Granite Outcrop Communities of the Piedmont Plateau in Georgia

Madeline P. Burbanck; Robert B. Platt

Ecology, Vol. 45, No. 2. (Apr., 1964), pp. 292-306.

Stable URL:

http://links.jstor.org/sici?sici=0012-9658%28196404%2945%3A2%3C292%3AGOCOTP%3E2.0.CO%3B2-Q

Ecology is currently published by Ecological Society of America.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtainedprior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content inthe JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/journals/esa.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. Formore information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.orgMon Jun 11 16:06:20 2007

M . P. BURBAKCK A N D R. B. PLATT Ecology, V O ~ . 45, NO. 2

GRANITE O U T C R O P COMMUNITIES O F T H E P I E D M O N T P L A T E A U I N GEORGIA

MADELINEP. BURBANCK B. PLATTAND ROBERT Department of Biology, Emory University, Atlanta, Georgia

Abstract. Communities in depressions, occurring as islands surrounded by naturally exposed granitic rock, were classified by the correlates of maximum soil depth and character- istic vegetation cover into: diamorpha communities, 2-6 cm; lichen-annual herb communities, 7-15 cm ; annual-perennial herb communities, 16-39 cm ; and herb-shrub communities, 40-50 cm. The total flora of the 40 island communities included 76 species, but only 39 of these occurred in five or more communities and were considered characteristic. The extreme environmental conditions of high light intensity and low soil moisture content were indicated by the fact that of the 20 plants present in over 50% of the communities, 18 were lichens, mosses, and flowering annuals and only two were perennial flowering plants. The flora of the island communities is distinct from that of border vegetation. Only three endemics are among the 39 characteristic plants. Seasonal aspects of the outcrop vegetation are striking. During the winter months, hTovember through February, mosses and lichens are most conspicuous, but seedlings of spring-blooming annuals are also present. From late February to May there is a continuous succession of plants in flower, the majority of which complete their life cycles before the hot dry months from May to August. In September the outcrops are covered with the orange-yellow flowers of Vigzliera Porteri. Spring-flowering annuals germinate in October at the time summer-flowering annuals are dying. The four types of island communities described appear to represent stages in plant succession directly related to edaphic conditions. Apparently the island con~munities have existed long enough in geologic rime for an endemic flora to have evolved. The life of any one community, however, is rneasured in hundreds rather than in thousands of years.

species are different, the types of plants and the This study is concerned with the characteristic stages of successiotl on exposed granite and ac-

plant communities which occur on outcroppings cumulated gravel at Enchanted Rock, Texas, as of granite gneisses and schists in the Atlanta area reported by Iihitehouse (1933), are similar to of the Piedmont Plateau of Georgia, with em- conditions in Georgia. Winterringer and Vestal phasis upon their seasonal aspects, the relative (1956) reported the presence and abundance of abundance of their component species, and the plants on rock ledges in southern Illinois and dis- role of succession in their composition. cussed succession of plants on three types of rock

Granite and granite gneisses are present in a surfaces. Similar studies of granitic outcrops in large part of this area (Watson 1902) and fre- the southeastertl United States have been made quently occur as extensive areas of exposed rock. by Oosting and Anderson (1939) and Keever, The best known is Stone Mocntain in DeKalb Oosting, and Anderson (1951) in North Carolina. County, a massive granite monadnock which rises Harper (1939) reported species of plants from to a height of 220 tn above the Piedmont Plateau granite outcrops in Chilton and Randolf Counties, (1,686 ft above sea level, Herrmann 1954) about Alabama, areas similar both physically and florist- 15 miles east of Atlanta. East and southeast of ically to outcrops in DeKalb County, Georgia. Stone Mountain are several lesser tnonadnocks AIcVaugh ( 1943) studied the occurrence and and small and large areas of flat or gently sloping origins of characteristic plants of the granitic flat exposures of granite gneiss. Crustose lichens rocks of North Carolina, South Carolina, Georgia, and mosses cover most of their exposed surfaces, and Alabama, but his floral lists were not com-while in temporary pools, shallow soil pockets, plete and some of the generalizations on succes-crevices, and on rubble heaps are plants which are sion on granite do not apply in the Georgia Pied- characteristic of and often restricted to such out- mont. Numerous studies have been made over the crops in this part of Georgia. past 25 years on outcrops in the Atlanta area.

Studies of rock vegetation have been made The seasonal flora of Stone Mountain was studied elsewhere in the United States. 111his study of by Smith (1938) and of Mount Panola by Mat- the forests of Isle Royale, Lake Superior, Cooper thews (1941). Two papers by Baker (1945, (1913) stressed the importance of crevice vegeta- 1956) reviewed the data accumulated by a number tion and heath mats in early stages of establish- of life history studies, and summarizedd&e,, vege- ment of plant communities on rock. Although the tation of the local outcrops. More recently's&eral

Spring 1964 GRANITE OUTCROP

papers habe appeared on the physiological ecology of indhi ~1 species (Cotter 1958, Lammers 1958, Cotter a1d Platt 1959, Wiggs and Platt 1962) and the effects of radiation on community struc- ture (McCormick and Platt 1962). Lists of mosses from Stone Mountain, Pine Mountain, and Mount Arabia have been published by Schornherst 1945, Schnooberger 1945, Clebsch 1954, and Breen and Purse11 1956.

This detailed analysis of the natural vegetation of granite outcrop comtnunities near Atlanta was undertaken to coordinate work on individual species. Of the several kinds of outcrop cotn-munities, those which occur as relatively isolated islands of vegetation surrounded by exposed, un- quarried rock offer a unique opportunity for the study of plant communities in geographically separated but physically similar habitats. These isolated plant communities may be divided into two types. One consists of "solution pits with intact rims" (Baker 1956) which retain water long enough to limit vegetation to plants which can survive submergence while rooted in shallow soil, and the other consists of soil-filled depres- sions with eroded rims which only rarely con-tain standing water. Both types were studied.

Experimental areas selected were similar geo- graphically, relatively undisturbed by quarrying, ungrazed, and were neither tourist attractions nor picnic spots. These factors ruled out Stone Mountain, long popular for sight-seeing and now a state park, extensively quarried Pine Moun-tain, and Mount Panola which is privately owned.

Of the other granite outcrops in the Atlanta region, Mount Arabia and Rock Chapel Mountain ("Rock Mountain" on some maps) are the most extensive and offer a variety of exposures. Two miles west of Rock Chapel Mountain above Swift Creek is another extensive area of exposed granite formerly called the North Georgia Quarries (Herrmann 1954) and more recently named Mount Rollaway (Cotter and Platt 1959). The greater part of the area has been quarried, but a northeast-facing slope has a vegetation pattern comparable to that of other undisturbed outcrops.

Mount Arabia, Rock Chapel Mountain, and Mount Rollaway provide a variety of habitats, namely : ( 1) mesophytic, usually forested areas adjacent to the rock outcrops; (2 ) marginal com- munities (ecotones) between the forest area and the cxposed rock; (3) exposed rock surfaces; (4) natural depressions in the rock containing soil; (5) rock crevices; and (6) rubble heaps. The crevices vary in size from a crack in which a

COMMUNTTIES 293

tuft of grass survives to a deep fissure in which a mature pine tree grows. The rubble heaps re-sulted from quarrying, and their flora varies de- pending upon the amount of soil present, the source of the rubble, and its proximity to other vegetation.

Attention was concentrated upon communities in depressions containing soil (type 4 above) en-tirely surrounded by exposed rock and therefore referred to as "island comtnunities." Many oc-cur on Mount Arabia and Rock Chapel Mountain but only a few on Mount Rollaway. Specific is- land communities on each of these three outcrops were chosen to include (1) different elevations, exposures, and habitats; (2) maximum number of characteristic species; and (3) areas large enough to support vegetation and yet small enough to present the total aspect at a glance.

The island communities had certain physical characteristics in common. The ~ l a n t s were rooted in pockets of soil accumulated in depres- sions. All but three were located on a sloping rather than a level part of the mait1 outcrop al- lowing precipitatiotlto run in and run off. T h e soil was deepest toward the center of each depres- sion and progressively shallower toward the pe- riphery, or in some, deeper toward the upper edge and shallower on the down-side. These soil pockets were usually irregularly circular or el-liptical with a maximum diameter of 9 m but more often 5 tn or less. A portion of the community usuallv occurred where the soil was shallow and level with the surrounding rock. Opposite this the rock might rise abruptly or gradually above the soil pocket from a few to 20 or 30 crn or even occasionallv to a meter. At each side, the rock sloped regularly or irregularly down to the level area (Fig. 1, C, D) : 111seven the soil level was equal to or even slightly above the surrounding rock (Fig. 2, A, B, C) .

In general, evaporation is very rapid, and per- manent pools of water are rare. In one island community, Arabia no. 10, water reltlained during all but the driest summer weather. and its flora was distinct from that of other communities. In two communities, water accumulated at times at one end, but the other end contained soil and characteristic vegetation (Fig. 2, F ) . Water oc- casionally stood in the shallow parts of some other communities.

Forty communities were selected and numbered on the adjacent rock with aluminum paint, 17 on Mount Arabia, 16 on Rock Chapel Mountain, and 7 on Mount Rollaway (Fig. 3 ) . Each of these was visited five times in 1957, and supplementary records made in 1958 and the fall of 1959. The

294 M . P.BURBASCK A N D R. B. PLATT Ecology, Val. 45, No. 2

presence and the abundance of each species of During July and August, size, depth of soil, moss, lichen, and seed plant were recorded. These depression depth, and slope of adjacent rock were records were timed to coincide with seasonal peaks measured, and notes taken on exposure, dominant or lows of growth otl the outcrops: March 27- herbs, drainage pattern, and distance to neighbor- April 2, spring-flowering plants; May 19-23, late ing communities. The outline of each community spring flowers ; July 2-10, xeric conditions ; Oc- was sketched on graph paper (scale of 1. in. = tober 8-10, fall aspect before frost; December 3-5, 1 m) in the field and a planimeter used to calculate winter aspect after frost. Abundance was based the areas. primarily on area covered, not on numbers of The life histories of characteristic outcrop plants, since such diverse plants as lichens, mosses, plants were compiled from data collected weekly. annual herbs, leafy perennials, and vines were A different outcrop was visited each week and compared. records were based on the outcrops as a whole, not

FIG.1. Island communities, Mount Arabia. Arrows indicate Dia~norphazones. A. Diamorpha community with plants in bud, April 10, 1962; B. Lichen-annual herb community no. 8, April 16, 1961 ; C. Early spring aspect of lichen-annual herb community no. 2, March 24, 1961; D. Fall aspect of community no. 2 with V i g u -iera in bloom and Dia+norpha dead, September 29, 1951 ; E. Arenaria and Senecio in bloom in an annual-perennial herb community, April 10, 1962; F. Annual-perennial herb community dominated by Se~j&o with last year's Andropogon and Viguiera stalks in center, April 10, 1962. , "'t,

295 Spring 1964 GRANITE OUTCRl 3~ COMMUNITIES

just o n G t h e plants in the 40 communities. The foll wing aspects of life cycles were recorded: dor3% v ( n o green parts evident), resumption of growth, seed germination, flower production, fruit production, and vegetative portion persistent.

Comwzunity types On the basis of maximum soil d e ~ t h correlated

with characteristic flora, island communities were divided into four categories as follows : diamorpha, lichen-annual herb, annual-perennial herb, and herb-shrub.

Diamorpha cor~vvcunities.-Those communities with the shallowest soil were designated dia-morpha communities (Fig. l, A ) . The four in this study, with maximum soil depths of 2, 4, 6, and 9 cm respectively, supported only one dicotyledonous plant, Diamorpha cymosa, and that in great abundance during its growing season. 111one community the only additional plant was a single clump of leafy lichens. The other three contained well-established plants of Cladonia leporina and Campylopus sp. The 9-cm depth community included Cladonia car/oliniana associ-ated with C . Zeporina, and a few plants of Viguiera Porteri, which died before flowering. For ap-proximately 18 weeks through the summer and early fall after the Dia~eorphadies, these habitats contain very few or no living plants.

Lichen-annual herb communities.-These com-munities were characterized by having a maximum soil depth of 7-15 cm inclusive and a flora of lichens, mosses, and annual herbs (Fig. 1, B, C, D) . Twelve species occurring in at least six of the ten lichen-annual herb communities studied and considered characteristic are : Cladonia leporina, Cladonia caroliniana, Campylopus sp., Agrostis Elliottiana, Bulbostylis capillaris, Are-naria brez~ifolia, Crotonopsis elliptica, Diawaorplza cymosa, Hypericum gentianoides, Linaria cana-densis, Rumex hastatulus, and Viguiera Porteri. A few established plants of Senecio tomentosus occurred in four of the lichen-annual herb com-munities and seedlings in a fifth, but this perennial is more characteristic of soil pockets with soil depths over 15 cm. Other species in two, three, or four communities were Cladonia rangiferina, Cla- donia spp., Ditrichum palliduwa, Bruchia Sulli-vanti, Polytriclzum commune, Juncus georgianus, Panirum meridionale, Cyperus granitophilus, Schoenolirion croceum, Krigia virginica, and Ta-linuwz teretifolium. Present in only one of the nine were Cladonia grayi, Isopterygizbm micans,

Aristida dichoto~~ta, Anzjhin~ttlzus pztsillus, Gnap- halium sp., and Opuntia humifusa.

Environmental conditions at the periphery of the lichen-annual herb communities were similar to those of diamorpha communities, there fre-quently being an outer zone of Diawzorpha only, with a more diverse flora on the deeper soil (Fig. 1, B) . In some lichen-annual herb communities, Cladonia leporina and C. caroliniana occupied almost the whole area with some Diamorpha at the edges and Cawzpylopus and annual seed plants scattered anlong the lichens. In such communities the gray of the lichens was a backdrop for the white and green of Arenaria and Agrostis respec-tively in the spring, the gray-green Crotonopsis and the brighter green, scale-leaved Hypericum in the summer, both with inconspicuous flowers. In the fall the bright yellow-orange flowers of Viguiera appeared. Yiguiera Porteri is very sus- ceptible to drought, and it was not unusual for seedlings to be common or abundant throughout a lichen-annual herb community in late spring and early summer but by flowering time there be only a few or no living Viguiera plants. Thus in contrast to seasonal diamorpha communities, much of the area of each lichen-annual herb com- munity was covered by persistent cladonias, sup- plemented by a year-round succession of annual herbs.

Annual-perennial herb cowzz.ttzunities.-Soil pockets with a maximum depth of from 14 to 39 cm support a mixed flora of lichens, mosses, an-nual and perennial herbs, and an occasional woody plant and are called annual-perennial herb com-munities (Fig. 1, E. F ) . The flora of the an-nual-~erennial herb communities is the most di-verse of any of the island communities. Most of the species listed in Table I are represented. Not all these species occurred in every annual-peren- nial herb community, however. The characteristic plants of the shallower soils were usually present at appropriate soil depths; occurrence of other lichens, mosses, and herbs varied with the excep- tion of Polvtrichum commune. Schoenolirion cvo-ceum, and Senecio tomentosus which were fre-quent and abundant in all annual-perennial herb communities. P. comwzune was present in every annual-perennial herb community but in only four lichen-annual herb communities. This moss covered over half the area of five annual-perennial herb communities at Rock Chapel Mountain and of three such communities at Mount Rollaway. The narrow, dark green blades of Schoenolirion leaves from a deeply buried corm were prominent from November until they yellowed and dieti early in June. Senecio towzentosus had green leaves all

296 &I. P. BURBASCK AND R. B. PLATT Ecology, Vol. 45, No. 2

year, and the plants tended to form clumps near cluently seen it1 annual-perennial herb communities the centers of the communities in the deepest and but not restricted to them. Although not present moistest soil (Fig. 1, F ) . in diamorpha communities and in only four lichen-

Other plants that were characteristic of annual- annual herb communities, the tiny moss, Bruchia perennial herb communities but which did not Sullivanti, grew at the edge of almost every an-achieve any degree of aspect dominance were nual-perennial herb community it1 a crust of soil Cladonia spp., Bruchia Sullivanti, Aulecomni.ttm even shallower than that which normally sup-palustre, Isopterygi~~wz micans, And7,opogon vir- ported Diamorpha A~~laconzniu~zcymosa. pal-ginicus, Juncus geo~,gianus, Gnaphalium sp., ustre occurred as scattered clumps while Isoptery-Krigia virginica, Polygala curtissii, and Rumex giuwz micans characteristically formed a flat mat hastatulus. Sod lichens, Cladonia spp., were fre- near the center of the communities under a shield-

FIG.2. Communities on Mt. Arabia, Rock Chapel Mt., and Mt. Rollaway. A. Dried Alzdropogolz stalks prominent in winter aspect of herb-shrub community no. 15, Mt. Arabia, February 14, 1960; B. Fall aspect of community no. 15, Mt. Arabia, September 29, 1961; C. Herb-shrub community no. 11, Rock Chapel, March 29, 1961; D. Midsummer dominance of Viguiera in herb-shrub community no. 2, Rock Chapel, July 29, 1960; E. Large weather pool on top of Mt. Arabia, April 10, 1962; F. Polytrichuril and Juncus georgiat~zts roo ed in soil, and inifihianfhur present and blooming among rocky debris in temporary pool basin, in water rudp(pll a 20" slope, community no. 6, Mt. Rollaway, March 25, 1960.

297 Spring 1964 GRANITE OUTCRI3~ COMMUNITIES

ing o,over of herbaceous plants and occupied only a smhllrprcentage of the total area. Juncus georgianus and Andropogon virginicus plants tended to form discrete clumps, the Juncus near a moist edge of a community and associated with Polytrichuvtt covrcvvcune (Fig. 2, F ) while the Andropogon was usually near the center or where the soil was deepest (Fig. 1, F ) . The small, slender plants of the annual Krigia virginica were showy ;Then in bloom, but this species, as well as Rumex hastatulus and Gnaphalium sp., are con- sidered weedy species and not typical of the out- crop flora. seedlings of Gnaphaliuvlz are evident during winter months but decrease in number as spring-flowering plants mature. Polygala curtissii is an annual which blooms during the summer and on into the fall, contributing scattered patches of bright pink to an otherwise rather colorless period. Additional plants which occurred in about one-half the annual-perennial herb communities were Ditrichum pallidufiz, Tradescantia olziensis, and Talinutuz teretifoliuw.

There were six species which occurred in less than half of the 20 annual-perennial herb com-munities. The endemic sedge, Cyperus grani- toplzilus, grew as scattered tufts in the shallower parts of eight; small cups of Cladonia grayi were identified in seven and may have been present in two others; seedlings and a few mature plants of Facelis retusa, were observed in six; Houstonia pusilla and Coreopsis grandifiora were present in four but were more abundant on the outcrops in other habitats; and isolated plants of Opuntia humifusa occurred in four.

Herb-shrub eomvlzunities.-Five communities with a maximum soil deuth of 40-50 cm and with small trees or stumps of trees in addition to herbs were called herb-shrub communities. Rock Chapel no. 2 contained a juniper stump 22 cm in diameter but no living woody plants (Fi,g. 2, D). The other four herb-shrub communities contained the woody vine Smilax glauea and at least one small tree, an Amelanchier arborea (3 m ) , Juniperus virginiana (3.8 m) (Fig. 2, C ) , or a Prunus sero- tina (1.8 and 2.5 m) (Fig. 2, A, B) . Other woody species present were Gelsevlziuvl% semper- virens, Parthenocissus quinquefolia, young Pinus taeda, Smilax Smallii, and Vaccinium ahoreum.

With the exception of the woody species, the floral list of the herb-shrub communities was simi- lar to that of the annual-perennial herb communi- ties with a few additional species, but the relative abundance of the species was very different. Physically these communities differed from the other types in that the soil was level with the sur- rounding rock or slightly above it with very little

area with shallow soil. Rather than a peripheral zone of Diafizorpha and lichens, masses of Poly-trichum frequently grew at the edges of the com- munities, and a larger proportion of the total area was occupied by perennial plants than in the communities with shallower soil. Several sedges and grasses such as Carex comvl%unis, Scleria oli- gantha, and Panicum species were more abundant in the herb-shrub communities than in the few annual-perennial herb communities in which they occurred. Two herbs recorded only from herb- shrub communities were Rhexia wzariana and Ascyruut?, Hypericoides.

Special communities.-One aquatic and several disturbed habitats were also studied. Water-filled solution pits, such as community no. 10 and others on top of Mount Arabia (Fig. 2, E), support a few species of aquatic plants, particularly the two endemics Isoetes vlzelanospora and Amphiantlzus pusillus. The former is dominant and perennial, often carpeting the floor of these pits with its bright green quill-like leaves, while the latter, an annual with small submerged and floating leaves, is scattered throughout. Conditions in these pits are often severe, since the water may dry up for long or short periods. When refilled by rain, however, the plants either begin growth anew or continue from the point where growth had ceased. There was water in community no. 10 at the time of each of the five floral surveys, but the absence of Amplzianthus on July 4 indicated an intervening dry period.

Community no. 16 at Mount Arabia developed in a disturbed area with soil and rubble from quarrying operations. I t was selected to include the endemic Oenothera fruticosa var. subglobosa. This species normally occurs as part of the marginal or ecotone vegetation at the base of the mountain where exposed rock alternates with wooded areas, as well as in accumulations of soil and rubble. Although community no. 16 con-tained species typical of annual-perennial herb communities, there was also an unusually large number of species characteristic of the border vegetation such as Coreopsis grandiflora, Geran- ium carolinianufiz, Houstonia pusilla, Oenothera fruticosa var. subglobosa, Parthenocissus quin-quefolia, Portulaca Smallii, and Saxifraga vir-giniensis.

Community no. 14 on Mount Arabia may also have been affected by quarrying operations since its site contained loose rocks and rubble and was located between an undisturbed slope and a partly quarried area. The presence of Coreopsis, Hou- stonia, and Portulaca in addition to the usual is- land community species is further evidence that

298 M. P. BURBANCK AND R. B. PLATT Ecology, Vol. 45, No. 2

such disturbed areas support a different flora than do undisturbed communities.

Size of community and soil depth Although communities were chosen not at ran-

dom but with an approximate upper size limit, the areas of individual communities varied from 1.3 to 56.9 sq. m (Fig. 3 ) . The maximum soil depth

3 E F T H OF S O L IN CM-GREG IN SO METERS

I " A w c R C QC R< $ RC R< k 8 b *< " & 8 $ RC R C $ $ R C *< A $ R C R R A R b k $ "C R C R C RC A k 2 1 ~ J i i * l ~ ' l ~ . l . . , . , , , . n l s l 2 1 8 9 8 , G0 < , # l ~ . R # *

Lm'*Oi/s*n L c * / * b**YbL " i R B AW,LbL PERLWW#bL "trim S"R"S

COMMUNITY DESIGNATION BY VEGETATION TYPE

FIG.3. Comparison of area with maximum soil depth for 39 communities (numbered as in Table I ) grouped according to vegetation type and arranged according to increasing soil depth. A = Mt. Arabia, R = Mt. Rolla- way, RC = Rock Chapel Mt.

ranged from 2 cm in a diamorpha community to 48 cm in an herb-shrub community. There was not a close correlation between area and soil depth for all communities (Fig. 3 ) , but the increase in average maximum soil depth of community types corresponded roughly with an increase in the average area for each type :

Type of Average Average maximum community area (m2) soil depth (cm)

Diamorpha 2.8 5.3 Lichen-annual herb 5.5 10.5 Annual-perennial herb 12.4 24.5 Herb-shrub 25.5 44.0

Presumably, as natural forces act on the underly- ing rock, disintegration does not proceed uni-formly under the soil, but most rapidly a t the deepest part of the concavity. Thus the increase in diameter is probably not always proportional to the increase in soil depth.

Flora of island communities Species recorded.-Presence and abundance

records were kept for the 76 species listed in Table I. As far as possible nomenclature follows that of Gray's Manual of Botany (Fernald 1950). Mushrooms, liverworts, and green algae were oc-

casionally seen but are not included. Lichens and mosses, however, are prominent and significant in island communities, and since complete lists have not been included in previous studies of the area, a special effort was made to identify them. Those most frequent, abundant, and readily recognized are included in Table I.

In addition to the five species of Cladorcia listed (Table I ) , three lichens occasionally collected from island communities were Cladonia subtenuis (Des Abb.) Evans, Cladonia furcata (Huds.) Schrad., and Cladonia sylvatica (L.) Hoffm. The listing Cladonia spp. represents semicircular, com- pact, gray masses, usually 2-6 cm in diameter, which occur on shallovr~ soil near the edges of com- munities. Two types were distinguished, those whose surfaces are somewhat leafy with the tips of the flakes spreading or recurved depending up- on moisture conditions, and those which are more compact with the upright flakes crowded closely together as in the pile of a rug. Positive identifi- cation could not be made (Dr . Mason E. Hale, private communication), but the specimens be-longed to the group known as sod lichens and some were probably Cladonia apodocarpa.

The listing Campylopus sp. includes the two species C. introj'lexus and C. flexuosus. Original-ly it was not known that there were two species present. Subsequent spot checks on all three outcrops indicated that the two species intermingle without showing dominance over each other and both occupy the same ecological niche. Therefore, Campylopus sp. is used with the assumption that either or both of the species may have been present.

Complete presence and abundance records were not kept for five mosses (addendum to Table I ) because field identification was difficult. Pohlia nutans (Hedw.) Lindb., reported here from Mount Arabia, was first found south of Tennessee in 1954 at Stone Mountain (Clebsch 1954). The species Bruchia Sullivanti Aust., frequent on Mount Arabia, Rock Chapel Mountain, and Mount Rollaway, is also newly reported from granite outcrops of Georgia.

Cuscuta was not seen in flower and fruit and hence could not be identified to species. Seedlings of Diodia and Gnaphalium never reached maturity in island communities although Gnaplzaliz~~~z pztr-pureum L. was identified in border vegetation. Panicutuz spp. represents several unidentified species both annual and perennial.

Observations on the outcrops bear out the con- clusion of McVaugh (1943) that Coreopsis saxi- cola Alexander probably is not "sufficie tly dis- tinct" from Coreopsis gmndiflora Hk$&l~$? be

Spr ing 1964 GRANITE OUTCROP COMMUNITIES 299

TABL?I . Species recorded f r o m 40 island communi t ies and t h e number o f communi t ies i n w h i c h t h e y occurred 11 3"-a

I l'. I

1 N u m b e r o f communi t ies Species

I M t . / R o c k I M t . I

Lichens Cladonia mistatella T u c k . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cladonia caroliniana T u c k . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cladonia leporina Fr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cladonia grayi Merrill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cladonia ranqiferina (L . ) W e b . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . Cladoniaspp. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Mosses Polytrichum commune Hedw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bruchia Sullivanti Aus t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ditrichum pallidum (Hedw.) Hampe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Campylopussp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aulacomnium palustre ( W e b . & Mohr) Schwaegr . . . . . . . . . . . . . . . . . . . . . . . . . Isopterygium micans (Sw.) Bro th . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sphagnumsp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Fern allies Isoetes melanospora Engelm .

Conifers P inus T a e d a L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Juniperus virginiana L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Flowering plants-Monocotyledons Festuca octojlora W a l t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Uniola laxa (L . ) B S P . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Agrostis Elliottiana Schult . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aristida dichotoma Michx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Panicum lanuginosum Ell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Panicum meridionale Ashe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Panicum scoparium L a m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Panicumspp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Andropogon virginicus L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cyperus granitophilus McVaugh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cyperus ovularis (Michx. ) Torr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bulbostylis capillaris (L . ) C .B .Clarke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Scleria oligantha Michx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carex communis Bailey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carex complanata Torr . & Hook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carex umbellata Schkuhr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Commelina erecta L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tradescantia hirsuticaulis Small . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tradescantia ohiensis R a f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Juncus georgianus Coville . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Al l ium Cuthbertii Small . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Erythronium americanum Ker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Schoenolirion croceum (Michx. ) A. G r a y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Smi laxg laucaWal t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Smilax Smallii Morong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Flowering plants-Dicotyledons Rumex hastatulus Baldw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Portulaca Smallii P . Wilson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tal inum teretifolium Pursh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Arenaria brevifoolia N u t t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Liriodendron Tulipifera L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Diamorpha cymosa ( H u t t . ) Bri t ton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Saxifraga virginiensis Michx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Amelanchier arborea ( M i c h x . f.) Fern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prunus serotina E h r h . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Geranium carolinianum L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Polygala curtissii A. Gray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Crotonopsis elliptica Wi l ld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A c e r r u b r u m L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Parthenocissus quinquefolia (L . ) Planch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ascyrum Hypericoides L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hypericum gentianoides (L . ) B S P . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Opuntia humifusa R a f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rhexia mariana L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

300 M. P. BURBAKCK A N D R. B. PLATT Ecology, V O ~ .45, NO. 2

Table I. (Continued)

I

1 N u m b e r o f communi t ies Species M t . Rock M t .

Arabia Chapel M t . Rol laway To ta l/ 1 1(17)' (16 ) ( 7 ) -

Oenothera fruticosa L . var . subglobosa Small . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vaccinium arboreum (Marsh.) N u t t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gelsemium sempervirens (L.)A i t . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cuscu tasp. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Linaria canadensis (L . ) D u m o n t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Amphianthus pusillus Torr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lindernia monticola Muhl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Plantago virginica L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Diod iasp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Houstonia pusilla Schoepf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gnapha l iumsp. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Facelis retusa (Lam. ) Sch. Bip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viguiera Porteri (A.G r a y ) B lake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CoreopsisgrandiJEoraHogg. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Senecio Smallii Br i t t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Senecio tomentosus Michx. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Krigia viryinica ( L . ) Wi l ld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Addendum-Presenc e or absence on ly indicated Pleuridium subulatum (Hedw. ) Lindb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Dicranum condensatum Hedw. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Dicranum scoparium H e d w . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -Weisia controversa H e d w . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Pohlia nutans (Hedw. ) Lindb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

I

1Number of experimental communiiies.

recognized as a separate species. Coreopsis, 2, E) and in a quarry pool at Mount Rollaway. moreover, does not occur "in immense numbers Amplzianthus pusillz~s grows in pools with Isoetes on all the outcrops" (McVaugh 1943). Tfiguicra, and also in certain conimunities on the sloping not Coreopsis, is the plant which covers the granite sides of the outcrops m,hich contain water for days outcrops of this area with yellow flowers in the or weeks. I11 &/larch 1960 there was no water late summer and fall. Coreopsis is not abundant standing in community no. 6, Mount Rollaway, in island communities but is more characteristic but plants of A?rzplziant/zus were in bloom among of disturbed areas and border vegetation. the "rubble" (Fig. 2, F ; the outline of the tem-

Endemics.-Although a high incidence of en- porary pool is indicated by the light area free of deniisni on granite outcrops has been stressed by Parz~teliarosettes.). McVaugh (1943) and Baker (1945, 1956), the Species distribzlfion and abundance.-The dis-list of plants which occur in island cornniunities tribution of species on the three outcrops indicated does not contain a large number of endemics. that there is a characteristic flora of the outcrops McVaugh stated that more than a third of the 44 of this region rather than a flora peculiar to any characteristic flat-rock species were endemic. Of one outcrop. The island communities had various the 39 species recorded from at least five island exposures, but the environmental conditions on communities, only three were endemics (Cyperzis these gentle slopes did not vary sufficiently to g?,aniiop/?ilzw, Juncms geo~,gianus, and T/iguiera influence either the floral make-up or the type po?*feri). Other endemics occasionally found of comniunity. Although 21 of the 76 plants were Isoetes 17lclanosfiora, Awzplziantkiis pltsillus, listed in Table I were recorded from only one Oenothera ft,ztticosa var. sz~bglobosa,and Portzt- outcrop, some were species with a very scattered laca S17tallii. Ocnotlzera, Portulaca, and many of distribution; some were species characteristic of the other ten endemics listed by McVaugh are disturbed areas; and some were species typical characteristic of the vegetation bordering the ex- of the border vegetation. Splzagnztm sp., Uniola posed rock rather than that of the island communi- laxa, Festzica octojlora, Isoetes 71zelanospora, ties. Isoetes 71zelanospora and Azqlpkiantlzzts pusil- Lindernia ~qzonticola,and Rlzen-in i?rarinnn were lzts are limited in distribution by special habitat recorded only from Mount Arabia, Jzcnipcrzts vir- requirements. The former occurs in persistent, giniana only from Rock Chapel Moun in and shallow pools on the top of Mount Arabia (Fig. Pimirun, scopnrizliil only from Mount 'AIllI{,way,

Spr111c I004 GRASITE OUTCROI' C C ) \ I ~ I U N I T I E S 301

P L A N T S A N D TOT&, N U M 2 F R O F C O M N J N I T ' E S ABUNDANCE RATINGS

N N H I C h E A C H C C C U 7 j

I IAMOHPHC 4 0 CL iIr'.i 3 8 I E P C R ' V L CCMPILOPUS 3 8 v G 1 1 l F R b. 3 7~. HYPERlCdM 35 CGROSTIS 'i. A R E N A R I C 3 4 CROTONOPSIS 3 2 'IA2OVl:, C i K C . h AU:, 3 0 CLAOOPI,A i? 3 0 L I N A R I A 7-9~

POLYTRICHUM i9 W L B O S T I L I S i-BRUCHIA is S O P T E R Y G I U M 2 5 RUMEX 2 5 S C H O E N O L I R O N ii. SLNECIO ? 3 S h A P H A L l d M 2 2 KR,u IA J U h C U I I 8 A U L C C O M h l b M 8 J I T R I C H U M 7 FO-Y'>ALA 6 i U 2 G C ? C u C Y 4 CLLC;hlG SQ:,Yl Ii T L L h L I M 2 C I P E R d i I ' TR;2FS'-il.T C O n l E V S l ~ C C i t " f , I i 8 O P U N T A CL,I2Ohl; R i U ~ ~ l ' E R l N i , PANIC: M M L Q : C O N A L E 7 FCCELI! 7 C A R i X CCMM,I l . lS 6 C L A D O N C CG S T C T E L L L 6 PINUS 6 L R I S T I S A 5 SMILAX ' b L i U i A 5

I . . 4 Coniparizon oi al~unrlance and irecjuenc!- (numher of comn~unities in which each species occurred) i~i'cliar;~ctc.ri~ticoutcroll ~~lant.. Al~untlance ratings : ; 4-abundant inI-rare ; 2-infrequent ; 3-con~n~on a local-ixc.11 are;{ I ; ~ tlva,t oni.-~lua~.terof total area) : .;-abundant (bazccl on the greatest abundance attained in a i. i \ c 11 c o i i i n i u ~ ~ i t y 1057'i.t l l ~ r ~ n g

11iit tlic,se ,sl)ccies \vpre seen in, o r lia\-e ljeen re-~ j o r t d frolli. similar 1ial)it:~ts on a t Icast o11e otlier of the thrcc ou tc ro l~s . \\-eed!- plants \vlnicli Ivere rccoi-tletl fro111 only one outcrop but \vliich ma!. occur near roads o r tlist11rl)ed areas o ~ i :dl three outcrops a r e Lliutiicr anit1 Citsi.iita. Seetlli~igsof i1ri.i. ritl,i,iliii aiid 12ii-iotl~~~~clTro~l c i a ~ e r cT1i1ipiJ7, fount1 at niore t l i a~ i one outcrop a s par t of tlic I~or t l r r 1-cgetation but recortletl f rom oiily on(. isla~itl conlmtunity. O t h e r s similarly rccortletl a re

characteristic p l ; ~ ~ l t s of island cominunjtics differ markedly f rom l r c \ - a u g h ' ~ list of characteristic illants of the flat rocks of southeastern L'nitetl States i h l c \ v a u g l ~19131 . O n l y 13 species a r e connilion to l l c \ ~ a u g l i ' s list of 4 1 characteristic l ) l a~ i t s a n d our list of 39 species. T h e species of the island coniniwnities appear to form a flora dis- tinct f rom the total flora of grani te outcrops.

In a large n u m l ~ e r of island com~nuni t i es , specica attained ail a b u ~ l d a n c e rat ing of only ra re o r in-

.i ~ii~-ln~ic.liic~i* .'Yo.t-ifrc7~gc7 ~,ii~~jiriicrzsis-, c71.71oi.c~ir. freciu(~11t( 1;ig. 4 ) ~ v h i c h suggests that : ' ~ -tloral

.\'(,ii(,(.io Siizallii, ant1 Ti.ncii~scc711tia 1iii.sittic-tritlis. conil)o"tio~i tcntls to I)e cliverse rather t h a r ~ ,.. - ., .

'I'l~u.;, of tile 76 sl)ecies. only 1rather t h a ~ i 21 \\.eri. 11:itetl I)!- onc o r tn-o species, except in the sha l lon Iiliiitc(1 to a single outcrop. I lven these, Cai,c~.l. soil c o ~ n ~ n u n i t i e s . A fe\v species tlo 1)ccollle c-oi~:pi~riiirtaatit1 .Srlci,ia ol i i ja~fl in from J I o u n t A\~-al)i;i .C't!rr.r f rom1r11~l,c~llc7ttr l l o u ~ i t Rolla~va!.. ant1 iT'(rizi(.itli! 1a~l!tiji~ios11111 Clnapplfro111 Tiocli l r e ) ~ i ~ ~ t : ~ i ~ i .IIIZIJ-wrll liave ~vicler clistriI~~itic)~i.

'I'lie 3') sl)ccies \vliicli occur in five o r niore oi thi. 1 0 c o m m ~ i n i t i c ~tiiclic.tl a r e consideretl tyl1ic;il ( Fig . 4 I . 'I'liey cc>nil)i-iqc. all(-tut 50C; of the tot:\\ flora. Of the 20 species present in more than half tli(> coliimu~iities only tivo a r e perei~nial floxvering i t s . l'lic others a r c fon-cring a11nu;lls. mosses. :111(1 l i c l i e~ i . wliicli :ire \\-ell atlal)tetl to tlie xeric

al111111:lalit in sollie con im~i~i i t i cs ,p:irticularly a t c?i-tai~l c,;isons or t l t ' v (~ lo l~~i le~ i t ;~ l ( F i g . 2 . ;t;lgc"; I)),and in general g rea t a l )unda~ice is correlatctl \\.it11 high frequency. T h r e e spccics were cscep-tions to this ge~neralization. C t r ~ i l p ~ d o p ~ t s sp , and Cltrrlotlin spp. occurred a s scattered patchm i l l

shnl lo~v soil ~ v i t h o u t ever l ~ e i ~ i g"al~untlant" \\.it11 recl)ect to the total area of a c o ~ n n i u ~ ~ i t y . I2iu(71.i(r c-ciiltrdc~.si.s, wit11 n-idesy~reatl distri1)ution on and off the grani te outcrops, appenrecl u n a l ~ l e to com- petc. very successfully with island community

sllallo\v soil co~iclitiolis tyl~ical of islnntl cc-~ni~iiuni- s l~ecies . tie. on gra~iitc ' o1ltcrol)s. Scnsorlal nspcct.c.-The different seasonal as-

?. tlit. 20 ~iiobt t ' r cc l~~e~i t 1 lie l i s t . ( ~ ~ i ant1 (li tlie 30 pects of the vegetation of the grani te outcroljs is

-------

AN11 K . B. I'LATT Ecology, \-01. 45, KO. 2

very str~king. This part of Georgia has mild n inter i , usually without snow. O n the outcrops, spring-blooming annuals germinate in the fall and o\ er\?inter as low-growing seedlings. l losses and lichens are the most conspicuous plants of the island communities during the months of Decem- ber, January, and February. During late Fehru- ary, hlarch, April, and May there is 3 profusion of hloom (Fig. 1 , E) . In the shallow, peripheral soil is the red-leaved, white-flowered Dia7ilorpha cyvlosn. Interior to it a ring of dark green Poly -t~ir11111i1co1117izlt~temay occur, mixed with the deli- cate, ~vhite-flo\?ered Aref luria brczifolia. In the center the taller, .ellow flowers of S c ~ t e c i o to71tcfi- toszds rise aljove their broad, gray-green leaves. Retneen the Sc~zcc ioand the Arenaria or near the periphery of island communities which do not ha1 e complete rings of Dia71rorplza and Polyirichzti~lare the tiny seeclings of later hlootning annuals, in- termingled with the dead stalks of the previous season ant1 tufts of the hright green grass, Agi+o-stis Elliottiic~za In tmany cotnmunities the quilled green lea\ es and yellow racemes of Schoi~nol ir ion crorelllil tlominate this zone for a hrief time in April

Summer is the least favorable season. Drought and high teml)eratures kill inany plants and cause the mosses and lichens to look dried and ihriv-elled. Yet there are always a feiv species l~looming such as scattered plants of Hypcricltill gllizfia-izoidcs, antl \vliere moisture is sufficient, great ~iumhers of ieedlings of l/icjzticm occtlr (F ig 2. D ) . By September, the T'igldicrtr. xvliich pro-duces scattered floivers as earl! as June, is in full l~loom, the gray-green, 10- to 12-inch-high plants of Croto~lop.ci.s are mature, and the tall green blatles of Aizdropogn~z ha le replaced the golden hro~vn ones so conspicuous during other seaion, (Fig. 2, A, R 1 .

Graphic representation of the life c\cle, of 22 of tlie most fretluently occurring seetl p la~l t t f ~ o mtlie granite outcrops are presented it1 Fiq. 5, grouped according to time of flo~veritlg and annual or ~xrenn ia l hal~it The data for these life c!clei represent com1)ined obser~at ions from all three outcrops ant1 from an! island cornmunit ieen For example, the height of 1)looming of Polygtrla cirrtissii \ \as in June ant1 J u l ~ . hut I)ecause of a few 1)lants which persi\tetl on ?IIount Arabia, Fig. 5 sho~vs the flowering time extentled ~ n t o earl! KO\ eml~er

The t1111e of germination of annual seedlings a t \\ell a i tinie of flowering of a~inuals antl

SPRING FLOWERING PLANTS ANNUALS

JAN I i r e I RAE I APE I M I " I ._NL I .ULI I i U O I 1 I 'OV I

i i C T l O S T 5 t L L . O T T A Y 1

iUYLI ML.STI7YLmI I

I I I I N L . R 1 B i i i V l i o L n

O,AYOT1P*Ii Y U 0 5 1

L,NL.RIb ChNhDENSIS

L R l i b V , i i C , N # i l

PERENNIALS

C A R E X cOUY_N8S

T R L . D E S C I N - # A O"8EN585

S C H O E N O L i i O * LROCEJM

mncus C E O R C A N Y S

S I N L C 1 0 TOYLN105.5

SUVVER -FLOWERING PLANTS ANNUALS

.A I & . a 1 I _ - x I U-Y I JJNE I JULY I nub I SEPT I O C T I no, I oat

CIPERUS C R I N T O P l L r i

B U L B O S T Y L S C I P l L L l i i S

P O L I C I L I C " R T S S 8 I

C R O T O N O P B S ~ H E i i i l S

H*PERICUY O E H T L N O D L S

Yl'UlERA PORTER1

PERENNIALS-PAH.CUU

U l R l D i O H I L L

IINOROPOBON V # R C # N I C J S

B Y l l P I C L I U C I

TL.L,NUU T L R L T F O L U U ._------------------_______.

CORFUPS1S ORINOIFLORI I

.SLLOLIHC* I I R U 5 N I - I D i n d W E E K N O:TOBLR

I . 5. Schematic reprewntation of life cycle. 11i cl~aracteristic outcrop plants. (;reell \-egetati\.e part.; ahove the ground a re represented hy a solid ha,e l ine: dormancy hy a broken base l ine; time o i flowering 1)y solid line a11oL.e base line; and time o i iruitilig hy wlid l i~ le a l ~ o ~ e floweritig line. Pc~tzicu?~~ llas t \ i o~ ~ ~ c r i d i ~ i l t u l ~ ~ flo\vering and fruiting periods per year.

ting-spritlg-floivering antluals are preparing to 11100111 or are completing their life cycles. One e s c e ~ ~ t i o nwas summer-flo~vering Hypcricttnz gclz- titr~loidcs which germinated in Octol~er, over- ~ v i ~ ~ t e r e dtiny seedlings, and growthas resumed [luring March. Flowers began appearing on Hjlprr.ic-11i1~~ ~ l a n t sin A l q , and flowers antl fruit were produced until the plants diet1 in I1ecenil)er. 13y that time, the seeds ~jrotl~icecl that summer

perennials separate the species 111to t \ \ o q r o t ~ l ~ s hat1 alreatly gern~inatetl mitl the seedlings ivere The sumlller-floweri~ig species ger11iiii;lte after \Tell established. This life span of more than 12 the first of the e a r nhen the fall-germina- mo~itlis is in sharl) contrast to that of h'/tlho.stj~li.s

Spring 1964 GRANITE O r T C R O P C O M M U K I T I E S 303

rapillaris, another summer-flowering plant ; 6 months (Fig. 5) is prohably longer than the life span of any individual plant of this species g r o w ing in the island communities and may represent several generations. Young seedlings ivere ob-serled in Ma!, June, July, August, 0c tof)er . and November X period of drought late it1 June k~lled many plants before they had bloo111ed I n .Augubt onl! a fev living mature specltnens \\ere observed, but there were many young seedlings which, it is presumed, completed their life cycles 11) Yovember

The ~ v o r d community has been used in this stutly to designate a group of plants liting in a restrictetl area, specifically, a soil-filled depres-sion st~rrounded 1)y exposed rock. The classifica- tion of the comtnunities is suggestive of sera1 stages in primary succession from the tlian~orpha to the herb-shruh community. There are itldi-cations that such a succession of vegetation does take place as the soil in the depression deepens 130th the deposition of organic and inorganic ma- terial on the surfaces of the comniu~iities and the tlecom~>osition of the underlying rock contribute. If the soil I~ecomes deeper and its water-holding capacity increaies, the floristic composition of the community can change.

Present findings tend to support the contet~tion of \\'interringer and lTestal (1956) that the con- ventional secluence on rock surfaces of crustose lichens, foliose lichens, fruticose lichens, mosses. herbs. ~voody plants, is not as important in plant iuccession as the accumulatiotl of soil through physical agencies. The depressions ill the surfaces of the granite outcrops are not nlerely the result of iucceeding generations of plants Depressions have I~een formed by the eufol~ation of the rock, 1)y erosion as na te r runs d o ~ v t ~ the outcrops, and 11) the intel-actions of wind. ~vater , and rock par- titles which may result in a scouring and deepen- illy of shallow clepressio~ls. Regardless of the geologic processes which niay have heen responsi- 1)le for their origins, there is a tendency for or-ganic and inorganic materials to accumulate in depressions. The growth of plants on such ac-cunlulations of coarse and fine sand appears more iianificant than the lichens and moss on the ad- jacent exposed rock in the ultimate establishment of a c l~ma\ cover of seed plants on these ot~tcrop- pings of granite. Similarly, Whitehouse ( 1933) notetl that iuccession was more rapid on acctumu- latetl gral el than on exposed granite at Enchanted Rock, Texas.

I t is tliffict~lt to e\.aluate tile role of Ditrl?zovplzrr

in succession. Dinnzorplza cywtosa has a metabo- lism peculiarly adapted to the extreme conditions of high light intensity and low moisture and min- eral content of the soil which exists in the shal- low, soil-filled depressions on the granite out-crops (Raker 1956, Lammers 1958, Il'iggs and Platt 1962). During the months of their growing season, the plants maintain extensive root systems and a low ground cover which aid in retaining soil particles. Even after death the desiccated plants niay intercept bits of \vindblo\vn and water- borne debris, but the midsummer loss of all mois- ture from the shallo~v soil urevents conversion of dead plant parts to humus as is usual in more mesic conditions. In writing of similar soil-filled depressions in exposed granite surfaces in North Carolina, Oosting and Anderson ( 1939) state that "Such dry depressions usually give no evi-dence of further successional development."

Although the shallolvest diamorpha comnluni-ties, such as Mount Rolla~vay no. 2 ~ i t h a maxi- Inurn soil depth of 2 cm, show no evidence of suc- cession, the number of species increases as maximum soil depth increases. A soil pocket covered with red-leaied D i a ~ ~ ~ o r p l l a aniay have central area of gray-white lichens and white Arctzat,ia blooms in the spring where the soil is 12 cm deep (Fig. 1, B 1 . Clndotiia lcporina, C . carolitziatza, Cawlpyloplts sp., and Rltlbostylis cn-pillaris are the plants most frequently seen as "in- vaders" of diamorpha communities Species of Cladonia ant1 apparently particularly C . / F ~ o P ~ ~ z ( I .

are adapted to survival on exposed, shallow soil 11y reason of an ability to ahsorb and store mois- ture in their ftungal elements, obtain nutrients from their captive algae, and maintain a vegetative year-round plant cmer , thus avoiding a critical stage in their life cycle such as seed plants may encounter. Observations of diamorpha and lich- en-annual herb communities at hlount Aral)ia, Rock Chapel Alountain, and hIount Rollaway suggest that the establishment of colonies of Cladonin lcporitza in island communities initiates a series of successional stages which correspond to the comn~unities listed earlier, i.e., lichen-an- nual herb ( F i g 1, C ) , annual-perennial her11 (Fig. 1, F),and herh-shrub (Fig . 2, B, C ) .

Once established. Cladonia lcporina influences further de~elopment. The exposed soil under the much hranched lichen cover is a microenviron-ment more suitahle for germination of moss spores and spermatophyte seeds than is corn-pletely bare and exposed soil or a seasonal dia- morpha community. Likelvise, the tangled mass of lichens is a barrier which intercepts debris the deposition of which increases soil depth. The

depositetl tlel~rih 111:~)- incltitle sl)ores, seeds, and l)l;lr~t 1);trth i ~ . l ~ i c ' l ~ newe~lal)le est:tl)lisl~nlent of members of tlie community. I\-ith tlie addition of Illosseh. 1)articularly Ctrlllpyloplis, and seed p l x ~ ~ t s ,the organic content and moisture-holding callacity of tlie sul)stratum is ~narkedly iticreased, hut t l~ir ing 1)eriods of tlrought, tlie soil of 1iche11- annual her11 cotnmtmities may become completely drietl, liniiting this stage in successio~i to annuals, lichens, ant1 drought-resistant mosses.

If tlie soil tleptli exceeds 15 cln, coml~lete clryi~iq out of tlie soil is less likely and pere~inials may iiivade the area. Sr - l locno l i r io~~r-rocc7liil~\vliich

irl~~crica~zliirl ,,'lscy1.1~711 Hypcr ico idrs , aiid Iilir.ritr ~lrnri~zrrto occur. l lToody vines can invatle such isla~ld comnlunities, and if moisture conditions are favorable for several consecutive years, seedlings of P i ~ z l i s Tncdtr, J~inipcrlrs ~'irginitrnir. ..31iii.lnt1- chicr nrborrn, P ~ I ~ T Z Z L Y sc~,ofit la,and 1 7accinilclll trr- borclrlrt nlay sur\-ive.

Parallelitlg the successional stages represented by types of communities, there are a l~pare~i t ly seral states lvithin the con~munities. T h e zones of platits, especially evidetit it1 annual-perennial herb conimunities in the spring, correspo~id roughly to tlie d ia~i ior l~ha ,lichen-annual herb, and annual-

has a deeply 1)urietl corm, and Scnccio t o l ~ ~ i ~ n t o s ~ ~ s perennial her11 comn~ut~it ies. Given a co~ntiiunity \vhich has a woody untlerground stelii and a tleep root system. are t\vo drought-resistant perennials which are amollg tlie first to appear. X sure sign that tile annual-perennial stage ill succession has 1)cen reached is an extensive niat of the hair call moss, I'ol~'fi-ic-ll~riii._\lthough occasio~ially occurring as scattered patches in lichen-annual herb conliilullities, this moss is frequelit and often a l~unt la~l tin deeper soil conlniunities. I n the sanie \Yay that species of C'ltrtlo~liashelter annual seeclings, I'olytr-ir-l~lililappears to be a biotic factor i n t l ~ i c ~ ~ t i a lin initiating tlie annual-perennial stage of succession. C'orrelatetl with the greater depth o i soil in this seral staqe are soil particles of a Illore uniform size, increased capacity of ~va t e r re- tention, :untl a co~n l~ le t el~ l an t cover tlistinctly ia!,c~retl except in sliallo~v l)cril~heral soil.

T h e increase in plant cover and diversity for each of tlie a l~ove succeetli~ig seral stages is ac-coml):~iietl by a corrcsl)onding ii1crc:ase i l l the orga~i ic cunteiit of tlie soil, and a tlecrcase ill acidity. Soils 2 cnl (lee11 llave an org:uiic content of al1o11t 37h and a p H of :tl)out 4.0: those 10 cm tlcc.1) :in organic contrnt of a l~ou t1'; ant1 a pH oi a l~out 4.2 2 :111(1 tliosc over 15 ~ 1 1 1 ;t11~ I c v ~ ) organic content of a l~ou tS? antl a 1113 oi nllout 1 .5 . ( 1 )ctailed data \\.ill be publislied sul~se-q"c11tly. ,

\\-it11 the increase in total vegetatio~i, it is as-~111iictltliat the soil in islantl com~iiunities grntltial- ly 1)econies tleel~cr 1)y tlel~osition of tlcl~ris, decay of 1)l;lnt parts. :ui(I tlisintegration of the ~iiltler-

\\-it11 a centrally located population of S r n r c i o t o i ~ c v z t o s ~ i s , brr.7';-ringed \\-ith a zone of L ' l r c n a ~ i n folia, lichens, and mosses, exterior to which is a pure stand of Dinl~zorphn i Fig. 1. E ) , tlie assump- tion is that tile central portion of a sliallo~v dia- niorpha co~iimunity was invaded first by lichens, mosses, and a~inuals , and secondly, as soil con-tinued to heconle deeper a t the center of the coni- mu~l i ty ,S r ~ l c r i oi~ivaded and colo~lized the deepest soil, displacing the second seral state ~vhich per- sisted 011 soil intermediate in depth betwee11 that necessay for S ~ v l r c i oa~l t l tlie very shallow periph- eral soil on \vIiich only Ditrilloi.pho: r-y~llosa call s ~ ~ r v i v e . T h e resultant zones are similar to tlie zonatioll ~vliich occurs a rou~ id a ~na r shy , fresli- ivater 11o11tl; in both instances.etlal)hic factors de- termine the floristic conipositio~l.

T h e :tl)ove description of st~ccessional zonation ill island conimunities is oversim~~lif ied. De- pressions may deepen irregularly ant1 a perfect l~atteri i of zones arranged as conce~itric rings is rare. T h e nu~iiber of zones ant1 their floristic con~posit io~ivaries from comnlunity to community ant1 from season to season, there freclue~~tly being niore t1i;ln three zones i ~ i aii~iual-~)erciiiiial l~c r l ) communities.

LTntler ~~rehel l t c l i~~ ia t i ccontlitions, tlie Iicr1,-hhrtll) stage appears to Ile all etlnl~liic sull-cliniaz. T h e lack of ~no i s t~ i r e , of co~ii l~eti- 11-liether I~ec:l~ise tion. i~isufficient rainfall, or sliallo\v soils, liniit.; tlie occurrelice of tree species aiid shortelis tlie lives of tlie fe\v trees \vhicli (lo beconie estal)lisl~ctl.

l!.i~lg rock. T h e thirtl seral staqe, an 1ierl)-~lirul) .As rece~itly as 1916, small stantls of AY~'s.cabifloi.o. co~iiniunit!-, i.q reaclietl \\-lien niasi~ii tun soil c le~) t l~ ant1 I'inlis Ttrrtltr g r e xJlr~iipri-11s ;-ii,gi~lic~lia, on is o\.ci- 10 cnl, t l ~ e soil level is eclual ~vitli o r S t o l ~ c 3Iou11taili (Cnn~l,l)ell lC121 r \\-here I I ~ \ \ . slightlj- :~l,o\.c thr surface of tlie surrountling t-ocli there ni-e 110 trees. D r . I\-. 13. E:~l<er (p~-i \ . :~tv with little or no margi~ial zone of shallo\v soil (I;i:,.. 2. 13 I , antl the tlominants arc perennial Iierl~s ant1 \\-ootly plants. Yucli a community acts as :L s13011gc, holding \v:~ter \vliicli runs in ant1 :~1-lo\vi~ig only a relatively smnll arno~lllt to run o11t. Conditions arc niesic enough for I?i.~fli i ,o~lii iri i

communicatio~i) of l<mor!- University stated tlint these trccs (lied after a very tlr>- summer. I t i h

reaso~i;ll)lcto supl>osc tliat similar events have oc- curred on other granite outcrops.

r .lhesc conimunities cannot I)e fully understootl without placilig then1 in their geological IJerspec-

Sprillg 1964 G R . \ S I T E OUTCROP C'O.\1IRlUI;II'IE~ 30..

1!

tive. T h e I'ietliiiont Plateau is comr>osed of a great sliieltl of granite scliists and gneisscs, which varies horizontally antl vertically in its resistance to \\.eatliering. 1)ecp soils overlie the more easily \vcntlieretl materials, whereas the protlucts of de- conil'osition \\-ash away a s fast a s they a r e formed 011 the more resistant areas , leaving tlie slightly weatlierctl rocli es l~ose t l , either a s lo\\- tlomcs o r a s f a t rocks, often several hundred acres in ex-tent. Stone l l o n u t a i n is a n extrusion of harder granite into the grani te schists and gneisses of the plnteau (luring the Permian I'criod. Su l~sequent -

the ,urrountling softer modified granites such :I> those of l l o u n t Arabia. Rock Chapel h loun-tain, ant1 1 I o u n t Rollaivay, ha\ e s lonly ncathcrct l a\va!. 1e:lving the more resistant mass a s a nio-natl~iocli standin:,. 200 m a1,ovc the surrounding plaiii. A s erosion continues in geologic time, tlie e x l ~ ~ s e t l ,liartlcr lenses eventually disappear with- out ever hnving hat1 a complete cover of soil a n d veqctation. lca\ iny exposed the softer, soil-pro- ducinq s trata . Concomitantly, atlditional harder lenses a r e lleing uncoveretl in other areas. Thu , . \vliile no one outcrop has persisted intlcfitiitelv. collectively the outcrops ha\-c been prcscnt long enough for the e v o l ~ ~ t i o i i en-of tlie many species deniic to them.

Tliis tl!n;uiiic concept of continual change em-plinii/es the fact that ~e a r e vie\\-ing these com-munitics a t x specific moment, ant1 that evcli a t this moment , sliorter cycles of change :1re going (111 \\.ithi11 tlie lax-gci- oncs of liuntlretls of thou-s:uitls of yeat-s in l(,ngtli.

I'cr11:1lv tlie ~ i ios t rapid erosional process is tlic conti~it~;,lexfoliation of the csposcd rock caused 11v thc. formation of exiolintion sliclls which vary 111 tliicl,~lc.>s from several centimeters to 2 or 3 111

antl \ a r j i l l tliametcr f rom a i e n to ;L huil t l~ c~tl 01-

iiiorc ~iic,tc~i-s (Hol)son 1'158). IYintei- cold aiitl s1uiiiiier hent help to loosen and fragnieiit the u11l)er shells \\-liicli qrndually \\-eather away. T h i s I I I - O C Y S H ~ ixfo1i;~tioii may destroy isla~ltl coiii- liltmitics \vliicli li:l\.e t;llieii lluntlreds o i years to tie\-clop.

I t is tliffic~llt to estimate the age o i island com- iiitmitics. T n - o soil sxmples iron1 tlie bottom of a comniunity at the to11 of l l o u n t ;li-al~~ix.\vIiere contlitio~is a r c niore s t a l ~ l r th:l~i 011 tlie slol)es, !.icltletl c:lrl~oii-tlxtctl ~iixtcrial n l ~ o u t 670 !.ears oltl. Ti1 terms of litmian ol)ser\-ation, this indi-

community. 1-egct:~tioiial changes in some islantl coni~iiunitics illav 1)e too slo\v for o l ) se r \a t io~ i of all!. s~~ccessiconalcllnnges in a 1ium:ln lifetitile.

Hence , tlie iilaiitl conimtmities have lleen per- iii:lneiitl! iiiarl\etl so tlint changes over 5. 10, 50,

o r even sevcral hundred years may be observctl o n a q ~ i a n t i t a t i \ e as ne l l a s a clualitative basis T h i s provides for one of the tnost \aluahle ant1 also inost difficult kinds of ficltl studies, n a m c l ~ . that of changes in our natural e n \ i r o n n ~ e n t through time.

\Ye \\.is11 to thank I l r . hInso11 E . Hale. S m i t h i o ~ ~ i ; ~ n Ini t i tut io~i , ior the identification oi liclic~i<. Tlr. T.e\~ii E . .L\nderson. Duke Univeriity, ior the itlcntificatio~i oi mos>ca, Dr. 11-ilhur H. Duncan, University of Georgia, ior hc l l~ \\-ith the idc~itification of some oi the grasseb. and Dr . S I e ~ ~ z i c 12al~ora-Stuivcr of the Geocliro~~onietr ic tory. Yale University, for the radiocarbon dates, a i sul)- portetl hy I S 1 2 grant G-19080. 11-c gratciully ; ~ c k ~ ~ o \ v l - cdgc the assi i ta~ice o i Robert Pedigo, College of 11.illian1 and Mary , in the collection o i tlic physical tlata, a~it i thank John T. SIcGinnis for the photographi taken in 1962. T h i i \\ark \\.a\ .uliliorted in part I)y colitl-act S o . A T (40-1 'I -2112 I\ it11 the Atornic Energy Corn~nia-sion.

Baker, W. B. 1915. Studies of the flora o i tile granite O L I ~ C ~ O I I . ' ~f Gcorgia. Etnory Univ. Quartel-ly 1 : 162-171.

---. 1956. Sorne i~i tcrest ir ig plants on tlie g r a ~ l i t c outcrol), of Gcorgia. Gcorgia Mineral S c \ v . ~ I c t t c ~ . 9: 10-19.

Breen, R. S., and R. A. Pursell. 1956,. I f o r e mo>.cc fro111 Stonc LIountain, Gcorgia, a n d vicinity. 1;1-y-ologist 59: 184-186.

Campbell, E. G. 1921. Some asnects o i Stonc LIoun-tain a n d its vegetation. Proc. Ind iana ;\cad. Sci. 31: 91-100.

Clebsch, A. 1953. Rryological riotc on S t o ~ ~ e1Iou11-tain, (;eorgia. Rryologi>t 57: -30.

Cooper, W. S. 1913. Tlie cl imax forest of Is le I<oy:ile. l.ahc Suprr ior , ant1 its 11. Tliet l e v c l o l ~ n ~ e t ~ t . .uc-cez,io~l.. J:ota~i. (;;Iz. 55: 115-110.

Cotter, D. J. 1958. Studic. on the ecological liiv hi.tory oi Po~ti i l (rc-cr .S'iiitrllii P . 11-ilson. 1'h.D. T l ~ c q i . ~ .Eniory Lni\er.ity. . i t l a~ i ta , Ga. 93 11.

Cotter, D. J., and R . B. Platt . 1959. Studies on the ecological life hiqtory of ,C?itiillii.I J o i ~ t i ~ / t r c n Ecolopy 40: i15l-hOS.

Fernald, M. L. 1950. ( ; ~ - a y ' s 111anual o i Ijotatly. 8th etl. -4rne1-ican 1:ook C'o.. S70rk. 1632 11.

Harper, R. M. 10.30. ( i ran i te outcrop vegetation in Alal~ati la . Tot-I-el a 39: 153-1 59.

Herrmann, L. A. 1953. ( ;ei~logy of tile Stonc lIou11- tain-1.ithonia d i > t ~ - i c t , (;col.gi;i. Geologic;il Survey of Georgia 1:ull. 61: 1-139.

Hopson, C. A. l<\ iol i :~t ion ant1 \vea t l~r r i~ ig- 1958. at Stone hlountain. (;eorgi:r, and the i r bearing on tlis-figurement o i the Coniederate S le~i ior ia l . (;eol-gia hlinel-a1 Scw>let tc .r 11(3) : 65-79.

Iceever, C., H. J. Oosting, and L. E. Anderson. 1951. P l a n t aucccqsion on cx l~oscd granite of Rocky F a c e h l o u ~ l t a i n , :Ilexa~itler C o u ~ l t y . Nor th C a r o l i ~ l a . Bu l l . Tor rey Rotan. C l ~ i b 78: 101-411.

Lammers, W. T. 1958. .4 s tudy of eel-tain e n v i r ~ n -

306 COI,BERT E. CUSHIXG, JR . Ecology, \*ol 35, KO. 3

riiental : ~ n d physiological fac tors influencing the adaptation of three granite outcrop e~idemics: Am-plliailtlizls pzrsillirs Tors. , Isoctrs iliclatiospovcz Eng-lem., and Diantorpiicz cyiriosa (Nut t . ) Britton. Ph .D. Thesis, Ernory University, Atlanta, (;a. 85 p.

Matthews, J. 1931. X survev of the flora of Mount Parlola. hl..A. Thesis, Emory University, Atlanta, Ga. 93 p.

McCormick, J. F., a n d R. B. Plat t . 1962. Effects of ionizing radiat ion on a na tura l plant community. Radiat ion Botany 2: 161-188.

McVaugh, R. 1913. T h e vegetat ion of the granit ic flat-rocks of the southeastern United States. Ecol . Monographs 13: 119-165.

Oosting, H. J., and L. E. Anderson. 1939. P l a n t suc-cession on gran i te rock in eastern Nor th Caroliria. Rotan. (;az. 100: 750-768.

Schnooberger, I. 1948. r iddit ional mosses of Stone Mountain, Georgia. Eryologist 51: 32.

Schornherst , R. 0. 1935. 3Iosses of Stone Mountain, Georgia. Bryologist 48: 29-33.

Smith, M. L. McC. 1938. -4 study of the flora of Stone Mountain and vicinity. hf.S. Thesis , Emory University, .Atlanta, Ga. 71 p.

Watson, T. L. 1902. h prel iminary report on a part of the granites and gneisses of Georgia. Geological Survey of Georgia Bull. 9-A: 125-143.

Whitehouse, E. 1933. P l a n t succession on central T e x a s granite . Ecology 14: 391-405.

Wiggs, D. N., a n d R. B. Plat t . 1962. Ecology of Dicz~ilorplza cyviosn. Ecology 4 3 : 653-670.

Winterr inger, G. S., and A. G. Vestal. 1956. Rock-ledge vegetat ion in southern Illinois. Ecol . Mono-graphs 26 : 105-130.

P L A N K T O N A N D W A T E R C H E M I S T R Y I N T H E M O N T R E A L RIVER L A K E - S T R E A M S Y S T E M , S A S K A T C H E W A N 1

.Ibstvact. The relations bet\veen dissolved solids, xvater exchange, and productivity were studied in a lake-stream system. Detached epilithic algae composed the plankton of the upper stream, but all other stations displayed a lentic plankton. Turbidity and unsuitable substrate prevented the development of sessile algae on the hottom of the slow-flo~ving river sections ( lakes) , but the macrophytes provided a surface for the development of these cornmunities. T h e concentrations of total dissolved solids and those ions essential for photosynthesis and algal nutrition decreased downstream coincident, in general, with a downstream increase of phyto- and zooplankton numhers. T h e different rates of the decrease of these tfutrients in the upper stream and lake-stream section a re probably related to the greater populations of autotrophic biota in the latter section. .Assuming that nutrients a re being added to the system throughout its course, then the downstream decrease indicates that the increasingly large autotrophic populations a re assimilating these nutrients a t a rate greater than they a re being added. Thus, the deciding factor regulating the accumulation o r removal of tiissolvcd nutrients in a lake-stream system appears to be the degree of autotrophic enrichnient of the systeni.

Studies of the relationshil~s between dissolved ions and related aquatic environn~ents usually con- cern either lakes or streams. Less attention has been directed toward the nature of the exchange of ions between lakes and streams or to the effects of these euchanges upon the productivity of a lake-stream system. Hynes (1960) suggests that lakes may act as "fertility traps." removing dissolved ions and depleting the supply of es-sential nutrients in the outlet stream. Alterna-tively, the ionic constituents mav increase down- stream due to progressive accumulation from the

' T h i s work was supported by ' .SF. Research Grant G-12428 and is taken from a thesis submitted t o the College of Graduate Studies, University of Saskatchewan, in partial fulfillment of the requirements for the degree of Doctor of Philosophy.' Present address : Biology Laboratory, Hanford Lab-

oratories. General Electric Company, Richland, IITash-ington.

drainage basin. Both theories have been ad-vanced to explain data from studies of lake-stream systems (Berg 1943, Reimers, Maciolek, and Pister 1955. Hynes 1960). The principal factors ~vhich influence ionic changes are ( 1 ) the geologi- cal substrate and size of the drainage basin ; ( 2 ) the quality and quantity of the biota, especially the producer organisms: (3 ) the size and degree of eutrophication of the lakei : (4) the pli! 4cal char- acteristics of the stream; and ( 5 ) ion exchange with the aquatic substrate.

The present study investigated the relationships hetween the dissolved ions, water exchange, and net plankton in the Montreal River and four associated lakes.

DESCRIPTIONOF STUDYAREA

Geology The drainage basin of the Ylontreal River is

situated immediately south of the Precambrian