23

THE FOREST VEGETATION OF WILSON MOUNTAIN,

TENNESSEE

Ho Ro DeSelmProfessor of Botany and Ecology

The University of TennesseeKnoxville, Tennessee 37916

Wo Ho Martin, 111Director, Division of Natural Areas andProfessor of General Science Studies

Eastern Kentucky UniversityRichmond_ Kentucky 40475

and

Eo ThorProfessor of Forestry and Ecology

The University of TennesseeKnoxville, Tennessee 37916

ABSTRACT

The vegetation of Wilson Mountain, part of Little BrushyMountain° Morgan County, Tennessee, has been investigated. Fivewidespread community types occurred; these were dominated byshortleaf pine on south ridges, tulip poplar in most north draws,northern red oak on the highest north slopes, white oak on mostlow elevation slopes but also dominating certain high slopes; andchestnut oak which was the dominant on most middle north and uppersouth facing slopes. Plots were placed on a two axis array usingelevation on the vertical axis and aspect slope position-slopeangle and A horizon thickness on the X axis. Basal area decreased47 percent, diversity decreased 40 percent and density increased31 percent across the meslc to xeric axis. Species distributedalong a continuum index axis exhibit distinct peaks. Othercharacteristics of species and communities are discussed.

This is a paper from the Botanical Laboratory of TheUniversity of Tennessee-Knoxville, N.S. 398.

24

INTRODUCTION

This reports several of the characteristics of the forestvegetation, soils and sites of Wilson Mountain. The need for suchinformation has arisen since the publication of accounts of similarvegetation in Kentucky (Braun 1935, 1942, 1950), since thedevelopment of various comparative procedures (cf. Whittaker 1956,Curtis and Mclntosh 1951, see also Shimwell 1971, andMueller-Dombois and Ellenberg 1974) and as an aid in managementof this diverse forest area.

Of particular use, since the time of Humbolt (1805) arediagramatic schemes representing distribution of landscapevegetation features. Lifezone distribution was shown on the SanFrancisco Mountains by Merriam (1890). Diagrams have beenpopularized in western mountains by Whittaker (1960) and Whittake_ _and Niering (1965) and in Southern Appalachian mountain vegetation(Whittaker 1956).

The writers acknowledge the assistanceof the late Mr. J.S.Kring, Mr. R.L. Duncan and Mr. Bob Kelly in the field and thecomputer programming of Ms. Virginia Patterson. The work received

support from the University of TennesseeDepartments of Forestryand Botany and the Graduate Program in Ecology.L

i!!_iiiiiili!iiiiiiiiiiii _

25

Wilson Mountain is the western end of Little Brushy Mountain,Camp Austin and Petros Quadrangles_ Morgan County_ Tennessee, at84 o 30 _ west longitude and 36 o 2 _ 30 '_ north latitude° It lies atthe southern end of the Cumberland Mountain portion of theAppalachian Plateau Province (Fenneman 1938), 40 km (25 mi)southwest of the Pine Mountain overthrust block which has mountainsto ].263 m (4145 ft) and within the Wartburg Basin with peaks to1006 m (3300 ft) (Wilson and Stearns 1958)o

The mountain here is a ridge with the long axis extendingnorthwesto.southeast to elevation over 671 m (2200 ft), 244 m<800 ft) above valley floors° Comparable ridges lie three km(two mi ) north and about 2°4 km (1.5 mi) south and southwest.The mountain_ an erosional feature produced by the Obed River,is underlain by horizontal Pennsylvanian rocks of the Pottsvilleseries composed of fine-grained sandstone layeralternatingwithbeds of shale° siltstone and coal (Wilson, et al., 1956; Wilsonand Stearns 1958). Soils have been reconnaissance mapped asMuskinghum but the deeper io1_1_5 m (3.5-5 ft) silt loam A, siltyclay loam B profiles on slopes are usually Shelocta_ and thesha!lower_ ridge position, sandy-loam to loam A, sandy loam tolight sandy clay loam B profiles are usually mapped as DeKalb(Springer 1966).

Climatic data fall into the humid mesothermal type with littleor no water deficiency in any season and thermal efficiency normal(B3B'2rb3) in the system of Thornthwaite (1948). Loca| temperaturevariations near Knoxville as seen on opposing slopes of a smallvalley _roduced mean variations of 1.7 C (3 F) equivalent to 290 m(950 ft) of elevation and 322 km (200 mi) latitude (Shanks andNorris L950)o Orographic precipitation and temperature changeson large nearby mountains are typical (Shanks 1954; Smallshaw1953)_ and similar climatic effects of aspect and elevation onWilson Mountain, while probable, have not been measured directly.

26

Settlement by European people, mainly of Swiss and Germanextraction, occurred in Morgan County in the ear}y 1800's (Freytagand Ott 1971). However, population levels in general have remainedlow--only 70 persons per square km (27 persons per square mi) in1960 (Bureau of the Census, 1961). Farming operations have beencarried out on the valley bottoms and in small plots on mountainbenches and crests. Probably scattered timber cutting on thestudy area occurred until the early 1900's when the Bryn MawrMining and Land Company acquired the land. Some timber was removeduntil 1937, and again by the University in 1948 when the presentUniversity Forestry Department management policy policies wentinto effect. Between 1953 and 1957 coal beds at 488 m (1600 ft),548 m (1800 ft) and 579 m (1900 ft ) were stripped (Anonymous1950). Some benches and outslopes have been planted to Pinustaeda plots. In 1953 a forest fire damaged the northwest cornerof the mountain outside the study area (Kring 1966) and someevidence of ground fire was seen on the plots established atlower elevations in the study area.

The vegetation of the Cumberland Mountains of Tennesseewas first described by Breazeale (1842) who said:

II

• . .Dense groves of large and towering pine, poplar,

chesnut, hickory and oak, cover the ridges, and highlands;

and along the margin of the riwers and large creeks,

beech, maple and sugar-tree are sometimes found; and in

many places groves of buckeye, linn and birch; but the

principal growth along the margin of the streams, is

tall white, and spruce pine, with dense laurel thickets,

and sometimes a narrow cane-brake, as an undergrowth."

Implicit in this are the upland pine and oak communities soprevalent in Southern Appalachian descriptions (cf. Cooper andHardin 1970, Whittaker 1956, Braun 1950, Martin 1971), and theravine mixed mesophytic forest segregates--some with hemlock (Braun

1950). His recognition of swamp forest is suggested by h_s mentionof cane. Of this area Killebrew and Safford (1874) note a groveof walnut timber...six feet in diameter .thickly set upon the soil,in .... ' ""company with massive whlte oaks. Later Killebrew (1877)reported pine and black oak nearWartburg and extensive white oakstands there. Northward he notes, "walnut in coves, poplars onnorth slopes .... also buckeye and ash...and in bottoms...beech."

z

27

Little ecological work on the Cumberlands appeared until thework of Braun in Kentucky (Braun 1935, et seq.) who describedextensive cove and north slope mixed mesophytic forests, extensiveupper north slope and south slope oak forests and pine and oak-pinestands on shallow soils chiefly with south arid southwest aspect.Yhe vegetation of this area is placed by Braun (1950) in the CliffSection of the Cumberland Plateau Section of the Mixed MesophyticForest Region although in 1942 Braun granted outlier status to themountains to the east and south of the Pine Mountain overthrustblock (the main Nixed Mesophytic Forest area)° They were treatedin 1950 as more nearly similar to plateau vegetation°

Two studies have recently been made within the Wartburg Basin.One is that of Cabrera (1969) who worked in a virgin northeast tonorthwest to southwest facing deciduous stand on Ash Log Mountain,567o-989 m (1860 to 3240 ft) elevation 32 km (20 mi) northeast ofWilson Mountain° On Ash Log Cabrera reports mixed mesophyticsegregates sugar maple-tuliptree-basswood_buckeye in north-facingdraws, a sugar mapleored oakotuliptreeoblack locust segregate onnorth and west leads, and west and northwest draws and flats. Lowde_isity of seedling and saplings of locust suggested its onlytemporary dominance_ A chestnut oak_black locust forest dominatedleads and draws of west and southwest aspect. Martin (1966) firstdescribed the Wilson Mountain forests° The forest reproductionportion of this study has already appeared (Thor, Martin, Deselm1969). The area has also recently been used in remote sensingresearch (Krumpe, DeSelm and Amundsen 1971 and DeSelm, Amundsenand Krumpe 1972) and in loblolly pine planting trials (Thor andKring 1964)o

METHODS

Transects extended from the base (elevation 415-429 m, 1360ft) to the top .of Wilson Mountain (elevation 693 m, 2260 ft) onridges and in draws on northeast and southwest exposures. Twentytransects were established--five each on ridges and draws on eachexposure° On each transect an average of ten sample plots .081 ha(0°2 A) in size were established_ these were separated by distancesallowing 24-30 m (80-I00 ft) rise in elevation° Areas within 30 m(100 ft) of the road and strip mine area were excluded (488, 548,579 m; 1600_ 1800, 1900 ft contours). In all, 199 sample locationswere seiected_ 102 on the northeast and 97 on the southwestexposure° The total sample was five percent of the forested area.

28

At each plot all trees over 12.7 cm (5 in) dbh were recordedby 5.1 cm (2 in) diameter class and selected soil and sitecharacteristics were measured. Aspect was determined by handcompass to the nearest few degrees; slope position by reference tothe topographic map and use of an altimeter; slope angle wasdetermined by clinometer and this value was used in adjustingplot size following a method described by Bryan (1956). Soilproperties determined in the field were type and thickness of the0 layer, thickness of the A horizon, depth to rock, and stonevolume; soil texture was estimated by "feel" (Soil Survey Staff,]_951); soil samples analyzed for pH values were a composite ofthree samples collected from the surface 15 cm (6).

Field determinations of tree taxa were aided by use of keysby Shanks and Sharp (1950) but taxonomic nomenclature followsFernal d (1950).

Relative density and basal area was calculated for alltaxa over 12.7 cm (5 in) dbh by plot; these were summed for animportance value (I.V.) of 200 for each plot. The taxa >100were designated leading dominants or codominants.

RESULTS AND DISCUSSION

Commu_Distribution

Two methods were used to represent the vegetationdiagramatically. The first (Figure 1) shows plot distribution onelevation on the ordinate (here a probable climatic differential,and also representingtopographic change. The large draws of thelower slopes dissipate into small ones above, and flat slopes,infrequent below, increase in area above). The abscissa representsaspect and topography but within each of the four cells is arepeating slope angle X thickness of.the A rating totalling 1-60,with flatand thickto steepand thin anglesand horizonsappearing left to right.

3O

Points are the location of each plot; communities indicated(dashed lines) are dominated by Tuliptree (Li tu), White oak (Qu al),Chestnut oak (Qu pr), Pine (Pi ec) and Northern red oak (Qu ru) asdetermined by leading dominant. The Tuliptree type area chiefly"tulip cove" vegetation included 12 plots dominated by other taxa,

prinus (4), Acer rubrum (4), Fraxinus americana (2),,,Tilia,T____ (1T; the t_ _s t hen__wo__"pure-_The lower White oak type is about 55 percent pure, the area alsocontaining plots dominated by Acer rubrum (3), Quercus prinus (2),

Pinus strobus (1) Liquid _), _]in_usechinata (1), _ (1),and Carya _is ilj. 'The Pi-ne type is 6-9 p-e-r_ _)ure, alsocontaining plots dominated by Quercus prinus (5), Quercus alba (3),Pinus virginiana (i), and Car_ tomentosa_ . The high Wh-iteoak area is without other type plbts,'as indicated. TheNorthern red oak type (14 plots) only includes six plots of thatdominant; Liriodendron (3), Robinia (I), Fraxinus (1), _.(i), Tilia-(i) andT_ ovalis (1) plots are als-o includea_ Themost widespread type, Che_ak,is 84 percent "pure" ,other plots aredominated by Acer rubrum (5), Quercus coccinea (31, Q. velutina(2), _. rubra-TiT,_ i;omenl;osa (1-)_ is (ZT, _dron(1), _. al---i"ba--(1),Pinus e_--(_l), P.--st_(Z) and Liqu_dambar(I). WhTl-e-the absolute location of type bO-un-cIa'ries(Figure 2)was subjective, the lines were drawn to include the greatestconcentration of the leading important dominant. The purity of aschematically shown type, here from 43 to I00 percent (consideringhigh White oak as a type) averaged 70 percent in this chieflysecondary vegetation. Whittaker (1956) indicated, in primarySmoky Mountain vegetation, a 90 percent expectancy in such ascheme.

A second method of illustrating vegetation relationshipsschematically (Figure 2) was obtained by deriving a ContinuumIndex Number for each plot and placing its locations at theintersection of the plots' elevation and Index Number. The latterwas obtained by summing the product of each species ImportanceValue by the species adaptation number, see Curtis and Mclntosh1951, and Cabrera (1969). The plot position and communityarrangement in Figure 2 is similar to that of Figure 1. Lateralplot location suggests that the most xeric communities (indexnumber ca. 200) occur here, but that the most mesic forest (numbers>1700), doubtless present regionally, are absent from the studyarea. Braun's assertion (1950) that mixed mesophytic forests may,when cut, be followed by much less mesic ones may be illustratedhere. The hiatus between the Tuli ptree and Chestnut oakcommunities suggests that more environments (as east facing) occurthan were actually sampled.

(:__

-rl

ELEVATION-FT.

XIOO

ELEVATION-FT

XIOO

w

_3_

o.....

ON

f,_3

c-0

-,o

_,•

•,,

,,.

.o

,,o

oO

o,,

"01

-,o

_,_,

.,_

....

,,o

o

"I

"$

11)

c_

oo

,.c m_

-',

xmo

_="-./..-"

,..

......

o,.o

4c:!.

......

.--

'-0

::_

..

._

"'_"

__,--.

(:}J

_.

.".

o._

.,..-

.I'.

-.

•:,,

'_o

_".

/'._

_,--

-.'_

"'_

.:_

"-.

,.__.

,K._

----

T,..

-_o

......

"-._

o\

=I

"_

l/}

\..

_-

(1)

L"I

_*,._.,-"'_'--"_

......

_,.--_"_'-'--"_

0e+

--::3

0"3

E.,.

,E

LE

VA

T)O

N-F

IX

!00

•--,o

;5._

__

"o"_

tl)E

LE

VA

TiO

N-H

UN

DR

ED

SO

FF

EE

T0

c+ -"'-"

;.•

i:,

_i

_,,

."o

oo_

........o

o,,

•-:

:y_

•o_

.,

o,---A.

-_

e

..--

y..

.....-.

:-

,,,.

..?

,%.L

:.__

:=__._f-...,..._--,--

,_.>,,":

.•._._",,.

\,-:.......

32

Community characters (solid lines and numbers, Figure 2)grade continuously across the diagram. Per acre density and basalarea are more or less inversely related, density, in more or lesshomogeneous areas, increases from 111.5 to 161.5 trees per acreand basal area in distinct but related areas decrease from 107 0.to 63 square feet per acre. These general data levels have been >

reported before (Tennessee Valley Authority 1973) and the inverserelationship is commonly seen. Diversity as indicated by numbers

of overstory taxa per plot decreases from 9,3 to 5.6 acrossapproximately the same axis, a decrease of 40 percent, the samepercent decrease reported in overstory taxa from mesic to xericforest types at the Marshall Forest (Lipps and DeSelm 19_9).

Species Behavior

Comparison of the "distribution" of taxa on the schematicbackground is of interest (Figures 1,2,3). A series of quitedifferent patterns of importance are suggested for the mesophyteTilia, the submesophytes Pinus strobus, Liriodendron, Quercusai'ba,and Q. rubra, and the xero--phytes_uercus 9rinus and Pinusechinata. These diagrams and those of other taxa suggest nointerspecific dependence but instead a strong interaction withthe microenvi ronment.

i

The bimodal nature of the importance of _alba_, found<

by Whittaker (1956) in the Great Smoky Mountains, is suggested ihere also. While not absent between the high and low areas ofconcentration, frequency and density are low at middle elevations, jThis phenomenon, occurring here on 244 m (800 ft) of topography,but to the east occurring on a 1372 m (4500 ft) axis isreminiscent of the downward distribution of life zones on lowmountains shown by Merriam (1890). Another possibility isexpression of two ecotypes with optimum development in differentsites although this phenomenon was not seen by Baranski (1975).

Species distribution across the environmental (directgradient) versus vegetation axes (indirect gradient) (Figures1 and 2) are similar and no good basis other than subjectivechoice has been found to justify one scheme over another. However,from the standpoint of illustrating probable cause mechanisms,the former is preferred. It is noted here that vertical andhorizontal banding occurs on both north and south aspects,although, on large scale topography in the Smokies, summertemperature banding is vertical on south-facing slopes and morenearly horizontal on north-facing slopes (Tanner, 1963).Environmental factors in addition to temperature are in operation

i

i4 o Rope

14-46 4-62000 1500 I000 500

CONTINUUMINDEXNUMBER

NyssoIi• .® D,#_ I_ t I¢3 J 22.2

e,

20 * Acru o _,,,_, """..o • coov -"

10 X oCato . .// '(L)oo_

= ,B,,.,,, "i" ,i ' /- X'

i_< '-° .,,,0 . I ii":'."= _> mlI.ld ,Pt ill . _

..,, w-,"""° it 7/0 * %" ' ,*' 'J

°t, ",-X---,:/" ............._ , ,,._" " ' °°°' " "" Qu pr . . o '*\\

14 o Rop= or TI he

Continuum index Number • . , • , = , i • , . , • t,.___j_..__-__ J..,..

..... 2000 1500 IO00 500

CONTINUUMINDEX NUMBER

Figure 3. Right diagrams are as in Figure 2. Left diagrams speciesare" 1. Ca_ Qv__a]is.,2. Acer rubrum, 3. _ glabra, 4.......... alba, 5 Fagus, 6. Li_quidam_bar_,_s_E__, 8. O__xy_clendron,9. Pinus

st--t-r-6-bus,i0.- Li riodendron, II. Fraxinus americana-,-l-2. _Querc__, 13. dug'ianslnigra,14. Tilia, 15. Sassafras,16. guercusp__nus, 17. Robinia pseudo-acacia, 18. Pinus echinata, 19. Quercusvelu-t{na,20. _0..._SEci'nea,_2T. _. 'ste]lata, 22. _q. _iandica. --

i_i>!!;17_ii!ii,i_il

i i_iii_ii_ii_i77

35

LITERATURE CITED

Anonymous. 1950o REPORT OF INITIAL TIMBER SALE, University ofTennessee lands, Morgan County. University of Tenn. Agr.Expt.Stao_ Tenn, Dep't of Conservation Forestry Div., T.V.A. Div.of For. Rel. Processed report. 5 ppo

Baranski_ MoJ° 1975. AN ANALYSISOF VARIATIONWITHIN WHITEOAK(q_v_#rcus alba L.). North Carolina Agric. Expt. Sta. Tech.Bullo No. -_2T6-o

Braun, E.Lo 1935. THE VEGETATION OF PINE MOUNTAIN, KENTUCKY. Amer.Mid. Nat. 16:517-565.

Braun, E.L° 1942. FORESTSOF THE CUMBERLANDMOUNTAINS. Ecol.Monog o 12: 413-447.

Braun, E.Lo 1950o DECIDUOUS FORESTSOF EASTERN NORTHAMERICA.Blakiston Co. Philadelphia. 596 pp.

Breazeale,J.W.M. 1842. LIFE AS IT IS i OR MATTERSAND THINGS INGENERAL, James Williams Printer. Knoxville. 256 pp.

Bryan, M.B. 1956. A SIMPLIFIED METHODOF CORRECTINGFORSLOPE ONCIRCULAR SAMPLE PLOTS. Jour. For. 54:442-445.

Bureau of the Census. 1961. UNITEDSTATESCENSUSOF POPULATION,1960, TENNESSEE, NUMBEROF INHABITANTS. U.S. Dept. ofCommerce. Washington, D.C.

Cabrera, H. 1969. PATTERNSOF SPECIES SEGREGATIONAS RELATED TOTOPOGRAPHIC FORMAND ASPECT. Thesis. Univ. of Tenn.Knoxville. 111 pp.

Cantlon, J.E. 1953. VEGETATION AND MICROCLIMATE ON NORTHANDSOUTH SLOPES OF CUCHETUNKMOUNTAIN, NEWJERSEY. Ecol. Monog.23:241-270.

Cooper, A.W. and J.W. Hardin. 1970. FLORISTICS AND VEGETATION OFTHE GORGESON THE SOUTHERNBLUE RIDGE ESCARPMENT. IN THEDISTRIBUTIONAL HISTORY OF THE BIOTA OF THE SOUTHERN'-"APPALACHIANS, PART II: FLORA. Virginia Polytechnic Instituteand State Univ. Research Div. Mong. No. 2:291-330.Blacksburg. P.C. Holt and R.A. Paterson (eds.).

36

Curtis, J.T. and R.P. Mclntosho 1951, AN UPLANDFORESTCONTINUUMIN THE PRAIRIE FOREST BORDERREGION OF WISCONSIN° Ecology32.476-4 96.

DeSelm, H.R., C.C. Amundsen and P.F. Krumpe. 1972. PREDICTION OFSITE AND COVERPARAMETERS. Proc. Tenth I.E.E.E. Confo" ppoM3-1throughM3-4. Knoxville, TN.

Dickson, R.R. 1959. SOMECLIMATIC-ALTITUDERELATIONSHIPSIN THESOUTHERNAPPALACHIAN MOUNTAIN REGION. Bull oAm.MetoSoc.40"352-359.

Fenneman, N.A. 1938. PHYSIOGRAPHYOF THE EASTERN UNITED STATES°McGraw-Hill. New York. 714 pp.

Fernald, M.L. 1950. GRAY'SMANUALOF BOTANY.Eighth Edo AmericanBook Co. New York. 1632 pp.

Freytag, Ethel and Glena K. Ott. 1971. A HISTORY OF MORGANCOUNTY,TENNESSEE. Specialty Printing Co. Knoxville. 379 ppo

Humbolt, A. von and A. Bonpland. 1805. ESSAI SUR LA GEOGRAPHIEDES PLANTES, Chez Levrault, Schoell et Compagnie, LibrairesoParis.

Killebrew, J.B. 1877. RESOURCESOF TENNESSEE ALONG THE CINCINNATIANDSOUTHERNRAILWAY.pp. 233-336, In Report of the Bureauof Agriculture, Statistics and Mines-T.Tavel, Eastman andHowellPrinters. Nashville, TN.

Killebrew, J.B. and J.M. Safford. 1874. INTRODUCTION TO THERESOURCESOF TENNESSEE.Tenn, Bureau of Agr. Tavel, Eastmanand Howell, Printers. Nashville. 1193pp.

Kring, J.S. 1966. Personal Communication.

Krumpe, P.F., H.R. DeSelm and C.C. Amundsen. 1971. AN ECOLOGICALANALYSIS OF FORESTLANDSCAPEPARAMETERSBY MULTIBANDREMOTESENSING. Proc. Seventh Int.Symp. Remote Sens. of Env.1-715-730.AnnArbor,Mich.

Lipps, E.L. and H.R. DeSelm. 1969. THE VASCULARFLORAOF THEMARSHALLFOREST,ROME,GEORGIA. Castanea 34:414-432.

Martin, W.H. 1966. SOMERELATIONSHIPSOF VEGETATIONTO SOIL ANDSITE FACTORSONWILSONMOUNTAIN,MORGANCOUNTY,TN. Thesis,Univ. of Tennessee,Knoxville.134 pp.

37

Martin_ WoHo 1971o FOREST COMMUNITIES OF THE GREAT VALLEY OF EASTTENNESSEE AND THEIR RELATIONSHIP TO SOIL AND TOPOGRAPHICPROPERTIES° PhoDo Disso Univo of Tenn.oKnoxville. 358 pp.

Merriam_ Coil° 1890, RESULTS OF A BIOLOGICAL SURVEY OF SANFRANCISCOMOUNTAINREGION AND THE DESERTOF LITTLE COLORADOARIZONA° No Amero Fauna 3:1-113_

Mueller-Dombois, Do and H, Ellenberg, 1974. AIMS AND METHODSOFVEGETATION ECOLOGY, Wiiey_ New York. 547 pp,

Shanks, RoEo 1954o CLIMATES OF THE GREAT SMOKYMOUNTAINS. ECOLOGY35:354-361o

Shanks_ RoEo and Folio Norris, 1950. MICROCLIMATIC VARIATION IN ASMALL VALLEY IN EAST TENNESSEE, Ecology 31:532-539.

Shanks, RoEo and AoJ. Sharp. 1950. SUMMERKEY TO TENNESSEETREES.Univ.of Tenno Press, Knoxville. 24 p.

Shimwell, D.W. 1971, DESCRIPTION AND CLASSIFICATION OF VEGETATION.Sidgwick and Jackson. London. 322 pp.

Smallshaw, J. 1953. SOMEPRECIPITATION-ALTITUDE STUDIES OF THETENNESSEEVALLEY AUTHORITY. Trans.American Geophysical Union.34: 583-588.

Soil Survey Staff. 1951. SOIL SURVEYMANUAL. U.S.D.A, Handbook No.18. Washington, D.C. 503 pp.

Springer, M.E° 1966. Personal Communication.

Tanner, J.T. 1963. MOUNTAIN TEMPERATURES IN THE SOUTHEASTERN ANDSOUTHWESTERNUNITED STATES DURING LATE SPRING AND EARLYSUMMER.Jour. Appl. Met. 2(4):473-483.

Tennessee Valley Authority. 1966. PRECIPITATION IN TENNESSEERIVER BASIN ANNUAL REPORT. Div. of Water Control Planning.Hydrolic Data Branch. Knoxville. 57 pp.

Tennessee Valley Authority. 1973. FOREST INVENTORY STATISTICS,MORGANCOUNTY UNIT, EAST TENNESSEE. Div. of Forestry,Fisheries and Wildlife Development, Norris, TN. 25 pp.

Thor, E. and J.S. Kring. 1964. PLANTING AND SEEDING OF LOBLOLLYPINE ON STEEP SPOIL BANKS. Jour, For. 62:575-576.

38

Thor, E., W.H. Martin and H.R. DeSelm. 1969. NATURALREPRODUCTIONONUPLANDSITES IN THE CUMBERLANDMOUNTAINSOF TENNESSEE°Jour. Tenn. Acad. of Sci. 44:96-100.

Thornthwaite, C.W. 1948. AN APPROACHTOWARDA RATIONALCLASSIFICATION OF CLIMATE. Geogr. Rev. 39(1):55-94o

Whittaker, R.H. 1951. A CRITICISM OF THE PLANT ASSOCIATION ANDCLIMATIC CLIMAX CONCEPTS. Northwest Science 25:17-31.

Whittaker, R.H. 1956. VEGETATION OF THE GREAT SMOKY MOUNTAINS.Ecol. Monog. 26:1-80.

Whittaker, R.H. 1960. THE VEGETATION OF THE SISKIYOU MOUNTAINS°OREGONAND CALIFORNIA. Ecol. Monog. 30:279-338.

Whittaker, R.H. 1966. FOREST DIMENSIONS AND PRODUCTION IN THEGREAT SMOKY MOUNTAINS. Ecology 47:103-121.

Whittaker, R.H. and W.A. Niering. 1965. VEGETATION OF THE SANTACATALINA MOUNTAINS, ARIZONA: A GRADIENT ANALYSIS OF THESOUTH SLOPE. Ecology 46:429-452.

Wilson, C.W., J.W. Jewell, and E.T. Luther. 1956. PENNSYLVANIANGEOLOGYOF THE CUMBERLANDPLATEAU. Tenn. Dept. of Cons.,Div. of Geology. 21 pp.

Wilson, C.W. and R.G. Stearns. 1958. STRUCTUREOF THE CUMBERLANDPLATEAU. Geol. Soc. Amer. Bull. 69:1283-1296.