Reference Review: The Monocot Tree

Dracaena draco, from Flora Canaria

Davis, J. I., D. W. Stevenson, G. Petersen, O. Seberg, L. M. Campbell, J. V. Freudenstein, D. H. Goldman, C. R. Hardy, F. A. Michelangeli, M. P. Simmons, C. D. Specht, F. Vergara-Silva & M. Gandolfo. 2004. A phylogeny of the monocots, as inferred from rbcL and atpA sequence variation, and a comparison of methods for calculating jackknife and bootstrap values. Systematic Botany 29 (3): 467-510.

Abstract: "A phylogenetic analysis of the monocots was conducted on the basis of nucleotide sequence variation in two genes (atpA, encoded in the mitochondrial genome, and rbcL, encoded in the plastid genome). The taxon sample of 218 angiosperm terminals included 177 monocots and 41 dicots. Among the major results of the analysis are the resolution of a clade comprising four magnoliid lineages (Canellales, Piperales, Magnoliales, and Laurales) as sister of the monocots, with the deepest branch within the monocots between a clade consisting of Araceae, Tofieldiaceae, Acorus, and Alismatales, and a clade that includes all other monocots. Nartheciaceae are placed as the sister of Pandanales, and Corsiaceae as the sister of Liliales. The Triuridaceae, represented by three genera, including Lacandonia, are resolved as monophyletic and placed in a range of positions, generally within Pandanales. Dasypogonaceae and Arecaceae diverge sequentially from a clade that includes all other commelinid taxa, and within the latter group Poales s. lat. are sister of a clade in which Zingiberales and Commelinales are sisters. Within Poales s. lat., Trithuria (Hydatellaceae) and Mayaca appear to be closely related to some or all elements of Xyridaceae. A comparison was conducted of jackknife and bootstrap values, as computed using strict-consensus (SC) and frequency-within-replicates (FWR) approaches. Jackknife values tend to be higher than bootstrap values, and for each of these methods support values obtained with the FWR approach tend to exceed those obtained with the SC approach."

A pleasing degree of consensus has developed in recent years in regards to many higher-level relationships among angiosperms (though one can't help wondering how much of this consensus is related to the fact that higher-level plant systematics has become an almost entirely molecular affair, and what would happen were a few sophisticated morphological datasets thrown into the mix. Anywho...) While relationships within the basal "magnoliids" are still a bit iffy, we have become reasonably certain about what are the major clades or grades within the flowering plants.

Many of you will have learnt in school about the differences between dicotyledons and monocotyledons, the supposed two major divisions of flowering plants, and I am sure that most basic biology textbooks still use this division. Technically, this is wrong. The characters that are ascribed to dicotyledons (two seed leaves, netted veins, stem with central xylem and external phloem, etc.) are actually the ancestral characters for flowering plants, and the dicots are paraphyletic with regard to the monocots (though a well-supported clade, the eudicots, does include all the traditional dicots except for the "magnoliids"). At least two "dicot" groups, the Nymphaeales and Aristolochiaceae, have some characters that are more traditionally associated with monocots.

The monocots, on the other hand, are a well-supported clade, both molecularly and morphologically. The Cronquist system of angiosperm classification recognised five subclasses within the monocots, the Alismatidae (aquatic and semi-aquatic forms), Arecidae (palms [Arecales], screw pines [Pandanales] and aroids [Arales]), Liliidae (lily-type plants), Commelinidae (grasses, rushes and allies) and Zingiberidae (gingers and bromeliads). Of these, the Arecidae and Zingiberidae are both currently regarded as polyphyletic, though Cronquist's "zingiberids" are all included in the commelinids*. It is the other three that I'll mainly be referring to here.

*Though the current Angiosperm Phylogeny Group classifications do still recognise higher clades based on the older subclasses, they prefer to use informal names such as "commelinid" or "rosid" rather than recognising them as formal subclasses such as Commelinidae or Rosidae.


Davis et al. (2004) came up with results for their analyses of monocot phylogeny that were fairly similar to what had and has been found elsewhere - basal alismatids (including Arales), then the liliids (including Pandanales) as a grade, then the monophyletic commelinids (including Arecales). Interestingly, Davis et al. in their combined analysis found alismatids as monophyletic - most other analyses have found them to be paraphyletic, with Acorus as the sister group to all other monocots, then a clade of most alismatids. Acorus (shown above in a photo from here) is a Holarctic semi-aquatic plant previously included in the Arales. When Davis et al. ran their analysis using rbcL data only, Acorus returned to its position outside the remaining monocots.



One other result of Davis et al. that definitely caught my eye is that Trithuria (Hydatellaceae) appears snugly nestled within the commelinids (albeit via a long-string of poorly-supported branches). One of the most interesting developments in plant systematics of the past year has been the publication of Saarela et al. (2007), which convincingly showed that the Hydatellaceae, a group of incredibly insignificant aquatic plants previously regarded as monocots (the picture above from Science Daily shows well how minute they are), are in fact part of the ANITA grade at the base of the angiosperms, sister to the Nymphaeales (waterlilies). While Saarela et al. didn't include as many taxa in their analysis as Davis et al., they did use considerably more loci (23 as opposed to only two) and showed that the relationship was also supported (albeit weakly) by morphological data, so I'm inlined to believe Saarela et al. over the earlier analysis. The really big question, then, is how did Davis et al. manage to get Hydatellaceae so deep within the monocots in the first place? What drew the previously used rbcL sequences towards the monocots? Were the sequences even Hydatellaceae at all, or had some sort of mix-up occurred? Regrettably, Saarela et al. gave no suggestions for how the previous analyses could have been so wrong**, and I'm completely stumped.

**Perhaps yet another case of the constrictions of the Nature format dooming us to a far-too-brief article. Humph.

The two papers also differ on the sister group of monocots. Davis et al. joined the monocots to the "core magnoliid" group, while Saarela et al. favoured the eudicots + Ceratophyllum***. While bootstrap support was better for Saarela et al., in neither case was it that great, so the question remains up in the air.

***Ceratophyllum is yet another unassuming aquatic taxon that seemingly exists only to give plant systematists splitting headaches****. For a while it seemed a strong contender for the position of basalmost living flowering plant, but was eventually booted from this position of power by the ANITA group. Since then, it has exacted its revenge for such a rude demotion by refusing to sit in the same place twice between analyses.

****If you're wondering what exactly is so evil about aquatic plants, their shift from a terrestrial to an aquatic habitat is generally associated with an accelerated evolutionary rate, giving them significant long branches. They also tend to develop simplified morphologies relative to their terrestrial relatives.

One thing I find particularly interesting about the state of monocot phylogenetics is the basal position of alismatids, especially if Acorus renders them paraphyletic (if you were wondering, Saarela et al. recovered the Acorus-as-basalmost-monocot topology). This makes me wonder whether monocots as a whole are derived from a semi-aquatic ancestor, which would make the "liliid" + commelinid clade secondarily terrestrial. Could this partially explain how monocots developed their highly derived morphology relative to other angiosperms? Interestingly, the aquatic Nymphaeales + Hydatellaceae clade, though indicated by molecular data to be quite distant from the monocots, shares a number of convergent features with monocots such as loss of the cambium.

One final point of interest about Davis et al. (2004) is their comparison of bootstrap supports calculated using different algorithms. While the differences were not huge, they were certainly there. For me, this definitely came in the "I did not know that" category, and will certainly have to be something I keep in mind in the future.

REFERENCES

Saarela, J. M., H. S. Rai, J. A. Doyle, P. K. Endress, S. Mathews, A. D. Marchant, B. G. Briggs & S. W. Graham. 2007. Hydatellaceae identified as a new branch near the base of the angiosperm phylogenetic tree. Nature 446: 312-315.