2010 NORTHEASTERN NATURALIST 17(4):659–666 Croton monanthogynus and Crotonopsis elliptica (Euphorbiaceae) in Ozark Rock Outcrop Communities: Abundance, Soil Depth, and Substrate Tolerance Stewart Ware* Abstract - The ecologically similar species Crotonopsis elliptica and Croton monanthogynus were described by Steyermark in his Flora of Missouri as inhabitants of acid rock outcrop communities (sandstone, granite, chert) in the Ozarks but not of limestone outcrops. However, field observations and transect data confirm that C. monanthogynus can be abundant in Ozark limestone outcrop communities, codominating with Isanthus brachiatus and Heliotropium tenellum in the 5 cm to 11 cm soil depth zone, while almost never occurring on sandstone outcrops. Greenhouse experiments revealed that C. elliptica was healthy on sandstone soil, but stunted and sickly on limestone outcrop soil, consistent with its absence from calcareous substrates. In contrast, C. monanthogynus was healthy on both soils, and had greater growth on limestone than on sandstone soil. Contrary to Steyermark’s characterization, C. monanthogynus grows abundantly on limestone soils, but is rare or absent on Ozark sandstone outcrops, not because of substrate intolerance, but probably because of competing species there. Introduction In portions of eastern North America where bedrock is at or near the soil surface, the shallow soil dries out quickly during the warm growing season and cannot support trees. Instead, natural vegetation is dominated by lichens, mosses, and herbaceous vascular plants (Baskin and Baskin 1988, Ware 2002). Extreme summer drought, wintertime soil saturation due to impermeable bedrock, and extreme soil chemistry caused by the closeness of the bedrock result in these rock outcrop habitats being different from treeless areas on deeper soils. Therefore, much of the usual weedy flora of roadsides, fields, and pastures are uncommon in natural rock outcrop communities; instead, the herbaceous vegetation is often dominated by plant species particularly adapted to the extremes of rock outcrop habitats (Ware 2002). Many of the latter species are largely confined to rock outcrop habitats and often are endemic to a particular rock outcrop system. The granite flatrocks of Georgia (Murdy 1968), the limestone cedar glades of middle Tennessee (Baskin and Baskin 1986), and the limestone outcrop communities of the Ozarks (Steyermark 1963) are well-known for endemic species confined to their particular rock outcrop systems. Many abundant plants of rock outcrop communities are neither endemics nor near-endemics. These include some widespread weeds of other open *Department of Biology, College of William and Mary, Williamsburg, VA 23187- 8795; saware@wm.edu. 660 Northeastern Naturalist Vol. 17, No. 4 areas that are also able to tolerate the environmental extremes of the rock outcrop habitat. Two such weeds that may be abundant in rock outcrop communities in the Ozarks of Missouri and Arkansas are Croton monanthogynus Michx. (Prairie Tea) and Crotonopsis elliptica Willd. (= Croton willdenowii G.L. Webster) (Wildenow's Croton), both summer annuals. Croton monanthogynus has a capsule that dehisces forcefully and throws the seeds away from the plant, while Crotonopsis elliptica has a single-seeded, indehiscent fruit that simply falls to the ground when mature. Despite noticeably narrower leaves in C. elliptica, adult plants of the two species growing in rock outcrop communities are similar in height and crown shape, and in general they seem to play similar ecological roles in both weedy sites and rock outcrop communities. Steyermark (1963) reported that Crotonopsis elliptica occurs on “acid soils overlying sandstone, chert, or granitic substrata, on rocky glades, sandy fields, and rocky open woods.” He described in detail its ecological position in rock outcrop communities, saying that it is “a characteristic plant of dry, sterile sandstone or chert glades and outcrops, commonly associated with Polygonum tenue, Oenothera linifolia, Hypericum gentianoides, Diodia teres, and other low-growing plants with reduced leaf-surfaces adapted to withstand the arid conditions prevailing on the mainly dry, sunny rock exposures.” Ladd and Nelson (1982) and Nelson (1985) also listed C. elliptica as a characteristic, sometimes dominant, component of both sandstone and granite outcrop communities in Missouri, and it is also commonly present in non-calcareous outcrop communities (granite, sandstone) in the southeastern United States (Murdy and Carter 2000, Quarterman et al. 1993). Steyermark (1963) described Croton monanthogynus as found “in usually acid soils, often overlying sandstone, chert, or granitic substrata”, and listed its habitats as rocky glades, dry upland prairies, open and waste ground, roadsides, along railroads, and, rarely, open woodlands. He did not list associates or describe any further its role in acidic outcrop communities. Neither Ladd and Nelson (1982) nor Nelson (1985) included this species in their lists of characteristic rock outcrop plants in Missouri on either acidic or calcareous substrates. Field studies of Ozark rock outcrop plants over the last three decades (Ware 2002) have confirmed Steyermark’s report that Crotonopsis elliptica is abundant only on non-calcareous rock outcrop communities, such as those on sandstone, granite, and chert. However, those same field studies have revealed that, contrary to Steyermark’s characterization, Croton monanthogynus is rarely found on sandstone or other acid outcrop communities of the Ozarks, although it occasionally occurs on roadsides in their vicinity. Further, it is often present in Ozark limestone outcrop communities, as it is in Tennessee limestone cedar glades (Cofer et al. 2008). Rock outcrop species that are confined to one substrate type may be excluded from a second substrate type by a physiological inability to thrive on that second substrate. Alternatively, they may be capable of growth on the second substrate, but be excluded from it by a superior competitor (Crow and Ware 2007). This study 2010 S. Ware 661 examines the composition of a limestone rock outcrop shallow soil community in which C. monanthogynus is abundant and compares the growth of both C. monanthogynus and Crotonopsis elliptica on their respective native outcrop type and on the substrate type of the other species. Field-site Description The study site where vegetation sampling was done is a gently sloping, southwest-facing limestone outcrop near the top of a saddle on a broad, otherwise forested ridge near Eagle Rock in Barry County, MO. This site is a typical “hillside glade” (Steyermark 1963, Ware 2002). Narrow bands of bare rock alternate with shallow soil. Soil depth can vary considerably over short distances because of uneven erosion (Ware 2002), but it generally increases upslope from one band of rock until the next higher band of bare rock is encountered. Sampled transects encompassed only the medium herb-annual graminoid vegetation zone (Ware 2002), where summer annuals like Croton monanthogynus and Crotonopsis elliptica would be expected to occur, and did not reach deeper soils, where perennial grasses usually dominate. Methods Field transects To quantify the importance of Croton monanthogynus in this limestone outcrop community, two parallel 25 cm-wide transects were run in July 2005 across an expanse of gradually deepening soil from the shallowest soil location of C. monanthogynus near an exposure of bare rock to the next higher rock shelf, a distance only a few centimeters shy of 6 m. Soil depth was measured at the base of the main stem of each plant of any species encountered along each transect. This measurement was made by pushing a knitting needle vertically downward into the moist soil immediately next to the stem until it would penetrate no further (presumably having struck rock underneath), extracting the needle, and measuring the length of the portion of the needle that had been below the soil surface. The transects provided data on relative abundance of C. monanthogynus and each of its associates, and on ranges and means of soil depth for all species within the transects. Taxonomic nomenclature in this paper follows Smith (1994), but commonly used synonyms are also given. Substrate tolerance Seeds of C. monanthogynus were collected at the study site, and those of Crotonopsis elliptica from an Ozark sandstone outcrop near Calico Rock in Stone County, AR. Soils were also collected from the same locations. Soil from the Barry County, MO study site had a pH of 8.3 and mineral content (in oxides) of Ca = 3350 ppm, Mg = 35 ppm, P = 6 ppm, and K = 50 ppm, while the Stone County, AR non-alkaline sandstone soil had a pH of 5.3 and a mineral content of Ca = 150 ppm, Mg = 15 ppm, P = 16 ppm, and K = 15 ppm (Crow and Ware 2009, Reinhard and Ware 1989). 662 Northeastern Naturalist Vol. 17, No. 4 Seeds of both species were given a moist cold treatment (4 °C) for at least eight weeks, and germinated thereafter on moist filter paper in petri plates. They were transplanted while in the cotyledon stage to plastic pots (8 cm in diameter) containing either their own outcrop soil or the other's native soil. Five seedlings of each species were planted per pot in three pots for each soil type (sandstone and limestone). Seedlings that died within the first week from presumed damage during transplanting were replaced, but those dying after one week were not. Pots were watered regularly throughout the experiments. Experiments were carried out in greenhouses at the University of Arkansas at Fayetteville, AR or at the College of William and Mary in Williamsburg, VA. Supplemental lighting was used during winter and early spring months. Throughout each experiment and at its termination, notes were taken on the general appearance and health of the plants, as well as on any mortality. At the end of each experiment, plants were harvested by clipping them at the base of the stem adjacent to the soil, taking only above-ground biomass (Crow and Ware 2009). Plants were then oven-dried at 100 °C for approximately 48 hours. Dry mass was recorded in milligrams. Results Field transects In the first limestone outcrop transect, Croton monanthogynus was the most abundant species, with 24.3% of 210 stems, followed closely by Heliotropium tenellum (Nutt.) Torrey (Pasture Heliotrope) and then Isanthus brachiatus (L.) BSP (= Trichostema brachiata L.) (False Pennyroyal) (Table 1). The fourth and fifth most abundant species were Sporobolus neglectus Nash (Small Dropseed) and Panicum capillare L. (Common Witchgrass or Panicgrass), although the latter was in a single patch near the upper end of the transect rather than widely spread along the transect like the other species. Croton monanthogynus was only slightly less abundant in the second transect (22.3% of 206 stems), but there it was exceeded by H. tenellum. Panicum capillare ranked fourth, though with a lower relative Table 1. Relative abundance and mean soil depth of species in a limestone rock outcrop community in Barry County, MO. Relative density and (in parentheses) number of plants of each species are given for each transect. Soil depth mean ± S.E. (cm) is given for each species for both transects combined. All species with at least 5 stems in both transects are listed. Transect 1 Transect 2 Both transects relative density relative density mean soil Species (number of plants) (number of plants) depth (cm) Heliotropium tenellum 23.8 (50) 28.6 (59) 7.88 ± 0.24 Croton monanthogynus 24.3 (51) 22.3 (46) 8.15 ± 0.25 Isanthus brachiatus 19.5 (41) 20.9 (43) 8.22 ± 0.30 Panicum capillare 11.4 (24) 8.7 (18) 7.42 ± 0.26 Sporobolus neglectus 13.3 (28) 4.4 (9) 8.43 ± 0.40 Palafoxia callosa 2.4 (5) 5.3 (11) 8.96 ± 0.47 All other species 5.2 (11) 9.7 (20) - Total 100.0 (210) 100.0 (206) - 2010 S. Ware 663 density than in the first transect, and again it was concentrated in a single dense patch toward the upper end of the transect, despite a lower mean soil depth than other species. Palafoxia callosa (Nutt.) Torr. & Gray (Small Palafox) ranked fifth, slightly exceeding S. neglectus. In the two transects combined, 23.3% of 416 stems were C. monanthogynus, 26.2% H. tenellum, and 20.2% I. brachiatus. Total density for the sparsely vegetated transects was 138 stems/m2. Soil depth for C. monanthogynus ranged from 3.5 cm to 11.7 cm, with a mean ± SE of 8.15 ± 0.24 cm. This was close to (and not statistically different from) the mean soil depth of H. tenellum (7.88 ± 0.24 cm) and of I. brachiatus (8.22 ± 0.30 cm) in these two transects (Table 1). In comparison, mean soil depths were 3.63 ± 0.13 cm for Talinum calycinum Engelm. (Rockpink) and 5.71 ± 0.20 cm for Evolvulus nuttallianus J.A. Schultes (Shaggy Dwarf Morningglory) measured elsewhere on the same outcrop complex (S. Ware and D.M.E. Ware, unpubl. data). Substrate tolerance Total dry mass in Crotonopsis elliptica on limestone soil was reduced almost 75% in comparison with its growth on sandstone soil (Table 2), and plants on limestone were generally small and misshapen with pale, yellow-splotched leaves. In contrast, Croton monanthogynus grew better on limestone soil than on sandstone soil, but in this case, plants on the non-native sandstone soil were normal in morphology and color, merely smaller than they were on their native limestone soil. Differences in survival of seedlings on different soils during the experiments were not statistically significant. Discussion Because of the extensive ecological information included in the species entries in Steyermark’s (1963) Flora of Missouri, this manual is destined to remain an important reference source, even as newer manuals supersede its taxonomic treatments. Steyermark (1963) described Croton monanthogynus as a species of acidic rather than calcareous substrates in the Ozarks, and neither Ladd and Nelson (1982) nor Nelson (1985) included it in lists of characteristic calcareous outcrop species. However, it was reported on Ozark limestone outcrops by Hicks (1981) and Logan (1992), whose unpublished theses were summarized by Baskin and Baskin (2000). Data in Table 1 show Table 2. Growth of Crotonopsis elliptica and Croton monanthogynus on native and foreign soil types (mean ± SE, dry mass, mg); * = different at P < 0.05, ** = different at P < 0.01, by t-tests. Soil type Species Sandstone Limestone Crotonopsis elliptica (97 d) 99.76 ±15.42 (n = 14) 25.92 ± 4.46** (n = 10) Croton monanthogynus (77 d) 45.83 ± 3.72 (n = 12) 60.65 ± 4.51* (n = 14) 664 Northeastern Naturalist Vol. 17, No. 4 that this species can be a major community dominant along with characteristic shallow soil calcareous outcrop species and is not just one of the various weedy species that can occur in deeper soils where perennial grasses predominate (Kucera and Martin 1957). Further, C. monanthogynus rarely occurs in acidic rock outcrop communities, contrary to Steyermark (1963). Crotonopsis elliptica of sandstone outcrops was severely stressed when growing on limestone soil, showing greatly inhibited growth, morphological abnormalities, and serious mineral deficiency symptoms. Thus, its absence from limestone outcrops is to be expected. In contrast, Croton monanthogynus grew quite normally on limestone as well as on sandstone soil, and it actually had more total growth on limestone soil. The percent growth reduction for C. monanthogynus on non-native sandstone substrate was much less than that of Crotonopsis elliptica growing on non-native limestone soil, so Croton monanthogynus might potentially occur regularly in both sandstone and limestone outcrop soils. On the other hand, the growth inhibition on sandstone presumably is sufficient to place this species at a competitive disadvantage with respect to other species it might encounter on sandstone soils. Such competitors could well include Crotonopsis elliptica, given the apparent close ecological equivalency of these two species on their respective preferred substrates. Such competitive pressure may explain the rarity of Croton monanthogynus on sandstone outcrop communities. Soil depth values should be taken as indicative of relative soil depth distributions for Croton monanthogynus and its associates in these transects, rather than absolute measures of their soil depth tolerances. Soil depth probes can encounter one of the larger fragments of limestone scattered through the soil, which stops the probe before the bedrock surface is reached, causing underestimation of soil depth (Baskin and Baskin 1999). Further, cracks in the bedrock may allow plants to extend their roots below the bedrock surface, with access to a much larger soil mass than a measure of depth-to-bedrock would suggest. In several places outside our transects, scattered meter-tall plants of the perennial Delphinium treleasei Bush ex K.C. Davis (Glade Larkspur), almost certainly rooted in cracks in the limestone, towered above stands of much smaller annual species rooted in soil less than 10 cm deep. In conclusion, consistent with the description of substrate preference by Steyermark (1963), Crotonopsis elliptica shows high constancy and fidelity to sandstone outcrops compared with limestone outcrops. Greenhouse experiments show that a physiological inability to grow normally on limestone soil is responsible for limiting it to acid substrates. However, contrary to information in Steyermark (1963), Croton monanthogynus can be an abundant species in the shallow soil zone of Ozark limestone outcrop communities, and is rarely found in sandstone outcrop communities. This rarity is not caused by physiological intolerance of non-calcareous substrates, but perhaps because its growth on sandstone is inhibited to the point that it is competitively excluded by other sandstone outcrops species—maybe including Crotonopsis elliptica. 2010 S. Ware 665 Acknowledgments This research was supported in part by a Faculty Research Award from the College of William and Mary. Thanks are extended to Dwight Talburt, Department of Biological Sciences, University of Arkansas, for providing office and greenhouse space, Donna M.E. Ware, College of William and Mary, for help with field work, and Jerry M. Baskin, University of Kentucky and two anonymous reviewers for helpful advice on the manuscript. Literature Cited Baskin, J.M., and C.C. Baskin. 1986. Distribution and geographical/evolutionary relationships of cedar glade endemics in southeastern United States. Association of Southeastern Biologists Bulletin 33:138–154. Baskin, J.M., and C.C. Baskin. 1988. Endemism in rock outcrop plant communities of unglaciated eastern United States: An evaluation of the roles of the edaphic, genetic, and light factors. Journal of Biogeography 15:829–840. Baskin, J.M., and C.C. Baskin. 1999. Cedar glades of southeastern United States. Pp. 206–219. In R.C. Anderson, J.S. Fralish, and J.M. Baskin (Eds.). Savannas, Barrens, and Rock Outcrop Communities of North America. Cambridge University Press, Cambridge, UK. Baskin, J.M., and C.C. Baskin. 2000. Vegetation of limestone and dolomite glades in the Ozarks and Midwest regions of the United States. Annals of the Missouri Botanical Garden 87:286–294. Cofer, M.S., J.L. Walck, and S.N. Hadayati. 2008. Species richness and exotic species invasion in middle Tennessee cedar glades in relation to abiotic and biotic stresses. Journal of the Torrey Botanical Society 135:540–553. Crow, S.E., and S. Ware. 2007. Soil-type tolerance in rock outcrop plants: species of non-calcareous substrates. Southwestern Naturalist 52:120–125. Crow, S.E., and S. Ware. 2009. Soil-type tolerance in rock outcrop plant communities: Satureja arkansana (Nutt.) Briq. (Lamiaceae) in the Ozarks. Journal of the Torrey Botanical Society 136:363–368. Hicks, J.L. 1981. A vegetative analysis of Hercules Glades Wilderness. M.Sc. Thesis. Southwest Missouri State University, Springfield, MO. 64 pp. Kucera, C.L., and S.C. Martin. 1957. Vegetation and soil relationships in the glade region of the southwestern Missouri Ozarks. Ecology 38:285–291. Ladd, D., and P. Nelson. 1982. Ecological synopsis of Missouri glades. Pp. 1–20, In E.A. McGinnes, Jr. (Ed.). Proceedings of the Cedar Glade Symposium. Occasional Papers of the Missouri Academy of Science 7. Logan, J.M. 1992. The glades of the Buffalo National River, Arkansas. M.Sc. Thesis. Iowa State University, Ames, IA. 120 pp. Murdy, W.H. 1968. Plant speciation associated with granite outcrop communities of the southeastern Piedmont. Rhodora 70:394–407 Murdy, W.H., and M.E.B. Carter. 2000. Guide to the Plants of Granite Outcrops. University of Georgia Press, Athens, GA. Nelson, P. 1985. The Terrestrial Natural Communities of Missouri. Missouri Department of Natural Resources, Jefferson City, MO. Quarterman. E., M.P. Burbanck, and D.J. Shure. 1993. Rock outcrop communities: Limestone, sandstone, and granite. Pp. 35–86. In W.H. Martin, S.G. Boyce, and A.C. Echternacht (Eds.). Biodiversity of the Southeastern United States: Upland Terrestrial Communities. John Wiley and Sons, New York, NY. 666 Northeastern Naturalist Vol. 17, No. 4 Reinhard, R.T., and S. Ware. 1989. Adaptation to substrate in rock outcrop plants: Interior Highlands Talinum (Portulacaceae). Botanical Gazette 150:449–453. Smith, E.B. 1994. Keys to the Flora of Arkansas. University of Arkansas Press, Fayetteville, AR. Steyermark, J.A. 1963. Flora of Missouri. Iowa State University Press, Ames, IA. Ware, S. 2002. Rock outcrop plant communities (glades) in the Ozarks: A synthesis. Southwestern Naturalist 47:585–597.