Bot. Bull. Acad. Sin. (2004) 45: 33-40

Chen et al. Evolution of apocarpy in Alismatidae

Evolution of apocarpy in Alismatidae using phylogenetic evidence from chloroplast rbcL gene sequence data

Jin-Ming Chen1, Dan Chen1, Wahiti Robert Gituru1.2, Qing-Feng Wang1,*, and You-Hao Guo1

1Laboratory of Plant Systematics and Evolutionary Biology, College of Life Sciences, Wuhan University, Wuhan, Hubei 430072, P.R. China

2Present address: Botany Department, Kenyatta University, P. O. Box 43844, Nairobi, Kenya

(Received November 26, 2002; Accepted July 22, 2003)

Abstract. The apocarpous groups in the monocotyledons are mainly concentrated in the subclass Alismatidae. The molecular phylogeny of Alismatidae based on analysis of chloroplast rbcL gene sequence data serves as a framework with which to evaluate character evolution with respect to the derivation of apocarpy in the group. 20 of the 27 genera in the subclass that display apocarpy have been included in our study. Our analysis indicates that apocarpy is polyphyletic within the subclass Alismatidae. Two independent origins of apocarpy in Alismatidae are explored in this study. Three separate origins of a single carpel and two separate origins of syncarpy in the subclass are also proposed. Basally connate carpel condition was the ancestral character in Alismatidae and evolved in two directions. It is possible for the unicarpellate condition to have been directly derived by reduction from syncarpy, and it could also be that the unicarpellate state has been derived from apocarpy by reduction in carpel number. The present results indicate that a progression has occurred in the evolution of carpels in Alismatidae from basally connate carpels through syncarpy or apocarpy to a single carpel.

Keywords: Alismatidae; Apocarpy; Basally connate carpels; RbcL gene; Single carpel; Syncarpy; Unicarpellate.


Apocarpy has been regarded as an ancestral character in the angiosperms (Bessey, 1915; Hutchinson, 1959). Doyle and Endress (2000) concluded that apocarpy is a primitive feature in angiosperms, but that in monocotyledons it represents a reversal from syncarpy. The apocarpous groups in the monocotyledons include almost all clades of the subclass Alismatidae (Cronquist, 1981) or Alismatiflorae (Stebbins, 1974; Thorne, 1976; Dahlgren and Clifford, 1982). Previous researchers hypothesized that the apocarpous groups in the Alismatidae were the most primitive monocotyledons (Hutchinson, 1959; Takhtajan, 1980; Cronquist, 1988). The origin of the monocotyledons from the Ranunculaceae was strongly advocated by Hutchinson (1973), who relied on the feature of apocarpy as evidence of their relationship. Although apocarpy is generally believed to represent the ancestral condition in Alismatidae (Cronquist, 1981), it has also been interpreted as derived by this subclass (Dahlgren and Rasmussen, 1983; Dahlgren et al., 1985).

The evolution of apocarpy is one of the most interesting evolutionary events in the history of monocotyledons. A better understanding of this evolution may provide valuable insight into the adaptive evolution in flowering plants and also assist in resolving the relationships among the apocarpous groups within the angiosperms. Several inves

tigations into the origin and evolution of carpel form and structure have been reported (Taylor and Hickey, 1996; Igersheim et al., 2001). Owing to the sampling limitations in most studies, the evolutionary relationships of apocarpy and other gynoecium conditions, such as syncarpous, basally connate, and unicarpellate conditions in monocotyledons, are still obscure. Here, we present a phylogenetic analysis of chloroplast rbcL gene sequences for 46 species from 46 genera representing all currently recognized families and 81% of the genera in the subclass Alismatidae. Delimitation of families and genera are according to Cook (1990), Cronquist (1981), and Tomlinson (1982). Les et al. (1997) constructed a phylogeny of the subclass Alismatidae, and the origins and evolution of marine angiosperms (seagrasses) and hydrophily in the subclass were also estimated; however, for our purpose only one species from each genus was selected because the condition of carpel fusion was constant in each genus. In the present analysis, 20 of the 27 genera that display apocarpy in the subclass are included. The molecular phylogeny of Alismatidae is used as a framework on which to evaluate character evolution with respect to the evolution of apocarpy in the group.

Materials and Methods

Taxa Sampling

RbcL sequences for 46 species of Alismatidae that were reported previously in Genbank were obtained. One rbcL sequence of Lemna (Lemnaceae) and five sequences of

*Corresponding author. Tel: +86-27-87682869; Fax: +86-27-87682869; E-mail:

Botanical Bulletin of Academia Sinica, Vol. 45, 2004

Araceae (Symplocarpus foetidus; Lasia spinosa; Xanthosoma sagittifolium; Orontium aquaticum and Gymnostachys anceps) reported in Genbank were used as

outgroup (Table 1). In Alismatidae not all the species are apocarpous as suggested by Dahlgren et al. (1985). Several sources were used to identify syncarpous, basally

Chen et al. Evolution of apocarpy in Alismatidae

connate, apocarpous, and unicarpellate species within the lineages depicted by the rbcL cladogram (Esau, 1977; Cronquist, 1981; Dahlgren and Clifford, 1982; Tomlinson, 1982; Dahlgren and Rasmussen, 1983; Dahlgren et al., 1985; Les and Haynes, 1995; Les and Schneider, 1995; Igersheim et al., 2001).

Tree Reconstruction

Due to difficulties in obtaining the complete sequences for the entire coding region of rbcL for all taxa, we restricted our analysis to 1178 nucleotides of sequence data (position 28-1205 of the coding region) that were complete for all taxa. Sequences were aligned using CLUSTALW (Thompson et al., 1994). Phylogenetic analysis of the complete data set of 52 sequences was conducted using the test version of PAUP 4.0b4a (Swofford, 2000), and heuristic searches were done with COLLAPSE, MULPARS and TBR branch-swapping options to save all of the equally most parsimonious trees. All characters were of the type "unordered" and had equal weight. Gaps were treated as "missing." Bootstrap analysis of 1000 replications was performed to show relative support for individual clades. The Alismatidae cladogram was rooted using taxa from Araceae and Lemnaceae, which two families had been shown previously as the appropriate outgroup (Les and Schneider, 1995; Les et al., 1997). Carpel conditions were added to the data matrix of molecular characters, and two optimization options of ACCTRAN (which favors reversals over parallelisms) and DELTRAN (which favors parallelisms over reversals) of MacClade, Version 3.06 (Maddison and Maddison, 1996) were used to study the evolution of carpel fusion.


Phylogenetic Structure of Alismatidae

Of 1178 total characters in the data matrix, 789 (67%) are constant characters; 94 (8%) are variable characters that are parsimony-uninformative; and 295 (25%) are parsimony-informative variable characters. Parsimony analysis of the data yielded 144 most parsimonious trees of 1285 steps with a consistency index (CI) of 0.426, retention index (RI) of 0.692, homoplasy index (HI) of 0.5743, and rescaled consistency index (RC) of 0.2945. The strict consensus tree of the 144 most parsimonious trees is presented in Figure 1.

The strict consensus of these trees indicates two major lineages in Alismatidae: One contains five families arranged in two subclades consisting of (1) Alismataceae and Limnocharitaceae and (2) Butomaceae, Hydrocharitaceae, and Najadaceae. The bootstrap value of this lineage is 73%. The other lineage includes ten families, in which: (1) Aponogetonaceae and Scheuchzeriaceae are basal; (2) one clade constitutes the families Lilaeaceae and Juncaginaceae; (3) one clade includes Cymodoceaceae, Posidoniaceae, and Ruppiaceae, and (4) another clade is comprised of Zosteraceae, Potamogetonaceae, and Zannichelliaceae. The bootstrap value of this lineage is

74%. The monophyly of Alismatidae is strongly supported with a bootstrap value of 99%. The strict consensus tree also indicates some unresolved polychotomies: (1) Alismataceae and Limnocharitaceae, (2) Lilaeaceae and Juncaginaceae, (3) Potamogetonaceae and Zannichelliaceae, and (4) Cymodoceaceae, Posidoniaceae, and Ruppiaceae (Figure 1).

Evolution of Carpel Fusion

Both ACCTRAN and DELTRAN gave similar results for all trees with respect to the evolution of apocarpy in the ingroup. Mapping the species with carpel fusion conditions onto the rbcL phylogeny indicates that in the first lineage of Alismatidae apocarpy has one independent origin in the clade Alismataceae and Limnocharitaceae with one reversal to basally connate in Damasonium (Alismataceae). In the second lineage the apocarpy has evolved independently at least once in the clade Potamogetonaceae, Zannichelliaceae, Zosteraceae, Ruppiaceae, Cymodoceaceae, and in Posidoniaceae. The syncarpy in Lilaea (Lilaceaceae) and Triglochin (Juncaginaceae) has possibly evolved once from apocarpy (Cycnogeton) while the syncarpy in Hydrocharitaceae also has an independent origin (Figure 2).

The rbcL phylogeny also indicates that the unicarpellate condition in Alismatidae has evolved independently in each of the three families Najadaceae, Zosteraceae, and Posidoniaceae. The basally connate condition is ancestral in Alismatidae. In addition, the unicarpellate condition in the outgroup may have had an independent origin, and the syncarpous condition has probably evolved twice: once in Xanthosoma and Lasia, and once in Symplocarpus (Figure 2).


Molecular Phylogeny of Alismatidae

Preliminary summaries of interfamilial phylogenetic relationships in Alismatidae have been discussed by Les et al. (1993), Les and Haynes (1995), Les et al. (1997), and the references therein. The analysis given here presents another estimation of relationships based on the majority of the genera in the subclass. The molecular phylogeny of Alismatidae served as a framework on which to evaluate the evolution of characters. Compared to the rbcL gene analysis presented earlier by Les et al. (1997), the results are overall similar in showing the presence of the same two major clades. However they differ in details such as the topology of the families in the unresolved clade (Cymodoceaceae, Posidoniaceae, and Ruppiaceae). The low bootstrap value of this clade in the earlier study and the unresolved polychotomies in the present study indicated that the rbcL gene is not capable of resolving the position of Cymodoceaceae, Posidoniaceae and Ruppiaceae and that additional gene sequences should be included in further studies. However, the analysis of the evolution of carpel fusion based on the phylogeny of Alismatidae was little affected by this problem.

Botanical Bulletin of Academia Sinica, Vol. 45, 2004

Figure 1. Maximum parsimony rbcL cladogram (strict consensus tree shown) of species from the 15 families typically recognized in the monocotyledon subclass Alismatidae with representatives from two outgroup families. The degree of internal support is indicated by bootstrap values. Values for nodes with <50% bootstrap support are excluded.

Chen et al. Evolution of apocarpy in Alismatidae

Figure 2. Example reconstruction of carpel fusion evolution for selected most-parsimonious phylogenetic trees based on the rbcL gene sequences analysis. The rbcL phylogeny indicates that apocarpy has evolved in two separate clades as indicated by the black branches. However, there is a reversal back to the basally connate carpels in Damasonium (Alismataceae).

Origins of Apocarpy in Alismatidae

Apocarpy as an ancestral character in angiosperms has never been seriously disputed (e.g., Hutchinson, 1973; Igersheim et al., 2001), but a question of degree still remains. Dahlgren and Rasmussen (1983) and Dahlgren et al. (1985) hypothesized that it must be a derived feature in monocotyledons, and Doyle and Endress (2000) also suggested that apocarpy in monocots represented a reversal. Mapping apocarpy on the rbcL cladogram indicates that apocarpy in Alismatidae has two independent origins (Figure 2). Apocarpy is likely to have evolved once in the apocarpous clade of Alismataceae and Limnocharitaceae (bootstrap value of 100%), except for Damasonium, the carpels of which are basally connate and probably represent a reversal. Another origin of apocarpy occurs in the clade

of Potamogetonaceae, Zannichelliaceae, Zosteraceae, Cymodoceaceae, Posidoniaceae, Ruppiaceae, Lilaeaceae, and Juncaginaceae with the high bootstrap value of 86%. However, apocarpy evolved in three clades: (1) in the clade of Lilaeaceae and Juncaginaceae; (1) in the clade of Potamogetonaceae, Zannichelliaceae and Zosteraceae; and (1) in the clade of Cymodoceaceae, Ruppiaceae, and Posidoniaceae. Basal carpel connation was the ancestral character in Alismatidae, as indicated by the rbcL cladogram, and basally connate carpels evolved in two clades: (1) in Butomaceae and (2) in the clade of Aponogetonaceae and Scheuchzeriaceae. Apocarpy might have evolved from basally connate carpels. The present study not only supports the results of Dahlgren and Rasmussen (1983), Dahlgren et al. (1985), Doyle and

Botanical Bulletin of Academia Sinica, Vol. 45, 2004

Endress (2000) and Igersheim et al. (2001), it also elaborates on the origins of apocarpy in monocotyledons, especially in Alismatidae.

Origins of a Single Carpel in Alismatidae

The relationship of Najadaceae (unicarpellate) and Hydrocharitaceae (syncarpous) has now been well established by cladistic analysis using anatomical, morphological, and molecular data (Miki, 1937; Shaffer-Fehre, 1991a, 1991b; Les et al., 1993; Les and Haynes, 1995; Tanaka et al., 1997). We support the merger of the families as suggested by Shaffer-Fehre (1991a, 1991b). It is possible for a single carpel to have been derived directly by reduction from syncarpy. It is also possible that a single carpel has been derived by reduction from apocarpy (e.g., in Zosteraceae and Posidoniaceae). Three origins of a single carpel in Najadaceae, Zosteraceae, and Posidoniaceae are indicated in the cladogram of Alismatidae. The unicarpellate condition in the outgroup has a separate origin.

Origins of Syncarpy in Alismatidae

Butomaceae (carpels are basally connate) are basal to the clade of Najadaceae and Hydrocharitaceae in the rbcL cladogram. The presence of syncarpy in Hydrocharitaceae is likely derived from basally connate carpels (Butomaceae). However, syncarpy in Lilaea and Triglochin may have an independent origin from apocarpy (Cycnogeton) in the clade of Lilaeaceae and Juncaginaceae (Figure 2). Syncarpy in the outgroup might have evolved independently twice from a single carpel in Symplocarpus and the clade of Xanthosoma and Lasia.

In conclusion, molecular phylogenetic studies of the subclass Alismatidae have provided insights into the evolution of carpel fusion. Several previous molecular phylogenetic analyses suggested that syncarpy was a derived feature in basal angiosperms and that it was probably polyphyletic (Doyle and Endress, 2000). Combining our results with those of earlier studies we conclude that apocarpy is an ancestral feature in angiosperms but that in monocotyledons it is a reversal (as Doyle and Endress, 2000), with possibly two independent origins of the apocarpy in Alismatidae. The present study does not support the hypothesis that apocarpy is ancestral in monocots, or that apocarpous Alismatidae are the most primitive monocots (Hutchinson, 1959; Takhtajan, 1980; Cronquist, 1988). Instead, the rbcL cladogram indicates that basally connate carpels are ancestral in Alismatidae. There has been a progression from basally connate carpels through syncarpy or apocarpy to a single carpel in the evolution of carpels in Alismatidae.

Several possible evolutionary advantages of syncarpy have been presented; (1) fused carpels mean that a single pollen delivery event can serve all carpels of a flower; (2) syncarpy permits an even distribution of pollen tubes among carpels; (3) the fusion of carpels means that pollen selection can occur in a single structure (Endress, 1982). However, the functional significance of the multiple ori

gins of apocarpy in the subclass Alismatidae is still unknown, further detailed investigation may furnish us with more exact data to explore its functional significance.

Acknowledgements. We are grateful to Spencer C.H. Barrett, Huang Shuang-Quan, Fu Cheng-Xin, Chen Zhi-Duan, Yang Chun-Feng and Kong Hang-Hui for their helpful comments on an earlier draft. This study was funded by the National Natural Science Foundation of China (Grant No. 30070055).

Literature Cited

Bessey, C.E. 1915. The phylogenetic taxonomy of flowering plants. Ann. Missouri Bot. Gard. 2: 109-164.

Cook, C.D.K. 1990. Aquatic Plant Book. SPB Academic Publishing, Netherland.

Cronquist, A. 1981. An Integrated System of Classification of Flowering Plants. Columbia University Press, New York.

Cronquist, A. 1988. The Evolution and Classification of Flowering Plants. 2nd edn. New York Botanical Garden, Bronx, New York.

Dahlgren, R.M.T. and H.T. Clifford. 1982. The Monocotyledons: A Comparative Study. Academic Press, New York.

Dahlgren, R.M.T, H.T. Clifford, and P.F. Yeo. 1985. The Families of the Monocotyledons. Springer-Verlag, Berlin.

Dahlgren, R.M.T. and F. Rasmussen. 1983. Monocotyledon evolution: characters and phylogenetic estimation. Evol. Biol. 16: 255-395.

Doyle, J.A. and P.K. Endress. 2000. Morphological phylogenetic analysis of basal angiosperms: comparison and combination with molecular data. Int. J. Plant Sci. 161(6 Suppl): S121-S153.

Endress, P.K. 1982. Syncarpy and alternative modes of escaping disadvantages of apocarpy in primitive angiosperms. Taxon 31: 48-52.

Esau, K. 1977. Anatomy of Seed Plants. 2nd edn. John Wiley, New York.

Gituru, W.R., Q.F. Wang, Y. Wang, and Y.H. Guo. 2002. Pollination ecology, breeding system, and conservation of Caldesia grandis (Alismataceae), an endangered marsh plant in China. Bot. Bull. Acad. Sin. 43: 231-240.

Hutchinson, J. 1959. The Families of Flowering Plants. 2nd edn. Clarendon Press, Oxford.

Hutchinson, J. 1973. The Families of Flowering Plants. 3nd edn. Clarendon Press, Oxford.

Igersheim, A., M. Buzgo, and P.K. Endress. 2001. Gynoecium diversity and systematics in basal monocots. Bot. J. Linn. Soc. 136: 1-65.

Les, D.H., M.A. Cleland, and M. Waycott. 1997. Phylogenetic studies in Alismatidae, II: evolution of marine angiosperms (seagrasses) and hydrophily. Syst. Bot. 22: 443-463.

Les, D.H., D.K. Garvin, and C.F. Wimpee. 1993. Phylogenetic studies in the monocot subclass Alismatidae: evidence for a reappraisal of the aquatic order Najadales. Molec. Phylogenet. Evol. 2: 304-314.

Les, D.H. and R.R. Haynes. 1995. Systematics of subclass Alismatidae: a synthesis of approaches. In P.J. Rudall, P.J. Cribb, D.F. Cutler, and C.J. Humphries (eds.),

Chen et al. Evolution of apocarpy in Alismatidae

Monocotyledons: Systematics and Evolution. Royal Botanic Gardens, Kew, pp. 353-377.

Les, D.H. and E.L. Schneider. 1995. Aquatic origin of monocotyledons. In P.J. Rudall, P.J. Cribb, D.F. Cutler, and C.J. Humphries (eds.), Monocotyledons: Systematics and Evolution. Royal Botanic Gardens, Kew, pp. 23-42.

Maddison, W.P. and D.R. Maddison. 1996. MacClade: analysis of phylogeny and character evolution. Version 3.06. Sinauer Associates, Sunderland, Massachusetts.

Miki, S. 1937. The origin of Najas and Potamogeton. Bot. Mag. (Tokyo) 51: 472-480.

Shaffer-Fehre, M. 1991a. The endotegmen tuberculae: an account of little known structure from the seed coat of the Hydrocharitoideae (Hydrocharitaceae) and Najas (Najadaceae). Bot. J. Linn. Soc. 107: 169-188.

Shaffer-Fehre, M. 1991b. The position of Najas within the subclass Alismatidae in the light of new evidence from seed coat structure in the Hydrocharitoideae (Hydrocharitaceae). Bot. J. Linn. Soc. 107: 189-209.

Stebbins, G.L. 1974. Flowering Plants: Evolution Above the Species Level. Harvard University Press, Cambridge,


Swofford, D.L. 2000. PAUP*. Phylogenetic Analysis Using Parsimony (*and other methods). Version 4. Sinauer Associates, Sunderland, Massachusetts.

Takhtajan, A, 1980. Outline of the classification of flowering plants (Magnoliophyta). Bot. Rev. 46: 225-359.

Tanaka, N., H. Setoguchi, and J. Murata. 1997. Phylogeny of the family Hydrocharitaceae inferred from rbcL and matk gene sequence data. J. Plant Res. 110: 329-337.

Taylor, D.W. and L.J. Hickey. 1996. Flowering Plant Origin, Evolution and Phylogeny. Chapman and Hall, New York.

Thompson, J.D., D.G. Higgins, and T.J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position specific gap penalties and weight matrix choice. Nucleic Acids Res. 22: 4673-4680.

Thorne, R.F. 1976. A phylogenetic classification of the Angiospermae. Evol. Biol. 9: 35-106.

Tomlinson, P.B. 1982. Anatomy of the Monocotyledons VII. Helobiae (Alismatidae). Clarendon Press, Oxford.

Botanical Bulletin of Academia Sinica, Vol. 45, 2004