Gymnarrhena micrantha

General description: 


[synonomy is available at the end of this section] [available protologues are attached as images; not all protologues are available at this time]

Tribe Gymnarrheneae Panero & VA Funk (subfamily Gymnarrhenoideae Panero & VA Funk).

Gymnarrhena Desfontaines, Mémoires du Museum d’histoire naturelle. Paris, 4: 1. (1818), T.: G. micrantha Desfontaines. Currently this genus has only one species.

Type: Gymnarrhena micrantha Desfontaines, Memoires du Museum d’histoire naturelle, Paris. iv (1818) 1. t. 1


Gymnarrhena balansae Coss. & Durieu ex Coss. & Kralik, Bulletin de la Societe Botanique de France 4: 180. (1857) [currently in synonomy with G. micrantha Desf.]

Annual amphicarpic herbs with a prostrate rosette, no more than a few cm high; no reports of milky sap. Leaves simple, forming a dense rosette, sessile, smooth, narrowly lanceolate to narrowly ovate, apex narrowly acute to attenuate, margins denticulate, base truncate, surfaces smooth and glabrous. Subterranean heads homogamous, female, cleistogamous, surrounded by the leaf bases; florets enclosed in involucral bracts, corolla vestigial; achenes relatively large, laterally flattened, blackish, sparsely hairy, remaining below the soil surface on the dead parent plant; pappus absent, vestigial, or of short, basally flattened, somewhat scale-like bristles. Aerial heads congested in the center of the leaf rosette, heterogamous, disciform; involucral bracts imbricate in several series, chartaceous, whitish, acute; receptacle convex, marginally bristly; with a few rows of chartaceous phyllaries. Functionally staminate florets in small groups, loosely connected on very short pedicels, interspersed among the small pistillate florets, corollas small 3-4 lobed, whitish; stamens 3-4, anthers calcarate, ecaudate, without apical appendage. Pistillate florets solitary, each enclosed in a prominent, stiff, white and green bract; corolla filiform, style arms long, with rounded apices. Achenes of functionally staminate florets vestigial, pappus of a few irregularly lacerate scales or absent; achenes of pistilate florets numerous, tiny, ovoid, ciliate, villous, with long twin hairs, cell walls thin, pappus of long-lanceolate, ciliate, acutely acuminate scales.

A second species G. balansae Cross. & Durieu was based on two collections from near the coast in Tunisia and Algeria and they appear to be somewhat different from the collections from the Middle East and further inland in Algeria. However, a more detailed study will have to be made to determine the validity of this taxon.

Gymnarrhena has been collected from North Africa to the Middle East. It is a winter annual and its flowering depends on the rain but usually takes place in March and April (May). According to Gutterman (1989) the life span of the plants is around 67 days out-of-doors (full sun) where it produces both types of heads and 140 days in the greenhouse (8 hours of light) where it produced only aerial heads.

Remaining Synonomy

Cryptadia Lindley ex Endlicher, Genera Plantarum secundum ordines naturales…. By Stephan L. Endlicher. Suppl. 1: 1381. Feb-Mar 1841.
T.: non designatus

Cryptadia euphratensis Lindley in Chesney, Narrative of the Euphrates Expedition carried on by Order of the British Government during the years 1835-1837. (1868) t. ad p. 441.

Frankia Steudel ex Steudel, Nomenclator botanicus: numerans ordine alphabetico nomina atque synonyma, tum generica tum specifica, et a Linnaeo et recentioribus de re botanica scriptoribus plantis phanerogamis imposita /auctore Ernesto Steudel. ed. 2. 1: 646, (1840); nom. inval.

Frankia schimperi Steud. Nomencl. Bot. [Steudel], ed. 2. 1: 646, (1840); nom. inval. in syn.


There are two different published chromosome counts; Murín & Chaudhri (1970) from Iraq and Kamel (1999) from Egypt; both report 2n = 20, each from a single plant. Nikulina & Kotseruba (1999) report 2n = 18.


The male florets are not always grouped in the center of the head as previously described, in fact some groups of male florets are found at the outer edge of the receptacle. It seems more likely that the aerial heads are actually groups of heads some of which are few flowered male heads and others are single flowered female heads so these could represent dioecious heads grouped on a common receptacle. The subterranean heads flower first, their petals protrude just above the soil surface. Later the aerial inflorescences appear. Likewise the fruits of the two types of heads of Gymnarrhena have different developments, the larger subterranean fruits develop first and the aerial fruits are produced later and only in wet years. The aerial fruits are wind dispersed while the subterranean fruits germinate underground. On the aerial heads the scales of the pappus as well as the bracts surrounding the achenes are hydrocastic (open when wet) and the achenes are dispursed by wind (Guterman, 2002).

Gymnarrhena exibits “dimorphic cleistogamy” or “true cleistogamy” in that it has two different flower types: chasmogamos and cleistogamous (Culley & Klooster, 2007). While about 228 genera have some type of cleistogamous flowers, only 168 are dimorphic. It is estimated that dimorphic cleistogamy has evolved less than 40 times and most of the lineages have very few species (Culley & Klooster, 2007).


The only anatomical data found is from Zamski et al (1983) who examined the taproot of Gymnarrhena and found that it starts to contract soon after emergence. Ultimately, this contraction causes the retraction of the main shoot apex from the soil surface to a depth of about 10 mm.

Taxon biology: 

Much of the material for this page is taken from Funk & Fragman-Sapir (2009).

Gymnarrhena is an unusual member of the Compositae (Asteraceae). It is an ephemeric, amphicarpic, dwarf desert annual. Amphicarpic plants have two types of flowers, in this case aerial casmogamous heads and subterranean cleistogamous ones, and the different flowers produce different fruits. In Gymnarrhena, the achenes produced from these two types of inflorescence and the seedlings that germinate from them, differ in size, morphology, physiology and ecology (Koller & Roth, 1964; Zamski et al., 1983). The plant is very small and has grass-like leaves, the aerial heads are clustered together and have functional male and female florets. The familiar parts of the Compositae head have been modified extensively and most of the usual identifying features are missing or altered (see images). Currently there is one species recognized, Gymnarrhena micrantha Desf., however there is some variation across the distribution and it should be investigated further.

Historical overview

As might be imagined the history of this taxon reflects its unusual morphology. Bentham (1873), Hoffmann (1894), and Cronquist (1955) put this genus in the Inuleae s.l. Hoffmann and Cronquist both mention the similarity to Geigeria; Bentham’s alternate choice was the Astereae. Small (1917-1979) considered the Inuleae to be linked to Centaurea of the Cardueae. Leins (1973) in his examination of the pollen of the Inuleae stated that Gymnarrhena did not belong in that tribe and suggested the Cynareae-Carlininae. Merxmuller et al. (1977) agareed that Gymnarrhena was not in the Inuleae and cited Leins (1973). Skvarla (1977) acknowledged a superficial resemblance between Gymnarrhena and the Cardueae but pointed out that it had Anthemoid type pollen (also found in the Senecioneae and other tribes) and did not belong in the Inuleae or Astereae. Skvarla further acknowledged that, based on the pollen, the genus was difficult to place. Bremer (1994) listed the genus as belonging to the Cichorioideae s.l. but as “unassigned to tribe” along with several other problem genera.

Systematics – Molecular Phylogeny

Anderberg et al. (2005) in a study of the Inuleae using ndhF determined that Gymnarrhena did not belong in the Asteroideae but rather was part of the then paraphyletic Cichorioideae s.l. or sister to the entire Asteroideae. In the most recent broad scale cladograms Gymnarrhena is in a clade by itself and is consistently located below the Cichoroideae s.s. – Corymbieae – Asterioideae clade and above the Carduoideae and Pertyoideae clades (Panero & Funk 2002, 2008; Funk et al. 2009). Its position means that it cannot be placed in any of the other suprageneric taxa and is now recognized as an independent lineage. This position is supported by the fact that Gymnarrhena lacks the ‘9 base pair deletion’ in the ndhF gene identified by Kim and Jansen (1995) and subsequently used by Bremer (1996) as a molecular characteristic in support of the recognition of the Cichorioideae-Chorymboideae/Asterioideae clade. In the most recent taxanomic overview for the family the tribe Gymnarrhenaeae was accepted by Jeffrey (2006) based on the molecular results reported above.


Gymnarrhena pollen was examined by Wortley et al. (2007; Fig. 1F-G) and they report that it is spheroidal, round in polar and equatorial view and tricolporate. The colpi are separate with acute ends. The grains are echinate and non-lophate, the spines unevenly-distributed, conical-pointed, 1-2 μm long, with bases slightly swollen and internal cavities. The tectum is microperforate, covering the whole surface of the grain. The infratectum comprises two distinct layers which are not clearly attached to one another; the outer layer is columellate or spongy, the inner supporting layer made up of thick, unbranched, solid columellae, evenly-distributed around the grain. The grains are ecaveate and the endexine is thinner than the foot layer. Despite the sparse spines and lack of a cavea, the ultrastructure of these grains, with unattached layers of evenly-distributed columellae, does not fit into the Cichorioideae (Anderberg et al., 2005) and is suggestive of the characteristics of Corymbieae, the sister group of Asteroideae (Wortley et al., 2007). This is not as far from the position on the molecular tree as it might seem since Gymnarrhena is the branch just below the Cichorioideae and Corymbium is nested one node higher than the Cichorioideae. Zhao et al (2006) commented that Gymnarrhenapollen shared the multilevel columellae with the Mutisieae s.l. but they noted that this character is found elsewhere in the family including the Anthemideae and Cardueae; they concluded that it did not belong in the Mutisieae s.l. because of its spinate pollen. So the pollen characters do not place the genus in any of the existing suprageneric taxa and are, therefore, consistent with recognizing it in its own tribe and subfamily.


Gymnarrhena is an amphicarpic herb (see images) of the Mediterranean biome of North Africa and the Middle East growing in dry, mostly bare, sandy areas and resembling a grass (see images). Research on the reproductive biology of this unusual plant is summarized by Koller and Roth (1964) and in the description above. There are three aspects to its reproductive biology that are most interesting, seed heteromorphism, the presence of aerial as well as subterranean heads, and the presence of both casmogamous and cleistogamous flowers.

Seed heteromorphism, that is the production of seeds with variable morphologies and ecological strategies, represents an allocation of different fractions of seed output to different ends and most examples are found in four flowering plant families, one of which is the Compositae (Harper, 1977). Seed heteromorphism appears to be largely restricted to relatively short-lived, fugitive species, particularly weeds. Venable and Burquez (1989) say that it might be a form of ‘bet hedging’ in response to environments that vary spatially or temporally. These morphological heteromorphisms are important because they may be associated with ecological strategies which have evolutionary significance, such as dispersal, dormancy, differential competitive performance, within or among year timing of germination, vulnerability to predators, seedling growth and survival or fecundity (see many references cited in Chmielewski, 1999).

The presence of aerial as well as subterranean heads has several different explanations. Koller and Roth (1964) reported that mean weight of Gymnarrhena aerial fruit was only 5-6% of the weight of a subterranean fruit and that 6-day-old subterranean seedlings were 6 times the weight of aerial seedlings. They reported that the survival was considerably greater for these subterranean seedlings and that plants under dry conditions may fail to produce aerial heads. On the other hand, according to Brown & Böer (2005) the species occurs on firm sometimes rocky substrates which are generally hostile to plant grown. The above ground fruits are small and possess a small pappus to aid in their dispersal so they can travel some distance from the parent plant while the 1-2 large underground fruits (nearly 20 times the weight of the aerial ones) lack a pappus and stay close to the parent plant. It is their opinion that the underground fruits ensure that when the mother plant dies later in the year the same favorable location is re-colonized. Gutterman (1989) thinks that the underground location protects the seeds from predation. Zeide (1978) described dual strategies for Gymnarrhena micrantha in the Negev desert of Israel. He determined that the subterranean fruits were produced according to a “pessimistic strategy” where fruit production begins as soon as possible concurrent with continued vegetative growth. The aerial fruits, on the other hand, were produce according to a more optimistic strategy (in order to maximize yield, the vegetative stage precedes a last and full switchover to heavy fruiting). This pessimistic strategy must therefore arise from other considerations such as unpredictable environments as it is at odds with optimality models of resource allocation in annuals (Cheplick & Quinn, 1982). They summarize by saying that there are three arguments that have been supported for Gymnarrhena: 1) the seedlings arising form the larger propagules are more tolerant to stress or competition, 2) these subterranean seedlings have a higher probability of surviving to produce seed, 3) genotypes with an early production of subterranean seed may be the only ones to produce seeds under stress (Evenari, 1963; Koller & Roth, 1964; Zeide, 1978). And another has been suggested (Cheplick and Quinn, 1982): 4) the larger subterranean seed may show a greater return on the energy invested than an aerial seed. Other explanations that have been provided are the protection afforded by burial against catastrophies and the placement of subterranean seed in the same microhabitat occupied by the parent plant, providing protection from the risks of random dispersal (Evenari, 1977; Koller & Roth, 1964). Cheplick & Quinn (1982) concluded that the early production of few larger seeds followed by the later production of many small seeds was presumably the result of the variable and unpredictable length of their growing periods and to their roles a fugitive species.

There are about 50 amphicarpic species worldwide, approximately one in every 5000 species of flowering plants; most of them occur in either frequently disturbed and or stressful habitats. Eight of these are found in Israel which has only 2500 species in its known flora, one in ca. 310 species, a very high ratio compared to flowering plants in general (Kaul et al., 2000; Lev-Yadun, 2000). Given the suggested reasons above, Lev-Yadun (2000) adds that since many of the amphicarpic annuals are found in disturbed areas that are the result of drought, fire and grazing. He thinks this indicates a long history of disturbances in the eastern Mediterranean region that pre-date human impact.

The cleistogamous aspect of the subterranean florets is more easily explained in that such flowers are invariably cleistogamous (Kaul et al., 2000).

Additional information can be found in the recent book by Gutterman (2002).


Gymnarrhena is listed on the web in the Global Compendium of Weeds as being on a list of weeds from Egypt (, however, an examination of the book (Boulos & el-Hadidi, 1984) yielded no mention of the species.

In a study of the areas of Kuwait that were damaged by oil towards the end of the Gulf War in 1991 the area covered by oil or by tar tracks is largely sterile. However, Gymnarrhena was one of the taxa found in this area growing in the control sites and in areas where the sand has blown out from under the tar tracks indicating that it has the ability to move into disturbed habitats (Brown & Porembski, 2000). But perhaps the most interesting study involving pollution and Gymnarrhena concerns the copper mining and smelting activities of the Nabatean, Roman and Byzantine periods in the southern Jordanian desert that has, for 2000 years, continued to exert an influence on the plants and animals, and no doubt the people, of the area (Pyatt et al., 2000). Khirbet Faynan (believed to be the Roman city of Phaino) was a major center of metal working in the ancient world and left huge deposits of slag containing copper and lead. Testing of recent soil, plants, and animals show enhanced concentrations of this lead and copper: one of the two plants with the highest concentrations was Gymnarrhena which is one of the favorite plants for the grazing goats in the area and the goats in turn also showed elevated levels of lead and copper (Pyatt et al., 2000). Past levels were certainly much higher and possibly impacted the plants, animals and humans for 2000 years. Anthropologists have a wonderful ability to meld interesting science with drama and so the final words of Pyatt et al., (2007) are worth repeating:

The copper and lead rich wadis of southern Jorden…saw industrial pollution on a scale which would have been familiar to the inhabitants of Victorian Sheffield.…Small wonder then that in Romano-Byzantine times the mines of Phaino were seen as a place to send recalcitrant criminals. Eusebius of Caesarea in his “Martyrs of Palestine” describes such a scene; “they demanded that he should be sent away to the mines, and not just any mines but to that of Phaino where even a condemned murderer is hardly able to live a few days.”

Nothing else has been found on either this topic or any other applied aspects of this species.

Mon, 2015-01-12 14:48 -- funkv
Scratchpads developed and conceived by (alphabetical): Ed Baker, Katherine Bouton Alice Heaton Dimitris Koureas, Laurence Livermore, Dave Roberts, Simon Rycroft, Ben Scott, Vince Smith