A peculiar habitat type found in the savannahs of Central Cuba, Villa Clara Province and characterised by the presence of a surface gravel layer of “perdigones”, an assemblage of small ferralitic concretions, upon the “mocarrero” soil prevailing in the area, is described. On sterile gravel patches, only one species grows: Elytraria serpens, a new species described and named here. It is noteworthy for possessing long and wide creeping, stoloniform subterranean peduncles with apical gemmae developing into rooting leaf rosettes enabling vegetative propagation. The new species is close to E. shaferi and considered to derive from the latter by adaptive evolution, enabling it to survive in its hostile habitat, sheltered from the competition of other plant species. Small soil insects, for example, ants, are believed to act as pollinators.

Se describe un hábitat particular encontrado en la provincia Villa Clara, caracterizado por la presencia de una capa superficial de perdigones (pequeñas concreciones ferralíticas) que recubre el suelo mocarrero de aquella área. En los parches estériles con grava solo crece Elytraria serpens, nueva especie aquí descrita, notable por tener pedúnculos subterráneos largamente rastreros y reproducirse vegetativamente por yemas apicales que forman rosetas satélites que luego se enraízan. La nueva especie es afín a E. shaferi y se considera derivada de esta por evolución adaptativa, permitiéndole la colonización de un hábitat desfavorable, abrigada de la competencia de otras especies vegetales. Se opina probable la polinización por pequeños insectos terrícolas, p. ej. hormigas.

Acanthaceae, Cuba, Elytraria, “mocarrero” soil, myrmecophyly, new species, “perdigón”, vegetative propagation, Villa Clara Province

Amongst many interesting Cuban localities that the authors visited in March 2019, under the expert guidance of botanists of the Botanical Garden, Universidad Central “Marta Abreu” of Las Villas, the one that we want to present here stands out, not only for hosting a species new to science that we are going to describe hereunder, but also by the particular aspect of its soil. Villa Clara botanists were already familiar with that very locality because of the presence in it of a small population of Paspalum edmondii León, a rare endemic species, listed as Critically Endangered (González Torres et al. 2016). This site is located in the Corralillo Municipality, at 22°50'55" latitude North and 80°28'22" longitude West, at an elevation of ca. 65 m a.s.l., between the Las Cañas and Clarita Rivers (or brooks). According to Borhidi’s (1991: 344) phytogeographic typology, it situated within the “Sagüense District” that encompasses the hills and plains of the northern coastal strip of Las Villas.

The soils of the area are of the “mocarrero” type, which, according to Bennett and Allison (1928: 67), is characterised by the presence of a 10–25 cm thick ferralitic rock layer (“ironstone”) at a depth of 25 to 100 cm, a layer that inhibits drainage and causes the lands, very arid in the dry season, to become swampy in the rainy season. This, added to the general soil oligotrophy, results in the prevailing of savannah plant formations (Bennett and Allison 1928; Borhidi 1991: 149). Borhidi (1991: 162) believes that the savannahs have replaced an original scrub vegetation, following an alternation of fire and grazing events.

“Mocarrero” soils cover large areas in the Cuba lowlands. The particular locality to which we refer is, moreover, characterised by the presence of a dense and homogeneous surface layer of ferralitic glomeruli, conditioning a very open vegetation interrupted by sterile gravel patches (Fig. 1). In Cuba, ± globular concretions of this type are found in various types of soil, including “mocarrero”. In pedology, they are known as “perdigón” or pellets (Bennett and Allison 1928: 4–6); they are thought of as matching the “mocarrero” ironstone’s mineral composition (Bennett and Allison 1928: 78–80) and they are considered as resulting from in situ concretion, not as alluvial deposits.

Figure 1. 

Elytraria serpens, a plant at the locus classicus, with six inflorescences (one of them with a flower, bottom right) emerging from the ground around it (photo R. A. Pérez Obregón).

It would be interesting, although exceeding the purpose of this paper, to study the microclimatic properties of the pellet surface layer in our locality. We surmise that the thermal and hydric conditions inside and below the gravel layer will be found to differ markedly from those at the soil surface and in the overlying air.

No exhaustive inventory of the flora present in that locality was made, but a good sample of the flora was collected, including the following taxa (listed alphabetically with, in parentheses, the collecting number of the voucher specimens [kept in the herbaria PAL-Gr, B, HAJB and ULV]): Angelonia pilosella J. Kickx f. (29683), Aniseia martinicensis (Jacq.) Choisy (29684), Byrsonima crassifolia (L.) Kunth (29696), Caesalpinia pinnata subsp. oblongifolia (Urb.) A. Barreto & Beyra (29681), Cuphea parsonsia (L.) R. Br. (29693), Evolvulus minimus Ooststr. (29685), Metastelma cubense Decne. (29686), Rauvolfia cubana A. DC. (29689), Paspalum edmondii (29695), Phyla stoechadifolia (L.) Small (29692), Pisonia rotundata Griseb. (29691), Sideroxylon salicifolium (L.) Lam. (29694), Stachytarpheta angustifolia (Mill.) Vahl (29682), Tabebuia lepidota (Kunth) Britton (29690) and Viguiera dentata (Cav.) Spreng. (29688). According to Borhidi’s (1991: 613) classification, that plant community would seem to belong to the class Curatello-Byrsonimetea Borhidi and perhaps to the order Byrsonimo-Andropogonetalia teneris Bal.-Tul.: low-grass savannahs with scattered palms, on seasonally flooded “mocarrero” soil.

Scattered on the gravel patches of that locality, we noticed a very curious plant, a species of the genus Elytraria Michx. (Acanthaceae) that did not match any of the known Cuban representatives of this genus: stemless perennial herbs, with all leaves flat on the ground, forming a basal rosette. The inflorescences, solitary or paired spikes, emerge from the soil at a distance from their rosette and are connected with it by a slender axis (the peduncle) winding below or inside the top layer of gravel. These peduncles, which mimic underground stolons, are pale, yellow or yellowish-brown in colour and covered with dense scales or sterile bracts. In the flowering season, they are devoid of adventitious roots, but at their apex, below the inflorescence(s), they frequently produce buds that develop into small foliar rosettes, destined to take root and give rise to new plants in the subsequent rainy season (Fig. 2). The spikes, surrounded by densely imbricate bracts, produce few tiny, white, unspectacular flowers that open one at a time. It is hard to believe that they are apt to attract pollinating flying insects; which is why it is likely that pollination, when performed, is the work of small terrestrial animals, for example, ants. It is obvious that reproduction is mostly vegetative, a hypothesis that explains the scarcity of well-developed fruits and the low efficiency of the seed dispersal mechanism.

Figure 2. 

Elytraria serpens, a plant at the locus classicus with a dozen unearthed stoloniform peduncles, showing several subapical buds (rosettes in formation) and one fully developed daughter rosette, already rooted (photo R. A. Pérez Obregón).

We are grateful to Rosalina Berazaín Iturralde for advice and information on ecology and habitat; to Idelfonso Castañeda Noa, who led us to the type locality of the new species and offered information on the locality studied; to Rafael Alejandro Pérez Obregón, who, at our behest, took for us the photographs of Fig. 1; and to Sigrid Liede-Schumann, Bayreuth, who identified our Metastelma gathering.