38% are perennial herbs, 23% are shrubs, 16% are trees, and 9% are shrubs/trees.
2. Flower Size
The majority of the species have small flowers (60%), falling into the very small (46%) or small (14%) ranges. Medium-sized flowers are found in 24% of the species of flora. Large (9%) and very large (7%) flowered species are uncommon.
3. Flower Shape
Most shape descriptors are obvious; the term "inconspicuous" is not. Faegri and van der Pijl (1979) and Bell (1991) use "inconspicuous" basically as a "shape" descriptor. However, this shape descriptor is defined as "with no optical attraction" by Faegri and van der Pijl (1979). Thus, flowers in this category would not seem to have a shape per se. Bowing to convention, we use the term "inconspicuous" as shape descriptor, but we also recognize that all such flowers are very small or small, where shape is essentially not a consideration.
Inconspicuous and dish-shaped flowers are widespread among the angiosperms of the archipelago, constituting 41% and 35% of the species, respectively. Tubular flowers characterize 15% of the species, and bell-shaped flowers only 6%. Finally, species with flag- (2%) and brush-shaped (1%) flowers are rare.
4. Flower Color
Green is the most widespread flower color among the species (41%). White and yellow flowers follow in importance, with 26% and 12% of the species, respectively. The remaining colors are present in much smaller percentages (red 9%, brown 7%, and blue 5%).
As expected, the majority of the species on the archipelago are hermaphroditic (about 70%). Some 9% of the species are dioecious, and 9% monoecious. Dioecious species are included in the dicot genera Coprosma, Fagara, Pernettya, and Robinsonia; Juania is the only dioecious monocot in the archipelago. Monoecious dicots are in the genera Boehmeria, Dysopsis, and Urtica, and there are several monoecious Cyperaceae. The remaining sexual systems are found exclusively among the dicots: Gynomonoecious species include 7% of the flora and are in Abrotanella, Cuminia, Chenopodium, Erigeron, and Lactoris; the one gynodioecious species is in Rhaphithamnus; polygamous taxa are scarce and included in Empetrum and Parietaria, as are andromonoecious taxa included in Cyperaceae and Asteraceae.
6. Separation of the Sexes
Because field studies of the floral biology are particularly meager, data on such characteristics as dichogamy and herkogamy are very limited. The combination of our direct observations and Skottsberg's observations (1928), together with generalizations from published data for some continental genera or families, indicate that around 30% of the species are or should be protandrous and 7% protogynous. Although these data are not complete, they do show that temporal separation of the sexes is not uncommon. Protandry was found mainly among members of Asteraceae, Campanulaceae, and Gunneraceae, with protogyny among the genera Drimys, Eryngium, Loctoris, Myrceugenia, and Plantago. The only orchid known in the archipelago, which grows on Alejandro Selkirk Island, is herkogamous.
Only about 14% (21 species) of the flora has been studied sufficiently to determine compatibility. Some 85% of the species studied (18) are SC, and only three are SI.
8. Floral Rewards
This information is also not exhaustive. Our direct observations (Bernardello et al., 2000; Anderson et al., 200la), combined with our general knowledge of related South American genera, indicate that at least 55% of the species do or are likely to offer nectar but that only about 2% offer pollen as a reward.
9. Floral Visitors
Overall, floral visitors are rare to uncommon. Combined, our direct observations and those of Skottsberg (1928) cover about 35% of the species. These data indicate that 29 species (54% of those studied) never had any floral visitors, even though those species have been carefully observed in the field for considerable periods of time. Both sexes of two hummingbird species (Sephanoides fernandensis, an endemic, and S. sephaniodes, native, but also found in Chile south of the Atacama Desert [Roy et al., 1998]) were recorded as regular visitors for 14 species of the total flora. We presume that they are regular pollinators for these, chiefly woody species in the genera Centaurodendron, Cuminia, Dendroseris, Eryngium, Escallonia, Greigia, Lobelia, Nicotiana, Notanthera, Ochagavia, Rhaphithamnus, and Sophora. Skottsberg (1928) reported pollen from three of those species (Rhaphithamnus venustus, Cuminia eriantha, Escallonia callcottiae) on the heads of captured hummingbirds.
Native insects (flies, moths, and beetles) have only rarely been recorded on flowers. Sporadic insect visitors were recorded for only 7% of the total flora; that is, 11 species in nine genera. Decreasing this total even further is the fact that two of the species visited by native insects, Escallonia callcottiae and Eryngium bupleuroides, are hummingbird pollinated. Furthermore, the insect visits we have documented only very doubtfully constitute effective pollination because of the casual nature of the visits and the lack of fidelity of the visitors (Anderson et al., 200la). More recent discoveries indicate that there are, in addition, an introduced ant (Linepithema humile, E. O. Wilson, pers. comm.; Anderson et al., 2001a) and an (doubtfully) endemic bee in the genus Lasioglossum (Anderson et al., 2000b; Engel, 2000.). Neither the bee nor the ants are yet important to the reproduction of the native flora.
From the data above, we can confidently conclude that about 9% of the total flora is bird pollinated. Given the length of study and the conspicuousness of these pollinators, we have reasonable confidence that this is, indeed, an accurate figure for the flora in toto. On the other hand, given the scant, irregular, and imprecise nature of native insect visitation, there is no certainty that any of the plant species any truly insect pollinated (with the possible exception of Libertia; see below). Based on floral morphology, floral biology, compatibility, and effective seed set in the absence of pollinators, we conclude that about 47% of the flora is likely wind pollinated. The pollination for the remaining 44% of the flora is unknown. Among the anemophilous species, some have most of the typical features, not surprisingly among the Chenopodiaceae, Cyperaceae, Juncaceae, Poaceae, Urticaceae, and the genera Coprosma, Dysopsis, Empetrum, Gunnera, Haloragis, Peperomia, Plantago, and Lactoris (Bernardello et al., 199 9). However, other species show combinations of adaptations to wind pollination and other features that would indicate biotic pollination (e.g., Azara, Drimys, Pernettya, and Robinsonia; Anderson et al., 2000a). A special case of wind as an indirect force of pollen transfer was found in Wahienbergia berteroi (Anderson et al., 2000b). A summary of the pollination is presented in section V, Discussion.
11. Hypothesized Pollination of Colonizers
Based on the pollination biology known for continental relatives, we estimate that about half of the colonizer species were primarily insect pollinated. Wind-pollinated colonizers likely constituted a comparatively high proportion as well (40%), whereas bird-pollinated colonizers and colonizers with mixed pollination account for much lower percentages (7% and 4%, respectively).
B. COMPARISONS OF HABIT AND FLORAL FEATURES
1. Habit and Flower Size
Species that are perennial herbs usually have diminutive flowers (70% very small, p <0.0001, Table II; 14% small). Tree species are at the extremes, in that they have mostly very large (36%) or very small (21%) flowers, but all size classes are represented. Other species with a perennial habit have mainly medium-sized flowers (62%).
2. Sex and Habit
Hermaphroditic species show all of the possible habits. Monoecious species are mostly perennial herbs or trees; all dioecious species are woody.
3. Sex and Flower Color
Most species are hermaphroditic, and there is not a dominant color, although the relatively dull white (38%, p = 0.0003, Table II) and green (31%) hermaphroditic flowers are the most frequent. More than 93% of the monoecious taxa have green flowers (p<0.0001, Table II), and green and yellow flowers are predominant in both gynomonoecious and dioecious species.
4. Flower Shape and Flower Color
Species with inconspicuous flowers possess predominantly dull colors: 71% are green (p <0.0001, Table II), and 10% are brown (Fig. 1A). Species with dish-shaped and tubular flowers show the full range of colors, but in both types white is most common (36% and 32%, respectively, Fig. 1A). Bell-shaped flowers are also most often white (85%, p 0.0005, Table II); a few species have either red or green flowers (Fig. 1A).
5. Flower Size and Flower Color
Most of the species (Fig. 1B) with very small flowers have green flowers (67%, p<0.000 1, Table II). Similarly, small-flowered species generally are not brightly colored: 56% are white, 19% brown, and 19% green (Fig. 1B). The species with very large flowers (Fig. 1B) are exclusively white (66%) or red (33%). The species with medium-sized flowers show the full range of colors, with white (41%) and yellow (33%) most common (Fig. 1B).
6. Flower Size and Sex
Hermaphroditic species show all of the possible flower sizes: About half (45%) have very small, 16% small, and 18% medium-sized flowers. Monoecious taxa are characterized almost exclusively (92%) by very small flowers. Dioecious species possess mainly medium-sized flowers (84%), and most gynomonoecious species bear medium-sized (55%) or very small (36%) flowers.
7. Flower Shape and Size
Not surprisingly, all of the substantial percentage (41%) of the flora (Fig. 2) that is characterized by the "inconspicuous" shape has flowers that are either very small (94%, p<0.0001, Table II) or small (6%). Species with dish-shaped flowers are commonly medium-sized (44%, Fig. 2). Species with tubular flowers have flowers more or less equally distributed by size: small (37%), medium-sized (37%), or large (21%, Fig. 2).
8. Flower Shape and Sex
Given that hermaphroditic flowered species are the majority in the flora, it is also not surprising that their flowers show all of the possible shapes. Similarly, species with inconspicuous and dish-shaped flowers are distributed over all of the sexual systems. Monoecious species bear primarily inconspicuous flowers (85%), but dioecious species generally possess dish-shaped flowers (61%).
C. COMPARISONS OF CURRENT POLLINATION WITH HABIT AND FLORAL TRAITS
We have good data on pollination systems for about 55% of the flora, and it is with these species that we make the following comparisons. As indicated previously, most of the species for which we have good information are either wind (47%) or bird (9%) pollinated. Only Libertia may be reliably insect pollinated.
All of the bird-pollinated species are perennial, ordinarily woody (shrubs and trees). Wind-pollinated species are perennial; the majority are herbs (60%, p = 0.0002, Table II).
2. Flower Size
The flowers of the bird-pollinated species are comparatively larger (large 46%, medium-sized 38%, very large 15%, Fig. 3A); when combined, large and medium-sized flowers proved to be associated with bird pollination (p <0.0001, Table II). On the contrary, most wind-pollinated species have very small (70%, p <0.0001, Table II) or small flowers (14%, Fig. 3A); both categories pooled were associated with wind pollination (p <0.0001, Table II).
3. Flower Shape
The floral features of most species show adaptations coincident with their pollination system (Fig. 3B). Thus, most bird-pollinated species have tubular flowers (57%). A substantial percentage (28%) are also dish-shaped flowers; however, these are all Asteraceae and, as required in Faegri and van der Pijl's classification (1979), are considered "dish" shaped, although each individual flower is obviously a diminutive tube. As would be predicted, wind-pollinated species have chiefly inconspicuous flowers (7l%,p <0.0001, Table II). However, there are also species with flower shapes that are not typically associated with wind pollination: bell (16%), tube (7%), or dish (5%).
4. Flower Color
Bird-pollinated species regularly have flowers with bright colors (red, orange, yellow, and mainly blue [36%]; Fig. 4A, colors combined: p <0.0001, Table II). On the other hand, flowers of wind-pollinated species are usually not showy (Fig. 4A), green (50%, p = 0.01, Table II) or brown (32%); both colors combined were associated with wind pollination (p <0.0001, Table II).
Some 80% of the bird-pollinated and 68% of the wind-pollinated species bear hermaphroditic flowers (Fig. 4B). The other sex forms are largely wind pollinated (Fig. 4B).
D. COMPARISONS OF THE HYPOTHESIZED POLLINATION OF COLONIZERS WITH HABIT AND FLORAL FEATURES
The following are obviously speculative, but the generalizations are informative in that they establish the likely evolutionary challenges faced by the colonizing species.
No typical habit seems to have been typical of insect-pollinated colonizers, but bird-pollinated colonizers likely had a perennial habit (from perennial herbs to trees). Within the species that are perennial herbs and shrubs, all pollination types were represented. Perennial herbs seem to have been primarily wind pollinated (53%) or insect pollinated (40%). Shrubs, however, were mostly (55%) insect pollinated, with no other class particularly strongly represented. Trees were mostly wind pollinated (60%, p = 0.03, Table II) or insect pollinated (37%).
2. Flower Size
The full range of flower sizes is found in insect-pollinated colonizers, although most would have been medium-sized or very small flowers (30% and 24%, respectively, Fig. 5A). As expected, very small flowers were typical of wind-pollinated (73%, p <0.0001, Table II) colonizers; large (50%) or medium-sized (33%) flowers were common in bird-pollinated colonizers (Fig. 5A).
3. Flower Shape
As expected, tubular flowers characterized all bird-pollinated colonizers, and inconspicuous flowers were common in almost all wind-pollinated species (85%, p <0.0001, Table II, Fig. 5B). The most frequent flower shapes among insect-pollinated colonizers were dish (50%, p <0.0001, Table II) and bell (20%, Fig. 5B).
4. Flower Color
The full range of floral colors was represented in insect-pollinated colonizers, with a predominance of white (56%, p <0.0001, Table II) or green (22%) colors (Fig. 6). As one might predict, showier flowers (blue 60%, red 20%, p <0.0001 for both colors combined, Table II) were typical of bird-pollinated colonizers (Fig. 6). The species with less showy green (82%, p <0.0001, Table II) or brown (9%) flowers were common in wind-pollinated colonizers (Fig. 6).
Given the sex distribution of the primary source flora, it is not surprising that hermaphroditic flowered species were the most common in the different pollination systems (78% of the insect-, 74% of the wind-, and 75% of the bird-pollinated species). However, dioecious colonizers seem to have been primarily insect pollinated (80%) or wind pollinated (20%). Monoecious colonizers are proposed as insect or wind pollinated (50% each), as are the few gynomonoecious species (40% insect and 40% wind).
E. CORRELATION BETWEEN CURRENT POLLINATION AND HYPOTHESIZED POLLINATION OF COLONIZERS
We assessed feasible changes in the pollination system for 39 genera (Table I), based on pollination system determination from our current studies together with our hypotheses of possible colonizers (see our discussion in the section above). About 70% of the genera seem to have retained the pollination system of the colonizers. Those that retained wind pollination include members of several monocot (Cyperaceae, Juncaceae, and Poaceae) and dicot families (Chenopodiaceae, Empetraceae, Euphorbiaceae [Fig. 7C], Gunneraceae, Haloragaceae, Urticaceae [Fig. 7D], Piperaceae, Plantaginaceae [Fig. 7A], Rubiaceae [Fig. 7B]). We propose the retention of hummingbird pollination for the Bromeliaceae (Greigia. Ochagavia), Campanulaceae (Lobelia), Fabaceae (Sophora), Lamiaceae (Cuminia [Fig. 8C]), Loranthaceae (Nothanthera), and Saxifragaceae (Escallonia [Fig. 8A]). The only species that seems to have retained insect pollination is Libertia.
At least some of the species of the other 30% of the genera bear a pollination system different from that of their presumed colonizers. For four genera, we suggest that the ancestral insect-pollination system switched to hummingbird pollination (Rhaphithamnus venustus, some Eryngium, Centaurodendron, and some Dendroseris [Fig. 8B]). Nicotiana cordifolia is now hummingbird pollinated, but the first colonizers could have been insect and/or hummingbird, so it is not included on the list of the species with changes. In addition, for seven other genera we suggest that the colonizers were likely insect pollinated and that on the islands they have become, of necessity, wind pollinated (Azara, Drimys [Fig. 9B], Juania, Pernettya, Robinsonia, Ugni [Fig. 9A]). Solanum species (Fig. 10B) may have changed from insect pollinated to pure autogamy, but autogamy is not uncommon in this genus, so this failsafe system may even have facilitated establishment of colonizers. In Wahlenbergia (Fig. 10A), we suggest that the ancesto rs were mainly insect pollinated, with a secondary system of autogamy, but that on the islands, and in the absence of insects, the selfing mechanism became the rule. The wind-aided mechanism of autogamy of W. berteroi must have arisen on the archipelago.
Like island systems as a whole (e.g., Darwin, 1859; Carlquist, 1965, 1974; Bramwell, 1972; Wagner et al., 1990), the Juan Fernandez Archipelago includes an inordinate number of perennial forms. There are shrubs/simple trees, for instance, in several Asteraceae endemic genera (e.g., Centaurodendron, Dendroseris, Robinsonia, Yunquea) and in individual species of Erigeron and Plantago whose presumed ancestors and relatives are smaller, herbaceous forms (Carlquist, 1965, 1974; Sanders et al., 1987). Molecular studies have provided confirmation of the evolution of arborescence in several plant groups from Hawaii (Wagner et al., 1990; Givnish et al., 1995; Baldwin, 1997; Sakai et al., 1997; Givnish, 1998) and the Canary Islands (Bohle et al., 1996; Kim et al., 1996; Francisco-Ortega et al., 1997; Panero et al., 1999).
Many hypotheses have been proposed to explain the preponderance of the woody habit, some with implications for reproductive biology. Wallace (1878) suggested that longevity (perennial habit) increased the opportunity for successful sexual reproduction when pollinators were scarce--as they are on some islands. Pollinators, other than hummingbirds, are certainly uncommon at best on the Juan Fernandez Islands. Bohle et al. (1996) went farther, claiming that Echium species from islands off the coast of Africa evolved the woody habit (from herbaceous perennial continental ancestors) to promote not just sexual reproduction but outcrossing--to reduce the inbreeding depression associated with the selfing of the initial colonizers. They also suggested that a secondary manifestation of outcrossing was rapid and extensive speciation on the islands. This latter pattern is perhaps foretold in the chromosomal races of the house mouse established on Madeira as well (Britton-Davidian et al., 2000).
B. FLORAL FEATURES
Angiosperms from oceanic islands are generally characterized by small, inconspicuous (recall = shape, "with no optical attraction") flowers versus showy, bright flowers with various shapes (e.g., Wallace, 1895; Carlquist, 1965, 1974), a trend particularly remarkable in the New Zealand flora (Godley, 1979; Lloyd, 1985; Webb & Kelly, 1993). The Juan Fernandez flora provides strong support for these generalizations: 60% of the flowers are very small/ small (and thus without "shape") and the most frequent flower colors are comparatively dull (green and then white, in decreasing order of frequency). Furthermore, this study, done species by species for the entire archipelago flora, is the first to statistically confirm the expected strong association among very small, inconspicuous, and green flowers. Finally, comparisons with the hypothesized colonizer flora imply that there has not been selection for change in flower color or size (selection neutral). This latter hypothesis implies that the small, inconspicuous f lowers characteristic of islands may better reflect selection for dispersal and establishment than adaptation in situ.
The distribution of flower colors on other islands (e.g., the Galapagos; McMullen, 1989, 1999) is comparable. This common pattern could be explained in several ways. For whatever reason, species with small green and white flowers may possess greater dispersability, or higher levels of SC. At least for the Juan Fernandez Archipelago, these species do not show a higher percentage of SC than do other elements of the flora. Another explanation is that these species with simple, nonshowy flowers may succeed by being serviced by a broader spectrum of pollinators. This hypothesis fails on the Juan Fernandez Islands because of the virtual lack of insect pollinators. Finally, species with these less distinctive flowers may be abiotically pollinated, a hypothesis supported by the high percentage of wind-pollinated species in the flora.
The bulk of the flora has hermaphroditic flowers (ca. 70%). Thus, as would be expected, of this majority condition among angiosperms, the island flora exhibits the range of habits and flower colors, sizes, and shapes. There are about twice as many monoecious (9%) and dioecious (9%) species in the Juan Fernandez Archipelago as there are in angiosperms as a whole (5% and 4% for monoecy and dioecy, respectively; Richards, 1997).
The Fernandezian monoecious species--included in the Euphorbiaceae, Urticaceae, and Cyperaceae--are largely herbaceous and have mostly very small, inconspicuous, green flowers. Because these families in general include many monoecious species (Cronquist, 1981), it is likely that the colonists in those groups were also monoecious. Most of these monoecious species are, and presumably were, wind pollinated--again a general feature of monoecious species (Richards, 1997).
The highest levels of dioecy among flowering plants are found in oceanic islands (Carlquist, 1974; Richards, 1997). Although dioecy in the Juan Fernandez flora is twice the world average, it is not as high as in some other islands, such as New Zealand or Hawaii (Godley, 1979; Sakai et al., 1995a, 1995b, respectively). Furthermore, Arroyo and Uslar (1993) determined that about 9% of the sclerophyllous montane flora of central Chile is dioecious. Thus, the frequency of dioecy in the Juan Fernandez Archipelago is similar to that in a latitudinally comparable flora--and the presumed main continental source. This association characterizes other island systems as well (cf. Baker & Cox, 1984). All of the Fernandezian dioecious species are woody, and most are shrubs. Thus, dioecy is related to woodiness, as it has been in other floras (e.g., Baker, 1959; Bawa, 1980; Sakai et al., 1995b). This association is attributed to strong selection for outcrossing in large, long-lived plants that might otherwise self-cross (cf. Bawa, 1980). Dioecy also is not equally distributed among the various habits in the continental source flora in Chile (Arroyo & Uslar, 1993). Some 62% of the dioecious species are long-lived, perennial forms, and 17% of the dioecious perennials are shrubs.
In addition, dioecy is overrepresented on the Juan Fernandez Islands in species with small green flowers, a pattern also reported in Hawaii (Sakai et al., 1995a, 1 995b). Such flowers are often abiotically pollinated, as is true for all of the Fernandezian dioecious species, and for many other floras as well (e.g., Conn et al., 1980; Freeman et al., 1980; Muenchow, 1987; Steiner, 1988), including that of the Hawaiian Islands (Sakai et al., 1995b).
As indicated above, most of the dioecious species likely had dioecious ancestors. Almost all of the species of Coprosma known worldwide are dioecious (Oliver, 1935). Thus, we presume that the original colonists in this genus were dioecious as well. The island progenitors of Juania and Fagara may have been, at least, incipiently dioecious or have a clear tendency to it, because there are dioecious Fagara species in Hawaii (Wagner et al., 1990), and there are many dioecious palms (Cronquist, 1981). In two genera, it seems likely that dioecy arose in situ. The dioecy in Robinsonia may have had SI ancestors (Crawford et al., 1998; Anderson et al., 200 la). Pernettya rigida is the only cryptically dioecious species known so far in the islands. Because of this, careful anatomical, morphological, and experimental studies were required to understand that the flowers of P. rigida, though complete, are functionally dioecious (Anderson et al., 2000a). Most continental species of Pernettya bear hermaphroditic flowers, bu t some have unisexual flowers (Arroyo & Squeo, 1987; Anderson et al., 2000a). Thus, the first colonists of Pernettya may have been hermaphroditic, but perhaps with incipient dioecy expressed (Anderson et al., 2000a).
About 7% of the flora is gynomonoecious, with representatives from the Asteraceae, Chenopodiaceae, Cuminia (Lamiaceae), and Lactoris (Lactoridaceae). A review of the sex distribution in the first two families indicates that gynomonoecy seems likely to have been present in the colonizers. On the other hand, in Cuminia, the variability of sex expressions both we and Skottsberg (1928) observed makes it more likely that the gynomonoecy arose in situ. Lactoris is a monotypic endemic with debatable relationships (Stuessy et al., 1998a), so it is impossible to confidently project its colonist relatives.
There are few examples of gynodioecy, polygamy, and andromonoecy in this archipelago. The only species studied carefully is Rhaphithamnus venustus (Sun et al., 1996; Anderson et al., 200 la). The gynodioecy in this species seems to have arisen in situ and to be incipient, given the variability in sex expression (e.g., fertile and infertile anthers are similar).
D. BREEDING SYSTEM
"Baker's law" (Baker, 1955, 1967; Stebbins, 1957) proposes that the species most likely to become established after long-distance dispersal will be SC. In addition, the subsequent acquisition of self-pollination would favor establishment (Baker, 1955, 1967; Carlquist, 1974;
Ehrendorfer, 1979; Barrett, 1998). Even though comprehensive surveys of the compatibility status of island plants are surprisingly few (Barrett, 1998), the data that do exist support this hypothesis (e.g., Stephens, 1964; Rick, 1966; Gillett & Lim, 1970; Strid, 1970; Carlquist, 1974; Kores, 1979; Pandey, 1979; Roelofs, 1979; Rabakonandrianina, 1980; Cory, 1984; Lowrey & Crawford, 1985; McMullen, 1987, 1990; Webb & Kelly, 1993). More than 85% of the Fernandezian angiosperms studied carefully are SC (Bernardello et al., 1999; Anderson et al., 2001a). Thus, although SC is likely frequent in the Juan Fernandez Archipelago, the level of autogamy (automatic selfing within the same flower) seems to be quite low, at least in comparison with the Galapagos flora, where 26 of the 29 native and endemic species examined experimentally showed some degree of automatic self-pollination (McMullen, 1987, 1990). The lower rate of autogamy on the Juan Fernandez Islands may be attributable to the significant percentage of species that have temporal separation of the sexes through dichogamy (mostly protandry). A significant level of dichogamy also characterizes the New Zealand flora (Godley, 1979). Nonetheless, self-crossing is common, but in the form of geitonogamy (pollen transfer among flowers of the same individual). Geitonogamy is the most frequent mechanism of pollen transfer among the cosexual Fernandezian species we studied experimentally. And, in general, geitonogamy is probably the most widespread mode of self-pollination, virtually inevitable in SC plants that produce a number of flowers open at the same time (Lloyd & Schoen, 1992). In the Faroes Islands, Hagerup (1951) indicated that species with large inflorescences tended to be geitonogamous. However, Lloyd (1992) also pointed out that most cosexual SC species may be unable to avoid mixed mating--as would most likely occur in all of the SC hummingbird-pollinated species on the Juan Fernandez Islands.
In some Wahlenbergia taxa (W. fernandeziana and a natural hybrid between this species and W. grahamiae; Anderson et al., 2000b), autogamy perhaps provides an alternative to the presumed entomophilous pollination system of colonizers. Features of the Fernandezian Wahlenbergia, like protandry, nectar, and secondary pollen presentation (by pollen-collecting hairs on the style), reduce self-pollination and suggest ancestral allogamy and biotic pollination. The pollination biology of Wahlenbergia elsewhere (Lloyd & Yates, 1982; Petterson, 1997) also supports these suppositions. Autogamy is implicated by a very low P/O ratio and is promoted by SC and partial overlapping of male and female phases at the end of each flower's lifetime and by recurvature of stigmatic branches all the way either back to the style or to the inner corolla surface.
Three SI species are confirmed so far for the Juan Fernandez Islands (Anderson et al., 2001a). In other island systems, a few SI species were found as well (e.g., Carpenter, 1976; Corn, 1979; Grant & Grant, 1981; Carr et al., 1986). At least for two of the SI Fernandezian taxa, the Dendroseris spp., the ancestors are proposed to have been SI, based on the ubiquity of this condition among its extant closest relatives (Crawford et al., 1998; Anderson et al., 200la).
Few species seem to offer pollen as a reward, but a significant number produce nectar (Skottsberg, 1928; Bernardello et al., 2000; Anderson et al., 2001a). However, with the exception of the two species of hummingbirds, there are virtually no floral visitors to collect this reward.
In general, nectary morphology of individual species follows the general pattern of the respective families, suggesting that nectaries came with the colonizers (Bernardello et al., 2000). In the small number of species in the archipelago that are hummingbird visited, nectar chemistry implies specialization for birds, again suggesting that the colonists were also ornithophilous. Likewise, the presence of nectaries and nectar in most non-hummingbird-pollinated species is most likely an indication of the pollination system of the first colonizers. Given the virtual lack of dedicated insect pollinators, the retention (or evolution of) nectar would not seem to reflect the extant biota. Particularly good examples of retention of elements of an ancestral system are the wind-pollinated, nectar-producing species, such as Pernettya rigida, Wahlenbergia berteroi, and the species of Robinsonia. The ancestors of P. rigida seem to have been insect pollinated, but today this cryptically dioecious species is wind pollinated, and visitors are extremely rare and ineffective (Anderson et al., 2000a). Wahlenbergia colonizers seem to have been entomophilous as well, but current pollination is largely autogamous (Anderson et al., 2000b). The situation in Robinsonia is not well understood. Their capitula are yellow/green and their flowers likely bear nectar, but visitors are particularly rare. In 1896, Johow reported flies visiting the genus, without citing the species. Skottsberg (1928) recorded no visitors, and in many hours of field observation in several expeditions, we never recorded any floral visitors either. Thus, given reasonable fruit set in the field, we postulated wind pollination (Anderson et al., 2001a), considering that shifts between insect and wind pollination have been reported to occur several times in this large family (Stebbins, 1970). Of course, fruits may be set apomictically as well, but at present we have no evidence of this.
In general, islands possess comparatively fewer animal species than do their source continents (MacArthur & Wilson, 1967). In consequence, their pollinator faunas are often smaller, with many groups completely absent (Carlquist, 1974; Woodell, 1979; McMullen, 1990, 1999; Inoue, 1993). The Juan Fernandez Islands can now become the exemplar of this pattern: Other than two species of hummingbird, floral visitors are absent or rare (Johow, 1896; Skottsberg, 1928; Bernardello et al., 1999; Anderson et al., 2001a). This small archipelago is notable because of the existence of the two hummingbird species, highlighted because one of them is the only endemic known on oceanic islands (Brooke, 1987; Meza, 1988; Colwell, 1989; Roy et al., 1998). The diet of the hummingbirds includes nectar from a total of 14 autochthonous plant species (Brooke, 1987; Meza, 1988; Colwell, 1989; Bernardello et al., 2000; Anderson et al., 2001a). In general, our analyses show that the ornithophilous species are woody and have tubular flower s with significant amounts of nectar. The chemical nectar composition of five of these species (Bernardello et al., 2000) indicates that they have a low concentration and high sucrose proportion, features common to the nectar of hummingbird-pollinated species (Pyke & Waser, 1981; Cruden et al., 1983; Baker & Baker, 1983a, 1983b, 1990). Rhaphithamnus venustus is the "most visited" ornithophilous endemic species (Brooke, 1987; Meza, 1988; Colwell, 1989; Sun et al., 1996; Anderson et al., 2001a). This species has a long flowering period and produces a large number of flowers with abundant, mostly sucrose-dominant nectar (Bernardello et al., 2000). In addition to the indigenous flora, these hummingbirds also visit flowers and feed on nectar from cultivated and adventitious species (Brooke, 1987; Meza, 1988; Colwell, 1989; Bernardello et al., 2000; Anderson et al., 2001a).
The Fernandezian insect fauna is small in general (cf. Johow, 1896; Kuschel, 1952; Wilson, 1973) and notably lacks species usually dedicated to floral visits (Skottsberg, 1928; Bernardello et al., 1999; Anderson et al., 2000a, 2000b). Consequently, it is not surprising that the records of insect visits to flowers are unusual. For instance, a few insects (beetles and moths) were observed on Berberis, Colletia, and Robinsonia species, either at the end of the nineteenth century (Johow, 1896) or early in the twentieth (Skottsberg, 1928); since then, there are no other records. It is worth mentioning that Skottsberg (1928) spent a total of at least six months in the field and recorded insects on only seven plant species (Table I). Although he gave no indication of the number of individual insects observed for each, based on his lack of citation of large numbers in an extensive text and the current dearth of flower visitors, we presume floral visitors have been rare. In more than 300 hours of observation in three field expeditions, we recorded only 23 floral visits by native insects. Furthermore, and with one exception (see below) the insects recorded are not specialists and, in general, should not be considered pollinators. Evidence of this comes from their behavior and from the absence or small amount of pollen on their bodies (Anderson et al., 2001a). They visit stamens, nectaries, corollas, leaves, other plant parts, and inanimate objects with equal interest and have very low or no fidelity to flowers in general (Anderson et al., 2000a, 2000b, 2001a). A possible exception is a newly described halictid bee species (Lasioglossum subgenus Dialictus; Engel, 2000). For the moment, this bee is a new endemic to the Robinson Crusoe Island. However, it is more than likely that it is a relatively new adventive to the island and that it represents an as yet uncollected and perhaps uncommon new continental species, rather than a continental waif that arrived on the island and evolved in situ (Anderson et al., 2001b).
Much more abundant were the ants we collected in the flowers on some endemics. Ant specialist B. O. Wilson identified the ants as a recently introduced, highly invasive species from South America (Linepithema humile, "Argentine ants"). Because of their presumed relatively recent introduction, their hard and small bodies, and the lack of directed movement among the flowers, we consider that they are not yet significant in transferring pollen and, more importantly, played no role in the evolution of the flora or its pollination.
G. CURRENT POLLINATION
Carl Skottsberg was the first to carefully study pollination biology in the context of his three-year study of the floristics of the Femandezian plants (1928). Nevertheless, many of his conclusions were inferred from flower morphology and pollination of closest continental relatives rather than from actual observations. For instance, of the 49 species he reported as entomophilous (1928), he actually observed insects on flowers for only 7. The species he proposed as anemophilous were based largely on knowledge of families from the continent. For the ornithophilous species, he did observe hummingbirds visiting flowers for 9 species, including, for 3 species, the observation of pollen on the faces of collected birds. Since Skottsberg's work (1928), there was virtually no study of the reproductive biology until very recently (brief comments as part of some broader study include Crawford et al., 1990; Ricci & Eaton, 1994; focused studies include Sun et al., 1996; Bemardello et al., 1999, 2000; Anderson et al., 200 0a, 2000b, 2001a).
1. Hummingbird Pollination
The Juan Fernandez Islands are notable for having hummingbird pollination. The species that hummingbirds visit have abundant nectar and possess a floral structure suited to bird pollination (Bernardello et al., 2000; Anderson et al., 2001a). Wallace (1878) suggested that the large and showy flowers on Robinson Crusoe Island are an evolutionary consequence of dependence on hummingbirds. He was probably correct in his observation, but not in his explanation. Given that the percentage of bird pollination is about the same in today's flora as we estimated for the first colonists, we suggest that very few species switched from some other pollination system to ornithophily. Rather, the proportion of bird-pollinated species represents the successful establishment of ornithophilous colonists. Certainly, the association of hummingbirds with flowers is long-standing; in their study of the origin of plant-animal mutualisms in the woody flora of the temperate forest of southern South America, Aizen and Ezeurra (1998) con cluded that the evolution of ornithophily appears to have taken place well before the climatic cooling and biogeographical isolation of that region in the Tertiary. Thus, as might be expected, most of the ornithophilous families or genera include hummingbird-pollinated species elsewhere. There is good information that some members of Bromeliaceae (Bernardello et al., 1991), Campanulaceae (Galetto et al., 1993), Loranthaceae (Smith Ramirez, 1993; Armesto et al., 1996; Rivera et al., 1996), and Saxifragaceae (Armesto et al., 1996), as well as of the genera Nicotiana (Hernandez, 1981), Rhaphithamnus (Smith Ramirez, 1993; Sun et al., 1996), and Sophora (Arroyo, 1981), are visited by hummingbirds or other kinds of birds in southern South America.
We do not know the timing of the arrival of either plants or hummingbirds, nor whether both may have arrived several times. This is, approximately, a kind of "chicken-and-egg" conundrum. Colwell (1989) considers these plant species unlikely to have been successful colonists of these islands without the presence of hummingbirds. But, without the bird flowers and their large supply of nectar, what provided nutrition for the hummingbirds? The hummingbird-pollinated species are perennial, mostly woody, and SC: Thus, at least some of them could have reproduced to some limited extent before the hummingbirds arrived.
However, even SC does not ensure reproduction. Some of the SC ornithophilous species (the gynomonoecious, gynodioecious, and dichogamous species) require at least interflower pollen transfer. In many of these species, the abundance of simultaneous flowers, the amount of nectar exuded, and the hummingbird behavior would favor pollen transfer among flowers of the same plant (geitonogamy) over interplant pollen transfer (xenogamy). There are also some ornithophilous species that seem to have involved shifts from insect pollination of the first colonizers to bird pollination. Some hummingbird-pollinated members of the genera Centaurodendron and Dendroseris (Asteraceae), Eryngium (Apiaceae), and Rhaphithamnus (Verbenaceae) have comparatively larger flowers (Skottsberg, 1957; Sanders et al., 1987; Skottsberg, 1953; Sun et al., 1996, respectively) than do island congeners of those genera pollinated otherwise (or whose pollination is doubtful). Thus, in these instances, flower length may have increased due to a speci alization to hummingbird pollination. Sun et al. (1996) suggested that in Rhaphithamnus venustus it is likely that ornithophily arose as a result of flower elongation from the only entomophilous continental species of the genus, R. spinosus. However, given that hummingbirds occasionally visit the Argentine and Chilean R. spinosus (Sun et al., 1996), the colonizing ancestors of R. venustus may have been preadapted to ornithophily.
2. Wind Pollination
Wind is considered a significant pollination agent for remote island floras (Carlquist, 1966; Whitehead, 1969, 1983; Regal, 1982; Barrett, 1998), with anemophily characterizing significant percentages of species in island floras (Thornton, 1971; Carlquist, 1974; Ehrendorfer, 1979). Indeed, wind pollination is important in the Juan Fernandez Archipelago, as reported here and in our previous studies (Bernardello et al., 1999; Anderson et al., 2000a, 2000b, 2001a). In addition to some experimental studies, we propose anemophily as important, based on extensive and intensive studies demonstrating the absence of significant native-insect-pollinating groups and the consequent nearly complete lack of flower visiting insects.
For a portion of the flora, anemophily is not a surprise, given the condition among presumed ancestors. For a significant proportion of other species, we propose that the lack of any other alternative means of pollen transfer has led to a dependence on dispersal of pollen by wind. Similar arguments have been made for elements of the continental flora isolated in habitats with little insect-pollinator service (e.g., Berry & Calvo, 1989). Hagerup (1932, 1951) has reported instances in which the rarity or inactivity of insects seemed to result in selfing, anemophily, or even rain pollination. A number of the Fernandezian anemophilous species lack the full suite of characters (cf., Regal, 1982; Whitehead, 1983) typically associated with wind pollination. Thus, we conclude that floral morphology does not always give a precise indication of the pollination mode in this evolutionary young flora. Instead, the floral features many times seem to reflect retention of the traits of the reproductive systems of the origina l colonizers. These features may be expressed because there has been relatively little time for evolution to modify the ancestral characters or because there has been little or no selection against the ancestral traits (Anderson et al., 2001a). The latter argument may be the stronger, given the observations by Britton-Davidian et al. (2000) that selection (at least for some features) can act quickly: At least six different chromosomal races of the house mouse were established on Madeira in 400 years.
For another portion of the flora, anemophily is not so clearly expected. Few species colonize islands, so that successful colonizers may move into niches new to them, both ecologically and evolutionarily. In the new environment and with fewer competitors (vis-a-vis continents), unusual forms (e.g., the habit of Dendroseris on the Juan Fernandez Islands) may be expressed (e.g., Carlquist, 1974), or ancestral conditions may persist because of the lack of selection against them. As a consequence, wind pollination of certain species may be obscured by retention of ancestral entomophilous features. Thus, detailed study of individual species may be required to reveal the actual pollination system. More extensive studies of the pollination biology of island floras may reveal anemophily for species in which the morphology would not have predicted it (Anderson et al., 2001 a). In the Galapagos, for instance, McMullen (1987, 1990) indicated that floral morphology conductive to wind pollination is rarely encountered (al so see Barrett, 1998). A subsequent study (McMullen & Close, 1993), which analyzed the amount of airborne pollen produced by six species, suggested that it was not of a magnitude sufficient for wind pollination. However, detailed study of the reproductive biology of the Galapagos has only included a small fraction (10-15%) of the native flora, and subsequent analysis may reveal anemophily, a pollination system operating perhaps by default, to be more prominent than flower morphology would imply.
An interesting and extreme example that shows the impact of wind transferring pollen was detected in Wahlenbergia berteroi (Anderson et al., 2000b), a derived species (Lammers, 1996) that produces nectar but is not visited by hummingbirds or pollinating insects. In the other Wahlenbergia species of the archipelago, autogamy occurs at the end of the flower lifetime, when the stigmata branches recurve almost 360[degrees] and contact pollen held on the style. However, in W. berteroi the degree of stigmatic recurvature is insufficient to reach the style and facilitate autogamy. Nevertheless, in the field most of the flowers observed did have pollen on the stigmatic surfaces. To explain how pollen transfer occurs, we suggested an unusual method of autogamy, based on our field studies (Anderson et al., 2000b). As anthers dehisce in bud stage, the corolla throat closely surrounds the anther cylinder and thus becomes coated with pollen. As in other Wahlenbergia, the pollen is actually presented by a stylar brush that pushes the pollen up near the open mouth of the now open corolla. The ring of pollen in the throat is obvious because its white color stands out against the pink corolla. Given that experimental studies imply that pollen stays viable throughout the receptive stage of the stigma, we hypothesized that transfer from corolla to stigmata occurs when flowers are shaken by the ever-present wind, promoting contact in the narrow throat between the stigmata and corolla. A similar situation was reported in Iris versicolor from Kent Island, New Brunswick, where facultative self-pollination was promoted by wind that forced reflexed stigmas to brush against petals of the same flowers where pollen had settled (Zink & Wheelwright, 1997).
H. HYPOTHESIZED POLLINATION OF COLONIZERS
The majority of the colonizers seem to have been pollinated either by insects or by wind. However, this association is likely related not to the successful colonization of species with these pollination systems but to enhanced chances of reproduction of established plants through SC or, possibly, to some other feature, such as greater dispersal ability. Obviously, anemophilous species are preadapted to successfully reproduce on islands, because they do not need biotic pollinators (Ehrendorfer, 1979). Bird-pollinated species that dispersed successfully would have done well, given that these islands eventually also included two hummingbirds. However, entomophilous species would have encountered serious reproductive problems on islands with no dedicated insect pollinators.
According to Carlquist (1974), long-distance dispersal is often not achieved by a single individual or disseminule. For instance, there is evidence for multiple disseminules per species in the colonization of Krakatau islets (Docters van Leeuwen, 1936) and on a newly emerged island near Iceland (Einarsson, 1967). Thus, it is possible to envision even the establishment of SI or dioecious colonizers when two or more compatible plants arrive. Furthermore, perennials have an advantage over short-lived species: Longevity increases the chances of securing sufficient pollination for effective sexual reproduction that establishes and maintains a species (Wallace, 1895). Recall that most Fernandezian species are woody. The longer life cycles also increase the chances of encountering, eventually, effective pollinators: for example, the arrival of hummingbirds on these islands.
Proportionally, there are many more recorded extinctions of vascular plants from islands than from continental areas (Reid & Miller, 1989; Smith et al., 1993; Frankham, 1997). Thus, it is even more important to invest extra effort in protecting the remaining endangered island species (Cronk, 1997; Carlquist, 1998; Raven, 1998). Conservation or restoration programs cannot be effective without an understanding of the reproductive biology (Hamrick et al., 1991; Karron, 1991; Weller, 1994). In addition, data on reproductive biology are an integral part of understanding and, especially, interpreting problems of genetic diversity generated via molecular studies (Karron, 1987; Crawford, 1990; Barrett & Kohn, 1991; Holsinger, 1991).
For the Juan Fernandez flora, even though insect pollinators are virtually absent, wind pollination serves a large percentage of the species, and pollen transfer--at some level--seems not be a strongly limiting factor. However, a large number of the anemophilous species lack some features that would make wind pollination most effective (cf., Regal, 1982; Whitehead, 1983). Among them, one is very important for conservation purposes: close proximity of con-specifics; that is, relatively close spacing of compatible plants. The rapidly spreading populations of invasives and the continued foraging by rabbits and feral goats on native species (Stuessy et al., 1997, 1998b) are significantly reducing and separating the already small populations of many species. As Weller (1994) pointed out, wholesale destruction of natural areas may have contributed far more to rarity than have evolutionary inadequacies of plant breeding systems or pollination biology. The intrusion of invasives exacerbates this situation. The pollin ation systems of island plants are fragile, and habitat changes may reduce rates of already insufficient pollination--obviously with negative consequences for maintenance of populations and species.
There is particular concern for the SI and dioecious species, especially those with few individuals left. For instance, in the SI Dendroseris neriifolia (with only three individuals in the wild in Quebrada Lapiz, plus a few individuals cultivated at GONAF gardens in San Juan Bautista; Stuessy et al., 1998b), examination of open-pollinated flowers on cultivated individuals showed no pollen grains on stigmata or pollen tubes in styles; that is, no natural pollination (Anderson et al., 2000b). Furthermore, the open-pollinated seed set of the small cultivated population of the SI D. pruinata was extremely low. Thus, for both of those examples, at least in part because of the reproductive system (coupled with rarity), the species are at severe risk of extinction in the near future if the existing very few plants are not protected, and if their pollination/reproduction is not enhanced.
Among the dioecious species, two of the seven species of Robinsonia are already rare: R. macrocephala has not been collected since 1917, and R. berteroi is known from one male specimen (Quebrada Villagra; Stuessy et al., 1998b). Thus, care must be taken to continue programs of propagation for these very rare species. For Coprosma and Fagara, although more field data are needed, the relatively low density of compatible plants seems to be a concern for their effective sexual reproduction. Some other species are more stable at present. For instance, Juania australis was saved from extinction because of its ability to grow on the high and inaccessible ridges (Stuessy et al., 1998b) beyond easy access--at least of humans. However, there are still no studies of the reproductive biology of this endemic genus, so its success in seed production is unknown. Pernettya rigida has vegetative reproduction through stolons, sexual seed production mediated by wind, and dispersal of berries by birds. All of these features comb ine to promote this species as a colonizer of new habitats. For the same reasons it thrives on both main islands of the archipelago and seems not to be endangered (Anderson et al., 2000a).
The ultimate fate of some species may depend on preserving the plant-hummingbird relationship (Buchmann & Nabban, 1996; Kearns et al., 1998; Nabhan et al., 1998; Roy et al., 1999), including the web of organisms that affect both plant and pollinator. The density of conspecifics is fundamental to favor higher levels of interplant pollen transfer, Reproductive success depends on the quantity and quality of effective pollination visits, and both of these are likely to depend on local abundance (Kunin, 1997). Here, again, invasives pose a serious problem. They outcompete endemic species for habitat, and if they also produce abundant nectar and are visited by hummingbirds (as is true in general), their prosperity may also lead both to a reduced visitation rate and to reduced effective pollination of the endemics (Brooke, 1987; Colwell, 1989; Bemardello et al., 2000). The highly invasive bramble (Rubus ulmifolius) is heavily visited as a nectar source by the native hummingbird (Brooke, 1987), and it seems to be bet ter suited than the endemic hummingbird to feed on this plant species (Colwell, 1989). This fact, among others, is postulated to have differentially favored the native hummingbird at the expense of the endemic species, whose populations are declining (Colwell, 1989).
The hummingbird-pollinated Nicotiana cordiflora is endemic to Alejandro Selkirk Island, where the endemic hummingbird (S. fernandensis) is now considered extinct (Brooke, 1987; Colwell, 1989). Fortunately for the ornithophilous plant species, since 1981 the native hummingbird (S. sephaniodes) has become established on Alejandro Selkirk Island (although it is not clear if they are a breeding population or only adventives from Robinson Crusoe Island; Brooke, 1987; Colwell, 1989). Given the number of ornithophilous species and their importance to the flora, the complete loss of the hummingbirds could have immense conservation implications. Regrettably, the goat-maintained grassland of the island's lower slopes, largely composed of introduced species, and the native fern forest of the uplands, combine to render major portions of this island poor hummingbird habitat (Brooke, 1987). Thus, to maintain this plant--animal interaction, the introduced goats should be eliminated and the native habitats restored.
It is also crucial to maintain the interaction of hummingbirds with Dendroseris litoralis. This species is known presently only from several plants on Morro Spartan and adjacent Santa Clara Island; hence it is nearly extinct in the wild. Fortunately, it is extensively cultivated on Robinson Crusoe Island (Stuessy et al., 1998b).
A recently discovered bee (Engel, 2000) is considered irrelevant for the evolution of the reproductive systems on the islands (Anderson et al., 2001b). Although currently a rare endemic, we propose that the bee is likely also to be discovered in continental South America. However, this bee may be important in future conservation or restoration programs on the island. The bees--scarce at present--were observed collecting pollen on the Wahlenbergia hybrid (Anderson et al., 2000b; Engel, 2000) in San Juan Bautista and its environs. Although likely recent introductions, they could become beneficial, promoting allogamy for this and other taxa. Given the lack of significant native insect fauna, these bees would not seem to pose a competitive danger to other insects; they will displace no native pollinator fauna. However, just as the more common continental species of hummingbird may be responsible for the decline of the endemic hummingbird (Colwell, 1989), the introduced bees may outcompete the hummingbirds for flo ral nectar and have a negative effect on the vigor of these island "banner species."
Finally, a new pressure has been added: Introduced locusts are widespread in the most arid and dry part of Robinson Crusoe Island, near the landing strip. They are voracious, much affecting the scarce native and endemic species adapted to live in the arid low-island conditions. We observed them virtually destroying a population of W. berteroi, eating stems, leaves, even fruits, and seeds.
In order to conserve these plants, programs must involve a combination of reproductive and environmental measures. The populations of introduced animals and weeds must be controlled. Experimentally produced allogamous seeds would do much to enhance diversity in restoration programs. In addition, the preservation of habitat seems to be the central challenge to indirectly protect the unique island species. The depauperate soil of these islands is mostly eroded, exposing bedrock to degradation and retaining less rainwater. Erosion is, in fact, a considerable problem in these islands, particularly on the steep volcanic terrain (Sanders et al., 1982).
The flora typically comprises perennials (ca. 90%). The majority of the species (60%) have comparatively insignificant flowers, either very small (46%) or small (14%). Inconspicuous flowers (a shape category for flowers with "no optical attraction") are widespread (41% of the taxa), as are dish-shaped flowers (35%). Green is the most frequent flower color (41% of the species), followed by white and yellow (26% and 12%, respectively). Most species (70%) are hermaphroditic, 9% are dioecious, and 9% are monoecious. Around 32% of the species are protandrous; 7%, protogynous. The few studies of compatibility of the flora (for 14% of the species) indicate that 85% of the species are self-compatible (SC) and that only 17% are self-incompatible (SI). Although SC is frequent, the level of autogamy is low. Nevertheless, geitonogamy, and thus selfing, would be the most frequent mechanism of pollen transfer. Outcrossing is mainly achieved through dioecy and SI, promoted by dichogamy, and facilitated by wind pollination.
About 55% of the species offer nectar as a reward; about 2%, pollen. Floral visitors are rare to uncommon. The activity of the two native hummingbird species (the endemic Sephanoides fernandensis and the native S. sephaniodes), recorded for 9% of the flora, is an exception to this generalization. Native insects (flies, moths, and beetles) were observed or recorded for only 11 species, but even on these, recorded floral visits were rare, In addition, the insects showed no marked preference for flowers over other plant parts and exhibited no observable species fidelity. Two species of introduced ants and a new endemic bee were recorded as well. It seems likely that the latter is a recent introduction.
We conclude that 9% of the flora is bird pollinated and 47% wind pollinated, including a special case of wind as an indirect force of pollen transfer in Wahlenbergia berteroi. Given the rare and imprecise nature of native-insect visitation, there is no certainty that any of the species are truly insect pollinated. This is a tribute to the paucity of the continental fauna, because we hypothesize that most of the original colonizers were either insect or wind pollinated.
Pairwise comparisons of many features were tested statistically. We confirmed the association between the following features known to be common for oceanic islands: green and very small flowers, green and inconspicuous flowers, and very small and inconspicuous flowers. Monoecious species are largely herbaceous and have mostly very small, inconspicuous, green flowers. Dioecious species are woody, chiefly with green and yellow, dish-shaped, medium-sized, wind-pollinated flowers. We propose that the progenitors of Coprosma, Juania, Fagara, and Pernettya already exhibited at least incipient dioecy; in Robinsonia dioecy seems to have arisen in situ. Both gynomonoecy in Cuminia and gynodioecy in Rhaphithamnus venustus likely arose in situ.
Hermaphroditic flowers and SC likely could have favored the reproduction and establishment of colonizers. There is a correlation between a number of floral features and the hypothesized pollination of colonizers. Therefore, the flower color, shape, and size of the extant flora may, to some large extent, express the pollination syndromes of colonizers rather than representing extant pollination. In addition, the presence of nectar in the extant flora does not necessarily indicate biotic pollination; for example, some clearly wind-pollinated species also bear nectar. Thus, studies of the reproductive biology of oceanic island plants need to be conducted species by species before broad, effective generalizations can be made, because the observed features can be misleading.
We assessed possible changes in the pollination system by comparison of species for which we have reliable data (for 39 genera) with the hypothesized pollination of their colonist progenitors. Most (70%) retained the pollination system of the colonizers, either by wind or hummingbirds. In 30% of genera, at least some species bear a different pollination system: In four genera the ancestral insect-pollination system seems to have switched to hummingbird pollination, and in seven other genera from insect to wind pollination. For the insect-to-bird pollination examples, flower length seems to have increased due to a specialization to hummingbird pollination. In Wahlenbergia, the ancestors were likely insect pollinated, with a secondary system of autogamy; but in the islands, with paucity of insect pollination, this selfing mechanism became the rule. A wind-aided mechanism of autogamy in W. berteroi must have arisen in situ. We also suspect that much of the remaining 44% of the flora for which pollination is unkn own is likely to be wind pollinated or autogamous (or both). The bird-pollinated species are mostly obvious, and insect pollination is mostly insignificant. Thus, it is likely that a much higher percentage of the flora switched from insect pollination of colonizers to abiotic or autogamous endemics.
The species that arrived having preadaptations to bird pollination seem to have retained those features. The hummingbird-pollinated species that were perennial and SC could have reproduced and become established even before the hummingbirds arrived. Considering the many simultaneous flowers produced by these species, the amount of nectar exuded, the hummingbird behavior, and the existence of SC, hummingbird visits likely favor geitonogamous rather than xenogamous pollen transfer.
Anemophily for a portion of the flora is expected, given the characteristics of the proposed progenitors; for other species, the lack of alternative means of pollination must have led to anemophily. For the latter, the features associated with wind pollination may not be so obvious. If there has not been sufficient evolutionary time or selection pressure, ancestral and misleading features of abiotic pollination may be retained. More extensive studies of the pollination biology of island floras may reveal anemophily for species in which features of the flowers would not have predicted it.
Given the many recorded extinctions of vascular plants from islands versus those from continental areas, it is imperative to invest additional effort in protecting the remaining island species. Conservation or restoration programs cannot be effective without a deep and broad understanding of the reproductive biology of the plants. In order to conserve these plants, programs must involve a combination of reproductive and environmental measures. The ultimate fate of some species may depend on preserving the plant--hummingbird relationship, including the web of organisms that affect both plant and pollinator. The populations of introduced animals and weeds must be controlled. Experimentally produced allogamous seeds would enhance diversity in restoration programs. In addition, the preservation of habitat seems to be the central challenge to indirectly protect the unique island species.
We thank the National Science Foundation, the University of Connecticut Research Foundation, the Department of Ecology and Evolutionary Biology, CONICET, CONICOR, and SECYT--Universidad Nacional de Cordoba (Argentina) for financial support and a sabbatical leave. We are grateful to Edward O. Wilson, Jane O'Donnell, Michael S. Engel, Haroldo Toro, and Charles Michener for insect identification, to Nalini Ravishanker for statistical advice, to Maryke Schlehofer, Leonardo Galetto, and Dick Jensen for helping in many ways, and to Virge A. Kask for help with the illustrations and Jim Hill with the manuscript. The assistance of CONAF (Chile) is sincerely acknowledged, especially Ivan Leiva (chief of the Juan Fernandez National Park) and park guides Jorge Angulo, Guillermo Araya, Oscar Chamorro, Bernardo Lopez, Manuel Recabarren, and Ramon Schiller, without whose invaluable help this work would have been impossible. Further assistance was provided by the Meteorological Service of Robinson Crusoe Island (especially O svaldo Jara and Alex Meneses) and by Fernandezians Jose M. Gutierrez, Valeria Salzmann, Jorge Palomino, and Juanita Lopez. We dedicate this article to the memory of G. L. Stebbins, for his stimulating work on plant breeding systems and evolution.
VIII. Literature Cited
Aizen, M. A. & C. Ezcurra. 1998. High incidence of plant-animal mutualisms in the woody flora of the temperate forest of Southern South America: Biogeographical origin and present ecological significance. Ecol. Austral 8: 217-236.
Allen, D. 1984. Threatened 'protected natural areas' of the world. Environ. Conservation 12: 77.
Anderson, G. J., G. Bernardello, P. Lopez S., T. F. Stuessy & D. J. Crawford. 2000a. Dioecy and wind pollination in Pernettya rigida (Ericaceae) of the Juan Fernandez Islands. Bot. J. Linn. Soc. 132: 121-141.
-----, -----, P. Lopez S., D. J. Crawford & T. D. Stuessy. 2000b. Reproductive biology of Wahlenbergia (Campanulaceae) endemic to Robinson Crusoe Island (Chile). P1. Syst. & Evol. 233: 109-123.
-----, -----, T. F. Stuessy & D. J. Crawford. 2001a. Breeding system and pollination of selected plants endemic to the Juan Fernandez Islands. Amer. J. Bot. 88: 220-233.
-----, -----, & M. S. Engel. 2001b. A newly discovered bee species on Isla Robinson Crusoe of the Juan Fernandez Islands (Chile): Conservation implications. Conservation Biol. 15: 803-805.
Armesto, J. J., C. Smith Ramirez & C. Sabag. 1996. The importance of plant-bird mutualism in the temperate rainforest of southern South America. Pp. 248-265 in R. G. Lawford, P. B. Alaback & E. R. Fuentes (eds.), High-latitude rainforests and associated ecosystems of the west coast of the Americas: Climate, hydrology, ecology, and conservation. Springer-Verlag, New York.
Arroyo, M. T. K. 1981. Breeding systems and pollination biology in Leguminosae. Pp. 723-769 in R. M. Polhill & P. H. Raven (eds.), Advances in Legume Systematics. Royal Botanic Gardens, Kew.
----- & F. A. Squeo. 1987. Experimental detection of anemophily in Pernettya mucronata (Ericaceae) in western Patagonia, Chile. Bot. Jahrb. Syst. 108: 537-546.
----- & P. Uslar. 1993. Breeding systems in a temperate Mediterranean-type climate montane sclerophyllous forest in central Chile. Bot. J. Linn. Soc. 111: 83-102.
Baker, H. G. 1955. Self-compatibility and establishment after "long-distance" dispersal. Evolution 9: 347-349.
-----. 1959. Reproductive methods as factors in speciation of flowering plants. Cold Spring Harbor Symp. Quant. Biol. 24: 177-191.
-----. 1967. Support for Baker's law as a rule. Evolution 21: 853-856.
----- & I. Baker. 1983a. A brief historical review of the chemistry of floral nectar. Pp. 126-152 in B. Bentley & T. S. Elias (eds.), The biology of nectaries. Columbia Univ. Press, New York.
----- & -----. 1983b. Floral nectar sugar constituents in relation to pollinator type. Pp. 117-141 in C. E. Jones & R. J. Little (eds.), Handbook of experimental pollination biology. Van Nostrand Reinhold, New York.
----- & -----. 1990. The predictive value of nectar chemistry to the recognition of pollinator types. Israel J. Bot. 39: 157-166.
----- & P. A. Cox. 1984. Further thoughts on dioecism and islands. Ann. Missouri Bot. Gard. 71: 244-253.
Baldwin, B. G. 1997. Adaptive radiation of the Hawaiian silversword alliance: Congruence and conflict of phylogenetic evidence from molecular and non-molecular investigations. Pp. 103-128 in T. J. Givnish & K. J. Systma (eds.), Molecular evolution and adaptive radiation. Cambridge Univ. Press, Cambridge.
Barrett, S. C. H. 1998. The reproductive biology and genetics of island plants. Pp. 18-34 in P. R. Grant (ed.), Evolution on islands. Oxford Univ. Press, Oxford.
----- & J. R. Kohn. 1991. Genetic and evolutionary consequences of small population size in plants: Implications for conservation. Pp. 3-30 in D. A. Falk & K. E. Holsinger (eds.), Genetics and conservation of rare plants. Oxford Univ. Press, New York.
Bawa, K. S. 1980. Evolution of dioecy in flowering plants. Ann. Rev. Ecol. Syst. 11: 15-39.
Bell, A. D. 1991. Plant form: An illustrated guide to flowering plant morphology. Oxford Univ. Press, Oxford.
Bernardello, G., L. Galetto & H. R. Juliani. 1991. Nectar and nectary structure in some Argentinean Bromeliaceae. Ann. Bot. (London) 67: 401-411.
-----, G. J. Anderson, P. Lopez S., M. A. Cleland, T. F. Stuessy & D. J. Crawford. 1999. Reproductive biology of Lactoris fernandeziana (Lactoridaceae). Amer. J. Bot. 86: 829-840.
-----, L. Galetto & G. J. Anderson. 2000. Floral nectary structure and nectar chemical composition of some species from Robinson Crusoe Island (Chile). Canad. J. Bot. 78: 862-872.
Berry, P. E. & R. N. Calvo. 1989. Wind pollination, self-incompatibility, and altitudinal shifts in pollination systems in the high Andean genus Espeletia (Asteraceae). Amer. J. Bot. 76: 1602-1614.
Bohle, U. R., H. H. Hilger & W. F. Martin. 1996. Island colonization and evolution of the insular woody habit in Echium L. (Boraginaceae). Proc. Natl. Acad. Sci. U.S.A. 93: 11740-11745.
Bourne, W. R. P., M. de L. Brooke, G. S. Clarke & T. Stone. 1992. Wildlife conservation problems in the Juan Fernandez Archipelago, Chile. Oryx 26: 43-51.
Bowman, R. I., M. Berson & A. E. Leviton (eds.). 1983. Patterns of evolution in Galapagos organisms. Pacific Division, AAAS, San Francisco.
Bramwell, D. 1972. Endemism in the flora of the Canary Islands. Pp. 141-159 in D. H. Valentine (ed.), Taxonomy, phytogeography, and evolution. Academic Press, London and New York.
----- (ed.). 1979. Plants and islands. Academic Press, London.
Britton-Davidian, J., J. Catalan, M. G. Ramalhinho, G. Ganem, J.-C. Auffray, R. Capela, M. Biscoito, J. B. Searle & M. L. Mathias. 2000. Rapid chromosomal evolution in island mice. Nature 405: 158.
Brooke, M. de L. 1987. The birds of the Juan Fernandez Islands, Chile. International Council for Bird Preservation, Fauna and Flora Preservation Society, World Wildlife Fund, Cambridge, England.
Buchmann, S. L. & G. P. Nabhan. 1996. The forgotten pollinators. Island Press, Washington, DC.
Carlquist, S. 1965. Island life: A natural history of the islands of the world. Natural History Press, Garden City, NY.
-----. 1966. The biota of long-distance dispersal, IV: Genetic systems in the floras of oceanic islands. Evolution 20: 433-455.
-----. 1974. Island biology. Columbia Univ. Press, New York.
-----. 1998. Shifting paradigms in island biology. Aliso 16: 85-88.
Carpenter, F. L. 1976. Plant-pollinator interactions in Hawaii: Pollination energetics of Metrosideros collina (Myrtaceae). Ecology 57: 1125-1144.
Carr, C. D., E. A. Powell & D. W. Kyhos. 1986. Self-incompatibility in the Hawaiian Madiinae (Compositae): An exception to Baker's rule. Evolution 40: 430-434.
Colwell, R. K. 1989. Hummingbirds of the Juan Fernandez Islands: Natural history, evolution and population status. Ibis 131: 548-566.
Conn, J. S., T. R. Wentworth & U. Blum. 1980. Patterns of dioecism in the flora of the Carolinas. Amer. Midl. Naturalist 103: 310-315.
Connor, H. E. 1988. Breeding systems in New Zealand grasses, X: Species at risk for conservation. New Zealand J. Bot. 26: 163-167.
Corn, C. A. 1979. Variation in Hawaiian Metrosideros. Ph.D. diss., Univ. of Hawaii, Honolulu.
Cory, C. 1984. Pollination biology of two species of Hawaiian Lobeliaceae (Clermontia kakeana and Cyanea angustifolia) and their presumed co-evolved relationship with native honeycreepers (Drepanididae). M.A. thesis, California State Univ., Fullerton.
Crawford, D. J. 1990. Plant molecular systematics: Macromolecular approaches. Wiley, New York.
-----, T. F. Stuessy, T. C. Lammers, M. Silva O. & P. Pacheco. 1990. Allozyme variation and evolutionary relationships among three species of Wahlenbergia (Campanulaceae) in the Juan Fernandez Islands, Bot. Gaz. 151: 119-124.
-----, T. Sang, T. F. Stuessy, S.-C. Kim & M. Silva O. 1998. Dendroseris (Asteraceac: Lactuceae) and Robinsonia (Asteraceae: Senecioneae) on the Juan Fernandez Islands: Similarities and differences in biology and phylogeny. Pp.97-119 in T. F. Stuessy & M. Ono (eds.), Evolution and speciation of island plants. Cambridge Univ. Press, Cambridge.
Cronk, Q. C. B. 1997. Islands: Stability, diversity, conservation. Biodiv. & Conserv. 6: 477-493.
Cronquist, A. 1981. An integrated system of classification of flowering plants. Columbia Univ. Press, New York.
Cruden, R. W., S. M. Hermann & S. Peterson. 1983. Patterns of nectar production and plant-pollinator coevolution. Pp. 80-125 in B. Bentley & T. S. Elias (eds.), The biology of nectaries. Columbia Univ. Press, New York.
Darwin, C. 1859. The origin of species by means of natural selection. John Murray, London.
Davis, S. D., V. H. Heywood & A. C. Hamilton (eds.). 1995. Centres of plant diversity: A guide and strategy for their conservation. World Wide Fund for Nature and IUCN-the World Conservation Union, Cambridge, England.
Docters van Leeuwen, W. M. 1936. Krakatau, 1883 to 1933. A. Botany. Ann. Jard. Bot. Buitenzorg 46-47: 1-506.
Ehrendorfer, F. 1979. Reproductive biology in island plants. Pp.293-306 in D. Bramwell (ed.), Plants and islands. Academic Press, London.
Einarsson, E. 1967. The colonization of Surtsey, the new volcanic island, by vascular plants. Aquilo, Ser. Botanica 6: 172-182.
Engel, M. S. 2000. A new Lasioglossum from the Juan Fernandez Islands (Hymenoptera: Halictidae). Rev. Chilena Entomol. 27: 5-10.
Faegri, K. & L. van der Pijl. 1979. The principles of pollination ecology. Ed. 3, rev. Pergamon Press, Oxford, England.
Francisco-Ortega, J., D. J. Crawford, A. Santos-Guerra & R. K. Jansen. 1997. Origin and evolution of Argyranthemum (Asteraceae: Anthemideae) in Macaronesia. Pp. 407-431 in T. J. Givnish & K. J. Systma (eds.), Molecular evolution and adaptive radiation. Cambridge Univ. Press, Cambridge.
Frankham, R. 1997. Do island populations have less genetic variation than mainland populations? Heredity 78: 311-327.
Freeman, D.C., K. T. Harper & W. K. Ostler. 1980. Ecology of plant dioecy in the intermountain region of western North America and California. Oecologia 44: 410-417.
Galetto, L., L. Bernardello & H. R. Juliani. 1993. Estructura del nectario, composicion quimica del nectar y mecanismo de polinizacion en tres especies de Siphocampylus (Campanulaceae). Kurtziana 22: 81-96.
Gay, C. 1845-1854. Historia fisica y politica de Chile segun documentos adquiridos en esta republica durante doce anos de residencia en ella y publicada bajo los auspicios del supremo gobierno: Botanica. Private publication, Paris.
Gillett. G. W. & E. K. S. Lim. 1970. An experimental study of the genus Bidens (Asteraceae) in the Hawaiian Islands. Univ. Calif. Publ. Bot. 56: 1-63.
Givnish, T. J. 1998. Adaptive plant evolution on islands: Classical patterns, molecular data, new insights. Pp. 281-304 in P. R. Grant (ed.), Evolution on islands. Oxford Univ. Press, Oxford.
-----, K. J. Sytsma, W. J. Hahn & J. F. Smith. 1995. Molecular evolution, adaptive radiation, and geographic speciation in Cyanea (Campanulaceae, Lobelioideae). Pp. 299-337 in W. L. Wagner & V. A. Funk (eds.), Hawaiian biogeography: Evolution on a hot-spot archipelago. Smithsonian Institution Press, Washington, DC.
Godley, E. J. 1979. Flower biology in New Zealand. New Zealand J. Bot. 17: 441-466.
Grant, B. R. & P. R. Grant. 1981. Exploitation of Opuntia cactus by birds in the Galapagos. Oecologia 49: 179-187.
Grant, P. R. (ed.). 1998. Evolution on islands. Oxford Univ. Press, Oxford.
Hagerup, O. 1932. On pollination in the extremely hot air in Timbuctu. Dansk Bot. Ark. 8: 1-20.
-----. 1951. Pollination in the Faroes--in spite of rain and poverty in insects. Biol. Meddel. Kongel. Danske Vidensk. Selsk. 18: 1-48.
Hamrick, J. L., M. J. W. Godt, D. A. Murrawski & M. D. Loveless. 1991. Correlations between species traits and allozyme diversity: Implications for conservation biology. Pp.75-86 in D. A. Falk & K. E. Holsinger (eds.), Genetics and conservation of rare plants. Oxford Univ. Press, New York.
Hemsley, W. B. 1884. Report on the botany of Juan Fernandez, the south-eastern Moluccas, and the Admiralty Islands. Rep. Sci. Results Voyage H.M.S. Challenger. Botany 1(3): 1-275.
Hernandez, H. M. 1981. Sobre la ecologia reproductiva de N. glauca Grah., una maleza de distribucion cosmopolita. Bol. Soc. Bot. Mexico 41: 47-73.
Holsinger, K. E. 1991. Conservation of genetic diversity in rare and endangered plants. Pp. 626-633 in E. C. Dudley (ed.), The unity of evolutionary biology. Dioscorides Press, Portland, OR.
Hooker, J. D. 1853. Introductory essay to the flora of New Zealand. Flora Novae-Zelandiae. Part I. L. Reeve, London.
Inoue, K. 1993. Evolution of mutualism in plant-pollinator interactions on islands. J. Biosci. 18: 525-536.
Jaksic, F. M. 1998. Vertebrate invaders and their ecological impacts in Chile. Biodiv. & Conserv. 7: 1427-1445.
Johow, F. 1896. Estudios sobre la flora de las islas de Juan Fernandez. Private publication, Santiago, Chile.
Juan, J. & A. de Ulloa. 1813. A voyage to South America. Pp. 313-696 in J. Pinkerton (ed.), A general collection of the best and most interesting voyages and travels in all parts of the world. Vol. 13. Longman, Hurst, Rees, and Orme, London.
Karron, J. D. 1987. A comparison of levels of genetic polymorphism and self-compatibility in geographically restricted and widespread plant congeners. Evol. Ecol. 1: 47-58.
-----.1991. Patterns of genetic variation and breeding systems in rare plant species. Pp. 87-98 in D. A. Falk & K. E. Holsinger (eds.), Genetics and conservation of rare plants. Oxford Univ. Press, New York.
Kearns, C. A., D. W. Inouye & N. M. Waser. 1998. Endangered mutualisms: The conservation of plant-pollinator interactions. Ann. Rev. Ecol. Syst. 29: 83-112.
Keast, A. & S. E. Miller. 1996. The origin and evolution of Pacific Island biotas, New Guinea to Eastern Polynesia: Patterns and processes. SPB Academic Publishing, Amsterdam.
Kim, S-C., D. J. Crawford, J. Francisco-Ortega & A. Santos Guerra. 1996. A common origin for woody Sonchus and five related genera in the Macaronesian Islands: Molecular evidence for extensive radiation. Proc. Natl. Acad. Sci. U.S.A. 93: 7743-7748.
Kores, P. J. 1979. Taxonomy and pollination in the wild Hawaiian orchids. M.S. thesis, Univ. of Hawaii, Honolulu.
Kunin, W. E. 1997. Population biology and rarity: On the complexity of density dependence in insectplant interactions. Pp. 150-173 in W. E. Kunin & K. J. Gaston (eds.), The biology of rarity: Causes and consequences of rare-common differences. Chapman & Hall, London.
Kuschel, G. 1952. Los insectos de las islas Juan Fernandez: Introduccion. Rev. Chilena Entomol. 2:3-6.
Lammers, T. G. 1996. Phylogeny, biogeography, and systematics of the Wahlenbergia fernandeziana complex (Campanulaceae: Campanuloideae). Syst. Bot. 21: 397-415.
Lawesson, J. E., O. Hamman, G. Rogers, G. Reck & H. Ochoa (eds.). 1990. Botanical research and management in Galapagos: Proceedings of the Workshop on Botanical Research and Management in Galapagos held 11-18 April 1987 at the Charles Darwin Research Station, Santa Cruz, Galapagos, Ecuador. Monogr. Syst. Bot. 32. Missouri Bot. Gard. 32, St. Louis, MO.
Linnaeus, C. 1747. Flora Zeylanica. Private publication, Stockholm.
Lloyd, D. G. 1985. Progress in understanding the natural history of New Zealand plants. New Zealand J. Bot. 23: 707-722.
-----. 1992. Self- and cross- fertilization in plants, II: The selection of self-fertilization. Int. J. Pl. Sci. 153: 370-380.
-----. & D. J. Schoen. 1992. Self- and cross-fertilization in plants, I: Functional dimensions. Int. J. Pl. Sci. 153: 358-369.
-----. & J. M. A. Yates. 1982. Intrasexual selection and the segregation of pollen and stigmas in hermaphrodite plants, exemplified by Wahlenbergia albomarginata (Campanulaceae). Evolution 36: 903-913.
Loope, L. L., O. Hamman & C. P. Stone. 1988. Comparative conservation biology of oceanic archipelagos. BioScience 38: 272-282.
Lowrey, T. K. & D. J. Crawford. 1985. Allozyme divergence and evolution in Tetramolopium (Compositae: Astereae) on the Hawaiian Islands. Syst. Bot. 10: 64-72.
MacArthur, R. H. & E. O. Wilson. 1967. The theory of island biogeography. Princeton Univ. Press, Princeton, NJ.
MacDonald, I. A. W. & J. Cooper. 1995. Insular lessons for global biodiversity conservation with particular reference to alien invasions. Pp. 189-203 in P.M. Vitousek, L. L. Loope & H. Adsersen (eds.), Islands: Biological diversity and ecosystem function. Ecological studies, 115. Springer-Verlag, Berlin.
Marticorena, C., T. F. Stuessy & C. M. Baeza. 1998. Catalogue of the vascular flora of the Robinson Crusoe or Juan Fernandez Islands, Chile. Gayana, Bot. 55: 187-211.
McMullen, C. K. 1987. Breeding systems of selected Galapagos Islands angiosperms. Amer. J. Bot. 74: 1694-1705.
-----. 1989. Flowering colonizers of the Galapagos Islands: Drab but not dull. Pl. Today 2: 192-196.
-----. 1990. Reproductive biology of Galapagos Islands angiosperms. Monogr. Syst. Bot. Missouri Bot. Gard. 32: 35-45.
-----. 1999. Flowering plants of the Galapagos. Cornell Univ. Press, Ithaca, NY.
-----. & D.D. Close. 1993. Wind pollination in the Galapagos Islands. Not. Galapagos 52: 12-17.
Meza H., J. 1988. Conservacion del picaflor de Juan Fernandez: Informe annual. CONAF, V Region, Valparaiso, Chile.
Mittermeier, R. A., N. Myers & C. Goetsch Mittermeier. 1999. Hotspots: Earth's biologically richest and most endangered terrestrial ecoregions. Cemex, Mexico City.
Moore, D. M. 1983. Flora of Tierra del Fuego. Oswestry, England, and Missouri Bot. Gard., Saint Louis, MO.
Mueller-Dombois, D., K. W. Bridges & H. L. Carson. 1981. Island ecosystems. Hutchinson Ross Publishing Co., Stroudsburg, PA.
Muenchow, G. E. 1987. Is dioecy associated with fleshy fruit? Amer. J. Bot. 74: 287-293.
Nabhan, G. P., G. Allen-Wardell, R. Bitner, A. Burquez, S. Buchmann, J. Cane, P. A. Cox, V. Dalton, P. Feinsinger, M. Ingram, D. Inouye, C. E. Jones, K. Kennedy, P. Kevan, H. Koopowitz, R. Medellin & S. Medellin-Morales. 1998. The potential consequences of pollinator declines on the conservation of biodiversity and stability of food crop yields. Conservation Biol. 12: 8-17.
Nunn, P. D. 1994. Oceanic islands. Blackwell Publishers, Oxford, England.
Oliver, W. R. 1935. The genus Coprosma. Bernice P. Bishop Mus. Bull. 132: 1-207.
Pandey, K. K. 1979. Long-distance dispersal and self-incompatibility. New Zealand J. Bot. 17: 225-226.
Panero, J. L., J. Francisco-Ortega, R. K. Jansen & A. Santos Guerra. 1999. Molecular evidence for multiple origins of woodiness and a New World biogeographic connection of the Macaronesian Island endemic Pericallis (Asteraceae: Senecioneae). Proc. Natl. Acad. Sci. U.S.A. 96: 13886-13891.
Perry, R. (ed.). 1984. Galapagos. Pergamon Press, Oxford.
Petterson, J. A. 1997. Revision of the genus Wahlenbergia (Campanulaceae) in New Zealand. New Zealand J. Bot. 35: 9-54.
Porter, D. M. 1983. Vascular plants of the Galapagos: Origins and dispersal. Pp. 33-96 in R. I. Bowman, M. Berson & A. E. Leviton (eds.), Patterns of evolution in Galapagos organisms. Pacific Division, AAAS, San Francisco.
Pyke, G. H. & N. M. Waser. 1981. The production of dilute nectars by hummingbird and honeyeater flowers. Biotropica 13: 260-270.
Rabakonandrianina, E. 1980. Infrageneric relationships and the origin of the Hawaiian endemic genus Lipochaeta (Compositae). Pacific Sci. 34: 29-39.
Ramanna, M. S. & G. T. Hermsen. 1981. Structural hybridity in the series Etuberosa of the genus Solanum and its bearing on crossability. Euphytica 30: 15-81.
Raven, P. H. 1998. Plant conservation in a changing world. Aliso 16: 121-126.
Regal, P. J. 1982. Pollination by wind and animals: Ecology of geographic patterns. Ann. Rev. Ecol. Syst. 13: 497-524.
Reiche, C. 1896-1911. Flora de Chile. Imprenta Cervantes, Santiago, Chile.
Reid, W. V. & K. R. Miller. 1989. Extinction: How serious a threat? Pp. 31-56 in W. V. Reid & K. R. Miller (eds.), Keeping options alive: The scientific basis for conserving biodiversity. World Resources Institute, Washington, DC.
Ricci, M. & L. Eaton. 1994. The rescue of Wahlenbergia larrainii in Robinson Crusoe Island, Chile. Biol. Conservation 68: 89-93.
Richards, A. J. 1997. Plant breeding systems. Ed. 2. Chapman & Hall, London.
Rick, C. M. 1966. Some plant-animal relations on the Galapagos Islands. Pp. 215-224 in R. I. Bowman (ed.), The Galapagos: Proceedings of the symposia of the Galapagos International Scientific Project. Univ. of California Press, Berkeley and Los Angeles.
Rivera, G. L., L. Galetto & L. Bernardello. 1996. Nectar secretion pattern, removal effects, and breeding system of Ligaria cuneifolia (Loranthaceae). Canad. J. Bot. 74: 1996-2001.
Roelofs, F. M. 1979. The reproductive biology of Cyrtandra grandiflora (Gesneriaceae) on Oahu. Pacific Sci. 33: 223-231.
Roy, M. S., J. C. Torres-Mura & F. Hertel. 1998. Evolution and history of hummingbirds (Aves: Trochilidae) from the Juan Fernandez Islands, Chile. Ibis 140: 265-273.
-----, -----, -----, M. Lemus & R. Sponer. 1999. Conservation of the Juan Fernandez firecrown and its island habitat. Oryx 33: 223-232.
Sakai, A., W. L. Wagner, D. M. Ferguson & D. R. Herbst. 1995a. Origins of dioecy in the Hawaiian flora. Ecology 76: 2517-2529.
-----, -----, -----, & -----. 1995b. Biogeographical and ecological correlates of dioecy in the Hawaiian flora. Ecology 76: 2530-2543.
-----, S. G. Weller, W. L. Wagner, P.S. Soltis & D. E. Soltis. 1997. Phylogenetic perspectives on the evolution of dioecy: Adaptive radiation in the endemic Hawaiian genera schiedea and Alsinidendron (Caryophyllaceae: Alsinoideae), Pp. 455-473 in T. J. Givnish & K. J. Systma (eds.), Molecular evolution and adaptive radiation. Cambridge Univ. Press, Cambridge.
Sanders, R. W., T. F. Stuessy & C. Marticorena. 1982. Recent changes in the flora of the Juan Fernandez Islands, Chile. Taxon 31: 284-289.
-----, -----, ----- & M. Silva O. 1987. Phytogeography and evolution of Dendroseris and Robinsonia, tree-Compositae of the Juan Fernandez Islands. Opera Bot. 92: 195-215.
Shelvocke, G. 1726. A voyage round the world by way of the great South Sea. J. Senex, London.
Skottsberg, C. 1921. The phanerogams of the Juan Fernandez islands. Pp.95-240 in C. Skottsberg (ed.), The natural history of Juan Fernandez and Easter Island. Vol. 2, Botany. Almqvist & Wiskell, Uppsala, Sweden.
-----. 1928. Pollinationsbiologie und Samenverbreitung auf den Juan Fernandez Inseln. Pp. 503-547 in C. Skottsberg (ed.), The natural history of Juan Fernandez and Easter Island. Vol. 2, Botany. Almqvist & Wiskell, Uppsala, Sweden.
-----. 1953. A supplement to the pteridophytes and phanerogams of Juan Fernandez and Easter Island. Pp.763-792 in C. Skottsberg (ed.), The natural history of Juan Fernandez and Easter Island. Vol.2, Botany. Almqvist & Wiskell, Uppsala. Sweden.
-----. 1957. Une seconde espece de Centaurodendron Johow. Bull. Jard. Bot. Etat 27: 585-589.
Sloane, H. 1707-1725. A voyage to the islands: Madeira, Barbados, Nieves, St. Christophers and Jamaica. Private publication, London.
Smith, F. D. M., R. M. May, R. Pellew, T. H. Johnson & K. R. Walter. 1993. How much do we know about the current extinction rate? Trends Ecol. Evol. 8: 375-378.
Smith Ramirez, C. 1993. Los picaflores y su recurso floral en el bosque templado de la isla Chiloe, Chile. Rev. Chilena Hist. Nat. 66: 65-73.
Stebbins, G. L. 1957. Self-fertilization and population variability in the higher plants. Amer. Naturalist 41: 337-354.
-----. 1970. Adaptive radiation of reproductive characteristics in angiosperms, I: Pollination mechanisms. Ann. Rev. Ecol. Syst. 1:307-326.
Steiner, K. E. 1988. Dioecism and its correlates in the Cape flora of South Africa. Amer. J. Bot. 75: 1742-1754.
Stephens, S. G. 1964. Native Hawaiian cotton (Gossypium tomentosum Nutt.). Pacific Sci. 18:385-398.
Stokes, M. E., C. S. Davis & G. G. Koch. 1995. Categorical data analysis using the SAS System. SAS Institute, Cary, NC.
Strid, A. 1970. Studies in the Aegean flora, XVI: Biosystematics of the Nigella arvensis complex. Opera Bot. 28: 1-169.
Stuessy, T. F. 1995. Juan Fernandez Islands, Chile. Pp.565-568 in S.D. Davis, V. H. Heywood & A. C. Hamilton (eds.), Centres of plant diversity: A guide and strategy for their conservation. World Wide Fund for Nature and IUCN-thc World Conservation Union, Cambridge, England.
----- & M. Ono (eds.). 1998. Evolution and speciation of island plants. Cambridge Univ. Press, Cambridge.
-----, K. A. Foland, J. F. Sutter, R. W. Sanders & M. Silva O. 1984. Botanical and geological significance of potassium-argon dates from the Juan Fernandez Islands, Science 225: 49-51.
-----, U. Swenson, D. J. Crawford, G. J. Anderson & M. Silva O. 1997. Plant conservation in the Juan Fernandez Islands. Aliso 16: 89-102.
-----, D. J. Crawford, G. J. Anderson & R. J. Jensen. 1998a. Systematics, biogeography and conservation of Lactoridaceae. Perspect. Pl. Ecol. Evol. & Syst. 1-2: 267-290.
-----, U. Swenson, C. Marticorena, U. Matthei & D. J. Crawford. 1998b. Loss of plant diversity and extinction on Robinson Crusoe Islands, Chile. Pp. 243-257 in C.-I Peng & P. P. Lowry II (eds.), Rare, threatened, and endangered floras of Asia and the Pacific Rim: Proceedings of the International Symposium on Rare, Threatened, and Endangered Floras of Asia and the Pacific Rim, April 30-May 4, Taipei, 1996. Monograph Series, 16. Institute of Botany, Academia Sinica. Taipei, Taiwan.
Sun, B. Y., T. F. Stuessy, A. M. Humana, M. Riveros G. & D. J. Crawford. 1996. Evolution of Rhaphithamnus venustus (Verbenaceae), a gynodioecious hummingbird-pollinated endemic of the Juan Fernandez Islands, Chile. Pacific Sci. 50: 55-65.
Thornton, I. 1971. Darwin's islands: A natural history of the Galapagos. Natural History Press, New York.
Vitousek, P.M., L. L. Loope & H. Adsersen (eds.). 1995. Islands: Biological diversity and ecosystem function. Ecological studies, 115. Springer-Verlag, Berlin.
Wagner, W. L. & V. A. Funk (eds.). 1995. Hawaiian biogeography. Smithsonian Institute Press, Washington, DC.
-----, D. R. Herbst & S. H. Sohmer. 1990. Manual of the flowering plants of Hawaii. Univ. of Hawaii Press, Honolulu.
Wallace, A. R. 1878. Tropical nature and other essays. Macmillan, London.
-----. 1880. Island life, or the phenomena and causes of insular faunas and floras, including a revision and attempted solution of the problem of geological climates. Macmillan, London.
-----. 1895. Natural selection and tropical nature: Essays on descriptive and theoretical biology. Macmillan, London and New York.
Webb, C. J. & D. Kelly. 1993. The reproductive biology of the New Zealand flora. Trends Ecol. Evol. 8: 442-447.
Weller, S. G. 1994. The relationship of rarity to plant reproductive biology. Pp. 90-117 in M. L. Bowles & C. J. Whelan (eds.), Restoration of endangered species: Conceptual issues, planning, and implementation. Cambridge Univ. Press, Cambridge.
Wester, L. 1991. Invasions and extinctions on Masatierra (Juan Fernandez Islands): A review of early historical evidence. J. Hist. Geog. 17: 18-34.
Whitehead, D. R. 1969. Wind pollination in the angiosperms: Evolutionary and environmental considerations. Evolution 23: 28-35.
-----. 1983. Wind pollination: Some ecological and evolutionary perspectives. Pp. 97-108 in L. Real (ed.), Pollination biology. Academic Press, Orlando, FL.
Whittaker, R. J. 1998. Island biogeography: Ecology, evolution, and conservation. Oxford Univ. Press, Oxford.
Williamson, M. 1981. Island populations. Oxford Univ. Press, Oxford.
Wilson, E. O. 1973. The ants of Easter Island and Juan Fernandez. Pacific Insects 15: 285-287.
Woodell, S. R. J. 1979. The role of unspecialized pollinators in the reproductive success of Aldabran plants. Philos. Trans., Ser. B 286: 99-108.
Woodward, R. L. 1969. Robinson Crusoe Island: A history of the Juan Fernandez Islands. Univ. of North Carolina Press, Chapel Hill.
Zink, R. A. & N. T. Wheelwright. 1997. Facultative self-pollination in island irises. Amer. Midl. Naturalist 137: 72-78.
[Figure 1 omitted]
[Figure 2 omitted]
[Figure 3 omitted]
[Figure 4 omitted]
[Figure 5 omitted]
[Figure 6 omitted]
Table I Habit, flower, and floral biology features, breeding system, floral visitors observed, and pollination of the Juan Fernandez autochthonous angiosperm flora Dicots Species (b) Family, genus (a) Native Apiaceae Apium A. australe Centella C. asiatica Eryngium Asteraceae (i) Abrotanella A. linearifolia Centaurodendron (k) Dendroseris (k) Erigeron Gamochaeta Lagenophora L. hariotti Robinsonia (k) Taraxacum T. fernandezianum T. subspathulatum Yunquea (k) Berberidaceae Berberis Boraginaceae Selkirkia (k) Brassicaceae Cardamine C. chenopodioides Campanulaceae Lobelia L. alata Wahlenbergia Caryophyllaceae Spergularia Chenopodiaceae Chenopodium Sarcocornia S. fruticosa Convolvulaceae Calystegia C. tuguriorum Dichondra D. sericea Empetraceae Empetrum E. rubrum Ericaceae Pernettya Euphorbiaceae Dysopsis Fabaceae Sophora Flacourtiaceae Azara Gunneraceae Gunnera Haloragaceae Haloragis Lamiaceae Cuminia (k) Lactoridaceae Lactoris (k) Loranthaceae Notanthera N. heterophylla Myrtaceae Myrceugenia Myteola M. nummularia Ugni Piperaceae Peperomia Plantaginaceae Plantago P. australis P. frima Ranunculaceae Ranunculus Rhamnaceae Colletia Rosacese Acaena Margyracaena Margyricarpus Rubus R. geoides Rubiacese Coprosma Galium Hedyotis H. salzmannii Nertera N. granadensis Rutaceae Fagara Santalaceae Santalum Saxifragaceae Escallonia Scrophulariaceae Euphrasia Mimulus M. glabratus Solanaceae Nicotiana Solanum S. pentlandii ssp. interandinum Urticaceae Boehmeria Parietaria P. debilis Urtica U. berteroana Verbenaceae Rhaphithamnus Winteraceae Drimys Species (b) Family, genus (a) Endemic Apiaceae Apium A. fernandezianum Centella Eryngium E. bupleuroides E. inaccessum E. sarcopphyllum E. x fernandezianum Asteraceae (i) Abrotanella Centaurodendron (k) C. dracaenoides C. palmiforme Dendroseris (k) D. berteroana D. gigantea D. litoralis D. macrantha D. macrophylla D. marginata D. micrantha D. neriifolia D. pinnata D. pruinata D. regia Erigeron E. fernandezianus E. ingae E. luteoviridis E. rupicola Gamochaeta G. fernandeziana Lagenophora Robinsonia (k) R. berteroi R. evenia R. gayana R. gracilis R. macrocephala R. masafuerae R. thurifera Taraxacum Yunquea (k) Y. tenzii Berberidaceae Berberis B. corymbosa B. masafuerana Boraginaceae Selkirkia (k) S. berteroi Brassicaceae Cardamine C. krusselii Campanulaceae Lobelia Wahlenbergia W. berteroi W. fernandeziana W. grahamiae W. masafuerae W. tuberosa W. fernandeziana x W. grahamiae (= hybrid) Caryophyllaceae Spergularia S. confertifolia S. masafuerana Chenopodiaceae Chenopodium Ch. crusoeanum Ch. nesodendron Ch. sanctae-clarae Sarcocornia Convolvulaceae Calystegia Dichondra Empetraceae Empetrum Ericaceae Pernettya P. rigida Euphorbiaceae Dysopsis D. hirsuta Fabaceae Sophora S. fernandeziana S. masafuerana Flacourtiaceae Azara A. serrata var. fernandeziana Gunneraceae Gunnera G. bracteata G. glabra G. masafuerae G. peltata G. bracteata x peltata Haloragaceae Haloragis H. masafuerana H. masatierrana Lamiaceae Cuminia (k) C. eriantha Lactoridaceae Lactoris (k) L. fernandeziana Loranthaceae Notanthera Myrtaceae Myrceugenia M. fernandeziana M. schulzei Myteola Ugni U. selkirkii Piperaceae Peperomia P. berteroana p. fernandeziana P. margaritifera P. skottsbergii Plantaginaceae Plantago P. fernandezia Ranunculaceae Ranunculus R. caprarum Rhamnaceae Colletia C. spartioides Rosacese Acaena A. masafuerana Margyracaena M. skottsbergii Margyricarpus M. digynus Rubus Rubiacese Coprosma C. oliveri C. pyrifolia Galium G. masafueranum Hedyotis Nertera Rutaceae Fagara F. externa F. mayu Santalaceae Santalum S. fernandezianum Saxifragaceae Escallonia E. callcottiae Scrophulariaceae Euphrasia E. formosissima Mimulus Solanaceae Nicotiana N. cordifolia Solanum S. fernandezianum Urticaceae Boehmeria B. excelsa Parietaria Urtica U. glomeruliflora U. masafuerae Verbenaceae Rhaphithamnus R. venustus Winteraceae Drimys D. confertifolia Family, genus (a) Habit Apiaceae Apium Perennial herb Centella Perennial herb Eryngium Shrub/tree Asteraceae (i) Abrotanella Perennial herb Centaurodendron (k) Tree Dendroseris (k) Shrub/tree Erigeron Shrub Gamochaeta Perennial herb Lagenophora Perennial herb Robinsonia (k) Shrub Taraxacum Perennial herb Yunquea (k) Tree Berberidaceae Berberis Shrub Boraginaceae Selkirkia (k) Shrub Brassicaceae Cardamine Annual or perennial herb Campanulaceae Lobelia Annual herb Wahlenbergia Shrub Caryophyllaceae Spergularia Perennial herb Chenopodiaceae Chenopodium Shrub Sarcocornia Perennial herb Convolvulaceae Calystegia Shrub Dichondra Perennial herb Empetraceae Empetrum Shrub Ericaceae Pernettya Shrub Euphorbiaceae Dysopsis Perennial herb Fabaceae Sophora Tree Flacourtiaceae Azara Tree Gunneraceae Gunnera Perennial herb Haloragaceae Haloragis Perennial herb Lamiaceae Cuminia (k) Shrub Lactoridaceae Lactoris (k) Shrub Loranthaceae Notanthera Shrub Myrtaceae Myrceugenia Tree Myteola Shrub Ugni Shrub Piperaceae Peperomia Perennial herb Plantaginaceae Plantago Shrub Ranunculaceae Ranunculus Perennial herb Rhamnaceae Colletia Shrub Rosacese Acaena Perennial herb Margyracaena Perennial herb Margyricarpus Perennial herb Rubus Perennial herb Rubiacese Coprosma Tree Galium Perennial herb Hedyotis Perennial herb Nertera Perennial herb Rutaceae Fagara Tree Santalaceae Santalum Tree Saxifragaceae Escallonia Shrub/tree Scrophulariaceae Euphrasia Shrub Mimulus Annual herb Solanaceae Nicotiana Shrub Solanum Perennial herb Urticaceae Boehmeria Tree Parietaria Annual herb Urtica Annual or perennial herb Verbenaceae Rhaphithamnus Tree Winteraceae Drimys Tree Flower Size, in mm Family, genus (a) (category) Apiaceae Apium 2 x 2 (very small) Centella 2 x 2 (very small) Eryngium 5 x 4 (small) Asteraceae (i) Abrotanella 3 x 2 (very small) Centaurodendron (k) 20 x 8 (large) Dendroseris (k) 22 x 27 (very large) Erigeron 8 x 8 (medium) Gamochaeta 3 x 3 (very small) Lagenophora 6 x 6 (medium) Robinsonia (k) 9 x 6 (medium) Taraxacum 30 x 8 (large) Yunquea (k) 30 x 15 (very large) Berberidaceae Berberis 6 x 8 (medium) Boraginaceae Selkirkia (k) 13 x 7 (medium) Brassicaceae Cardamine 6 x 4 (small) Campanulaceae Lobelia 11 x 5 (medium) Wahlenbergia 12 x 8 (medium) Caryophyllaceae Spergularia 2.5 x 2.5 (very small) Chenopodiaceae Chenopodium 1 x 1.5 (very small) Sarcocornia 2 x 2 (very small) Convolvulaceae Calystegia 40 x 50 (very large) Dichondra 2.5 x 2.5 (very small) Empetraceae Empetrum 3 x 2 (very small) Ericaceae Pernettya 8 x 4 (medium) Euphorbiaceae Dysopsis 2 x 2 (very small) Fabaceae Sophora 30 x 5 (large) Flacourtiaceae Azara 4 x 6 (small) Gunneraceae Gunnera 1.2 x 1.2 (very small) Haloragaceae Haloragis 4.3 x 7 (medium) Lamiaceae Cuminia (k) 15 x 5 (medium) Lactoridaceae Lactoris (k) 3.5 x 3 (small) Loranthaceae Notanthera 12 x 2 (small) Myrtaceae Myrceugenia 2.5 x 2.5 (very small) Myteola 8 x 6 (medium) Ugni 8 x 8 (medium) Piperaceae Peperomia 1 x 1 (very small) Plantaginaceae Plantago 5 x 2 (small) Ranunculaceae Ranunculus 8 x 2 (small) Rhamnaceae Colletia 5 x 3 (small) Rosacese Acaena 2 x 1.5 (very small) Margyracaena 2 x 1.5 (very small) Margyricarpus 1.5 x 1.5 (very small) Rubus 20 x 15 (large) Rubiacese Coprosma 7 x 6 (medium) Galium 2 x 3 (very small) Hedyotis 5 x 2 (small) Nertera 2 x 2 (very small) Rutaceae Fagara 4 x 4 (small) Santalaceae Santalum 4 x 4 (small) Saxifragaceae Escallonia 15 x 5 (medium) Scrophulariaceae Euphrasia 12 x 6 (medium) Mimulus 25 x 10 (large) Solanaceae Nicotiana 25 x 4 (medium) Solanum 10 x 8 (medium) Urticaceae Boehmeria 1.5 x 1.5 (very ssmall) Parietaria 1 x 1 (very small) Urtica 1.5 x 1 (very small) Verbenaceae Rhaphithamnus 29 x 6 (large) Winteraceae Drimys 4 x 23 (medium) Flower Family, genus (a) Shape (c) Color Apiaceae Apium Dish White/green Centella Dish White/purplish Eryngium Dish Green Violet Green Violet Asteraceae (i) Abrotanella Dish (discoid) Brown/red Centaurodendron (k) Dish (discoid) Blue, dark purple Dendroseris (k) Dish (radiate) White White Orange Orange Orange Orange White White White White White Erigeron Dish (radiate) White/yellow Gamochaeta Dish (discoid) White Lagenophora Dish (radiate) White/pink Robinsonia (k) Dish (radiate) Yellow Yellow Yellowish green Yellow Yellow Green Yellowish green Taraxacum Dish (ligulate) Yellow Yunquea (k) Dish Berberidaceae Berberis Dish Yellow Boraginaceae Selkirkia (k) Bell White Brassicaceae Cardamine Dish White Campanulaceae Lobelia Tube Blue Wahlenbergia Dish, tube Pink White White White White White Caryophyllaceae Spergularia Inconspicuous White, white/pink Chenopodiaceae Chenopodium Inconspicuous Green Sarcocornia Inconspicuous Green Convolvulaceae Calystegia Bell White Dichondra Bell Green Empetraceae Empetrum Bell Red Ericaceae Pernettya Bell White Euphorbiaceae Dysopsis Dish Green Fabaceae Sophora Flag Yellow Flacourtiaceae Azara Brush Yellow Gunneraceae Gunnera Inconspicuous Red Haloragaceae Haloragis Dish Green Lamiaceae Cuminia (k) Tube Violet Lactoridaceae Lactoris (k) Inconspicuous Green Loranthaceae Notanthera Tube White Myrtaceae Myrceugenia Bell White Myteola Bell White Ugni Bell White Piperaceae Peperomia Inconspicuous Green Plantaginaceae Plantago Tube Brown Ranunculaceae Ranunculus Tube White/pink Rhamnaceae Colletia Tube White/pink Rosacese Acaena Inconspicuous Green Margyracaena Inconspicuous Green Margyricarpus Inconspicuous Green Rubus Dish White Rubiacese Coprosma Tube Green/brown Galium Tube White/green Hedyotis Tube White Nertera Tube White/green Rutaceae Fagara Inconspicuous Brown Santalaceae Santalum Dish White Saxifragaceae Escallonia Tube Red Scrophulariaceae Euphrasia Tube White Mimulus Tube Yellow Solanaceae Nicotiana Tube Dark red Solanum Dish Violet,yellow /white Urticaceae Boehmeria Inconspicuous Green Parietaria Inconspicuous Green Urtica Inconspicuous Green Verbenaceae Rhaphithamnus Tube Violet Winteraceae Drimys Dish White Family, genus (a) Sex Apiaceae Apium Hermaphroditic Centella Hermaphroditic Eryngium Hermaphroditic Asteraceae (i) Abrotanella Gynomonecious Centaurodendron (k) Andromonecious Dendroseris (k) Hermaphroditic Erigeron Gynomonecious Gamochaeta Hermaphroditic Lagenophora Monoecious/ gynomonecious Robinsonia (k) Dioecious Taraxacum Hermaphroditic/ gynomonoe cious Yunquea (k) Hermaphroditic Berberidaceae Berberis Hermaphroditic Boraginaceae Selkirkia (k) Hermaphroditic Brassicaceae Cardamine Hermaphroditic Campanulaceae Lobelia Hermaphroditic Wahlenbergia Hermaphroditic Caryophyllaceae Spergularia Hermaphroditic Chenopodiaceae Chenopodium Gynomonoecious Sarcocornia Hermaphroditic/ polygamous Convolvulaceae Calystegia Hermaphroditic Dichondra Hermaphroditic Empetraceae Empetrum Polygamous Ericaceae Pernettya Dioecious Euphorbiaceae Dysopsis Monoecious Fabaceae Sophora Hermaphroditic Flacourtiaceae Azara Hermaphroditic Gunneraceae Gunnera Hermaphroditic Haloragaceae Haloragis Hermaphroditic Lamiaceae Cuminia (k) Gynomonoecious Lactoridaceae Lactoris (k) Gynomonoecious Loranthaceae Notanthera Hermaphroditic Myrtaceae Myrceugenia Hermaphroditic Myteola Hermaphroditic Ugni Hermaphroditic Piperaceae Peperomia Hermaphroditic Plantaginaceae Plantago Hermaphroditic Ranunculaceae Ranunculus Hermaphroditic Rhamnaceae Colletia Hermaphroditic Rosacese Acaena Hermaphroditic Margyracaena Hermaphroditic Margyricarpus Hermaphroditic Rubus Hermaphroditic Rubiacese Coprosma Dioecious Galium Hermaphroditic Hedyotis Hermaphroditic Nertera Hermaphroditic Rutaceae Fagara Hermaphroditic Santalaceae Santalum Hermaphroditic Saxifragaceae Escallonia Hermaphroditic Scrophulariaceae Euphrasia Hermaphroditic Mimulus Hermaphroditic Solanaceae Nicotiana Hermaphroditic Solanum Hermaphroditic Urticaceae Boehmeria Monoecious Parietaria Polygamous Urtica Monoecious Verbenaceae Rhaphithamnus Gynodioecious Winteraceae Drimys Hermaphroditic Family, genus (a) Dichogamy (d) Apiaceae Apium Centella Eryngium Protogyny Asteraceae (i) Abrotanella Protandry Centaurodendron (k) Protandry Dendroseris (k) Protandry Erigeron Protandry Gamochaeta Protandry Lagenophora Protandry Robinsonia (k) Protandry Taraxacum Protandry Yunquea (k) Protandry Berberidaceae Berberis Boraginaceae Selkirkia (k) Brassicaceae Cardamine Campanulaceae Lobelia Protandry Wahlenbergia Protandry Caryophyllaceae Spergularia Chenopodiaceae Chenopodium Sarcocornia Convolvulaceae Calystegia Dichondra Empetraceae Empetrum Ericaceae Pernettya Euphorbiaceae Dysopsis Fabaceae Sophora Flacourtiaceae Azara Gunneraceae Gunnera Protandry Haloragaceae Haloragis Protandry Lamiaceae Cuminia (k) Lactoridaceae Lactoris (k) Protogyny Loranthaceae Notanthera Myrtaceae Myrceugenia Protogyny Myteola Ugni Protandry Piperaceae Peperomia Plantaginaceae Plantago Protogyny Ranunculaceae Ranunculus Rhamnaceae Colletia Rosacese Acaena Margyracaena Margyricarpus Rubus Rubiacese Coprosma Galium Hedyotis Nertera Rutaceae Fagara Santalaceae Santalum Saxifragaceae Escallonia Protandry Scrophulariaceae Euphrasia Mimulus Solanaceae Nicotiana Protandry Solanum Protandry Urticaceae Boehmeria Parietaria Urtica Verbenaceae Rhaphithamnus Protandry Winteraceae Drimys Protogyny Breeding system Floral Family, genus (a) determined (d) reward Apiaceae Apium Nectar Centella Nectar Eryngium Nectar Asteraceae (i) Abrotanella Nectar Centaurodendron (k) Nectar Dendroseris (k) Nectar Self-compatible Self-incompatible Erigeron Nectar Gamochaeta Nectar Lagenophora Nectar Robinsonia (k) Nectar Taraxacum Agamospermous Nectar Yunquea (k) Nectar Berberidaceae Berberis Nectar Self-incompatible Boraginaceae Selkirkia (k) Nectar Brassicaceae Cardamine Nectar Campanulaceae Lobelia Nectar Wahlenbergia Nectar Self-compatible Self-compatible Self-compatible Caryophyllaceae Spergularia Nectar Chenopodiaceae Chenopodium Sarcocornia Convolvulaceae Calystegia Nectar Dichondra Nectar Empetraceae Empetrum Ericaceae Pernettya Nectar Euphorbiaceae Dysopsis Fabaceae Sophora Nectar Flacourtiaceae Azara Self-compatible Nectar Gunneraceae Gunnera Haloragaceae Haloragis Self-compatible Lamiaceae Cuminia (k) Nectar Lactoridaceae Lactoris (k) Self-compatible Loranthaceae Notanthera Nectar Myrtaceae Myrceugenia Myteola Self-compatible (n) Nectar Ugni Nectar Piperaceae Peperomia Plantaginaceae Plantago Self-compatible Ranunculaceae Ranunculus Nectar Rhamnaceae Colletia Nectar Rosacese Acaena Nectar Margyracaena Nectar Margyricarpus Nectar Rubus Self-compatible (a) Nectar Rubiacese Coprosma Nectar Galium Nectar Hedyotis Nectar Nertera Nectar Rutaceae Fagara Nectar Santalaceae Santalum Saxifragaceae Escallonia Self-compatible, selfer Nectar Scrophulariaceae Euphrasia Nectar Mimulus Nectar Solanaceae Nicotiana Self-compatible Nectar Solanum Pollen Self-compatible, (o) self Urticaceae Boehmeria Self-compatible Parietaria Urtica Verbenaceae Rhaphithamnus Self-compatible Nectar Winteraceae Drimys Pollen Floral visitors Family, genus (a) observed (d) Apiaceae Apium Centella Eryngium Hummingbirds, (g, h) flies (g, i) Asteraceae (i) Abrotanella Centaurodendron (k) Hummingbirds (m) Dendroseris (k) Flies (i) Hummiogbirds (g, h, l, m) Flies & moths (l) Erigeron None (l) Gamochaeta Lagenophora Robinsonia (k) Flies (i) None (g, l) None (g, l) None (g, l) None (g, l) Taraxacum Yunquea (k) Berberidaceae Berberis None (l) Beetles (g) Boraginaceae Selkirkia (k) Brassicaceae Cardamine Campanulaceae Lobelia Wahlenbergia Ants & flies (l) Months & anys (l) Moths (g) Moths, ants, and bee (l) Caryophyllaceae Spergularia Chenopodiaceae Chenopodium None (l) Sarcocornia Convolvulaceae Calystegia Dichondra Empetraceae Empetrum Ericaceae Pernettya Flies & beetles, (l) moths (g) Euphorbiaceae Dysopsis None (l) Fabaceae Sophora Hummingbirds (g, h) Flacourtiaceae Azara None (g, l) Gunneraceae Gunnera None (l) Not observed (l) Haloragaceae Haloragis None (l) Lamiaceae Cuminia (k) Hummingbirds (g, l) Lactoridaceae Lactoris (k) None (l) Loranthaceae Notanthera Hummingbirds (g) Myrtaceae Myrceugenia None (g) Myteola Ugni None (g, l) Piperaceae Peperomia None (l) Plantaginaceae Plantago None (l) Ranunculaceae Ranunculus Rhamnaceae Colletia Moths (g) Rosacese Acaena Margyracaena None (l) Margyricarpus None (l) Rubus Rubiacese Coprosma None (l) Galium Hedyotis Nertera Rutaceae Fagara Santalaceae Santalum Saxifragaceae Escallonia Hummingbirds, (g, h, l) flies, (l) moths (g, l) Scrophulariaceae Euphrasia Mimulus None (g) Solanaceae Nicotiana Hummingbirds (g, l) Solanum Urticaceae Boehmeria None (l) Parietaria Urtica Verbenaceae Rhaphithamnus Hummingbirds (g, h, l, m) Winteraceae Drimys None (l) Family, genus (a) Pollination (d, e) Apiaceae Apium Centella Eryngium Hummingbird Asteraceae (i) Abrotanella Centaurodendron (k) Hummingbird Dendroseris (k) Hummingbird Erigeron Gamochaeta Lagenophora Robinsonia (k) Taraxacum Yunquea (k) Berberidaceae Berberis Boraginaceae Selkirkia (k) Brassicaceae Cardamine Campanulaceae Lobelia Hummingbird Wahlenbergia Selfer, wind Caryophyllaceae Spergularia Chenopodiaceae Chenopodium Wind Sarcocornia Wind Convolvulaceae Calystegia Dichondra Empetraceae Empetrum Wind Ericaceae Pernettya Wind Euphorbiaceae Dysopsis Wind Fabaceae Sophora Hummingbird Flacourtiaceae Azara Wind Gunneraceae Gunnera Wind Haloragaceae Haloragis Wind Lamiaceae Cuminia (k) Hummingbird Lactoridaceae Lactoris (k) Wind Loranthaceae Notanthera Hummingbird Myrtaceae Myrceugenia Myteola Ugni Wind Piperaceae Peperomia Wind Plantaginaceae Plantago Wind Ranunculaceae Ranunculus Rhamnaceae Colletia Rosacese Acaena Margyracaena Margyricarpus Rubus Rubiacese Coprosma Wind Galium Hedyotis Nertera Wind Rutaceae Fagara Santalaceae Santalum Saxifragaceae Escallonia Hummingbird, selfer Scrophulariaceae Euphrasia Mimulus Solanaceae Nicotiana Hummingbird Solanum Selfer Urticaceae Boehmeria Wind Parietaria Wind Urtica Wind Verbenaceae Rhaphithamnus Hummingbird Winteraceae Drimys Wind Pollination of Family, genus (a) colonizers (f) Apiaceae Apium Insect Centella Insect Eryngium Insect Asteraceae (i) Abrotanella Insect Centaurodendron (k) Insect Dendroseris (k) Insect Erigeron Insect Gamochaeta Insect Lagenophora Insect Robinsonia (k) Insect Taraxacum Insect Yunquea (k) Insect Berberidaceae Berberis Insect Boraginaceae Selkirkia (k) Insect Brassicaceae Cardamine Insect Campanulaceae Lobelia Hummingbird Wahlenbergia Insect Caryophyllaceae Spergularia Insect Chenopodiaceae Chenopodium Wind Sarcocornia Wind Convolvulaceae Calystegia Insect Dichondra Insect Empetraceae Empetrum Wind Ericaceae Pernettya Insect Euphorbiaceae Dysopsis Insect Fabaceae Sophora Hummingbird Flacourtiaceae Azara Insect Gunneraceae Gunnera Wind Haloragaceae Haloragis Wind Lamiaceae Cuminia (k) Hummingbird Lactoridaceae Lactoris (k) Wind Loranthaceae Notanthera Hummingbird Myrtaceae Myrceugenia Insect Myteola Insect Ugni Insect Piperaceae Peperomia Insect Plantaginaceae Plantago Wind Ranunculaceae Ranunculus Insect Rhamnaceae Colletia Insect Rosacese Acaena Insect Margyracaena Wind Margyricarpus Wind Rubus Insect Rubiacese Coprosma Wind Galium Insect Hedyotis Insect Nertera Wind Rutaceae Fagara Insect Santalaceae Santalum Insect Saxifragaceae Escallonia Hummingbird, insect Scrophulariaceae Euphrasia Insect Mimulus Insect Solanaceae Nicotiana Hummingbird, insect Solanum Insect Urticaceae Boehmeria Wind Parietaria Wind Urtica Wind Verbenaceae Rhaphithamnus Insect Winteraceae Drimys Insect Monocots Species (b) Family, genus (a) Native Arecaceae Juania (k) Bromeliaceae Greigia Ochagavia (k) Cyperaceae Carex, Cyperus, Eleocharis, 9 spp. Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia L. chilensis Juncaceae Juncus J. cappillaceus J. imbricatus J. pallescens J. planifolius J. procerus Luzula [_NOT REPRODUCIBLE IN ASCII] Orchidaceae Gavilea Poaceae Agrostis, Bromus, Chaetotropis, 10 spp. Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Species (b) Family, genus (a) Endemic Arecaceae Juania (k) J. australis Bromeliaceae Greigia G. berteroi Ochagavia (k) O. elegans Cyperaceae Carex, Cyperus, Eleocharis, 4 spp. Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia Juncaceae Juncus Luzula L. masafuerana Orchidaceae Gavilea G. insularis Poaceae Agrostis, Bromus, Chaetotropis, 5 spp. Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Family, genus (a) Habit Arecaceae Juania (k) Tree Bromeliaceae Greigia Shrub Ochagavia (k) Shrub Cyperaceae Carex, Cyperus, Eleocharis, Perennial herb Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia Perennial herb Juncaceae Juncus Perennial herb Luzula Perennial herb Orchidaceae Gavilea Perennial herb Poaceae Agrostis, Bromus, Chaetotropis, Annual or perennial herb Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Flower Size, in mm Family, genus (a) (category) Arecaceae Juania (k) 10 x 4 (medium) Bromeliaceae Greigia 30 x 10 (large) Ochagavia (k) 35 x 10 (large) Cyperaceae Carex, Cyperus, Eleocharis, 2.5 x 1.5 (very small) Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia 15 x 20 (large) Juncaceae Juncus 4 x 2 (very small) Luzula 2.5 x 2 (very small) Orchidaceae Gavilea 15 x 10 (large) Poaceae Agrostis, Bromus, Chaetotropis, 2.5 x 1.5 (very small) Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Flower Family, genus (a) Shape (c) Color Arecaceae Juania (k) Dish Green Bromeliaceae Greigia Tube Ochagavia (k) Tube Red/violet Cyperaceae Carex, Cyperus, Eleocharis, Inconspicuous Green/brown Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia Dish White Juncaceae Juncus Inconspicuous Green/brown Luzula Inconspicuous Green/brown Orchidaceae Gavilea Flag Yellow Poaceae Agrostis, Bromus, Chaetotropis, Inconspicuous Green/brown Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Family, genus (a) Sex Arecaceae Juania (k) Dioecious Bromeliaceae Greigia Hermaphroditic Ochagavia (k) Hermaphroditic Cyperaceae Carex, Cyperus, Eleocharis, Hermaphroditic, 6 spp.; Macherina, Oreobolus, Scirpus, monoecious, 7 spp.; Uncinia andromonecious, 1 sp. Iridaceae Libertia Hermaphroditic Juncaceae Juncus Hermaphroditic Luzula Hermaphroditic Orchidaceae Gavilea Hermaphroditic Poaceae Agrostis, Bromus, Chaetotropis, Hermaphroditic, 10 spp. Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Family, genus (a) Dichogamy (d) Arecaceae Juania (k) Bromeliaceae Greigia Ochagavia (k) Protandry Cyperaceae Carex, Cyperus, Eleocharis, Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia Juncaceae Juncus Luzula Orchidaceae Gavilea Poaceae Agrostis, Bromus, Chaetotropis, Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Breeding system Floral Family, genus (a) determined (d) reward Arecaceae Juania (k) Bromeliaceae Greigia Nectar Ochagavia (k) Nectar Cyperaceae Carex, Cyperus, Eleocharis, Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia Nectar Juncaceae Juncus Luzula Orchidaceae Gavilea Nectar Poaceae Agrostis, Bromus, Chaetotropis, Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Floral visitors Family, genus (a) observed (d) Arecaceae Juania (k) Bromeliaceae Greigia Hummingbirds Ochagavia (k) Hummingbird (g) Cyperaceae Carex, Cyperus, Eleocharis, Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia Files, (g, l) moths (g) Juncaceae Juncus Luzula Orchidaceae Gavilea Poaceae Agrostis, Bromus, Chaetotropis, Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Family, genus (a) Pollination (d, e) Arecaceae Juania (k) Wind Bromeliaceae Greigia Hummingbird Ochagavia (k) Hummingbird Cyperaceae Carex, Cyperus, Eleocharis, Wind Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia Insect Juncaceae Juncus Wind Luzula Wind Orchidaceae Gavilea Poaceae Agrostis, Bromus, Chaetotropis, Wind Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum Pollination of Family, genus (a) colonizers (f) Arecaceae Juania (k) Insect Bromeliaceae Greigia Hummingbird Ochagavia (k) Hummingbird Cyperaceae Carex, Cyperus, Eleocharis, Wind Macherina, Oreobolus, Scirpus, Uncinia Iridaceae Libertia Insect Juncaceae Juncus Wind Luzula Wind Orchidaceae Gavilea Insect Poaceae Agrostis, Bromus, Chaetotropis, Wind Chusquea, Danthonia, Leptophyllachloa, Megalachne, (k) Nassella, Piptochaetium, Podophorus, (k) Trisetum (a)Characteristics are recorded at the genus level unless noted. When there is more than one species per genus, the general character state or average is given; however, the distinctly different species are also noted. (b)In the species columns, empty cells indicate that no species of the kind are known for the respective genus; varieties are only included when they are the only representative of a taxon in the archipelago. (c)Flower-shape categories are after Faegri and van der Pijl (1979). For the Asteraceae, the type of capitulum is also shown, in parentheses. (d)Empty cells indicate missing information. (e)Includes conclusions for species for which data or literature offer reasonable evidence. (f)Hypothetical, based on what is known from the literature on the pollination of close relatives. (g)Observed by Skottsberg (1928). (h)Observed by Meza (1988). (i)Observed by Johow (1896). (j)For the Asteraceae, data represent the size of the capitulum. (k)Endemic genera. (l)Observed by the authors. (m)Observed by Brooke (1987). (n)From Moore (1983). (o)From Ramanna and Hermsen (1981). Table II Statistical tests (chi-square and general association coefficient) of features of the Juan Fernandez flora that were positively associated. In all cases degrees of freedom = 1. p = probability for both chi-square ([x.sup.2]) and general association coefficient. General association Feature [x.sup.2] coefficient p Very small flowers and perennial 31.65 31.44 <.0001 herbs Hermaphroditic and white flowers 13.17 13.07 0.0003 Monoecious and green flowers 17.21 17.09 <.0001 Inconspicuous and green flowers 48.50 48.17 <.0001 Bell-shaped and white flowers 12.07 11.95 0.0005 Very small and green flowers 50.09 49.77 <.0001 Inconspicuous and very small 118.47 117.69 <.0001 flowers Perennial herbs and current wind 13.96 13.80 0.0002 pollination Large- or medium-sized flowers and 25.50 25.20 <.0001 and current bird pollination Very small flowers and current 24.49 24.21 <.0001 wind pollination Very small or small flowers and 40.27 39.81 <.0001 current wind pollination Inconspicuous flowers and current 27.41 27.10 <.0001 wind pollination Bright-colored flowers and current 59.16 58.44 <.0001 bird pollination Green flowers and current wind 6.22 6.14 0.01 pollination Green or brown flowers and current 47.03 46.53 <.0001 wind pollination Trees and wind pollination of 4.76 4.72 0.03 colonizers Very small flowers and wind 36.31 35.88 <.0001 pollination of colonizers Inconspicuous flowers and wind 48.43 47.86 <.0001 pollination of colonizers Dish-shaped flowers and insect 26.51 26.20 <.0001 pollination of colonizers White flowers and insect 24.77 24.46 <.0001 pollination of colonizers Bright-colored flowers and bird 44.82 44.27 <.0001 pollination of colonizers Green flowers and wind pollination 31.02 30.65 <.0001 of colonizers
|Printer friendly Cite/link Email Feedback|
|Author:||Bernardello, Gabriel; Anderson, Gregory J.; Stuessy, Tod F.; Crawford, Daniel J.|
|Publication:||The Botanical Review|
|Date:||Jul 1, 2001|
|Previous Article:||Senna reticulata, a Pioneer Tree from Amazonian Varzea Floodplains.|
|Next Article:||Remains of palms (palmae) at archaeological sites in the new world: A review.|