DUDLEYA NUBIGENA (Brandegee) Britton & Rose
Abstract. D. nubigena (including D. xanti Rose) occurs from sea level to at least 1800m in Baja California south of 24°50'N. Typically, the calyx is distinctive in the genus for its deep sinuses and for the way its sepal tips stand out from the corolla. In ssp. nubigena, despite low nectar production in the one population reported, the pink-to-red inflorescence, open cymes, long slender pedicels, and red-marked corolla with rather long tube, all suggest some adaptation to hummingbird pollination. In ssp. cerralvensis Moran, of Isla Cerralvo, the less colored inflorescence, denser cymes and yellow corolla suggest less adaptation.
Dudleya nubigena is a little-known plant of southern Baja California, Mexico - not mentioned in the only flora (Wiggins 1980). It is nearly confined to the Cape Region, the geographically and biologically distinctive lower seventh of the peninsula, below La Paz, which apparently was an island during the Pliocene. There it is the least rare Dudleya in a region where Dudleya is seldom seen: it grows in a dozen known places, from Arroyo San Antonio de Coyote and Isla Espiritu Santo, north of La Paz, to Cabo San Lucas, at the southern tip, and from sea level to at least 1800 meters -with a weak subspecies on Isla Cerralvo.
T. S. Brandegee, first botanist to explore the mountains of the Cape Region, made the first of his six trips in January 1890, from Todos Santos into the Sierra de la Laguna (Moran 1952). He found D. nubigena flowering on rocks of the summits of the Sierra and named it the next year (Brandegee 1891). Though "nubigena" suggests cloudy summits, Brandegee later identified with D. nubigena plants from sea level at San Jose del Cabo (Brandegee in 1892, UC).
J. N. Rose (Britton and Rose 1903: 23) separated the lowland plant as D. xanti, named for L. J. Xantus de Vesey, who first found it. From 1859 to 1861, while a tidal observer at Cabo San Lucas for the U.S. Coast Survey, Xantus also traveled about and collected natural history specimens, some of the first from this region (see Zwinger 1986). Among them was a specimen of Dudleya from Cabo San Lucas. In his account of the Xantus plant collections, Asa Gray (1861) reported this one as Echeveria farinosa Lindl. [=Dudleya farinosa (Lindl.) Britt. & Rose], which is a very different plant, of coastal central California to southern Oregon. Rose named the Xantus specimen type of D. xanti.
From the key and descriptions (Britton and Rose 1905: 34, 42), D. xanti would seem to be a smaller plant than D. nubigena, with shorter leaves, floral stems, and pedicels, and to differ further in having obtuse calyx lobes, shorter than the corolla tube. However, the several specimens now at hand do not sort out on this basis into two species, and I include D. xanti in D. nubigena. In particular, the height of 4-5dm that Britton and Rose quoted for the floral stem of D. nubigena is unsubstantiated and questionable (Moran 1987: p. 191).
Uhl and Moran (1953: 495) reported chromosome counts of n=17 for three collections of D. nubigena, including a topotype of D. xanti. Dr. Uhl tells me of three more counts: A. Carter 5429, n=17; Moran 7135, n=17 prob.; and Moran 11798, n=17. He also found n=17 for the type collection, and for a topotype, of D. nubigena cerralvensis (Moran 1969: 271). These eight collections represent six localities, from the northernmost known to the southernmost and from sea level to 1800m. With 17 the basic number for Dudleya, D. nubigena is thus diploid. The calyx of D. nubigena generally differs a little from the rather uniform pattern of the other species, making the flower slightly more Echeveria-like (see especially figs. 2, 9). The somewhat U-shaped sinusescommonly are so deep that the calyx disk, or concrete part, is narrower than the corolla base. And in an-thesis the segments usually are erect or somewhat ascending, with the tips 1-2mm from the corolla, whereas in all other species they are almost invariably appressed. The tidy taxonomist is disappointed to find, however, that D. nubigena is not entirely consistent in these and other generally distinctive characters.
Fig. 1. Flowering plant of D. nubigena nubigena, from Arroyo del Salto (M7135); 13 December 1959; (x 0.33).
With expansion of the rosette, aging floral stems of D. nubigena decline from erect to spreading; meanwhile, the pedicels continue to bend upward, tending to keep the old flowers, or fruits, erect. As a result, if a detached old floral stem on an herbarium sheet is taken to have been erect, the flowers may appear, deceptively, to have been spreading.
Fig. 2. Cyme of D. nubigena nubigena; same plant and date as fig. 1. (x 1.5).
Dudleya nubigena nubigena
Although the type is from the summits of the Sierra, my knowledge of D. nubigena nubigena comes mainly from eight living collections from lower elevations. The lowland plant, including D. xanti, has solitary subsessile rosettes of mostly farinose leaves, which in winter send up erect floral stems with erect cymes (figs. 1, 3). The mostly 2-3 cincinni are circinate at first but straighten and become erect to spreading as the flowers open (figs. 1, 2, etc.). At least the first flowers, on their long slender pedicels, are erect throughout anthesis, and some later ones become erect during anthesis. The corolla is orange to coral red and about 8-10mm long (fig. 2).
Although several collections besides the type show D. nubigena nubigena in the Sierra de la Laguna, I had only one living one (M11798), brought by George Lindsay in 1961 from a rocky canyonside just below La Laguna. His plants, flowering several times in San Diego (fig. 4), were generally like lowland plants and like them were diploid; and I noted no consistent difference. However, they varied in some characters that seem more constant in lowland plants. Thus the caudex in some plants was branching (as also in one plant of Gentry 4413, from La Laguna). Young floral stems were sometimes erect from the start, as in lowland plants, but were sometimes at first nodding; and the flowers sometimes opened erect but sometimes were pendent (fig. 5). Unfortunately, I did not keep close enough track to be sure that the plant of fig. 5 was, as I suppose, a different individual from that of fig. 4. In any case, this is a population the tidy taxonomist would rather forget: it doesn't seem to fit, and it spoils some otherwise distinctive characters. As he would say, it needs more study. Herbarium specimens do not show whether other Sierran populations, including the type, vary similarly. So there is more to learn.
Fig. 3. Flowering plant of D. nubigena nubigena, from NW of Cerro Mechudo (E. Larson ex A. Carter 5429); 18 February 1971 (x 0.65).
At Arroyo del Salto, according to An-netta Carter (@2588), D. nubigena nubi-gena was called "mescalito", and to it was ascribed some value in the treatment of rabies. For a nearby locality, southeast of La Paz, Robert Peters (60) reported the same common name - which is diminutive for "mezcal", used for various species of Agave.
Dudleya nubigena cerralvensis
Dudleya nubigena ssp. cerralvensis Moran (1969) is known only from one spot due east of La Paz, on Isla Cerralvo in the Gulf of California. There it grows on a north-facing granitic cliff in a canyon on the southwest side of the island. Exploration in several other parts of the island suggests that it is not common but very likely is confined to such relatively cool and shady places, which must be scarce.
The ssp. cerralvensis is very close to ssp. nubigena. Its cymes tend to branch more and so to be denser (figs. 6, 9). And it differs conspicuously in its yellow flowers and its green or slightly glaucous but not farinose leaves. These differences do not seem impressive considering how such characters vary within other species or even populations. However, the question is how well these apparent differences will hold up with further field study.
It is noteworthy, though I don't know how significant, that ssp. cerralvensis, so far as known, is endemic to an island. The islands of the Gulf are rather close to shore, and the rate of endemism among their vascular plants is low (Moran 1983). Of 143 species known from Isla Cerralvo, Ferocactus diguetii (Weber) Britt. & Rose is endemic to Cerralvo and seven other islands of the Gulf, and two other species, aside from the Dudleya, are endemic to Cerralvo alone: Mammillaria cerralboa (Britt. & Rose) Orcutt and Marina occulata (Rydb.) Barneby. And of 168 species known from Isla Espiritu Santo, where ssp. nubigena makes its only insular appearance, one species (Opuntia bre-vispina H. Gates) and one variety (Dryopetalon crenatum var. racemosum Rollins) appear to be endemic. However, Isla Cerralvo is separated from the peninsula by a channel 11km wide and ca. 500m deep and is supposed to have been isolated since the Pliocene; whereas Isla Espiritu Santo is separated by a channel 6km wide but only ca. 10m deep and probably became separated much more recently, in the Pleistocene (Gastil, Minch, and Phillips 1983).
Now let me digress, with the excuse of using D. n. cerralvensis as an example. Although Moran (1969) may have said it better, I'll put this in my own words for those who may have overlooked his learned but obscure publication: a plant of Dudleya brought into cultivation may look more different from its former self in the wild than plants of two different species, grown side by side, sometimes look from each other. For example, the difference between the spare plants of D. n. cerralvensis photographed in the field (fig. 6) and a bloated one flowering unseasonably in San Diego (fig. 7) is, to say the least, striking. Actually, plants of 9542 flowering in San Diego at the normal time of year (fig. 8) differed less from the wild plants. I don't mean to single out D. n. cerralvensis for special comment but am wandering from the immediate subject here to make this general point simply because figs. 6 and 7 show it so well. The plasticity that the two extremes suggest is a real source of difficulty for taxonomists working with such succulent plants as Dudleya; and herbarium specimens from greenhouse plants are sometimes especially hard to harmonize with wild plants.
Surveying reproductive systems of Dudleya subgenus DUDLEYA, Levin and Mulroy (1985) studied one collection of D. nubigena, from Cabo San Lucas - a topotype of D. xanti, as it happens. They plotted the phenology of its nectar production as representative for the sub-genus (their fig. 3). They found high auto-fertility and low nectar production, which together place this collection in their Group I (their fig. 7), with D. cymosa (Lem.) Britt. & Rose (3 subspecies) and D. linearis (Greene) Britt. & Rose. They thought the plants of Group I probably were pollinated mainly by insects such as bees and flies. However, Dr. Levin writes me that red-flowered populations of D. cymosa cymosa are visited also by hummingbirds, the level of visitation varying considerably from one time and place to another. They supposed that environmental unpredictability and pollinator unreliability might explain the apparent trend in Group I towards autogamy.
In some species, Levin and Mulroy calculated maximum energy content of nectar, and measured auto-fertility, for more than one collection. Two collections of D. arizonica Rose, and likewise two of D. cultrata Rose, differed in maximum energy content of nectar by a factor of ca. 3 (their table 1). One collection of D. brittonii Johansen showed ca. 2 percent seed set by auto-fertilization, and another ca. 31 percent (their fig. 6). In fact, although one collection of D. brittonii falls in Group II, another seems to fall in Group I. (Dr. Levin writes me that both bees and hummingbirds visit D. brittonii.) These examples show that species, or what are regarded as species, are not necessarily consistent throughout in such traits. In fact, we might expect the values in different populations to change as relationships with pollinators evolve. Thus what Levin and Mulroy found for D. nu-bigena from a semidesert shore is not necessarily true for plants of other places, such as high in the Sierra.
Fig. 4. Flowering plant of D. nubigena nubigena, from La Laguna (G. Lindsay M11798); 3 February 1962 (x 0.6).
Grant and Grant (1968) saw hummingbirds visiting the flowers of D. cy-mosa minor (Rose) Moran and D. lanceo-lata (Nutt.) Britt. & Rose. They mentioned D. cymosa minor among some plants visited and pollinated by hummingbirds and by some other agents, usually bees; and they wrote that the flowers show "some of the characteristics of hummingbird flowers in a rudimentary state of development .... Here we may be witnessing a transitional stage in the evolutionary development of hummingbird flowers." Very likely this is as true of other species seen visited by hummingbirds and of others not known to be visited. Moldenke (1976) thought hummingbirds important pollinators of Dudleya, not saying for what species but perhaps implying for many (presumably of subgenus DUDLEYA!).
Fig. 5. Cyme with pendent flowers; same collection as fig. 4, 25 March 1964 (x 1.3).
What can we say of D. nubigena, whose visitors and pollinators are unknown? It does not have the very short corolla tube or usually the dense cyme of D. cymosa, in Group I. And in some traits it approaches the pattern elucidated by the Grants (1968) for hummingbird-pollinated plants of western North America -a pattern more closely approached by D. anthonyi Rose and D. pulverulenta (Nutt.) Britt. & Rose (Levin and Mulroy 1985, Moran 1987). The cyme, of few and often spreading cincinni, may be fairly open (fig. 1), making the flowers easily accessible. The pedicels are rather long and slender; in fact, the longest pedicel I have seen in Dudleya, 4cm long, was in a cultivated plant of this species. In one montane population, flowers may sometimes be pendent. The usually red and yellow or orange corolla might be visible to some insects as well as to birds, but the variously pink-to-red floral stems, cyme branches, pedicels, and calyx in some populations seem to offer a target more uniquely for birds.
Fig. 6. Freshly collected plants of D. nu-bigena cerralvensis; part of type collection (M3618), 4 April 1952 (x 0.175).
Although the corolla of D. nubigena is small (mostly 8-l0mm long), the tube is fairly long in proportion (2.5-5.5mm). Levin and Mulroy (1985: figs. 4, 5) found both volume and maximum energy level of nectar in Dudleya subgenus DUDLEYA strongly correlated with corolla tube length. However, in their collection of D. nubigena both nectar values are well below average, even for the relatively short (3.2mm) corolla tube of that collection. We may wonder whether flowers of other populations produce enough nectar to interest a hummingbird. Do hummingbirds ever visit and pollinate flowers of D. nu-bigenal If so, do populations differ in amount of hummingbird visitation and pollination, and are they evolving differently as a result? Are D. nubigena and D. xanti, for example, diverging? Has the population of D. xanti sampled by Levin and Mulroy lost some of the nectar production implied by the tube length of D. nubigena as a whole or even some of that implied by the shorter tube length of that collection? Or has it ever produced more nectar? Thus with no field data on pollinators of D. nubigena, I have many questions but no real answers. All things considered, however, at least some hummingbird pollination seems likely.
The ssp. cerralvensis seems to have fewer hummingbird traits than does ssp. nubigena. Not only are the flowers yellow, but the inflorescence is less colored and the cyme tends to be denser—though both the pedicels and the corolla tube still are about as long. If the species is trending towards hummingbird pollination, is ssp. cerralvensis perhaps lagging behind, retaining more of its original characters; and if so, why? For there is no lack of hummingbirds on the island: according to Banks (1963), Costa Hummingbird (Calypte costae) is widely distributed on Cerralvo, and Xantus Hummingbird (Hylocharis xantusii), which prefers higher parts of the island, seems to be only slightly less abundant.
Fig. 7. Flowering plant of D. nubigena cerralvensis; topotype (M9542), 28 October 1962 (x 0.65).
Dudleya nubigena (Brandegee) Britt. & Rose
Caudex rarely 7cm high, 1-1.5 (-2.5)cm thick, commonly unbranched, in montane plants sometimes few-branched. Rosettes thus subsessile and mostly solitary, flat-tish, 4-18cm wide, of 10-22 crowded leaves, or to 30 in cult. Rosette leaves green to (commonly) farinose, triangular-lanceolate to oblong-obovate, acute to acuminate, often apiculate, 3-10 (-15)cm long, 1.5-3 (-3.5)cm wide, 2-4 (-6)mm thick, flattish or mostly channeled ven-trally, the base l-2cm wide, the margins acute near base, obtuse to rounded above. Floral stems in winter, mostly erect from the start, rarely with tips at first nodding (montane), in age declining with expansion of rosette, 1-3 (0.5-4)dm tall (to cyme), 1.5-5mm thick, often pink to red and commonly glaucous, bare in lower 2-12cm, with 5-23 leaves above, the leaves ascending, triangular-ovate to -lanceolate, acute, the lower 4-30mm long, 3-12mm wide. Cyme commonly pinkish and glaucous, of 2-3 branches that bifurcate 0-2 times, the cincinni 1-ca. 8, circinate becoming erect to spreading, 2-15 (-30)cm long, with 4-24 flowers, in cultivation the flowers opening at intervals of 2-5 days (average 3+), remaining open 5-6 days. Pedicels often pinkish, the first mostly erect through anthesis, rarely pendent (montane), 8-25mm long (to 40mm in cult.), 1-1.5mm thick at base, thicker upward, later ones erect through anthesis or spreading at least at first, shorter than first, all +/- erect in fruit, bending up if floral stem declined. Flowers December through April, or in cultivation as early as October. Calyx green to red, 3-6mm long, 4-6 (-8)mm wide, subtruncate to rounded below, the disk 4-5mm wide, the sinuses sometimes U-shaped at anthesis, later V-shaped to closed, the segments commonly erect or ascending at anthesis, with tips 0.5-2mm from corolla, later ap-pressed, uncommonly appressed at anthesis (montane), deltoid, acute, 2-5mm long, 2-3.5mm wide. Corolla pyramidal with sides nearly flat, 7-10 (-12)mm long, elongating by 1-1.5mm during anthesis, 4-6mm wide at base and thus often wider than calyx disk, 2.5-4mm wide at apex, yellow to mostly orange or coral red, the keels glaucous; tube 2.5-5.5 (mostly 3.5-4.5)mm long; segments oblong, obtuse to broadly acute, 1.5-2.5mm wide. Filaments subterete, light yellow, 5-8mm long from corolla base, adnate 2-4mm; anthers oblong-lanceolate, 1.5-2mm long, 0.6-0.7mm wide, light yellow. Nectar glands white, ca. 0.25mm high, 1-1.5mm wide, 0.3-0.5mm thick radially. Gynoecium 5-9mm high, 2.5-3mm thick, the pistils erect, appressed, connate ca. 0.5-l.Omm, greenish, the ovaries oblong, tapering rather abruptly to yellowish or reddish styles 1-1.5mm long; ovules 50-110, ca. 0.5mm long and 0.2mm thick. Seeds brown, narrowly ovoid, acute, longitudinally striate, ca. 0.5-0.7mm long. Chromosomes: n=17 (diploid).
Fig. 8. Flowering plant of D. nubigena cerralvensis; topotype (M9542), 15 March 1964 (x 0.4).
Dudleya nubigena (Brandegee) Britt. & Rose ssp. nubigena
Cotyledon nubigena Brandegee, Proc. Calif. Acad. Sci. ser. 2, 3:136. 1891.
Dudleya nubigena Britt. & Rose, Bull. N.Y. Bot. Gard. 3(9): 22. 1903.
D. xanti Rose in Britt. & Rose, op. cit. 23.
Cotyledon xanti Fedde, Bot. Jahresber. Just. 31*: 826. 1904.
Echeveria nubigena Berger in Engler, Nat. Pflanzenfam. ed. 2, 18a: 479. 1930.
E. xanti Berger in Engler, loc. cit.
Rosette leaves glaucous to mostly farinose; floral stems, cyme branches, pedicels, and calyx variously +/- pink to red; cyme branches mostly simple, the cincinni mostly 2-3; corolla yellow at very base, marked with red above, especially on keels or nearly throughout, and thus appearing orange to coral red.
TYPE COLLECTION: Upon rocks of the summits of the Sierra de la Laguna, Baja California Sur, Mexico, January 24, 1890, T.S. Brandegee s.n.: holotype UC124956.
DISTRIBUTION: Open rocky places or often on cliffs, Baja California Sur south of 24°50'N: from Cerro Mechudo to Cabo San Lucas, and from near sea level to at least 1800m. Collections: Aguaje de Arroyo San Antonio de Coyote, NW of Cerro Mechudo, 600m, E. Larson ex A. Carter 5429 (SD, UC); steep slope beneath north cliffs, head of arroyo above Ense-nada Ballena, Isla Espiritu Santo, 300m, Moran 9649 (SD), ex hort. (SD); steep cliffs, Arroyo del Salto, E of La Paz, 150m, A. Carter 2588, "mescalito--para la rabia" (SD, UC); crevices in granite, north slope above fall, Arroyo del Salto, 100m, Moran 7135 (SD); weathered granite in canyon bottom, Los Planes road 15 mi SE of La Paz, 1800', R. Peters 60, "mescalito" (UC); Sierra de la Laguna, T.S. Brandegee "in 1889" (CAS), in 1892 (UC), M.E. Jones 24297 (PENN, POM); granite rocks, oak-pine forest, La Laguna, Sierra de la Laguna, 6000', H.S. Gentry 4413 (DS, MO); rocky canyon just below La Laguna, G. Lindsay ex Moran 11798 ex hort. (SD); La Laguna, A.J. Sloan in 1965 (SD); San Bernardo to El Sauz, Sierra de la Laguna, 3500', Nelson & Goldman 7428 (US); decomposed granite, San Bernardo, A. Carter 2686 (CU, UC); north-facing cliff at mouth of San Lazaro Canyon, 100m, Moran 7322 (SD); San Jose del Cabo, T.S. Brandegee in 1892 (GH, PH, UC, US), C. Grabendorffer in 1898 (UC), C.A. Purpus in 1902 ex hort. Berger (NY); Cabo San Lucas, Xantus 39 (type of D. xanti, US48362; GH, NY), J. Rose 16359 (NY, US), J.T. Howell in 1932 ex hort. (CAS), Moran 4145 ex hort. (CU), 7055 (SD).
Dudleya nubigena ssp. cerralvensis Moran, Trans. San Diego Soc. Nat. Hist. 15: 269, fig. 3. 1969.
Rosette leaves green or slightly glaucous; floral stems, cyme branches, pedicels, and calyx light green; cyme branches 0-2 times bifurcate, the cincinni 2-8; corolla clear yellow, not marked with red.
TYPE COLLECTION: Locally common on north-facing granitic cliffs at 120m elevation, in arroyo ca. 1 mi. upstream (east) from abandoned Ruffo Ranch, SW side of Isla Cerralvo, Baja California Sur (near 24°11'N, 109°51'W), April 4, 1952, Moran 3618: holotype SD69717; isotypes BH, DS, UC, US, etc.?
DISTRIBUTION: Known only from the type locality. Other collection: Moran 9542 (SD), ex hort. (SD).
Fig. 9. Cyme of D. nubigena cerralvensis, topotype (M9542), 21 February 1965 (x 1.3).
As usual, I am indebted to Annetta Carter, Billie Meeder, and Geoff Levin, who have patiently helped with the questions that so constantly remind me of how easy it used to be to step into the herbarium or library. Geoff also has reviewed, and made helpful suggestions for, the section on the reproductive system. And Charles Uhl has, as usual, provided the cytological information. I thank them all for their help.
Banks, R.C. 1963. Birds of Cerralvo Island, Baja California. Condor 65:300-312.
Brandegee, T.S. 1891. Flora of the Cape Region of Baja California. Proc. Calif. Acad. Sci. ser. 2, 3: 108-182.
Britton, N.L., and J.N. Rose. 1903. New or noteworthy North American Crassulaceae. Bull. N.Y. Bot. Gard. 3(9): l-45.
------ and -------. 1905. Crassulaceae. N. Amer. Fl. 22: 7-74.
Gastil, G., J. Minch, and R. Phillips. 1983. The geology and ages of the islands. In: T.J. Case and M.L. Cody, eds., Island biogeography in the Sea of Cortez, Chapt. 2, pp. 13-25. Univ. California Press, Berkeley &c.
Grant, K.A., and V. Grant. 1968. Hummingbirds and their flowers. Columbia Univ. Press, New York.
Gray, Asa. 1861. Enumeration of a collection of dried plants made by L.J. Xantus at Cape San Lucas &c. in Lower California between August 1859 and February 1860 and communicated to the Smithsonian Institution. Proc. Amer. Acad. Arts. Sci. 5: 153-173.
Levin, G.A., and T.W. Mulroy. 1985. Floral morphology, nectar production, and breeding systems in Dudleya sub-genus Dudleya (Crassulaceae). Trans. San Diego Soc. Nat. Hist. 21: 57-70.
Moldenke, A.R. 1976. California pollination ecology and vegetation types. Phytologia 34: 305-361.
Moran, R. 1952. The Mexican itineraries of T.S. Brandegee. Madrono 11: 253-262.
------ 1969. Twelve new Dicots from Baja California, Mexico. Trans. San Diego Soc. Nat. Hist. 15: 265-295.
------ 1983. Vascular plants of the Gulf Islands. In: T.J. Case and M.L. Cody, eds., island biogeography in the Sea of Cortez, Append. 4.1, pp. 348-381. Univ. California Press, Berkeley &c.
------ 1987. Dudleya rigida Rose. Cact. Succ. J. 59: 187-194.
Uhl, C.H., and R. Moran. 1953. The cytotaxonomy of Dudleya and Hasseanthus. Amer. J. Bot. 40:492-502.
Wiggins, I.L. 1980. Flora of Baja California. Stanford Univ. Press.
Zwinger, A.H. 1986. The letters of John Xantus to Spencer Fullerton Baird, from San Francisco and Cabo San Lucas, 1859-1861. Introduction, notes, and illustrations by Ann H. Zwinger. Dawson's Book Shop, Los Angeles.
Cactus & Succulent Journal of America