Were the first carpels plicate or ascidiate?

[This article extends the discussion begun in my post "What's so primitive about Amborella?"]

The carpel is the distinctive seed chamber of the angiosperms, or flowering plants.  It is in fact the definitive feature of this major group.  When the first carpel evolved, the first angiosperm came into existence.  The carpel encloses, protects, and facilitates the fertilization of the ovules, which then mature as seeds. The carpel then becomes the fruit, and participates in the dispersal of the seeds.  In the flower, carpels occupy the center and are surrounded by stamens and tepals (petals and sepals).  In most modern angiosperms the carpels are joined together into a compound pistil (see "Were the first monocots syncarpous?)"

In those flowers in which the carpels remain separate, there are two fundamental shapes: plicate - resembling leaves that have folded with the opposite edges sealed together, and ascidiate - shaped like a vase or an urn.

The ascidiate carpels of Amborella
contain a single seed, and have a
large, folded stigma. A. the carpel
at the time of pollination.  B. the
 mature fruit, which is a drupe.
Drawing from Bailey and Swamy,
 1948. 

It is now clear that the three clades of the ANITA grade (Amborella, Nymphaeales and Austrobaileyales) are the most ancient branches of the flowering plants.  We assume that whatever characteristics they have in common were inherited from their common ancestor.   The carpels in this group are mostly ascidiate, which is variously described as vase-, urn-, or bottle-shaped.  The wall of the ascidiate carpel is smooth and seamless, tightly enclosing the contents but open at the top in the stigmatic region.  It resembles a sock pulled up around a foot.

Above the opening, which is blocked only by a drop of fluid, the stigma is typically prolonged along what can be described as the backbone of the carpel,  and ovules are attached in a line on the opposite side (i.e. in what might be called the "belly"), within the urn-shaped base.  Sometimes, as in Amborella, the stigma  is folded at the backbone, forming narrow flaps along each side.  During the growth of the carpel, tissues at the base (“a meristematic cross-zone between the primordium margins” – Endress & Doyle 2009) push the wall upward around the ovules.

After pollination the carpels of
Austrobaileya spread apart as they
swell with the developing seeds
within. Photo courtesy Dennis
Stevenson.

The carpels of Austrobaileya are ascidiate, but
contain two rows of ovules opposite the
backbone.  The stigmas are pushed together
to form a common head for receiving pollen.
Photo courtesy Dennis Stevenson.

Ascidiate carpels contrast with carpels described as plicate, or folded.  A folded carpel, like a pea pod has the structure of a folded leaf.  Opposite the midrib or backbone of the carpel, the  margins of the hypothetical leaf blade are figuratively sewn together in a suture.  The suture is actually more like a zipper, as it forms as alternating cells from the two margins  expand and interlock with one another.  Many plicate carpels dry out as they mature, and split open along the suture to release their seeds, though just about any kind of fruit can develop from them. Plicate carpels are common among the higher branches of angiosperms: the magnolids, monocots, and eudicots.

Simple plicate carlpels, called follicles,
 can be seen most readily in the
 Ranunculaceae,  in genera like 
Aquilegia (columbine), Delphinium, 
Eranthis, and others. Drawing from 
Asa Gray's Botanical Textbook, 1879.
.

The pea pod is somewhat more specialized
than a follicle, as it has adapted to split
along both the suture and the backbone
 when it opens to release its seeds. The seeds
alternate along the two margins.  Drawing 
from Thomé 1877, Textbook of Structural and
Physiological Botany.

Follicles of the genus Eranthis
(Ranunculaceae) are quite leaf-like.

Because the earliest branching angiosperms have primarily ascidiate carpels, the simplest (most parsimonious) interpretation about the earliest carpels is that they were ascidiate as well (Endress & Doyle 2009, and others).  The corollary of this interpretation is that  the plicate carpel of the magnolids, eudicots and monocots must have evolved from one of these ancient ascidiate carpels.

As I have argued in earlier posts, however, these theoretical conclusions based on probability need to be tesed in the adaptive arena. In other worids, do they make sense in terms of "adaptive parsimony?" (see "Were the first monocots syncarpous?" for an explanation of this term)

In ancient seed ferns, such as this
Sphenopteris, seeds were
borne directly on large, frond-
like leaves. Drawing from Brown,
1935, The Plant Kingdom.

Ovules were originally borne directly on the leaves of ancient seed ferns, and on modified seed-leaves (megasporophylls) in later gymnosperms such as the cycads. According to traditional theory, the first carpel came about as a simple blade-like megasporophyll, with ovules along both edges, rolled or folded together, enclosing the ovules within a protective chamber.  The structure would have been very similar to a pea pod or a follicle.

An interesting alternate idea is the "mostly male hypothesis" (see Frohlich and Chase 2007)  in which early blade-like stamens became carpels by the genetic accident of ovules popping up where pollen sacs should have been.  Such things do happen, and are reminders that leaves and seed-leaves were originally one and the same.

The third possibility raised by the current phylogentic conclusions is that, instead of simply folding around the ovules, the first carpels formed by an ascidiate growth pattern, i.e. the base of the ancient ovule-bearing structure, or a leaf below it, formed a cup-like base that grew up around a group of ovules (or conceivably a single ovule as in Amborella, but I have already argued agains that in "What's so primitive about Amborella?").  At the same time, the backside of this cup-like structure would develop as a strong backbone, resembling the mid-rib of a leaf, the open top  developed a folded structure, and the ovules would come to be placed in two rows opposite the backbone. In terms of genetic and developmental processes, this seems to be a much more complex scenario than simply folding a leaf together. If this is indeed what happened, we need fossil evidence and/or genetic-developmental evidence to confirm it.

If there were a selective pressure for enclosing ancient ovules, the principle of evolution along the lines of least resistance (Stebbins 1974) would clearly favor the easier path of a folding leaf.  (see "G. L. Stebbins and the process of adaptive modification" for a full and detailed explanation of this evolutionary principle).

Though not directly ancestral to the angiosperms, the seed-leaves of the living gymnosperm genus Cycas illustrate the kinds of structures that might have folded together to form the first carpels.  Drawing from Asa Gray, 1879.

So is there a clash between cladistic parsimony and adaptive parsimony with respect the first carpels?  Yes and no.  First the "no" part.  We must remember that the cladistic studies that place ascidiate carpels at the base of the angiosperm tree were based on living angiosperms -  the crown group, and so have no bearing on what might have been happening in the stem group (the extinct angiosperms that preceded the common ancestor of all living angiosperms - see "the birthplace of the angiosperms").  There were probably carpels among extinct angiosperm ancestors long before the crown group ancestor evolved.  Therefore, there is no conflict between the idea that the first carpels were folded and the idea that the common ancestor of the living angiosperms had ascidiate carpels.  We are free to choose the folded leaf model for the first carpels.

Ascidiate carpels most likely evolved among early crown group angiosperms, and presumably evolved for a reason. Most ascidiate carpels, at least those in Amborella and most Austrobaileyales, mature as drupes or berries.  These brightly colored fleshy fruits may have been adaptations for improved dispersal by birds in shady forest environments, where these archaic plants survive today.

A number of gymnosperms, such as this yew (Taxus) has
a fruit-like layer that grows up around each seed. Similar
features can be found widely among different
angiosperms, including magnolids, eudicots and
 monocots, and may have been present in the
earliest angiosperms.  Photo by
Didier Descouens, posted on Wikipedia.

In the the first hypothetical leaf-like carpels, the unsealed edges probably simply reopened to release the mature seeds.  Quite possibly, these seeds were covered with colorful fruit-like layers called arils.  These arils were comsumed by birds, who in the process dispersed the seeds. Many gymnosperms, including cycads, junipers, podocarps, and yews have similar adaptations. As fleshy fruits evolved among the early members of the crown group, the fruity function of the aril was genetically transferred to the wall of the carpel itself.

In this scenario, the evolution of the ascidiate growth form was an adaptation to embed the ovules more securely within a uniform, sealed wall. The lower tissues could have grown together while leaving the open folded region just at the top.  Something similar happened in the evolution of both roses and apples, where tissues of the receptacle were extended up and around the separate carpels.

Now I return to the apparent re-evolution of plicate carpels from ascidiate carpels, as predicted by cladistic analysis. It is odd that simple, leaf-like carpels with marginal rows of ovules would have evolved from the decidedly less leaf-like ascidiate structure, rather than directly from an ancient carpel of essentially the same design.  The nature of the suture in modern plicate carpel strongly suggests the joining of opposite edges, and that ovules were attached in rows along those edges.

The carpels of Illicium are folded and split open
to release seeds, but only one seed is produced
per carpel, rather than a row along each margin.
They may have evolved from an ascidiate
carpel through expansion of the folded stigmatic
region. Drawing from Kerner & Oliver, 1895. The
Natural History of Plants.

 One possible scenario is suggested by an exceptional member of the Austrobaileyales, Illicium (star anise), which does have plicate carpels. If these evolved from ascidiate carpels, it is conceivable that the ascidiate portion contracted while the folded stigmatic region expanded.  Though the edges are pressed together and partially fused, this does not seem to be the same suture structure as found in other plicate angiosperms (Robertson and Tucker, 1979).  Also, there is only one seed in each Illicium carpel.  Evolution of carpel with marginal rows of ovules would require a different path, i.e. extending the stigmatic split down between two rows of ovules in something like Austrobaileya,  followed by union of the margins into a suture.  This again is a rather cumbersome scenario with no apparent adaptive value.

A simpler adaptive scenario is that a folded carpel with marginal rows of ovules was retained in some ancient crown group angiosperm, and this evolved directly into the more advanced form of plicate carpel with sutures found in the higher angiosperms.  The folded nature of the stigmatic region in Amborella  may in fact be a remnant of the earliest folded carpels, and the two rows of ovules in Austrobaileya another remnant.  So the pieces of the earliest folded carpels are still present among the ANITA grade, and may have been still together in the ancestor of magnolids, monocots, and eudicots.


References: 

Bailey, I. W. and G. L. Swamy. 1948.  Amborella trichopoda Baill., J. Arnold Arbor. 23:245-254, plus plates.

Endress, P. &  J. Doyle. 2009.  Reconstructing the ancestral angiosperm flower and its initial specializations. Am. J. Bot. 96(1): 22-66.  

Frohlich, M. W. & M. W. Chase, 2007.  After a dozen years of progress the origin of angiosperms is still a great mystery.  Nature 450: 1184-1189.

Robertson and Tucker, 1979.  Floral ontogeny of Illicium floridanum, with emphasis on stamen and carpel development.  Amer. J. Bot. 66(6): 605-617.

Stebbins, G. L.  1974.  Flowering Plants.  Evolution above the species level.  Belknap Press of Harvard University Press.  Cambridge, MA.

Mosses of Central Florida 8. Entodon seductrix

The small ovate leaves of Entodon seductrix are
pressed against the stem when dry, giving the stems
a rope-like appearance.  The sporangia are upright
and cylindrical, with two rows of teeth around the
mouth.

[For other mosses in this series, see the Table of Contents]

Entodon seductrix (Hedw.) Müll. Hal. (Entodontaceae) is a common, straggling moss with horizontal stems covered with short leaves, and with sporophytes arising along the sides.  This species is found throughout eastern North America, as far north as Ontario, and extending westward from Florida to Texas.  Our collections are from the middle of the state northward, so it's not clear how far south it extends.  It characteristically forms lush mats at the bases of trees.

This species is easily confused with Isopterygium tenerum. Like that species, leaf cells are elongate and slightly curved, and there is no midrib (costa).   The short leaves of E. seductrix, however, press closely to the stem when dry, while those of Isopterygium are spread out.  The two species differ most obviously when sporangia are present.  In E. seductrix they are upright, nearly symmetrical, and cylindrical, in contrast with the strongly curved sporangia Isopterygium.

Another species of Entodon in our state is E. macropodus, which has slightly larger leaves that remain spread out when dry, but the sporangia are still upright and cylindrical. This species is more southern in its distribution, occurring northward to Virginia and southward into tropical America.  It is also found in Japan.

The slender leaf cells of Entodon are like tiny worms, with
thick, clear walls between them. Toward the base, they
become more rectangular, but not conspicuously
inflated.

Were the first monocots syncarpous?

In my recent post on Acorus, I suggested that one of the ways this genus is specialized is that the three carpels within each flower are joined together into a single pistil (syncarpous).  Acorus is at the end of the earliest branch of the monocot tree -  the sister clade to all other monocots. Its various characteristics are therefore of great interest when discussing the nature of the first monocots.  Carpels are mostly free of one another (apocarpous) in many members of the Alismatales, the next most ancient branch of the monocot tree after Acorus.  Because carpels are also free in the archaic angiosperms of the ANITA grade, many magnolids, and some basal Eudicots, it seems most logical that the common ancestor of Acorus and the rest of the monocots also had free carpels.  That logical assumption, however, has recently been challenged by Sokoloff et al. (2013).

Carpels in ancient angiosperms were separate structures, each of which had to receive pollen individually from a visiting insect. Butomus, on the left, is a monocot in the Alismatales that retains separate carpels.  In more advanced angiosperms, such as the tulip on the right, also a monocot, carpels are fused together, with a common stigmatic area where a single deposit of pollen can fertilize all the ovules in the ovary.  Left photo by Sten Porse, right photo by Bernd Haynold,  both posted on Wikimedia Commons

Tradition and conventional wisdom hold that the fusion of carpels has adaptive value and is a more-or-less irreversible process. This was confirmed by Armbruster et al. (2002), who explored the advantages of shifting from apocarpy to syncarpy and estimated 17-26 separate instances of this shift among angiosperms. The fusion of carpels brings stigmas together in such a way that pollen is deposited in a single central location by a visiting insect.  Pollen tubes can then pass through the common style and enter into any of the carpel chambers, which is said to increase competition among pollen tubes, but also insures more even fertilization among the ovules in the ovary as a whole. Carpel fusion also reduces material needs as only the outer walls need to be fortified for protection of the developing ovules. Monocots were not analyzed in the Armbruster study, but apparently were assumed to be fundamentally syncarpous. 

The general trend among seed plants is for gradually tighter and deeper enclosure of ovules within protective stuructures,including the tighter closure and fusion of carpels.  Because of the adaptive advantage attached to this trend, reversals back to apocarpy are considered unlikely:

ovules on open, leaf-like structures (ancient seed ferns)----->
     ovules on specialized leaf-like, cone-like, or shoot-like structures (gymnosperms) ------>
          ovules within loosely-closed, leaf-like carpels (stem angiosperms) ------->
                carpels apocarpous (early crown-group angiosperms and basal magnolids, eudicots, and monocots)                                   ------->
                     carpels syncarpous or unicarpellate, and sometimes surrounded by tissues of the receptacle                                                               ("inferior ovary") (most advanced angiosperms)

Armbruster and colleagues did detect two probable reversals from syncarpy to apocarpy among  eudicots (in the genus Crossosoma and members of the Saxifragales), and speculated that the advantage might be to extend the period of ovule fertilization so as to receive pollen from different sources. It's not clear that such an extension happens in these examples, however, as each only has a few carpels.  Likewise, in archaic monocots like Butomus (in the Alismatales), the small number of carpels are receptive at the same time and for only one day (Bhardwaj & Eckert 2001).   In Sagittaria, also in the Alismatales, the carpels are more numerous and physically spread out, but again are receptive at the same time for only one day.  If multiple visitors fail to arrive during that window, it potentially leaves carpels unfertilized. Multiple insect visitors are possible in one day, but does the potential advantage of genetically different pollen arriving in the same flower outweigh the usual advantages of syncarpy? The idea needs fleshing out with stronger selective arguments and actual examples of it working.   

From another perspective, if there is an advantage in spreading out the fertilization of ovules in either time or space, a much simpler way to effect that advantage in syncarpous flowers is to make the flowers smaller, with just one or a few ovules in each, and produce a series of them.  This has in fact happened many times, as in the Saururaceae, Piperaceae, Asteraceae, palms, aroids, etc. I am not aware of any series of carpels in an apocarpous flower functioning this way, let alone a syncarpous ovary splitting apart to do so.

 Despite the foregoing, Sokoloff et al. (2013) concluded that the earliest monocots were syncarpous and that apocarpous flowers evolved several times among them.  In some cases, according to this scenario, syncarpy re-evolved a short time later from newly-apocarpous ancestors. What is the basis for this counter-intuitive proposal?  It appears not to be based simply on the fact that the most ancient monocot lineage (Acorus) is syncarpous, but on a more extensive theoretical exercise involving cladistic analysis plus an additional step of "optimized parsimony analysis."

DNA-based cladistic analysis provides a clear, and increasingly accurate picture of the ancestral branching patterns of groups of organisms - - i.e. a phylogenetic tree.  When only DNA information is used, the tree tells us nothing about when and where new adaptive traits arose.  For this, we can "map" such physical characteristics onto the tree. For example, we can make a little mark on each branch containing only species known to be syncarpous.  If two adjacent (sister) branches have syncarpous flowers, we assume their common ancestor had the same.  It is also possible, however, that syncarpy evolved  independently on each branch from an ancestor that was apocarpous.  But this interpretation is less parsimonius, because it involves more independent evolutionary changes (syncarpy evolved twice instead of only once).  

Optimized parsimony  analysis takes into consideration a wide variety of tests, assumptions, taxa lists, character definitions, etc. for the same set of data and determines the most parsimonious interpretation of the evolution of particular characteristics.  Some of the tests  performed by Sokoloff et al. were ambiguous about the nature of the ancestors, but overall they suggested that syncarpous flowers were present first in the monocots.  Quite possibly, the basal position of syncarpous Acorus tilted the final results in this favor, raising again the questions I raised in my previous post.

So from this cladistic perspective, multiple evolution of apocarpous flowers from a syncarpous ancestor is more parsimonious than multiple origins of syncarpy from an apocarpous ancestor.  But does evolution necessarily follow the most parsimonious path?  In the real world, perhaps one evolutionary trend, because of its adaptive value, is more likely to occur multiple times than the opposite.  So to evaluate this proposal further, we need to consider the adaptive basis for each trend.   Sokoloff and colleagues in fact raised the question: “Assessing the functional and adaptive significance of evolutionary transformations is clearly important" ( p. 75). They proceed to reiterate the advantages of syncarpy, but made no suggestions as to why reversals might occur. They made another disturbing statement:  "Interestingly, in the monocot order Alismatales, congenital intercarpellary fusion was first lost and then re-appeared in three independent clades according to this scenario" (p. 64). One of those reversals would include the Butomus pictured above.

I recently outlined the principles advocated by Stebbins for evaluating alternate evolutionary scenarios (G. L.Stebbins and the process of adaptive modification).  These principles result in the "other parsimony," the parsimony in which sequences of adaptive changes are assumed to proceed along the simplest paths, or "along the lines of least resistance."  My example above, in which shifting to a series of small flowers, instead of splitting the ovary to make a series of separate carpels, is an example of such a simpler path.  Further, once a developmentally complex and highly functional structure like syncarpy evolves, one would need a rather powerful selective pressure to undo it, something giving an advantage to apocarpy strong enough to cancel out the documented benefits of  the syncarpous ovary.   No one has yet offered such an adaptive scenario.

Therefore, "cladistic parsimony" must be balanced against "adaptive parsimony."  Seventeen independent transformations from apocarpy to syncarpy may be more reasonable in view of selective pressures we know about than even one reversal.  I think it is therefore premature to dismiss apocarpy in the ancestral monocots.  Despite the fused carpels and other specializations of Acorus, apocarpy and looser forms of syncarpy (due to post-genital fusion) are widespread among the Alismatales, which are nearly as old as Acorus, and possibly more conservative with respect to the condition of their carpels.  In the palms as well, separate carpels occur in more archaic groups (Nypa and the Coryphoideae), though the first branch, the lepidocaryoid palms, are syncarpous. This is a more complex situation, however, as the palms appear to have originated among clades that were already syncarpous.  Still a scenario is needed for why fused carpels might become separate again in these palms.  Does such a scenario make sense in the real world of adaptive pressures?

References:

Armbruster, W. S.  , E. M. Debevec & M. F. Willson. 2002. Evolution of syncarpy in angiosperms: theoretical and phylogenetic analyses of the effects of carpel fusion on offspring quantity and quality. Journal of Evolutionary Biology 15 (4): 657-672.

Bhardwaj, M., Eckert, C. G. 2001. Functional analysis of synchronous dichogamy in flowering rush, Butomus umbellatus (Butomaceae). Am. J. Bot. 88(12):2204-13.

Sokoloff, D. D., M. V. Remizowa and P. J. Rudall. 2013. Is syncarpy an ancestral condition in monocots and core eudicots? in Early Events in Monocot Evolution, Eds.  P. Wilkin & S. J. Mayo. Cambridge University Press.