New home for tiny aquatics

A shake-up of current thinking about the evolution of the angiosperms — the flowering plants — is a consequence of the relocation of a hitherto obscure branch on the angiosperm evolutionary tree.

Many new constellations in the angiosperm evolutionary tree have come to light as relationships within the flowering plants have been explored using molecular analyses1,2. Some of these changes were not unexpected. But others have resulted in fundamental reinterpretations of angiosperm evolution3,4.

On page 312 of this issue, Saarela et al.5 report one of the most striking realignments required so far. They find that the diminutive, moss-like, aquatic plants Hydatella and Trithuria, which are members of the family Hydatellaceae and were thought to be monocots close to grasses, are actually the closest living relatives of water-lilies and their allies (Fig. 1). This new-found position for the Hydatellaceae, near the point at which water-lilies (order Nymphaeales) diverged from other flowering plants, could scarcely be more attention-grabbing. On the evidence of Saarela and colleagues' analyses5, and given other generally accepted relationships at the base of the angiosperm evolutionary tree2,3, only a single angiosperm species, Amborella trichopoda, diverged from other flowering plants below this point.

Figure 1: A new branch near the root of the angiosperm evolutionary tree.

Amborella remains the sister group to all other flowering plants, but the Hydatellaceae join the Nymphaeales (water-lilies) as the next diverging subsidiary branch of the tree. This new alignment5 raises questions about the origin of classic bisexual flowers in the Nymphaeales and other angiosperms. The unisexual flowers of Hydatellaceae — with male and female flowers consisting of a single organ only (stamen and carpel, respectively), aggregated in dense unisexual or bisexual inflorescences — contrast with the solitary, bisexual flowers of the Cabombaceae, represented here by Cabomba, which were previously regarded as the basic condition in the Nymphaeales. More generally, simple and often unisexual flowers occur among several early-diverging angiosperm lineages, for example in the Trimeniaceae (Piptocalyx, Trimenia) and Chloranthaceae (Sarcandra, Hedyosmum), as well as early-diverging eudicots (Buxus). But whether this simplicity reflects the basic phylogenetic condition or ecological adaptation is an open question. Flowers of Amborella are also small, and functionally unisexual, but female flowers have staminodes, indicating a basic bisexual organization. In this situation, interpretations of character evolution depend on the position of the root of the phylogenetic tree, and may also be sensitive to the addition of new taxonomic groups. Infl, inflorescence (flower cluster). (Drawings by P. von Knorring.)

But is this new discovery simply a minor matter of phylogenetic tidying up? After all, the Hydatellaceae are a small family; their position has long been uncertain6; and the unexpected is bound to crop up when relationships in a group the size of flowering plants (some 350,000–400,000 extant species) are looked at more carefully.

At one level, as Saarela et al.5 point out, repositioning of the Hydatellaceae conforms to, rather than overthrows, current ideas of relationships among the extant representatives of early angiosperm lineages. These ideas have come to look increasingly secure, as more genes and more plants have been incorporated into molecular phylogenetic analyses. The Hydatellaceae associate with the Nymphaeales based on the new molecular data, and the relationship seems to be well corroborated. Many of the morphological characteristics of the Hydatellaceae, at least in so far as they are known, also make more sense in the context of the Nymphaeales than in the context of grasses and their relatives. And it is convenient that the Hydatellaceae, as a family of aquatic plants, link to the Nymphaeales rather than to the other, mainly woody, plants that make up the root of the angiosperm evolutionary tree. In this position they do not further complicate ideas about the evolution of life in an aquatic habitat.

For other traits, however, repositioning the Hydatellaceae raises questions that add to an already long list of unresolved issues in early angiosperm evolution, particularly with regard to features that are perhaps too easily interpreted as 'reduced', 'lost' or 'absent'. At some point it may become more straightforward to infer that some early angiosperms never had certain features, rather than had them and then lost them.

For example, does the seemingly simple floral morphology of the Hydatellaceae reflect 'reduction', or might it represent an early 'prefloral' stage of angiosperm evolution in which the classic bisexual flower, with its whorls of three or more different organs, was not yet fully formed? Flowers of the Hydatellaceae — along with those of other early-diverging angiosperm lineages (for example, fossil and living Hedyosmum7 and other members of the Chloranthaceae) — could hardly be more simple (Fig. 1). It is possible, in the case of the Hydatellaceae, as well as Ceratophyllum and perhaps the fossil Archaefructus8,9, that the simple flowers reflect loss of floral parts associated with life in a submerged habitat, as has happened in other aquatic angiosperm lineages. But here, near the base of the angiosperm phylogenetic tree, it is difficult to be sure whether floral simplicity signals ecological adaptation or the basic phylogenetic condition. If it is the latter, it would have profound implications for ideas on the early evolution of the classic angiosperm flower.

It will take some time to digest all the implications of suddenly introducing a new plant family into discussions of early angiosperm evolution. Hydatella and Trithuria (Fig. 2) have not been on the radar screen of most specialists working on the subject, and there are many pieces of this puzzle still to work out. Certainly, repositioning the Hydatellaceae as the closest living relatives to the Nymphaeales will dramatically influence ideas about the early evolution of water-lilies and their allies, and it will modify important details of character evolution at the base of the angiosperm tree. Whether it will also affect the rooting of the angiosperm tree as a whole, perhaps in ways that would displace Amborella to a less prominent position, remains to be seen. Identification of angiosperm precursors in the fossil record would probably have an even greater impact.

Figure 2


Trithuria — not previously on the radar screen of most specialists.

It is also relevant that several species of Hydatellaceae have been described only in the past 25 years. It seems likely that more will come to light, and there may also be revelations among the several hundred plant species that are described as new to science every year. Examples such as the Hydatellaceae, and the Australian conifer genus Wollemia10 discovered a little over a decade ago as a 'living fossil', illustrate the inadequacy of our knowledge of plant diversity at a time when so much is being destroyed or is under serious threat.

Hydatella and Trithuria will inevitably be the subject of detailed investigation in the coming years. But whatever the outcomes of these studies, the radical realignment discovered by Saarela et al.5 should remind us not to become too comfortable with the current picture of early angiosperm relationships, and especially with the details of character evolution that they imply. There will be more surprises as new plants are added to the mix. They will come not just from our gradually improving knowledge of living plants, but more especially from our exploration of the riches of the plant fossil record — both for early angiosperms and for their elusive relatives among other seed plants.


  1. 1

    Chase, M. W. et al. Ann. Missouri Bot. Gard. 80, 528–580 (1993).

  2. 2

    Soltis, D. E., Soltis, P. S., Endress, P. K. & Chase, M. W. Phylogeny and Evolution of Angiosperms (Sinauer, Sunderland, MA, 2005).

  3. 3

    Qiu, Y.-L. et al. Nature 402, 404–407 (1999).

  4. 4

    Endress, P. K. & Igersheim, A. Int. J. Plant Sci. 161, S211–S223 (2000).

  5. 5

    Saarela, J. M. et al. Nature 446, 312–315 (2007).

  6. 6

    Dahlgren, R. M. T., Clifford, H. T. & Yeo, P. F. The Families of the Monocotyledons: Structure, Evolution, and Taxonomy (Springer, Berlin, 1985).

  7. 7

    Friis, E. M., Pedersen, K. R. & Crane, P. R. Palaeogeogr. Palaeoclimatol. Palaeoecol. 232, 251–293 (2006).

  8. 8

    Sun, G. et al. Science 296, 899–904 (2002).

  9. 9

    Friis, E. M., Doyle, J. A., Endress, P. K. & Leng, Q. Trends Plant Sci. 8, 369–373 (2003).

  10. 10

    Jones, W. G. et al. Telopea 6, 173–176 (1995).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Friis, E., Crane, P. New home for tiny aquatics. Nature 446, 269–270 (2007).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.