The Cordia Clade

The tropics are home to a wide diversity of plant species, many of them belonging to groups less familiar in cooler regions of the world. Prominent among these are members of the family Cordiaceae, a group of about 350 known species of mostly trees and shrubs. The Cordiaceae (alternatively treated as the subfamily Cordioideae of the family Boraginaceae) are a well distinguished clade both molecularly and morphologically. Most members of the clade have flowers with the stigma divided between four lobes, fruits with an undivided endocarp, and plicate cotyledons (Miller & Gottschling 2007).

Beach cordia Cordia subcordata, copyright Tauʻolunga.

Historically, most members of the clade have been assigned to a single genus, Cordia. This arrangement was revised by Miller & Gottschling (2007) who recognised the separate genus Varronia for about 100 species of multi-stemmed shrubs native to the New World. The remaining 250 or so species, most of them single-trunked trees, remained in the pantropical Cordia. The two genera also generally differ in their leaves (most Varronia have leaves with serrate margins whereas Cordia have entire margins) and inflorescences (most Cordia have broad cymose inflorescences whereas Varronia have smaller, more compact inflorescences). Few species of Cordiaceae are not assigned to either Cordia or Varronia. Three previously recognised small genera, Auxemma, Patagonula and Saccellium, are now synonymised with Cordia. The small African genus Hoplestigma and the prostrate annual herb Coldenia procumbens are placed in Cordiaceae primarily on the basis of molecular data (Miller & Gottschling 2007; Weigend et al. 2014).

Black sage Varronia curassavica, copyright Mauricio Mercadante.

A number of Cordia species are grown for their wood, with South American species providing timbers known as bocote, freijo (C. alliodora), and ziricote (C. dodecandra). These are only moderately strong woods but strikingly patterned and are more often used for aesthetic rather than structural purposes (such as cabinet veneers and musical instruments). Cordia alliodora has become an invasive in regions where it has been planted outside its native range such as Africa and Vanuatu. Various species are also grown for their edible fruits, such as the Assyrian plum C. myxa and the fragrant manjack C. dichotoma. These fruits are decidedly gooey when ripe and are often given names reflecting this fact such as glue berries, clammy cherries or, here in Australia, snotty gobbles (though this name is more widely used for fruits of the unrelated genus Persoonia). Pulp from unripe fruits of C. myxa can supposedly also be used as a type of glue. Your office reports may not be informative but they will at least be tasty!

REFERENCES

Miller, J. S., & M. Gottschling. 2007. Generic classification in the Cordiaceae (Boraginales): resurrection of the genus Varronia P. Br. Taxon 56 (1): 163–169.

Weigend, M., F. Luebert, M. Gottschling, T. L. P. Couvreur, H. H. Hilger & J. S. Miller. 2014. From capsules to nutlets—phylogenetic relationships in the Boraginales. Cladistics 30: 508–518.

The New Centaury

In an earlier post, I described the South American flowering herbs known as the Coutoubeinae. In this post, I'm going to take a step back and look at a clade of which the coutoubeines form a part, the Chironieae.

Seaside centaury Centaurium littorale, copyright Anne Burgess.

The Chironieae are one of the major tribes of the flowering plant family Gentianaceae, including about 160 known species. Representatives are found in most parts of the world, though as part of the native flora in Australasia they do not extend past the north of Australia (some exotic species have been introduced further south). The Chironieae seem to primarily be supported as a clade on the basis of molecular data (Struwe et al. 2002). All members are herbs, from annuals to short-lived perennials. Most have an erect growing habit; members of the Caribbean genus Bisgoeppertia are annual climbers and some species of the Mexican genus Geniostemon are creeping perennials. There may or may not be a basal rosette of leaves, and a number of genera have winged stems. Flowers are solitary or borne in cymose or racemose inflorescences. These flowers are most commonly salver-shaped (that is, shaped like a flat dish) or tubular, and usually have four or five petals (some species may have up to twelve). The calyx is usually comprised of fused sepals and is unwinged and tubular. The fruit is usually a septicidal capsule (splitting along the septa between carpels), more rarely a berry.

Yellow centaury Cicendia filiformis, copyright Hajotthu.

Members of the Chironieae are divided between three subtribes that are mostly distinct both morphologically and biogeographically. As described in the previous post, the Neotropical Coutoubeinae are characterised by releasing their pollen in tetrads whereas the other subtribes shed individual pollen grains. The Canscorinae are mostly found in the Old World tropics and have white or cream-coloured flowers (less commonly yellow, pink or purple) with the calyx tube longer than the calyx lobes. The Chironiinae mostly includes found in northern temperate regions, as well as the southern African genera Chironia and probably the South American Zygostigma. Their flowers come in a range of colours—pink, yellow, purple or blue, but less commonly white or cream-coloured—and may have calyx lobes longer than the tube. Many chironiine flowers also have anthers that become spirally twisted after releasing pollen whereas those of Canscorinae are always straight. Molecular data usually support the monophyly of the three subtribes and the majority view seems to be that the temperate Chironiinae represent the sister group of a tropical clade of Canscorinae and Coutoubeinae.

Cultivated Eustoma, copyright Rameshng.

Perhaps the best known members of the Chironieae are the centauries of the genus Centaurium. Historically, about fifty species across the Holarctic have been included in this genus. However, phylogenetic studies have demonstrated that this broad sense of the genus is polyphyletic and thus it has been cut down to a group of about twenty species found in Europe and western Asia. The name 'centaury' refers to the use of common centaury Centaurium erythraea as a medicinal herb, after the legendary centaur healer Chiron. Other Old World species are now placed in the genus Schenkia whereas North American species form the genera Gyrandra and Zeltnera. The yellow centauries of Cicendia are small, filiform annuals native to Europe and the Americas that have been introduced to Australia. The rose gentians Sabatia of North America bear pinkish-purple flowers, often in lax cymes. There are also the prairie gentians of the genus Eustoma. Native to southern North America, these plants bear large, showy flowers that have become popular in cultivation. Commercially, they are labelled as lisianthus. This is not to confused with Lisianthius, a distinct genus of Gentianaceae, or Lisyanthus, a name that has been used in the past for members of yet another gentianaceous genus. Both of these belong to completely different tribes in the family, and may be subjects for another day.

REFERENCE

Struwe, L., J. W. Kadereit, J. Klackenberg, S. Nilsson, M. Thiv, K. B. von Hagen & V. A. Albert. 2002. Systematics, character evolution, and biogeography of Gentianaceae, including a new tribal and subtribal classification. In: Struwe, L., & V. A. Albert (eds) Gentianaceae: Systematics and Natural History pp. 21–309. Cambridge University Press: Cambridge.

The Coutoubeines

Members of the family Gentianaceae, the gentians, are for the greater part associated with cooler climes. Residents of areas subject to heavy snowfalls have often commented on the appearance of their showy flowers with warming weather in the spring. But not all subgroups of the gentians are so temperate: some, such as the Coutoubeinae, are inhabitants of the tropics.

Schultesia guianensis, copyright João de Deus Medeiros.

The Coutoubeinae are a group of about thirty known species divided between five genera found in Central and South America (Struwe et al. 2002). A single species, Schultesia stenophylla, is found in western Africa but, as it is also found in Brazil alongside related species, it can be reasonably presumed to be a recent immigrant to that region. Like most other members of the Gentianaceae, species of the Coutoubeinae are low herbs, often found growing in open habitats. With the exception of the genus Deianira, most lack a basal rosette of leaves. Flowers are usually white or pink, and are quadrimerous (with four corolla lobes) in the majority of species (one species, Schultesia pachyphylla, has blue pentamerous flowers; Guimarães et al. 2013). Perhaps the most characteristic feature of the group is that pollen is released in tetrads (clumps of four). I haven't come across any specific comments on the functional significance (if any) of this feature in coutoubeines but it has been suggested that pollen clumping in plants may correlate with visits from pollinators being relatively uncommon (and getting a decent amount of pollen transported at a time becomes more important than increasing the chance of pollen being transported to multiple targets).

Coutoubea spicata, copyright Alex Popovkin.

The largest genus of coutoubeines is Schultesia, including about twenty species. Schultesia species are annual herbs with long-lanceolate leaves and tube-shaped, usually pink (occasionally yellow or blue) flowers with the calyx tube at least as long as the lanceolate corolla lobes. The species Xestaea lisianthoides, sometimes included in Schultesia, differs from Schultesia in the arrangement of stamens (inserted unevenly in the corolla rather than in the upper part of the tube) and the shape of the stigmatic lobes (oblong rather than rounded). Coutoubea species have white, salver-shaped flowers with triangular corolla lobes. Symphyllophyton caprifolium, a rare species restricted to southern Brazil, is a short-lived perennial with perfoliate leaves and yellow to cream salver-shaped flowers with the calyx tube shorter than the corolla lobes. Finally, Deianira includes suffrutescent herbs with a basal rosette of leaves and salver-shaped flowers with a short calyx tube.

REFERENCES

Guimarães, E. F., V. C. Dalvi & A. A. Azevedo. 2013. Morphoanatomy of Schultesia pachyphylla (Gentianaceae): a discordant pattern in the genus. Botany 91: 830–839.

Struwe, L., J. W. Kadereit, J. Klackenberg, S. Nilsson, M. Thiv, K. B. von Hagen & V. A. Albert. 2002. Systematics, character evolution, and biogeography of Gentianaceae, including a new tribal and subtribal classification. In: Struwe, L., & V. A. Albert (eds) Gentianaceae: Systematics and Natural History pp. 21–309. Cambridge University Press: Cambridge.

Of Shrimp Plants and Bear's Breeches

For today's semi-random post, I drew the plant subfamily Acanthoideae. As recognised by Scotland & Vollesen (2000), the Acanthoideae is the largest of the subfamilies of the Acanthaceae by a considerable margin, including about 95% of the family's 2500+ species. Though perhaps not hugely familiar to readers in more temperate climes, the Acanthoideae are one of the dominant groups of herbs and shrubs in tropical parts of the world.

Golden shrimp plant Pachystachys lutea, copyright Dryas.

The Acanthoideae have been recognised as a morphological group since the late 1800s and their integrity has been confirmed by more recent molecular studies. They are distinguished from related plants (within the Lamiales, the order that also includes such plants as the mints and snapdragons) by having capsular fruits that dehisce explosively when mature to scatter their seeds. The seeds are attached within the capsule by hook-shaped stalks called retinacula that presumably play a role in determining how the seeds are released. A classification of Acanthaceae published in 1965 by Bremekamp restricted the family to species with explosive fruits and retinacula, dividing them between two subfamilies, the Acanthoideae and Ruellioideae, based on the absence or presence, respectively, of cystoliths. These are outgrowths of the epidermal cell walls that are impregnated with calcium carbonate. They are visible in the stems and leaves, at least in dried specimens, as hard white streaks. As phylogenetic studies have supported division of Acanthoideae in the broad sense between a cystolith-possessing and a cystolith-lacking clade, the decision whether to recognise 'Ruellioideae' as a separate subfamily comes down to a ranking choice only. At lower levels, the classification of Acanthoideae is less straightforward. Over two hundred genera of Acanthoideae are recognised but just three of those—Justicia, Strobilanthes and Ruellia—account for about half the total number of species. Each of these mega-genera is morphologically diverse and likely to be para- or polyphyletic with regard to related taxa, raising the distinct likelihood of future revisions.

Spiny bear's breeches Acanthus spinosus, copyright Magnus Manske.
Economically, few of the Acanthoideae are of great significance except for a number of species being grown ornamentally. One such species is Acanthus mollis, which goes by the vernacular name of 'bear's breeches' (why, I have absolutely no idea). Acanthus was a popular decorative motif in classical Greece and forms the basis for the design of Corinthian columns. Its use as an ornamental has lead to it becoming regarded as an invasive weed in some regions, largely because this is one of those garden plants that Just Will Not Die, spreading easily from seeds and tubers. We've got some in a pot outside that is currently flourishing despite having been burnt down to a nub by the searing Perth summer sun, metaphorically shouting its defiance at an uncaring world.

REFERENCE

Scotland, R. W., & K. Vollesen. 2000. Classification of Acanthaceae. Kew Bulletin 55 (3): 513–589.

Sorting Monkeys: Dissecting Mimulus

New Zealand musk Thyridia repens, one of the species previously included in Mimulus section Paradanthus, copyright Jon Sullivan.

Long-time readers of this site will know that often, when I pull out the name from the virtual hat of my subject taxon for the week, I find that the greatest challenge lies in determining just what that name applies to. I knew that I was in for another one of these challenges as soon as I established that this week's post would be on the section Paradanthus of the plant genus Mimulus.

Mimulus, as it is most commonly recognised, is a genus of about 120 species found in many parts of the world but with by far the greatest diversity in western North America (where about three-quarters of the recognised species are found). They are commonly known as monkeyflowers or muskflowers in reference to the appearance and scent of the flowers of some species. Because of their diversity (in a well-studied part of the world), monkeyflowers have attracted a fair bit of interest as a model system for studying processes of evolution and speciation. Most Mimulus species are herbs though some are small shrubs. However, even shrubby forms do not have extensive root systems, and they are mostly annual or seasonal; perennial forms usually die off above ground over winter, growing back from the rootstock when conditions improve. These perennial forms may also propagate vegetatively through drooping stems putting down new rootstock where they contact the ground (Grant 1924). Mimulus species primarily grow in damp habitats; some will even grow in standing water.

Nepal monkeyflower Erythranthe nepalensis, copyright Qwert1234.

Until fairly recently, Mimulus was primarily classified in the family Scrophulariaceae. Members of this family (including Mimulus) were united by the possession of a flower type referred to as a 'scroph'; examples of plants with scroph flowers that may be familiar to you include snapdragons or foxgloves. Characteristic features of a scroph include having the calyx and corolla each fused basally. The corolla is hence more-or-less tubular at the base, then divided towards the top into five out-turned lobes corresponding to the petals (usually with two above and three below; the flower is therefore zygomorphic or bilaterally symmetrical). This flower type is primarily adapted to pollination by insects such as bees that use the lower lobe (or lip) as a 'landing ramp' when visiting the flower. After pollination, the flower develops into a dehiscent, many-seeded capsule. The advent of molecular phylogenetic analysis has established, however, that scroph-flowered plants belong to several lineages within the order Lamiales; the scroph has apparently evolved (and been lost) on a number of occasions. As a result, Mimulus is now placed in a separate family Phrymaceae with a handful of small, closely related genera.

Muskflower Erythranthe moschata, copyright Nick Moyes.

Mimulus was divided into two subgenera and ten sections by Grant (1924); with minor modifications, her system remained in place until recently. The two subgenera, Synplacus (or Mimulus proper) and Schizoplacus, were recognised based on whether the flower's placenta was united or divided, respectively. Four of Grant's sections were placed in the subgenus Mimulus; one of these was Paradanthus. Grant's system was not entirely phylogenetic as we would understand the term today: she did provide a diagram of suggested relationships between the sections but it did not cleanly separate the two subgenera. The section Paradanthus, in particular, was explicitly established as a convenient holding-place for a number of "small closely allied associations which, however, were not sufficiently distinct to warrant being placed in sections by themselves". Most had relatively unspecialised flowers, mostly (but not always) with funnel-shaped corollas and more or less equal lobes. Grant also somewhat unceremoniously dumped the majority of non-American Mimulus species into Paradanthus. She confessed that this section was "a collection of groups not necessarily related to one another and in all probability most of them have been derived from members of the other sections".

Phak taptao Mimulus orbicularis, one of the few true Mimulus species under the system of Barker et al. (2012), from here.

It therefore came as no surprise whatsoever when later phylogenetic analyses did not uphold the section Paradanthus as monophyletic. Instead, the primary division within Mimulus found by Beardsley & Olmstead (2002) was between a clade centred on western North America and one containing the majority of species from elsewhere in the world, with the Paradanthus species falling in either clade depending on their distribution (many of Grant's other sections, in contrast, do correspond to monophyletic groups). Perhaps more unexpected was the finding that Mimulus as a whole was not monophyletic. Instead, other genera of Phrymaceae were nested in Mimulus, including Phryma, a genus of one or two species found in eastern North America and eastern Asia in which the fruit is a single-seeded achene instead of a multi-seeded capsule, and Leucocarpus, a Central American genus in which the fruit is a fleshy berry.

The non-monophyly of Mimulus raised the question of whether these other genera should be subsumed within a broadened Mimulus (in which case the genus Mimulus and the family Phrymaceae would potentially become identical in content). An alternative tack was proposed by Barker et al. (2012) who divided the 'traditional' Mimulus into several genera corresponding to monophyletic clades separated by the previously recognised segregate genera. Of the 120 or so original species, only seven remain in the restricted Mimulus (species of Grant's section Paradanthus end up divided between no less than four genera). None of the western North American species remain in Mimulus; instead, species from this region are divided between the genera Diplacus and Erythranthe.

Only time will tell whether this proposed reorganisation will gain acceptance. I can see there being a lot of resistance to the idea that many of the most familiar 'Mimulus' species should no longer be included in Mimulus, particularly in non-academic circles. Nobody likes being made a monkey by monkeyflowers.

REFERENCES

Barker, W. R., G. L. Nesom, P. M. Beardsley & N. S. Fraga. 2012. A taxonomic conspectus of Phrymaceae: a narrowed circumscription for Mimulus, new and resurrected genera, and new names and combinations. Phytoneuron 2012-39: 1–60.

Beardsley, P. M., & R. G. Olmstead. 2002. Redefining Phrymaceae: the placement of Mimulus, tribe Mimuleae, and Phryma. American Journal of Botany 89 (7): 1093–1102.

Grant, A. L. 1924. A monograph of the genus Mimulus. Annals of the Missouri Botanical Garden 11 (2–3): 99–388.