Systematics and Distribution of Hexalectris spicata var. arizonica (Orchidaceae)

Paul M. Catling and Victor S. Engel

Abstract

The new combination Hexalectris spicata var. arizonica , based on Corallorhiza arizonica , is provided for the auto-pollinating race of H. spicata . Variety arizonica differs from var. spicata by the smaller flowers with smaller keels less than 0.7 mm high on the central veins of the lip, by the column 12-14 mm long which is without a rostellum, and by the perianth being connivent or only slightly spreading. Variety arizonica is known from Arizona, Texas and northern Mexico where it occurs in juniper, pine and oak forests over limestone. The possible origin and relationships of var. arizonica are discussed and a description, illustrations and a distribution map are included.

METHODS -- A description was prepared on the basis of both the living material from the Dallas and Austin areas of Texas and material found in several herbarium collections (see citations). The herbaria examined included AMES, BAYLU, F, MICH, NY, SEL, SMU, SR, TEX, TUC, and US (acronyms from Holmgren et al., 1990). Pollen from material collected in Dallas and Austin was stained to facilitate determination of viability using a method described by Owczarzak (1952).

RESULTS AND DISCUSSION -- Nomenclature -- There is no specimen of Arethusa spicata Walter in the Fraser folio of Walter's herbarium fragments at BM (Blake, 1915; Fernald & Schubert, 1948, pers. obs. of photos at GH). There is, however, little doubt that the name accompanying Walter's description is correctly applied to the plant currently known as Hexalectris spicata , due to his reference to a succulent leafless plant, about 45 cm tall (" sesquipedale ") and with yellow petals having purple lines, etc. Arethusa spicata was undoubtedly collected in eastern North America, but examination of herbarium material from throughout North America suggested that the auto-pollinating taxon is confined to the southwestern United States (Fig. 1). Thus it seems very unlikely that Walter's plant was the autopollinating race.

Similarly Nuttall's Bletia aphylla , while clearly a synonym of Hexalectris spicata based on examination of the type at PH and on his reference to a leafless plant from Carolina with brownish-purple, spurless flowers, is not likely to apply to the auto-pollinating race since it was based on material in the Muhlenberg and Baldwin herbaria, thus probably also having originated in the eastern United States. There is a specimen in the Muhlenberg collection at PH, but it is without flowers or any label information. Another sheet in the general type collection at PH is labeled " Bletia aphylla " in Nuttall's hand. This sheet, appropriately considered to be the type of Bletia aphylla , has three stem portions and two flowers. Examination of one of these flowers confirmed that it is the outbreeding race of H. spicata , the column having a rostellum and the lip with the highest keels 0.8 mm high.

Unlike the two preceding names, Watson's Corallorhiza arizonica was described from material collected in the southwest and examination of the type material (NY) suggests that this name is in fact based on the auto-pollinating race in question. Although the type is in poor condition, there is affixed to the sheet a packet with whole flowers and floral parts, and this material (Fig. 2) is adequate to establish that the specimen had a column lacking a rostellum and relatively small flowers with lips having relatively low keels 0.4-0.6 mm high, thus differing from the widespread Hexalectris spicata , but corresponding to the confusing material from Austin and Dallas, and the herbarium material from various parts of the southwest. Watson did not compare C. arizonica with other species of Hexalectris or Corallorhiza.

Basis for variatal rank -- There is currently insufficient evidence in terms of pollen and seed sterility (or other evidence) to consider the auto-pollinating plants as spontaneous hybrids. A small sample of pollen from Dallas plants did appear abnormal in having a high proportion of cells with collapsed cytoplasm. However, there was substantial variation in the proportion of this abnormality and there was overlap in proportion with pollen from flowers of H. nitida and H. spicata . Flowers from three plants from Austin did not have a higher percentage of abnormal pollen than putative parents collected nearby. Additionally, H. spicata var. arizonica is widespread, well established and evidently fully capable of reproduction by auto-pollination. Its close resemblance to H. spicata var. spicata suggests that it is an auto-pollinating race derived from the latter. It may also have originated from H. revoluta , but this seems unlikely because the latter taxon has a relatively longer lip with relatively longer lateral lip lobes (Correll, 1941), and extensive changes in lip shape, other than those associated with peloria, are unnecessary for the development of an auto-pollinating race. Furthermore the auto-pollinating race derived from H. revoluta may be H. nitida , a taxon which includes very similar auto-pollinating and outbreeding races.

The concept of varietal rank employed here involves a very close relationship with another taxon, type typical variety, but a clear discontinuity in a few characters and bimodality in other characters associated with the discontinuity. This concept has given rise to the suggestion of treating auto-pollinating races of orchids as varieties of outbreeding species (Catling, 1990). This has the advantage of reflecting a close relationship but at the same time maximizing the descriptive value of the classification by using the discontinuity to draw attention to probable differences in patterns of variation, selection pressure, and related phenomena, at the same time avoiding misconceptions that sometimes arise regarding the significance of structural modification (e.g. Catling, 1990).

Evidence for a hybrid origin -- Auto-pollinating races derived from different species may appear very similar due to response to a similar auto-pollination selection pressure (Catling, 1990). Thus it can sometimes be difficult to determine from what species an auto-pollinating race is derived. Hybridization also involves morphological convergence and the possibility of auto-pollinating hybrids can add another dimension of complexity to relationship and rank. Hexalectris spicata var. arizonica is a case in point. Although varietal rank is here proposed, a review of the evidence for a possible hybrid origin is appropriate.

The lip of H. spicata var. arizonica has the purple veins and purplish midlobe characteristic of H. spicata var. spicata but with some pale yellow centrally as in H. nitida . The petals have veins prominently marked unlike H. nitida but not to the extent of H. spicata var. spicata . The petals and sepals are of a paler color than those of H. spicata var. spicata and the petals in their pale pink or whitish tips are more like those of H. nitida . Consequently, the flower color of H. spicata var. arizonica combines features of both H. nitida and H. spicata var. spicata .

The lip does not have midveins as prominent as in H. spicata var. spicata , but much more like those of H. nitida . The overall lip shape however is more like that of H. spicata var. spicata , being relatively broad with lateral lobes not extending far beyond the base of the midlobe. The lack of a rostellum promoting auto-pollination is a feature shared with the wide-spread race of H. nitida . The size of floral parts is intermediate between the relatively large-flowered H. spicata and the relatively small-flowered H. nitida .

Two other lines of evidence also suggest a hybrid origin. Firstly, at the Dallas site, H. spicata var. arizonica occurs within an area of oak-juniper woodland that is occupied by both H. nitida and H. spicata var. spicata , whereas at the Austin site it is accompanied by H. nitida . Secondly, H. spicata var. arizonica flowers in June and July, later than H. spicata var. spicata which flowers primarily in May (occasionally to early June) in eastern Texas, but earlier that the majority of plants of H. nitida which flower from late June through July.

Since H. revoluta is very similar to H. nitida , there is a possibility that the new taxon is a result of a cross of H. revoluta and H. spicata . This seems unlikely, however, because larger flowers with relatively longer lips having relatively longer lateral lobes would be expected from such a cross.

Hexalectris warnockii seems unlikely to be involved in a hybrid origin of H. spicata var. arizonica because the latter is without characteristics of H. warnockii , such as a lip with a relatively short midlobe, distally prominent and irregular or crisped keels, and relatively narrow and acute sepals and petals.

Terrestrial saprophytic (possibly parasitic), glabrous herbs. Rhizomes with circular sheathing bracts, the scars of which separate the rhizomes into short segments. Leaves reduced to sheathing scales. Inflorescence a few- to many- flowered raceme. Flowers Column straight or curved; anther terminal, with four waxy pollinia in each cell; pollen yellow, in tetrads; stigma subterminal. Capsules elliptic, pendent. Seeds 0.5-0.7 mm.

Species 7, mostly Mexican, five species occurring in the United States, where all but the wide-ranging H. spicata are confined to the states of Texas and Arizona.

Hexalectris spicata (Walter) Barnhart

Rhizome segments 0.3-1 mm long. Stems yellowish-brown, pinkish-red, pinkish-yellow, pinkish-white, or purple, (12) 15-70 (80) cm tall, 4-8 mm in diameter, with (2) 3-5 (6) sheathing bracts 0.5-2 cm. Inflorescence a many-flowered raceme 8-30 cm long, floral bracts lanceolate, 3-10 mm. Flowers 5-25, with pedicillate ovaries 10-16 (20) mm long' sepals and petals yellowish- or purplish-brown, sometimes very pale or whitish, with or without a pinkish cast, reflexed or remaining closed, the petals with prominent purple or brown veins; dorsal sepal lanceolate or oblong, 15-25 x 2.7-8 mm; lateral sepals falcate-lanceolate, 13-20 x 4-9 mm; petals elliptic-falcate or obovate-falcate, 14-21 (23) x 4-9 mm; lip white, yellowish-white or purplish-white, with or without the veins and keels purple or brown, 12-20 x (9) 10-15 (16) mm, with 5 keels prominent at the base of the midlobe and including the midvein, the highest of the keels 0.4-1 mm above the lip surface, becoming obscure apically but with the midvein somewhat inflated or apically keeled, lip three-lobed with the lateral lobes extending 0-1/3 the length of the midlobe; column white or creamy-white above, flat or arcuate, dilated at the apex and winged or not below the apex, 11-18 mm, rostellum present or absent. Capsule 16-30 x 8-20 mm, the stalk 8-20 mm.

1a. The 5 central veins of the lip with their highest keels raised (0.4) 0.7-1 mm above the lip surface; column with a rostellar flap separating the pollen masses from the stigmatic surface ... H. spicata var. spicata

1b. The 5 central veins of the lip with their highest keels raised 0.4-0.7 mm above the lip surface; column without a rostellar flap separating the pollen masses from the stigmatic surface ... H. spicata var. arizonica

1a. Hexalectris spicata (Walter) Barnhart var. spicata , Torreya 4: 121. 1904.
Arethusa spicata Walter, Fl. Carol. 222. 1788. TYPE: (Holotype, n.v.).
Bletia aphylla Nuttall, Gen. 2: 194. 1818. TYPE: (Holotype, PH!).
Hexalectris aphylla (Nuttall) Rafinesque, Flora Tellur. 4: 48. 1838.
Corallorhiza spicata (Walter) Tidestrom in Tidestrom & Kittel, Fl. Ariz. N. Mex. 733. 1941.

DISTRIBUTION: United States (Ala., Ariz., Ark., Fla., Ga., Ill., Ind., Ky., La., Miss., Mo., N.Mex., N.C., Ohio, S.C., Tenn., Tex., Va., W.Va.) and northern Mexico (Coahuila).

HABITAT: Calcareous sandy or organic souls in oak, hickory or conifer woods.

FLOWERING TIME: April in the south to August in the north.

1b. Hexalectris spicata var. arizonica (S. Watson) Catling & Engel, stat et comb. nov.
BASIONYM: Corallorhiza arizonica S. Watson, Proc. Am. Acad. 17: 379. 1882. TYPE: UNITED STATES. Arizona : in rocky places on the Santa Rita Mountains, July 1881, C. G. Pringle (Holotype: NY!).

Stems pinkish-red or pale pinkish-yellow or pinkish-white, 15-70 cm tall. Flowers with sepals and petals pale creamy yellow or dull, pale pinkish-purple, with dull pinkish-red, pinkish-brown or pinkish-purple veins and with whitish margins connivent or spreading apically; dorsal sepal lanceolate or oblong, 15-18 x 2.7-4.5 mm; lateral sepals 13-15 x 4-7 mm; petals mostly creamy-white with pinkish-red or pinkish-brown veins which are most prominent toward the apex on the outside, obovate-falcate, 14-16 x 4-5 mm; lip mostly white or creamy-white, 12-14.5 x 10-12 mm, with the 5 central veins with keels 0.2-0.7 above the lip surface, the central vein also keeled toward the apex, the midlobe pinkish-purple with an irregular whitish margin and darkened pinkish-red or reddish-purple veins and pale yellowish centrally, the lateral lobes white or creamy-white with veins dark pinkish-red or reddish-purple, extending 1/5 - 1/3 the length of the midlobe; column white or creamy-white, arcuate, winged or not below the apex, 11-14 mm long, rostellum absent; anther compartments creamy-white, the anterior dorsal surface somewhat dehydrated, reddish-purple and pointed or green and fleshy. Capsule 16-18 (25) x 8-12 (15) mm, their stalks 8-12 (15) mm.

MATERIAL EXAMINED: MEXICO. Coahuila : Municipio de Muzquiz, eastern slope of Sierra de San Manuel, Rancho Agua Dulce, 30 June 1936, F. L. Wynd and C. H. Mueller 371 (US). Municipio do Villa Acuna, Sierra del Carmen, Canyon de Sentenela, Hacienda Piedra Blanca, 6 July 1936, F.L. Wynd and C. H. Mueller 530 (US).

UNITED STATES. Arizona: Cochise Co. : Chiricahua Mountains, lower Pinery Canyon at junction of Pine Canyon Road, 5 August 1958, J. McCormick 1958-434 (TUC). Pima Co. : Rincon Mountains, Rincon Peak Trail, 6,000 ft. elev., 1 August 1982, J. E. Bowers and S. P. McLaughlin R435 (TUC) Santa Cruz Co. : trail from Gardner Canyon to Baldy Springs, 6,200 ft elev. 4 July 1978, J. Kaiser 1200 (TUC).

TEXAS:Anderson Co. : 8 mi. E of Palestine, 16 July 1946, D. S. Correll 13275 (SMU). Brewster Co. : Chisos Mountains, 30 July 1932, C. H. Mueller s.n. (MICH). lower Oak Creek, basin of the Chisos Mountains, 5,200 ft elev. 7 Sept. 1964, B. H. Warnock 20469 (SR). Culberson Co. : Lost Peak, upper reaches of West Dog Canyon, 7,500 ft elev., 19 Aug. 1981, B.H. Warnock 21693 (SR). Dallas Co. : near the Dallas Nature Center on the SW side of Dallas mid-June 1988, Victor S. Engel s.n. (AMES). Palo Pinto Co. : 4 mi. NW of Mineral Wells near Turkey Creek, 2 June 1946, D. S. Correll and H. B. Correll 12339 (SMU). Travis Co. : in juniper woodland above Barton Creek, SW Austin, 19 July 1992, Victor S. Engel s.n. (AMES).

DISTRIBUTION: Arizona, Texas and northern Mexico.

HABITAT: Oak, pine, or juniper woods over limestone in rotting wood or leaf litter.

FLOWERING TIME: June and July.

COMMENTS: Most of the plants seen in the Dallas area and all of those seen so far in the Austin area have closed flowers that never open, but some plants with open flowers (Figs. 3 & 4) have been found at the Dallas site. The closed flowers tend to have more truncated perianth parts than the open flowers. In addition the anthers of the open flowers have pointed and dehydrated, purplish-red tips, whereas the anther tips in closed flowered plants examined have distended, fleshy, green tips. Both the expanded anther tips and the terminally expanded midvein may contribute to forcing the pollinaria onto the stigmatic surface as the column of closed flowers extends and buckles. Such mechanisms promoting auto-pollination have not been well documented (Catling, 1990). There may be an opportunity for cross pollination in open flowers but this has yet to be determined.

CONSERVATION STATUS: Although it may be more widespread than is currently known, Hexalectris spicata var. arizonica is nevertheless a geographically restricted taxon based on current information. Fortunately, it is protected with other native flora in some of the sites where it occurs. Any intended collecting of plants should be coordinated through local botanists or natural resource officials.

Blake, S.F. 1915. Some neglected names in Walter's Flora Caroliniana. Rhodora 17(199): 129-137.

Catling,P.M. 1990. Auto-pollination in the Orchidaceae, pp. 121-158 in J. Arditti, (Ed.), Orchid Biology, Reviews and Perspectives, V . Timber Press, Portland, Oregon.

Correll,D.S. 1941. Studies in Isochilus, Mormodes, and Hexalectris. Bot. Mus. Leafl. 10: 1-20.

Fernald, M.L., and B.G.Schubert. 1948. Studies of American types in British herbaria, part IV. Some species of Thomas Walter. Rhodora 50: 190-208, 218-233.

Holmgren,P.K., N.H.Holmgren, and L.C.Barnett. 1990. Index Herbariorum, part 1:The herbaria of the world . New York Botanical Garden. Bronx, New York.

Owczarzak,A. 1952. A rapid method for mounting pollen grains. Stain Technol. 27: 249-251.