What's so primitive about Acorus?

So first of all what is Acorus, and why would anyone think it primitive?

Acorus is a marginal aquatic with sword-shaped leaves
typical of monocots, and flowers densely packed onto
a thick spike resembling the spadix of an aroid.  Photo
by H. Zell, posted in Wikimedia commons.

Acorus, commonly known as Sweet Flag, is a widespread marginal aquatic plant native to North America and Asia.  A marginal aquatic is essentially one in which the roots are in soil that is waterlogged most of the time. So they occupy shallow water at the edges of streams, lakes and swamps, as well as marshes and intermittently wet prairies. The leaves of Acorus are flattened into a fan-like arrangement at right angles to the rhizome, similar to many members of the Iris family.  It's flowers, however, are small and crowded onto a dense spike resembling the spadix of an aroid (e.g. Calla lily or Anthurium).  It was in fact classified as an aroid until recently, generally as its most archaic member, but now has been placed in its own family and order.

In this simplified phylogenetic tree of the monocots,
Acorus is the first clade to branch off.  It is the
 sister clade to the common ancestor of all the
 remaining monocots.  Alismatales includes
mostly aquatic plants, but also the mostly
terrestrial aroids (Araceae).  The Commelinoids
 include palms, gingers, and grasses (among others),
while the Lilioides include lilies, irises and orchids.

This new placement is based on recent phylogenetic studies which suggest that Acorus is the sole representative of the most ancient lineage of monocots.  In technical parlance, we say that the single genus Acorus, in its own family and order, is sister group to all the rest of the monocots.   In other words, the ancestor of Acorus split from from the common ancestor of all other monocots 120 million years ago or more.  So we look at it and we say, "so this is what the first monocots looked like!"

Well, probably not.  The truth is that we know very little about what the first members of the Acorus clade looked like, or what kind of habitat they lived in, let alone earlier monocots.  We have no fossils of any of them. The modern genus Acorus is the end points of over 100 million years of evolution, and the lineage certainly has changed in some way over that time, possibly with earlier genera now extinct.

As an analogy, Homo sapiens is the sole surviving species of the family Hominidae, but does that mean that the first members of the family looked or behaved just like us?  We happen to have a pretty good fossil record of our own history, and know that we were preceded by Australopithecus and even more ancient forms.  These early hominids walked upright, but other than that looked a lot more like chimpanzees that like us, and we are separated from them by a mere 5-6 million years.

What we can say is that the common ancestor of the Acorus clade, as it split from the main monocot line, already had the basic features of all monocots:  leaves with parallel veins that grow from the base, a single seedling leaf ("cotyledon") in the embryo, a clonal herbaceous growth form lacking in secondary growth ("wood"), and an entirely adventitious root system (no taproot or other permanent root system).  (See How the Grass Leaf Got its Stripes, January 26, 2012).  Monocots mostly also have flower parts in 3s - 3 sepals, 3 petals, 6 stamens, and 3 carpels. This also appears to have been established by the time the Acorales split off the family tree.

We know even less about the "stem monocots," early monocots that preceded the origin of the Acorus clade and bridged the transition from dicotyledonous plants.   Phylogenetic studies indicate that the monocots originated in or near the Magnolid clade. This clade includes diverse kinds of plants, from the lofty Magnolia trees to the herbaceous or semi-herbaceous members of the Aristolochiaceae and Piperaceae, but no aquatics similar to Acorus.  So there was a kind of "evolutionary wormhole" into which went plants with broad, dicotyledonous leaves and embryos with two cotyledons, and out of which emerged true monocots with sword-shaped leaves and a single cotyledon.

So in what ways might the modern Acorus species be like stem monocots?  And in what ways are they likely more specialized?  Were the first Acoroids  marginal aquatics, like the modern species?  They likely were, since the next clade to branch off the monocot  tree, the Alismatales, consists largely of  aquatics as well. This vast assemblage ranges from marginal wetland plants like Sagittaria to fully-submerged sea grasses.  It includes also the aroids, which range from marginal aquatics to rain forest epiphytes.

The sword-shaped monocot leaf, which pushes upward from a basal growth zone (basal intercalary meristem) appears, however, to have been shaped by a strongly seasonal environment subject to grazing and fires.  It is a leaf adapted for rapid regeneration from an underground stem system.  If the leaf tips of young grass leaves are burnt or bitten off, for example, the basal growth zone can reactivate and restore the blade.

The  unifacial leaves of Iris, Acorus, and many
other monocots are "folded" with the two
halves  in the  lower part pressed together to
forming a flat, narrow sheath.  The upper part
is solid, as if the two sides were glued
together.  Each new leaf emerges through
the sheaths of adjacent leaves.

Also, as monocot seeds germinate, the single cotyledon usually remains within the seed.  It lengthens at its base like monocot leaves in genera, but this serves to push the rest of the embryo deeper into the soil, rather than to push itself into the light.  This suggested to the great evolutionary botanist G.L. Stebbins (1974) that the first monocots were adapted to environments that were alternately very wet and very dry, like a modern savanna.  Seeds germinated when the soil was wet, with the embryo being "planted" deeply in the mud.  The embryo was thus protected from desiccation when the soil later dried out.  So most likely the early monocots evolved within a range of savanna to marshy habitats, where a great many monocots, from grasses to cat-tails, live today.

In what other ways might Acorus either resemble or differ from early monocots?  Most likely the flowers were larger, fewer in number, and more loosely arranged than they are in the present dense spike.   The trend from loose clusters of flowers to dense spikes, as far as we can see, is a one-way trip.  The spike is an efficient reproductive structure, a "super-flower" if you like, that becomes more and more specialized.  It often becomes surrounded by other structures, like the spathe of the aroids, that not only protect the flowers, but also serve to attract pollinators or manipulate their movement over the stamens and stigmas.  It is hard to imagine any selective pressure that would cause that trend to reverse.  Families that contain small flowers in dense spikes include the Araceae, some Arecaceae (palms), Cyclanthaceace, Pandanaceae, Piperaceae, Saururaceae, and to a smaller degree many other families.  In none of these has the trend ever reversed.

Also, we might expect that the carpels (the chambers in which the seeds develop) of the ancestral monocots were separate from one another (apocarpous).  Most ancient angiosperms, including most Magnolids, and most Alismatales (other than aroids), have separate carpels. The three carpels of Acorus, however, are solidly fused together and share a common stigma, the way it is in more advanced angiosperms in general.  Once carpels are fused together, they rarely revert back to separate entities, except sometimes to produce specialized fruit types like the schizocarps of the Apiaceae (carrot family) or the follicles of Asclepiads (milkweeds).  So in this way modern Acorus is more advanced than the Magnolids that preceded it and the Alismatales that followed it.

Tofieldia, in the Alismatales, grows in moist meadows. Its
carpels are only loosely joined together, retain separate
stigmas, and separate as they mature into seed capsules
 (follicles).

What about the fan-shaped clusters of leaves?  The leaves of Acorus are unifacial, which means that the leaf blades appear as if they were folded and fused together, so that both sides are really the back side (abaxial) of a normal leaf.  The apparent fusion of the two sides of the leaf occurs just above the opening of the narrow sheath through which newer leaves emerge.  Leaves like this are common in the iris family, in Butomus and Tofieldia in the Alismatales, and scattered among other monocot families.  Butomus and Tofieldia incidentally both have carpels more separate from one another than in Acorus, and so in that sense are more archaic.

The flowers of Butomus, in the Alismatales, are larger, in
looser clusters, and have separate carpels, and so
are probably more like the ancient monocot flowers than
are those of Acorus.

It is possible that the common ancestor of Acorus and other living monocots had unifacial leaves.  More ordinary open leaves, as found in grasses, cat-tails, daylilies, etc., could have evolved from them by increasing the growth of the open sheath region and reducing the growth of the unifacial blade.  Or the reverse could have occurred, as it evidently has happened multiple times in different families, perhaps as an adaptation to the marginal aquatic environment.  The stiff fan-shaped cluster of leaves may be more sturdy in the face of moving water or flooding.

 Monocot leaf development is highly flexible, as growth emphasis can easily be shifted from one part of the leaf to another, resulting in the vast array of monocot leaves from grass blades to palm fronds, paddle-shaped banana leaves, and heart-shaped yam (Dioscorea) leaves.  This perhaps more than anything, is the key to the great diversity of monocots today.

A very young monocot leaf resembles the hood
of a sweatshirt.   The tissues surrounding the opening
represent the young leaf sheath, with the next younger leaf and
apical meristem (AM) visible within.
The tissues around the opening will form the sheath of
the mature leaf.  A zone of dividing growing cells
(meristem) that forms around the base of zone A will
 form a  cylindrical leaf sheath.  If a meristem  forms
 in zone B, but not zone C, an open leaf sheath flattening
into a long, grass-like blade will develop.  But if growth
 is largely focused at the base of the solid tip (C),
 a long unifacial blade will form.  

A mixture of primitive and specialized features is common is to be expected in surviving members of ancient lineages, the specialized features no doubt the reason why these plants are still around!  The dense flower spike and fused carpels of Acorus, and possibly the unifacial leaf blade, are likely specialized features that have allowed it to not only survive, but still flourish today.



Reference:  Stebbins, G. L.  1974.  Flowering Plants, Evolution above the Species Level.  Belknap Press of Harvard University.  Cambridge, MA.