Tillandsia ixioides, Rossado & Donadío & Bonifacino, 2024

Rossado, Andrés J., Donadío, Sabina & Bonifacino, José M., 2024, A taxonomic revision of Tillandsia (Tillandsioideae, Bromeliaceae) from Uruguay, Phytotaxa 667 (1), pp. 1-90 : 1-90

7. Tillandsia ixioides Grisebach (1879: 333) . Figs. 5B , 23 , and 24.

Type (lectotype designated here):—[ ARGENTINA]. [Entre Ríos]: Concepción del Uruguay, “In uferwaldungen, selten”, 30 October 1875, Lorentz in Flora uruguensis 494 ( GOET000436 [online image!]; isolectotype: CORD00004758 [online image!]) .

= Tillandsia meridionalis Baker (1888: 15) . Type (lectotype designated by Smith & Downs 1977: 819 [as “ holotype ”]):— URUGUAY: “Banda oriental”, without exact locality, s.d., Tweedie s.n. (K [K000308666, online image!]).

Plants short-stemmed to long caulescent, commonly much branched forming globose clumps up to 35 cm in diameter when epiphyte or up to 1 m in diameter when epilithic. Roots 0.5–1 mm diameter, only present at the base of the stem. Stems up to 70 cm long but generally shorter (much shorter in epiphytic plants). Leaves spirally arranged, clustered forming rosettes, grayish; sheaths 13–31 mm wide, distinct from the blade, passing gradually into the blade, very widely triangular-ovate to very widely ovate, glabrous except for the lepidote apical portion; blades 55–175 × 6–16 mm, suberect to spreading, very narrowly to narrowly triangular, channelled from the base to almost the apex, rigid, densely appressed to subappressed-lepidote throughout, trichomes nearly symmetric. Inflorescences simple; peduncles wholly covered by bracts or sometimes apically visible, 3–23 cm long, ca. 2.5 mm in diameter, glabrous; peduncle bracts 5 to 9 in number (the basal ones leaf-like), erect, evenly distributed along the peduncle, densely imbricate (sheath reaching at least the basal half of the following bract); upper peduncle bract with a short blade or bladeless, 25–65 × 11–17 mm, narrowly elliptic to narrowly ovate-elliptic, stramineous to reddish-pink, subdensely to densely lepidote except for the glabrous base, or sometimes glabrous up to the middle portion. Raceme spirally and (sub)densely (2)3 to 9-flowered, (25–) 30–95 mm long, ellipsoid; rachis from totally hidden by the floral bracts to wholly exposed, reddish-pink to greenish, glabrous. Floral bracts densely distributed (1.5–8 times longer than the internodes), generally not imbricate, clasping the basal half of the flower, narrowly elliptic-ovate, ecarinate, reddish-pink and apically stramineous or totally stramineous, subdensely to densely lepidote at least at the apex; basal floral bracts 21–53 × 7.7–15.2 mm, slightly surpassing to equaling the sepals, or sometimes shorter, commonly with a very short blade (less than 4.5 mm long and not clearly differentiated from the sheath), or sometimes bladeless; upper floral bracts shorter, commonly bladeless and not surpassing the sepals. Flowers 22–40 mm long, strongly fragrant; pedicel 2–17 mm long (in basal flowers always longer than 4.5 mm); sepals partially visible, (14–)15.5–28 × 6–10(–11.5) mm, evenly very short connate at the base for 0.2–1.1 mm (appearing free), ovate, sometimes somewhat asymmetric, cream and generally pinkish over the apex and carinas, semitransparent, glabrous or with trichomes towards the apex and/or over the carina; abaxial sepal ecarinate; adaxial sepals strongly carinate, and generally longer than the abaxial one; petals 25–40 mm long, spatulate; claw 2.7–3.5 mm wide, white and apically yellow; limb 5.4–8(–9) mm wide, spreading, elliptic, intensely yellow, margins irregular or slightly crenate; stamens 15–25 mm long, included and visible (reaching the throat of the corolla), generally longer than the pistil; filaments 12–21 mm long, not plicate; pollen yellow; pistil 11–17 mm long, included, not visible; ovary 4.5–6 × 2–3.5 mm, ellipsoid, tapering into the style; style 6–8 mm long, 1.2–1.7 times as long as the ovary, whitish; stigmas simple-erect. Capsules 26–38 × 4.5–6 mm, exceeding its respective floral bract, cylindrical-prismatic to very narrowly ellipsoid-prismatic, apex obtuse to obtuse-acute and short-beaked.

Vernacular name:— Azahar del campo ( Herter 1949 –1956), Clavel del aire amarillo ( Herter 1949 –1956; Brito & Llano 2008).

Classification:— Tillandsia ixioides belongs to T. subg. Anoplophytum s.s. ( Smith & Downs 1977; Barfuss et al. 2016). Two subspecies are described for T. ixioides : T. ixioides subsp. ixioides and T. ixioides subsp. viridiflora ( Rauh 1979: 13) Gouda (2010: 76) . Of the two subspecies, the typical one is the most widespread and the only one occurring in Uruguay.

Distribution and habitat: — Bolivia, Paraguay, Brazil, Argentina, and Uruguay. In Brazil, Tillandsia ixioides is only known for “Parque estadual do Espinilho”, in southwestern Rio Grande do Sul state (southern Brazil), near the border with Argentina and Uruguay. In Argentina, this species is widely distributed from the northwest to the center of the country, mainly inhabiting open dry forests of the Chaco and Espinal phytogeographic provinces.

In Uruguay, T. ixioides has a restricted distribution ( Fig. 24 ). This species mainly occurs on the plains alongside the Uruguay River, growing as epiphyte in open thorn forest (mainly on Aspidosperma quebracho-blanco Schltdl. and Neltuma spp. ; Fig. 23D ), and palm forest of Butia yatay (Mart.) Becc. ( Fig. 23B ). Additionally, an exclusively epilithic population is found in San José department ( Fig. 23A ). The individuals of this epilithic population have some morphological differences from those occurring in western Uruguay (see Note I), and tend to form large clumps that grow on more or less horizontal areas of big rocky blocks, in a region of xeric hillside forest and xeric shrubland. This population represents the southern limit of distribution of T. ixioides .

Phenology:— In Uruguay, Tillandsia ixioides flowers from mid-winter to mid-spring (mainly in August, September, and October; Fig. 4 ). Fruits begin to develop as of September and October, and they continue to grow in length for about one month. Then, they remain without growth for one to two months until seed dispersal in late spring–early summer (December and January). In T. ixioides the development of the inflorescence takes from three to four months, and the whole reproductive cycle between six and seven months. It should be noted that the northern populations tend to bloom earlier than the southern ones.

Reproductive observations:— Multiple flowers within each inflorescence are simultaneously open ( Fig. 23E ). These flowers emit a strong fragrance throughout the day, and once they wither the petals turn darker yellow.According to Bianchi & Vesprini (2014), Tillandsia ixioides is a self-incompatible species whose flowers remain open from four to six days.

Conservation:— Tillandsia ixioides was categorized as Threatened in Uruguay. This species is considered as a priority for conservation in Uruguay by Marchesi et al. (2013), but it is absent in national protected areas. However, one of the examined specimens is from the buffer area of the ‘Parque Nacional Esteros de Farrapos e Islas del Río Uruguay’. Therefore, at this site, the species could potentially benefit from some form of protection. Additionally, we consider particularly important the protection of the population that grows epilithically. This population exhibits morphological differences from the rest of the epiphytic populations and has been registered in only one locality (see Note I).

Diagnostic characters:— In Uruguay, Tillandsia ixioides is the only species of Tillandsia with flowers that have conspicuous and clearly differentiated pedicels. The pedicels from the basal flowers are longer than 4.5 mm. In the rest of the species, the pedicels (if any) are shorter than 4 mm and they are not clearly differentiated from the receptacle. This character allows for an easy identification of this species both in the field and in herbarium. Furthermore, in the field it is clearly distinguishable from other species as it is the only one with spirally arranged yellow flowers. When sterile, T. ixioides may be confused with T. aëranthos , which occurs in sympatry. The former has notoriously rigid leaves, generally with pungent tip, and trichomes with divergent wings, while T. aëranthos has somewhat stiff leaves, apex not pungent, and trichomes with appressed wings.

Background in Uruguay:— Tillandsia ixioides was first cited for the Uruguayan flora by André in 1905, although without specifying reference material. Subsequently, several works mentioned this species for Uruguay ( Herter 1930, 1949 –1956; Smith 1933) but it was Smith, in 1972, the first to cite a herbarium specimen.

Note I: —We have identified two morphotypes of Tillandsia ixioides growing in Uruguay. The most common morphotype exclusively grows as epiphytic, forming globose, dense clumps (occasionally not-branched plants) with short stems (generally not over 15 cm long; Fig. 23B–D ). The other morphotype exclusively grows as epilithic and forms extensive clumps that grow on more or less horizontal areas of rocks ( Fig. 23A ). The latter morphotype has very long stems (generally over 50 cm long) that rest on the rocks for most of their length and ascend apically. The differences between these two morphotypes are also reflected in the size of both vegetative and reproductive structures. In this regard, the epilithic morphotype shows larger leaves, longer peduncles and peduncle bracts, longer inflorescences, and larger floral bracts and flowers. Likewise, the inflorescences of the epilithic morphotype are sturdy and erect, whereas those of the epiphytic ones are generally recurved and less sturdy. Despite these two morphotypes being clearly distinguished, our analysis of T. ixioides specimens from Argentina revealed the presence of morphologically intermediate individuals between the two morphotypes observed in Uruguay. Therefore, we suggest that treating these two morphotypes as distinct entities may not be appropriate. Future genetic studies could offer valuable insights into whether they are indeed different entities or not.

Note II: — Tillandsia ixioides was validly published by Grisebach (1879) in Symbolae ad Floram Argentinam. Unfortunately, in this publication Grisebach did not specify the collector, nor collection number, nor date of the specimen/s in which he based his descriptions. However, Grisebach included in the protologue data of the place and habitat where the specimens were collected: “E. [Entre Ríos, Argentina]: pr. Concepcion del Uruguay, raro in fruticetis ripariis”. According to Hunziker (1960), the specimens studied by Grisebach (1879) from Entre Ríos are based on collections of P.G. Lorentz obtained between November 1872 and December 1873. Considering all this information, we identify Lorentz (Flora uruguensis) 494 as original material for T. ixioides , as the labels of the specimens of this gathering, besides indicating the locality “Concepcion del Uruguay ” also include “In Uferwaldungen, selten [in riverside forest, rare]”, which matches exactly the habitat information and abundance indicated by Grisebach (1879; “raro in fruticetis ripariis”) on the protologue. Further, “ Tillandsia ixioides n. sp. ” is written on its label in Grisebach’s handwriting, as a reference to a new species.

The gathering Lorentz (Flora uruguensis) 494 is composed of at least two specimens, i.e. GOET000436 and CORD00004758. Here, we designate the one housed at the herbarium of the University of Goettingen (GOET) as the lectotype for T. ixioides , as it was the place where Grisebach, the author of the species, studied the Argentinian material ( Hunziker 1960, Stafleu & Cowan 1976). Additionally, it is the only one that on its label has “ Tillandsia ixioides n. sp. ” written in Grisebach’s handwriting [based on the comparison of the calligraphy with other labels that have confirmed handwritten notes by Grisebach, e.g. Lorentz & Hieronymus 85 (Planchuela & Ariza Espinar 2015)]. Further, on the sheet of Lorentz 494 (GOET) there is a drawing of the flower parts with indication of some measurements, which are an exact match to those presented in the protologue of T. ixioides ( Grisebach 1879) .

Note III: — Smith & Downs (1977; and later works based on these authors) cited as the holotype for Tillandsia ixioides the gathering Lorentz (Flora entreriana) 1196 (B), thus making a potential inadvertent lectotypification. Lorentz 1196 consists of material of the same entity that Lorentz 494 (under our point of view), and its label also includes “Concepción del Uruguay ” as the place of the collection. However, the other information included on Lorentz 1196 labels: “in algarrobenwalde am ufer des Gualeguaychú, Estancia Galarza [in Algarrobo forest at the shore of the Gualeguaychú, Estancia Galarza]” is not consistent with that included in the protologue (“raro in fruticetis ripariis”; Grisebach 1879). Therefore, and with no further evidence that Grisebach based his original description in Lorentz 1196 also, we discard this gathering as type material for T. ixioides , and consequently Smith & Downs (1977) inadvertent lectotypification is ineffective.

Note IV— Shortly before Tillandsia ixioides was validly published by Grisebach (1879), Lorentz (1878) published “ Tillandsia ixioides ” as the designation for a new species. In that work, the designation is clearly ascribed to Grisebach and the information cited under the name is in accordance with the information on the labels of the gathering Lorentz 1196 (e.g. B100243515). Nevertheless, Lorentz (1878) did not provide a description nor diagnosis or illustration with analysis, or reference to a previously published one, and therefore the designation was not validly published (ICN Art. 38.1, Turland et al. 2018). It should be noted that Lorentz (1878) stated that the new species contained in this publication would be later described by Grisebach, which did happen a year later in Symbolae ad Floram argentinam ( Grisebach 1879).

Note V: — Tillandsia meridionalis was published by Baker (1888), who stated in the protologue “Hab. Uraguay [ Uruguay], Tweedie!”. There is a specimen hosted at K (K000308666) whose morphological characteristics and data on the label agree with the information on the protologue of T. meridionalis . Further, “ Tillandsia meridionalis ” is written on the sheet in Baker’s handwriting ( Till & Luther 1995). Smith & Downs (1977) cited this specimen as the holotype and therefore performed an inadvertent lectotypification.

It is worth mentioning that specimen K000308666 was identified by C. Mez in 1894 as T. ixioides , as also did W. Till almost a century later in 1983, when he signaled it as the type for T. meridionalis (annotations on the type sheet). Later on, in 1995 W. Till revisited his identification and stated that this specimen did not correspond to T. ixioides and could be a synonym of T. aëranthos or even belong to a different entity ( Till & Luther 1995). The determination of the identity of this specimen is not easy as it is represented just by the upper leaves of one individual and its inflorescence. However, based on the fact that the material is from Uruguay, we do consider T. meridionalis as a synonym of T. ixioides (as does Gouda 2010), as it is the only Tillandsia species within the country that distinctively shows pedicellate flowers, or spirally arranged flowers in combination with basal floral bracts without foliaceous apex (characters easily observable in the type specimen).

It should be noted that the name Tillandsia meridionalis has been misapplied by several authors ( Castellanos 1945, Smith 1972, Smith & Downs 1977, Zuloaga et al. 2019) when referring to what is now known as T. recurvifolia ( Till & Luther 1995) .

Representative specimens examined:— URUGUAY. Artigas: Cuareim ( Santa Rosa ), 24 September 1901, Berro 1657 ( MVFA) . Maldonado: Cerro de los Leones , 31 October 1910, Berro 5961 ( MVFA) . Rio Negro: Farrapos, 22 October 2011 (cultivated in Montevideo, pressed 12 November 2012), Rossado 84 ( MVFA) . Salto: Arroyo Espinillar, cerca Villa Constitución, 26 December 2017, Rossado et al. 495 ( MVFA) . San José: Sierra de Mahoma, 27 October 1940, Rosengurtt B-3171E ( MVFA) .

Figures

FIGURE 4. Phenology of Tillandsia species occurring in Uruguay. Divisions within each month indicate intervals of approximately 10 days, identified as the beginning, middle, and end of each month. Tillandsia loliacea is excluded due to lack of enough information. * Denotes species with cleistogamous flowers. ** Denotes species with insufficient data to ascertain the timing of fruiting or fruit dehiscence.

FIGURE 5. Illustrations of floral parts. A. Tillandsia aëranthos. B. T. ixioides. C. T. × marchesii. D. T. stricta. E. T. recurvifolia. F. T.geminiflora. From left to right: floral bract (stretched view), sepals (stretched view; adaxial sepals to the right), petal (stretched view), and pistil.A based on Rossado et al. 29 (MVFA; floral bract and sepals) and Haretche 807 (MVFA; petal and pistil); B based on Rossado et al. 455 (MVFA); C based on Rossado 463 (MVFA); D based on Rossado et al. 200 (MVFA; floral bract, petal and pistil) and Rossado 412 (MVJB; sepals); E based on Rossado et al. 366 (MVJB); F based on Rossado 17 (MVFA; floral bract, petal, and pistil) and MVFA 440 (sepals).

FIGURE 23. Tillandsia ixioides. A. Habitat; Sierra de Mahoma (San José, Uruguay). The big clump bearing inflorescences on middle ground corresponds to T. ixioides while the isolated plants on both background and foreground correspond to T. xiphioides. B. Flowering plant on Butia yatay in a palm forest near the Guaviyú River (Paysandú, Uruguay). C. Habit (near Mocoretá, Entre Ríos, Argentina). D. Habit (near the Román Grande River, Río Negro, Uruguay). E. Inflorescence. F. Close up of the fertile part of the inflorescence. G. Front view of the flower. H. Lateral view of the flower (floral bract removed). I. Detail of immature capsule. Photos by Andrés Rossado.

FIGURE 24. Distribution map of Tillandsia ixioides in Uruguay and neighboring areas.