| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Systematics |
Department of Plant Biology, University of Minnesota, St. Paul, Minnesota 55108 USA
Received for publication August 7, 2003. Accepted for publication January 8, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: breeding systems • Ficus • molecular dating • Moraceae • phylogenetic classification • pollination
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Cook and Rasplus, 2003
; Jousselin et al., 2003
). However, the origin of the fig inflorescence has remained a mystery because its unusual morphology is not easily related to other Moraceae inflorescences (Fig. 1). Different evolutionary pathways to the syconium have been suggested based on the diversity of inflorescences in the family. Corner (1978)
speculated that the fig evolved from an urn-shaped receptacle resembling Antiaropsis or Sparattosyce (Fig. 1), while Berg (1989)
hypothesized a cymose ancestor. Recent molecular studies suggested a close relationship with Poulsenia (Herre et al., 1996
) or Castilla (Sytsma et al., 2002
). However, taxon sampling in these studies was not sufficient to resolve the sister group to the figs. Investigating the fig origin requires a detailed phylogenetic hypothesis that has been lacking for the family until now.
|
; Wiegrefe et al., 1998
; Sytsma et al., 2002
). The Urticalean rosids differ from most other rosids in the presence of solitary ovules, lactifers, cystoliths, paired inflorescences in leaf axils, and unisexual flowers. Urticaceae plus Cecropiaceae are sister to Moraceae, distinguished from the latter in having lactifers only in the bark, clear latex, and orthotropous ovules (Sytsma et al., 2002
).
Moraceae are characterized by milky latex in all parenchymatous tissue, unisexual flowers, anatropous ovules, and aggregated drupes or achenes. Growth forms include trees, shrubs, hemiepiphytes, climbers, and herbs. Flowers are reduced and, when present, the perianth is four- or five-merous, tepaloid, and often membranous. Filaments are either inflexed in bud or straight. Inflexed stamens, often referred to as "urticaceous," are associated with a pistillode against which the anthers are appressed in bud. These stamens, springing outward at anthesis to release their pollen, are indicative of wind pollination (Corner, 1962
; Berg, 2001
). Straight filaments are often but not always associated with pollination involving insects. The perianths of carpellate flowers are often connate or adnate to the receptacle (Berg, 2001
), a condition hypothesized to protect flowers against phytophagous insects (Berg, 1989
, 1990
).
Moraceae have been divided into five tribes (Table 1; Rohwer, 1993
). Ficeae are monotypic with a pantropical distribution and 
750 species. Plants can be either monoecious with bisexual inflorescences or gynodioecious but functionally dioecious (Weiblen, 2000
). Artocarpeae are represented by 12 genera and 87 species, including the economically important Artocarpus (jackfruit, breadfruit). Species are either monoecious or dioecious, with unisexual inflorescences of variable architecture including racemes, spikes, capitula, globes, discs, and solitary flowers (Jarrett, 1959
; Berg, 1988
). Berg (1988
, 2001
) recognized that the Artocarpeae lack the homogeneity of other Moraceae tribes and suggested that it might be subdivided into three subtribes on account of the morphological variation.
|
, 1977a
, 2001
). Species are monoecious, dioecious, or androdioecious (Berg, 1972
, 1977a
, 2001
).
The eight African and neotropical genera of Dorstenieae are dominated by 100 species of Dorstenia. The tribe is defined by discoid or globose bisexual inflorescences with the carpellate flowers often embedded within the receptacle, peltate interfloral bracts, and uncinate hairs in many species. Growth form varies from arborescent in most genera to herbaceous or suffructescent in Dorstenia (Berg and Hijman, 1999
). Although most species are monoecious, dioecy and androdioecy also occur (Berg, 1988
).
The eight genera of Moreae include approximately 70 species. Species are mostly dioecious, although a few species are monoecious with bisexual inflorescences. Inflorescences are relatively simple racemes, spikes, or globose heads. Most species have urticaceous stamens, apparently related to anemophily. However, there is a great deal of variation in vegetative and floral morphology, and generic delimitation is somewhat uncertain (Berg, 2001
). Up to 40 generic names have been recognized at one time or another in this group (Berg, 2001
). Several genera have been enlarged in an attempt to devise a more satisfactory and cohesive classification (Corner, 1962
, 1975
; Berg, 1982
, 1986
, 1988
, 2001
). It has been suggested that taxonomic uncertainty within Moreae stems from the retention of plesiomorphic features (Berg, 1989
, 2001
).
We examined Moraceae phylogenetic relationships, reproductive character evolution, and the origin of the fig pollination mutualism. We employed sequences from the chloroplast gene ndhF because it has proven useful for examining ordinal and familial relationships among plants (Olmstead et al., 2001
; Sytsma et al., 2002
). Moraceae chloroplast DNA phylogeny provided insights into (1) tribal relationships, including the sister group to the figs, (2) ancestral changes in breeding systems and pollination syndromes, (3) ancestral traits associated with the origin of the syconium, and (4) molecular divergence time estimates for the origin of fig pollination.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1200 bp each using primers from Olmstead and Sweere (1994)
including ndhF8f, ndhF972, ndhF1318r, and ndhF2110r. Amplification conditions were as follows: 20 ng genomic DNA, 1x TaKaRa Ex Taq buffer (2 mmol/L MgCl2), 160 nmol/L each primer, 0.2 mmol/L each dNTP, 1.25 unit TaKaRa Ex Taq DNA polymerase (Otsu, Shiga, Japan). Thermal cycling was performed in 25 cycles of 94°C for 30 s, 48°C for 60 s, 68°C for 90 s, and a final extension at 72°C for 7 min. The PCR products were cleaned using the QIAquick or MinElute PCR purification spin columns (Qiagen, Valencia, California, USA). Clean PCR products were quantified using Hoechst 33258 fluorescent dye (Acros Organics, Morris Plains, New Jersey, USA) in a Turner Quantech fluorometer (Barnstead-Thermolyne, Dubuque, Iowa, USA). Ten- microliter sequencing reactions were performed with Big Dye sequencing reagents and protocols (versions 2, 3; Applied Biosystems, Foster City, California, USA) and data were collected using an ABI 377 Automated DNA sequencer (Applied Biosystems). Between seven and nine sequencing primers were used for each taxon, including the aforementioned primers, ndhF972r, ndhF1318, and ndhF1603r. In addition, we designed two primers for Urticales, ndhF84f (5'-TCT TCG CCG TAT AGT GGG TTT TTC C-3'), and ndhF713r (5'-ATC RGG TAA CCA TAC ATG AAG RGG-3'). Sequences were edited using Sequencher version 3.0 (Gene Codes, Ann Arbor, Michigan, USA). Sequence alignment was approximated with ClustalX (Thompson et al., 1997
), followed by manual alignment to maintain open reading frames for the entire portion of the gene.
Parsimony analyses
Tree searches were performed using PAUP* version 4.0b10 (Swofford, 2002
) with the tree-bisection-reconnection (TBR) branch- swapping algorithm and 10 000 random addition sequence replicates. Maxtrees was set to increase without limit. Support was assessed with 1000 bootstrap replicates (10 addition sequence replicates per bootstrap replicate) and maxtrees set at 10 000. The decay index (Bremer, 1988
, 1994
) was assessed using TreeRot version 2 (Sorenson, 1999
) with 20 random addition sequences per replicate and maxtrees set at 10 000. All clade support and tree length calculations were conducted with uninformative characters excluded.
Likelihood analyses
Modeltest (Posada and Crandall, 1998
) was used to identify the best fitting model of sequence evolution with the fewest additional parameters. Tree-bisection-reconnection branch swapping was performed under the likelihood criterion using PAUP* 4.0b10 on a starting tree generated by neighbor joining with the parameters obtained from Modeltest. After the search located the tree with the highest likelihood, model parameters were estimated for that tree. These parameters set a new round of branch swapping on the maximum likelihood (ML) tree. Branch swapping and parameter estimation were iterated until analyses converged on the same likelihood score and model parameters.
Parametric bootstrapping
Particular phylogenetic hypotheses were also evaluated by parametric bootstrapping (Huelsenbeck and Hillis, 1996
). We focused on support for the monophyly of Ficeae plus Castilleae. Parsimony searches were performed in which Ficeae plus Castilleae were constrained to be non-monophyletic. Likelihood parameters and branch lengths were estimated from one of the parsimony trees resulting from the reverse constraint search. The tree postulating the non-monophyly of Ficeae plus Castilleae and its likelihood parameters were used to simulate 100 replicate data sets using SEQ-GEN (Rambaut and Grassly, 1997
). Maximum parsimony trees were inferred for each simulated data set, and –lnL scores given the aforementioned parameters were compared between an MP tree and the tree used to simulate the data. The log-likelihood difference of the constrained (null) and unconstrained (best) topologies provided a distribution of the test statistic under the null hypothesis that systematic error accounted for the monophyly of Ficeae + Castilleae. The log likelihood difference of the empirical data yielded the probability of obtaining this clade in error.
Character evolution
The following four reproductive traits were scored for each taxon based on information from the taxonomic literature, herbarium specimens, and observations in the field. Pollination was scored as (0) anemophilous or (1) entomophilous. Taxa with "urticaceous" stamens were scored for wind pollination because no species with this morphology are known to be insect pollinated. Carpellate inflorescences were scored as (0) lacking an involucre of bracts around the carpels or (1) with an involucre of bracts surrounding carpels. Breeding system was scored as (0) monoecious, (1) dioecious, (2) androdioecious, or (3) gynodioecious. Monoecious, androdioecious, and gynodioecious plants were further scored for (0) unisexual inflorescences or (1) bisexual inflorescences. Ancestral state reconstruction was performed on the maximum likelihood tree using MacClade version 3.0 (Maddison and Maddison, 2000
).
Correlated change in pollination syndrome, breeding system, and the presence of an involucre of bracts around the carpels on a randomly resolved ML tree was examined using the concentrated changes test (Maddison, 1990
). Correlated evolution was also examined in a likelihood framework using Discrete (Pagel, 1994
, 1997
). Omnibus tests evaluated the hypothesis of correlated evolution in two characters by examining the fit of an independent model of character evolution to a dependent model. Likelihood tests were conducted on a pruned data set of only species for which pollination syndrome is known (63 of 102 taxa) because pollination syndromes are unknown for many species. Branch lengths were calculated using likelihood parameters estimated on the ML tree.
Molecular dating
A likelihood ratio (LR) test compared the likelihood of the data with and without the assumption of a molecular clock (Felsenstein, 1988
), where LR = 2 (lnLclock – lnLno clock) was assumed to be 
2 distributed with the degrees of freedom equal to n taxa minus two. Because rates of evolution varied greatly across Moraceae, we performed dating procedures using a penalized likelihood approach (Sanderson, 2002
) as implemented in the program r8s version 1.6 (Sanderson, 2003
). Penalized likelihood is a semiparametric method that allows substitution rates to vary among lineages according to a smoothing parameter. The optimal smoothing parameter was chosen by means of a data-driven cross validation procedure in which taxa were sequentially pruned from the tree. The r8s program calculated the parameter estimate that best predicted the removed data (i.e., minimized the 2 error). Cross-validation and initial dating were performed with the age of the root node fixed at one. Nodes in the resulting ultrametric tree were constrained with the minimum ages of Moraceae fossils including Ficus achenes aged at 60 mya, Morus leaves dated at 40 mya, and Moraceae fruits dated at 90 mya (Collinson, 1989
). The root node of the tree was also set to a maximum age of 135 mya based on the oldest known angiosperms fossil (Magallon et al., 1999
).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
Modeltest identified a model with equal substitution rates for transitions and unequal rates for transversions and a parameter for heterogeneity in rates of substitution across sites (TVM + 
; Posada and Crandall, 1998
) as the most appropriate model of sequence evolution with the fewest additional parameters. After three iterations, the analysis converged on the same likelihood tree and parameter estimates. Branch swapping yielded two equally likely trees with a score of –lnL 17 928.474. The rate matrix for the tree shown in Fig. 3 was 1.69381, 2.02679, 0.39974, 1.17943, 2.06549, with 
= 0.688340. Base frequencies were A = 0.303690, C = 0.142979, G = 0.159156, and T = 0.394175. The topology of the two trees differed only with respect to the position of the Bleekrodea madagascariense plus Streblus elongatus clade.
|
|
An involucre of bracts surrounding the developing carpels also occurs in Castilleae and Ficus (Fig. 4). The involucre encircles the developing flowers in Castilleae, and the bracts are pushed apart as the receptacle expands. By contrast, the involucre in Ficus forms an ostiole that limits access to the flowers, traps most pollinators inside of the syconium, and seals the infructescence as well. Outside of the Castilleae plus Ficus clade, involucral bracts surrounding the carpels are present only in Trophis caucana (Pittier) C.C.Berg. A few Dorstenieae and Artocarpeae have small bracts associated with the receptacle but these do not enclose the flowers at any stage of development.
The evolution of involucral bracts was not significantly concentrated in insect-pollinated lineages (P = 0.50), despite the perfect correlation of entomophily and bracts in the Castilleae plus Ficeae clade. The concentrated changes test is sensitive to limited numbers of evolutionary events, in this case, a single origin of involucral bracts. By contrast, the correlation between bracts and insect pollination was significant according to the omnibus test (LR = 17.0; df = 4; P < 0.05).
The concentrated changes test indicated that changes to monoecy were not significantly concentrated in wind-pollinated lineages (P = 0.11), when breeding system was considered to be dependent on pollination syndrome. Likewise, changes to wind pollination were not significantly concentrated in monoecious lineages (P = 1.00). The omnibus test indicated an overall association between breeding system and pollination syndrome (LR = 14.66; df = 4; P < 0.05). However, models restricting the probability of change in pollination syndrome to be dependent on the state of the breeding system (LR = 0.4; df = 1), and vice versa (LR = 0.6; df = 1), were not significantly more likely than the independent model.
Molecular dating
Even with a parameter-rich model of nucleotide substitution, the data strongly rejected the assumption of a molecular clock (LR = 709.42; df = 100; P < 0.001). Therefore, a penalized likelihood analysis assuming nonclock-like evolution was performed. Cross-validation analysis estimated the optimal smoothing parameter at 1.12. Based on a maximum age constraint of 135 mya for the root of Urticalean rosids and minimum age constraints for Ficus (60 mya), Morus (40 mya), and Moraceae (90 mya), the minimum divergence for the stem lineage of Ficeae plus Castilleae was 83 mya. These estimates indicated that Moraceae diverged at least 99.4 mya while the Cecropiaceae plus Urticaceae diverged at least 98 mya, which is consistent with fossil Urticaceae fruits dated at 90 mya (Collinson, 1989
).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Machado et al., 2001
). Previous studies based on limited sampling of Moraceae suggested that some members of Castilleae are close relatives of the figs (Herre et al., 1996
; Sytsma et al., 2002
). We demonstrate for the first time that the entire Castilleae plus some Artocarpeae are sister to Ficus, and we propose taxonomic changes in accord with these findings.
Taxonomic implications
We present a new classification of Moraceae to reflect the evolutionary relationships of the tribes (Appendix; see Supplemental Data accompanying the online version of this article). Minor revisions in the placement of particular genera result in four monophyletic tribes. Moreae remain paraphyletic in our classification pending more detailed sampling of the group, which encompasses extreme morphological heterogeneity and is recognized on the basis of plesiomorphic features of Urticalean rosids, including inflexed stamens and wind pollination (Sytsma et al., 2002
). Our results point to a Moreae sensu stricto (Figs. 3–4) but further study of Moreae genera is warranted. Trophis and Streblus are especially problematic and have been divided into many monotypic or oligotypic genera by various authors. These genera display tremendous variation in growth form, breeding system, and inflorescence morphology. Berg (1988)
broadly circumscribed these genera based on the presence of free tepals in Streblus vs. connate tepals in Trophis. Streblus is further divided into five sections, two of which are monotypic. We sampled three representatives of sect. Paratrophis and found that S. pendulinus and S. glaber are sister, while the position of S. smithii remains uncertain. In addition, only sect. Sloetia was represented in our analysis by S. elongatus, which is sister to Bleekrodea. Even with limited sampling, our results suggest the disintegration of the genus.
Trophis is divided into six sections, four of which are monotypic. Trophis racemosa (sect. Trophis) and T. involucrata (sect. Echinocarpa) are a part of the Moreae sensu stricto but T. scandens (sect. Malaisia) is more closely related to Broussonetia and Dorstenieae. As with Streblus, the non-monophyly of this genus is not surprising given its morphological complexity, and further investigation is warranted.
Artocarpeae as circumscribed by Rohwer (1993)
is polyphyletic, and we recognize a more restricted Artocarpeae by excluding five genera from the tribe. Antiaropsis, Poulsenia, and Sparattosyce are transferred to Castilleae, and Bagassa and Sorocea are transferred to Moreae sensu stricto (Appendix). This delimitation is strongly supported by ndhF (Fig. 2) and nuclear ribosomal DNA (G. D. Weiblen, unpublished data) and is not contradicted by morphology (Fig. 4). All genera in the revised Artocarpeae have a reduction in stamen number (four stamens) and have large seeds that lack endosperm. Although Hullettia and Treculia were not sampled, we include these genera in the revised Artocarpeae based on morphology.
There is moderate support for the Dorstenieae based on ndhF sequences although the position of Trilepisium remains uncertain (Fig. 2). African Scyphosyce and Bosqueiopsis were unavailable, and the inclusion of these genera might influence relationships in the tribe. We expect, however, that they are closely allied with Dorstenieae based on morphology. Many closely allied species of Moreae and Dorstenieae are monoecious with bisexual inflorescences. Dorstenieae are distinguished by the presence of one to several carpellate flowers embedded in the receptacle and surrounded by staminate flowers.
Sequence data support the inclusion of Sparattosyce, Antiaropsis, and Poulsenia in a broadened Castilleae, characterized by self-pruning branches (Berg, 1977a
) and unisexual inflorescences with involucral bracts. Interestingly, these three genera were the only members of the Artocarpeae sensu Rohwer (1993)
with involucrate inflorescences. Berg (1990)
hypothesized that both involucral and interfloral bracts represent adaptations preventing insects from feeding on the developing carpels and stamens. The early development of the involucre prior to floral differentiation is consistent with a protective role, although flowers are not completely enclosed at receptivity or anthesis in many Castilleae. These bracts are free in most species but are fused along most of their length in Sparattosyce, resembling an urn-shaped receptacle.
Breeding system evolution
Contrary to the traditional view that dioecy evolves from monoecy (Renner and Ricklefs, 1995
; Weiblen et al., 2000
), our results suggest that dioecy is the ancestral condition in Moraceae and that monoecy has evolved independently between two and four times in the family. However, if gains of dioecy are assigned twice the cost of losses, then monoecy is reconstructed as the ancestral condition under parsimony. Monoecious lineages are neatly divided into those with unisexual inflorescences (Artocarpeae and Castilleae) and those with bisexual inflorescences (Dorstenieae and Ficeae). Based on these differences in inflorescence architecture, monoecy may have evolved four times in the family (Fig. 4). Functional dioecy in Ficus, resulting from the interaction of gynodioecious inflorescences with pollinating seed eaters, appears to have evolved at least twice with several reversals to monoecy (Weiblen, 2000
; Jousselin et al., 2003
). Androdioecy, a very rare breeding system in angiosperms (Renner and Ricklefs, 1995
), has been documented in Castilla (Castilleae) and Helianthostylis (Dorstenieae), indicating two origins within Moraceae. Androdioecy and gynodioecy might be more common in the family but incomplete collections make this assessment difficult.
Pollination syndrome evolution
An apparently plesiomorphic condition in Moreae is the presence of stamens that are inflexed in bud, generally associated with a pistillode. As the filament elongates, the anthers are oriented along the pistillode in bud such that the stamens spring back explosively and release pollen into the air as the flower opens. This adaptation to wind pollination is common in the Urticaceae. Some members of Moreae have straight filaments in bud, including Bagassa, Sorocea, and Maclura sect. Cudrania, suggesting one or two losses of "urticaceous stamens" in the Moreae sensu lato. Bagassa has pendant, staminate inflorescences that are often associated with anemophily, but Sorocea has no obvious features of either wind- or insect-pollination. Pluricellular trichomes on the carpellate inflorescences exude a nutritious substrate that can serve as a medium for fungal mycelium. This exudate and the associated fungus may serve as a reward for pollinators (Berg, 2001
).
Many species of Dorstenieae are hypothesized to be insect pollinated based on floral structure and scents, and beetles have been observed visiting some African species (Berg and Hijman, 1999
). However, seed set in the absence of pollinators has been recorded in Dorstenia (Berg and Hijman, 1999
). In Trilepisium, staminate inflorescences with a strong odor are known to attract beetles that may breed in the inflorescences (Berg, 1977b
). Among the neotropical Dorstenieae, Brosimum alicastrum has been reported to produce clouds of pollen from staminate inflorescences suggestive of wind pollination (Berg, 2001
). However, visitation by "small diverse insects" has also been reported (Bawa et al., 1985
; Kress and Beach, 1994
), and other authors suggest that inflorescences of Brosimum are adapted for insect pollination (Croat, 1978
; Berg, 1990
).
In Artocarpus integer, pollination appears to be mediated by gall midges feeding on fungal parasites of staminate inflorescences (Sakai et al., 2000
). Gall midges feed on an exudate from the mycelium of the fungus and breed in the mycelium. Insects may be attracted to the inflorescences by a strong, sweet smell that is emitted at night (Sakai et al., 2000
). In the process of feeding and breeding on staminate inflorescences, insects pick up sticky pollen grains and occasionally transport them to carpellate inflorescences. Gall midges have been observed at low frequency on carpellate inflorescences, possibly attracted by the floral fragrance that is emitted. However, the fungus on which midges feed is not found on carpellate inflorescences, and short visits by gall midges to these inflorescences suggest pollination by deceit. Fungal growth may be limited by the structure of the Artocarpus inflorescence. The perianths of adjacent flowers fuse to form an exterior surface through which the stigmas protrude through a small opening at the apex. This structure may limit entry of phytophagous insects to the delicate tissue of the gynoecium (Berg, 1990
). We hypothesize that, in contrast to the paleotropical Artocarpeae, the pendulous, catkin-like staminate inflorescences of the neotropical genera are adapted to wind pollination. On the other hand, some authors allude to wind pollination in Artocarpus and nothing is known of syndromes in closely allied Parartocarpus and Prainea (Jarrett, 1959
; Berg, 2001
). Field observations are needed to test these predictions based on phylogeny.
Berg (1990)
hypothesized that the involucral bracts of Castilleae may be an adaptation to prevent phytophagous insects from feeding on the flowers. Pollination syndrome is unknown for about half of the genera of Castilleae, but thrips pollination has been reported in Castilla, Antiaropsis, Naucleopsis, Perebea, and Poulsenia (Sakai, 2001
; G. D. Weiblen, unpublished data). The same species of thrips, Frankliniella diversa (Thripidae; Thysanoptera), has been observed visiting Castilla and Poulsenia (Sakai, 2001
). Thrips pollination appears to be associated with insects breeding in staminate inflorescences. Pollen serves as the primary food source for both nymphs and adults. Females also visit carpellate inflorescences, but for shorter periods of time. Similar to the situation in Artocarpus, thrips may be attracted to carpellate flowers by a floral scent that mimics the staminate inflorescences (Sakai, 2001
). Further investigation of the generality of thrips pollination in the Castilleae is needed, but our results indicate that traits associated with pollination in the Castilleae are critical to understanding the evolution of fig pollination.
Origin of the fig
The sister relationship of thrips-pollinated Castilleae and Ficus suggests that insect pollination was a feature shared by the common ancestor of these lineages. Maximum likelihood analysis of correlated evolution further detected a significant association between entomophily and the presence of involucral bracts that encircle the floral primordia in Ficus and Castilleae (Berg, 1972
; Verkerke, 1989
). During fig development, the receptacle expands to accommodate the flowers, while the tightly appressed bracts form the ostiole. The involucral bracts of many Castilleae encircle the carpels and form a ring through which the receptive stigmas protrude. Even partial enclosure of the carpels could provide protection from phytophagous insects and play a role in pollinator specificity (Berg, 1990
). The staminate inflorescences of Castilleae provide breeding sites for pollen-feeding thrips, which are deceived into visiting the carpellate inflorescences by means of floral scent (Sakai, 2001
). In contrast, ostiolar bracts that completely seal the flowers (and the fruit) inside the receptacle serve to trap fig pollinators. We interpret the precondition of the syconium as a partially closed, involucrate inflorescence similar to that of the modern Castilleae. The complete enclosure of the fig by the ostiole was likely associated with pollination by specialized, parasitic Hymenoptera and led to the origin of the obligate mutualism. At present, we can only speculate as to how Agaonids became Moraceae specialists. Cospeciation between specialized pollinators and their hosts might account for the extreme species diversity of Ficus relative to its sister group (Weiblen and Bush, 2002
), and the notion that pollinator specialization was the engine of speciation in Ficus can now be explored through the phylogenetic comparison of diversification rates (Sanderson and Donoghue, 1996
).
Molecular dating using penalized likelihood suggests that Ficus diverged from Castilleae at least 83 mya, close to the date obtained from fig wasp DNA sequences, suggesting that the mutualism originated 87.5 mya (Machado et al., 2001
). However, divergence time estimates from fig wasps pointed to the crown group radiation of Ficus, while our date refers to the origin of the stem lineage. It is noteworthy that Machado et al. (2001)
based their divergence time estimates on an analysis in which taxa with nonclock-like substitution rates were discarded from a group that demonstrates a great deal of rate heterogeneity. Furthermore, the analysis was calibrated with a fossil fig wasp dated at 18 mya. Dates based on clock-like substitution rates are most accurate near the calibration point and tend to overestimate the timing of ancient events (Arbogast et al., 2002
). Although Ficus appears to have a Cretaceous origin, the crown radiation of Ficus may have occurred more recently than suggested by Machado et al. (2001)
. Molecular biogeographic studies that integrate knowledge of figs, pollinators, and their relatives are the next logical step in this line of investigation.
Conclusions
The origin of the fig has not figured in the extensive literature on coevolutionary complexity of the fig/ pollinator interaction (reviewed in Cook and Rasplus, 2003
; Weiblen, 2002
). Our revised classification of the Moraceae based on ndhF sequences provides a phylogenetic framework for understanding the evolution of fig pollination. The sister relationship of Castilleae to Ficus provides molecular corroboration of the view that insect pollination and involucral bracts encircling the flowers in a Cretaceous ancestor were likely associated with the origin of the remarkable pollination mutualism. However, other aspects of the fig origin remain a mystery. How did the extreme protogyny evolve that established the synchrony of fig and pollinator life cycles? When and where did the Agaonidae become fig specialists? It is hoped that developmental studies and molecular biogeography may provide further insights into these questions.
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Bawa K. S. S. H. Bullock D. R. Perry R. E. Colville M. H. Grayum 1985 Reproductive biology of tropical lowland rain forest trees. II. Pollination systems. American Journal of Botany 72: 346-356[CrossRef][Web of Science]
Berg C. C. 1972 Olmedieae Brosimeae (Moraceae). Hafner, New York, New York, USA
Berg C. C. 1977a The Castilleae, a tribe of the Moraceae, renamed and redefined due to the exclusion of the type genus Olmedia from the Olmedieae. Acta Botanica Neerlandica 26: 73-82[Web of Science]
Berg C. C. 1977b Revisions of African Moraceae (excluding Dorstenia, Ficus, Musanga, and Myrianthus). Bulletin du Jardin botanique de l'Etat, Bruxelles 47: 267-407
Berg C. C. 1982 The reinstatement of the genus Milicia Sim (Moraceae). Bulletin du Jardin botanique de l'Etat, Bruxelles 52: 225-229
Berg C. C. 1986 The delimitation and subdivision of the genus Maclura (Moraceae). Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, Amsterdam, the Netherlands, series C 89: 241-247
Berg C. C. 1988 The genera Trophis and Streblus (Moraceae) remodeled. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, Amsterdam, the Netherlands, series C 91: 345-362
Berg C. C. 1989 Systematics and phylogeny of the Urticales. In P. R. Crane and S. Blackmore [eds.], Evolution, systematics and fossil history of the Hamamelidae, vol. 2, 193–220. Clarendon, Oxford, UK
Berg C. C. 1990 Differentiation of flowers and inflorescences of Urticales in relation to their protection against breeding insects and to pollination. Sommerfeltia 11: 13-24
Berg C. C. 2001 Moreae, Artocarpeae, and Dorstenia (Moraceae) with introductions to the family and Ficus and with additions and corrections to Flora Neotropica Monograph 7. New York Botanical Garden Press, New York, New York, USA
Berg C. C. M. E. E. Hijman 1999 The genus Dorstenia (Moraceae). Ilicifolia 2: 1-211
Bremer K. 1988 The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42: 795-803[CrossRef][Web of Science]
Bremer K. 1994 Branch support and tree stability. Cladistics 10: 295-304[CrossRef][Web of Science]
Collinson M. E. 1989 The fossil history of the Moraceae, Urticaceae (including Cecropiaceae), and Cannabaceae. In P. R. Crane and S. Blackmore [eds.], Evolution, systematics and fossil history of the Hamamelidae, vol. 2, 319–339. Clarendon, Oxford, UK
Cook J. M. J. Y. Rasplus 2003 Mutualists with attitude: coevolving fig wasps and figs. Trends in Ecology and Evolution 18: 241-248[CrossRef]
Corner E. J. H. 1962 The classification of Moraceae. Gardens Bulletin Singapore 19: 187-252
Corner E. J. H. 1975 The evolution of Streblus Lour. (Moraceae): with a new species of sect. Bleekrodea. Phytomorphology 25: 1-12
Corner E. J. H. 1978 Ficus dammaropsis B1. and the pedunculate species of Ficus subgen. Sycocarpus. Philosophical Transactions of the Royal Society of London, series B 281: 373-406[CrossRef]
Croat T. B. 1978 Flora of Barro Colorado Island. Stanford University Press, Stanford, California, USA
Felsenstein J. 1988 Phylogenies from molecular sequences: inference and reliability. Annual Review of Genetics 22: 521-565[CrossRef][Web of Science][Medline]
Herre E. A. C. A. Machado E. Bermingham J. D. Nason D. M. Windsor S. S. McCafferty W. Van Houten K. Bachmann 1996 Molecular phylogenies of figs and their pollinating wasps. Journal of Biogeography 23: 521-530[CrossRef][Web of Science]
Huelsenbeck J. P. D. M. Hillis 1996 Parametric bootstrapping in molecular phylogenetics: applications and performance. In J. D. Ferraris and S. L. Palumbi [eds.], Molecular zoology: advances, strategies, and protocols, 19–45. Wiley-Liss, New York, New York, USA
Jarrett F. M. 1959 Studies in Artocarpus and allied genera, I. General considerations. Journal of the Arnold Arboretum 40: 1-37
Jousselin E. J. Y. Rasplus F. Kjellberg 2003 Convergence and coevolution in a mutualism: evidence from a molecular phylogeny of Ficus. Evolution 57: 1255-1269[CrossRef][Web of Science][Medline]
Kress W. J. J. H. Beach 1994 Flowering plant reproductive systems. In L. A. McDade, K. S. Bawa, H. A. Hespenheide, and G. S. Hartshorn [eds.], La selva: ecology and natural history of a neotropical rain forest. University of Chicago Press, Chicago, Illinois, USA
Machado C. A. E. Jousselin F. Kjellberg S. G. Compton E. A. Herre 2001 Phylogenetic relationships, historical biogeography and character evolution of fig—pollinating wasps. Proceedings of the Royal Society of London, series B 268: 685-694[Medline]
Maddison W. P. 1990 A method for testing the correlated evolution of two binary characters: are gains or losses concentrated on certain branches of a phylogenetic tree?. Evolution 44: 539-557[CrossRef][Web of Science]
Maddison W. P. D. R. Maddison 2000 MacClade: analysis of phylogeny and character evolution, version 4. Sinauer, Sunderland, Massachusetts, USA
Magallon S. P. R. Crane P. S. Herendeen 1999 Phylogenetic pattern, diversity, and diversification of eudicots. Annals of the Missouri Botanical Garden 86: 297-372[CrossRef][Web of Science]
Olmstead R. G. C. W. dePamphilis A. D. Wolfe N. D. Young W. J. Elisens P. A. Reeves 2001 Disintegration of the Scrophulariaceae. American Journal of Botany 88: 348-361
Olmstead R. G. J. A. Sweere 1994 Combining data in phylogenetic systematics: an empirical approach using three molecular data sets in the Solanaceae. Systematic Biology 43: 467-481[CrossRef][Web of Science]
Pagel M. 1994 Detecting correlated evolution on phylogenies: a general method for the comparative analysis of discrete characters. Proceedings of the Royal Society of London, series B 255: 37-45
Pagel M. 1997 Inferring evolutionary processes from phylogenies. Zoologica Scripta 26: 331-348[CrossRef][Web of Science]
Posada D. K. A. Crandall 1998 Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817-818
Rambaut A. N. C. Grassly 1997 Seq-Gen: an application for the Monte Carlo simulation of DNA sequence evolution along phylogenetic trees. Computer Applications in Bioscience 13: 235-238
Renner S. S. R. E. Ricklefs 1995 Dioecy and its correlates in the flowering plants. American Journal of Botany 82: 596-606[CrossRef][Web of Science]
Rohwer J. G. 1993 Moraceae. In K. Kubitzki, J. G. Rohwer, and V. Bittrich [eds.], The families and genera of vascular plants, 438–453. Springer- Verlag, Berlin, Germany
Sakai S. 2001 Thrips pollination of androdioecious Castilla elastica (Moraceae) in a seasonal tropical forest. American Journal of Botany 88: 1527-1534
Sakai S. M. Kato H. Nagamasu 2000 Artocarpus (Moraceae)-gall midge pollination mutualism mediated by a male-flower parasitic fungus. American Journal of Botany 87: 440-445
Sanderson M. J. 2002 Estimating absolute rates of molecular evolution and divergence times: a penalized likelihood approach. Molecular Biology and Evolution 19: 101-109
Sanderson M. J. 2003 r8s: inferring absolute rates of molecular evolution and divergence times in the absence of a molecular clock. Bioinformatics 19: 301-302
Sanderson M. J. M. J. Donoghue 1996 Reconstructing shifts in diversification rates on phylogenetic trees. Trends in Ecology and Evolution 11: 15-20
Sorenson M. D. 1999 TreeRot, version 2. Boston University, Boston, Massachusetts, USA
Swofford D. L. 2002 PAUP*: phylogenetic analysis using parsimony (*and other methods), version 4. Sinauer Associates, Sunderland, Massachusetts, USA
Sytsma K. J. J. Morawetz J. C. Pires M. Nepokroeff E. Conti M. Zjhra J. C. Hall M. W. Chase 2002 Urticalean rosids: circumscription, rosid ancestry, and phylogenetics based on rbcL, trnL-F, and ndhF sequences. American Journal of Botany 89: 1531-1546
Thompson J. D. T. J. Gibson F. Plewniak F. Jeanmougin D. G. Higgins 1997 The CLUSTAL-X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 25: 4876-4882
Verkerke W. 1989 Structure and function of the fig. Experientia 45: 612-622[CrossRef][Web of Science]
Weiblen G. D. 2000 Phylogenetic relationships of functionally dioecious Ficus (Moraceae) based on ribosomal DNA sequences and morphology. American Journal of Botany 87: 1342-1357
Weiblen G. D. 2002 How to be a fig wasp. Annual Reviews of Entomology 47: 299-330[CrossRef][Web of Science][Medline]
Weiblen G. D. G. L. Bush 2002 Speciation in fig pollinators and parasites. Molecular Ecology 11: 1573-1578[CrossRef][Medline]
Weiblen G. D. R. K. Oyama M. J. Donoghue 2000 Phylogenetic analysis of dioecy in Monocotyledons. American Naturalist 155: 46-58[CrossRef][Medline]
Wiegrefe S. J. K. J. Sytsma R. P. Guries 1998 The Ulmaceae, one family or two? Evidence from chloroplast DNA restriction site mapping. Plant Systematics and Evolution 210: 247-279
Zavada M. S. M. Kim 1996 Phylogenetic analysis of Ulmaceae. Plant Systematics and Evolution 200: 13-20[CrossRef][Web of Science]
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  M. D. Marks, L. Tian, J. P. Wenger, S. N. Omburo, W. Soto-Fuentes, J. He, D. R. Gang, G. D. Weiblen, and R. A. Dixon Identification of candidate genes affecting {Delta}9-tetrahydrocannabinol biosynthesis in Cannabis sativa J. Exp. Bot., September 1, 2009; 60(13): 3715 - 3726. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Duangjai, B. Wallnofer, R. Samuel, J. Munzinger, and M. W. Chase Generic delimitation and relationships in Ebenaceae sensu lato: evidence from six plastid DNA regions Am. J. Botany, December 1, 2006; 93(12): 1808 - 1827. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K Monro The revision of species-rich genera: a phylogenetic framework for the strategic revision of Pilea (Urticaceae) based on cpDNA, nrDNA, and morphology Am. J. Botany, March 1, 2006; 93(3): 426 - 441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Ronsted, G. D Weiblen, J. M Cook, N. Salamin, C. A Machado, and V. Savolainen 60 million years of co-divergence in the fig-wasp symbiosis Proc R Soc B, December 22, 2005; 272(1581): 2593 - 2599. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Machado, N. Robbins, M. T. P. Gilbert, and E. A. Herre Critical review of host specificity and its coevolutionary implications in the fig/fig-wasp mutualism PNAS, May 3, 2005; 102(suppl_1): 6558 - 6565. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
