J,.. TreePhysiology 14,735-749 @ 1994 Heron Publishing-Victoria, Canada

Shade, leaf growth and crown development of Quercus ruhra, Quercus velutina, Prunus serotina and Acer rubrum seedlings

KURT W. GOTTSCHALK us Department of Agriculture, Forest Service,Northeastern Forest Experiment Station,180 Canfield Street,Morgantown, WV 26505-3101, USA

Received July 19, 1993

Summary The study was conducted in an open field to detennine the optimum irradiance for establishmentand growth of two oak speciesand two major associatedwoody species.Half-sib seedlings of black cherry (Prunus serotina Ehrh.), red maple (Acer rubrum L.), northernred oak (Quercusrubra L.) and black oak (Q. velutina Lam.) were grown for two years under shade-clothtents. Eight shadetreatments (94, 70, 57, 45, 37, 27, 20 and 8% of full sunlight) with three replications eachwere used. Measurementswere made on seedlingsharvested at the end of the first and secondgrowing seasons.In the second year, shading significantly decreasedthe number of leaves for all species except black cherry, but only significantly decreasedleaf area in northern red oak. Shading significantly decreasedaverage leaf size of red maple. Average leaf size of black cherry was largest in the intennediate shade treatments and decreased significantly with increased and decreasedshade. Leaf weight/leaf area (mg cm-2) increased signifi- cantly in a quadratic pattern with decreasing shade for all four species. Leaf area ratio (cm2 g-l) decreasedsignificantly with decreasingshade for all speciesexcept red maple in the first year and black oak in the secondyear. Total branch developmentincreased significantly with decreasingshade in red maple and northern red oak, whereasindetenninate branchesincreased significantly with decreasing shade only in black cherry, and short branchesincreased significantly with decreasingshade only in red maple. Keywords: black cherry, black oak, branch development,growth analysis, leaf area, leafarea ratio, leafweight/leaf area, northern red oak, red maple.

Introduction Successfulregeneration of mixed oak (Quercus)forests after clear-cutting depends on the numberand size of oak seedlingsand saplingspresent in the understoryat the time of harvest(Sander 1972,1977, Sanderetal. 1976, 1984).Germination and early growth of oak seedlingsare not limited by heavyshade beneath hardwood canopies, but many seedlings fail to survive in deep shade once cotyledonary reservesare exhausted(Tryon and Carvell 1958, Carvell and Tryon 1961). Shelterwoodcutting, or thinning, increasesirradiance below the canopy,enhances the survival of hard- wood seedlings and usually increases height growth, root development and root/shootratios (Jarvis 1964,McGee 1968,Musselman and Gatherum 1969, Loach 1970,Phares 1971). On many mixed oak sites in the Allegheny Plateau region there are few oak seedlingsand saplings present beneaththe canopy.These standsare replaced after harvestby standsof red maple (Acer rubrum L.) or a mixture of red maple and black 736 GO'n'SCHALK

chelTY (Prunus serotina Ehrh.). Partial removal of the canopy may promote the advance regeneration of oaks allowing them to compete more successfully with maple and cherry following complete canopy removal (Loftis 1990). The present study was undertakento determine more precisely the relationship betweenirradi- anceand the growth of oak, red maple and black cherry seedlings.The work, which analyzesgrowth as a function of leaf area,extends previous researchin which height and diameter growth, root/shoot ratio, biomass production and partitioning of the sametrees were examined(Gottschalk 1985, 1987).The results indicate the value of shelterwood cutting to promote oak regeneration.

and methods A randomized block design of eight shadetreatments with three replications each was used. Twenty-four 2.4 x 2.4 m plots were covered with a 1.8-m-tall tent covered with one of eight gradesof Saranshade-cloth. The light transmittancein the different treatmentswas 94, 70, 57, 45, 37, 27, 20 and 8% of full sunlight. The study areais an abandonedagricultural site on the flood plain of the Allegheny River nearWarren, Pennsylvania.The plots were not watered or fertilized. Although there were minor differences in soil water availability among the shadetreatments, the differences were not significant becauseof the frequent and abundantrainfall during the two growing seasons.Each plot was roto-tilled beforeplanting and a border of aluminum flashing was installed to a depth of 15 cm to preventdamage by rodents.Open-pol- linated seedsfrom one tree eachof northernred oak, black oak and black cherrywere sown in November 1980. Open-pollinated seeds from one red maple tree were collected and sown in June 1981. Each specieswas planted in one row of five with aO.3-m spacingbetween and within rows (1.2 x 1.5 m). Northernred oak, black oak, and black cherry seedgerminated in May and grewunshaded until June 15-17, 1981. Shade-clothtents were then erected,and height, diameterat 2 cm aboveground and leaf number were measured.Red maple seeddid not germinate until mid- to late June, after the tents had been installed. No mortality occurred in black cherry and northern red oak during the fIrst year, but one black oak seedlingand 10 red maple seedlingsdied (no more than two seedlingsfrom anyone treatmentplot). During the secondyear, one black oak seedling and three red maple seedlings died. All four specieswere grown in the sameshade treatment plots, but were analyzedas separate experimentsby speciesbecause of the heterogeneityof variancesamong species. In mid-October 1981, the height and diameter of all seedlingswere measured.In addition, one seedling from each speciesx shade block combination was selected randomly for harvest. These seedlingswere excavatedby hand to recoveras many roots as possible. The harvestedseedlings were partitioned into stem, leaves and roots, dried at 75 °C, and weighed to the nearestmg. The leaveswere photocopied before drying and leaf areas were measured with an integrating planimeter. The shade-clothtents were dismantled in November after leaf fall was complete. The shade-clothtents were erected again in April 1982 before bud break. In late September-early October 1982, the height and diameter of all seedlings were

Materials SHADE, LEAF GROWTH AND CROWN DEVELOPMENT 737

measured.By this time, some black cherry seedlingshad reachedthe tops of the tents and the study was terminated. All remainingseedlings (n = 2, 3 or 4) were harvested in the same way as in the fIrst year, except that branches were partitioned into separatecategories and leaf area was measured with a Li-Cor leaf area meter. Branches were classified by growth habit as indeterminate,determinate and short (Gottschalk 1984). Short branchesconsist of only preformed leaves in the bud with little or no internode elongation. Determinate branches consist of only preformed leaves in the bud with normal internode elongation. Indeterminate brancheshave normal internode elongationand consist of preformed leaves in the bud and leaves formed after bud break.

Data analysis Leaf numbers,leaf areasand dry weights of the 2-year-old seedlingswere averaged for eachplot and the plot averagesused in the analysis of variance.Average leaf size was calculated by dividing the total leaf area of a seedling by the total number of leaves. Leaf weight per unit leaf area (W, mg cm-1 was calculated by dividing total seedling leaf dry weight by total seedling leaf area. Leaf area per unit seedling weight, or leaf area ratio (LAR, cm2 g-l), was calculated by dividing the total leaf area of a seedling by the total seedlingdry weight. Analysis of variance was based on blocks and shadingtreatments. Orthogonal contrastsbased on single degreesof freedomtested the linear, quadratic,cubic and quartic polynomials for a quantitative relationship betweenshade treatment and crown variables.

Results

Number of leaves In the secondyear, the shadetreatments significantly affected the number of leaves produced for all species except black cherry (Table 1). Although black cherry exhibited the largest absolutedifferences and a linear increasein leaf number with decreasingshade, none of the differenceswere statistically significant (Figure 1). In

Table 1. Analysis of variance and polynomial orthogonal contrasts for leafnwnbe~ of seedlingsof four tree speciesgrown in different shadetreatments for two years.

Species Source of R2 df P (F-ratio) P (polynomial orthogonal contrasts) variation Linear Quadratic Cubic Quartic Red maple Block 2 0.108 Shade treatment 0.672 7 0.026 0.006 0.207 0.236 0.990 Northern red oak Block 2 0.009 Shade treatment 0.763 7 0.008 0.010 0.003 0.617 0.020 Black oak Block 2 0.190 Shade treatment 0.637 7 0.039 0.009 0.227 0.018 0.314 Black cherry Block 2 0.045 Shade treatment 0.585 7 0.189 0.036 0.181 0.916 0.237 738 GO1TSCHALK

Figure 1. Mean number of leaves (:t standard error of mean)of seedlings of four tree speciesgrown in different shade treatmentsfor two years. Significant orthogonal polynomial contrasts areplotted for the four speciesas follows: A. red maple--linear, B. northernred oak-quadratic, C. black oak-cubic, and D. black cherry-linear.

8% of full sunlight, significantly fewer leaveswere producedby red maple, northern red oak and black oak than in other shadetreatments (Figure 1). With decreasing shade,leaf production in red maple increasedlinearly, whereasin northernred oak and black oak, it increasedquadratically and cubically, respectively(Figure 1). The analysis of variance explained 58 to 76% of the variation in number of leaves.

Leaf area Shadeaffected leaf area significantly (P < 0.05) only in northern red oak; however, the effects of shade on red maple and black oak approachedthe 5% level of significance (Table 2). In northern red oak, the leaf area produced in 8% of full sunlight was significantly less than in the other shadetreatments (Figure 2). Black oak and red maple also had a lower leaf area in 8% of full sunlight than in the other treatments,but the differences were not significant. Although the shadetreatments resulted in large differences in leaf area of black cherry, the differences were not statistically significant (Figure 2). Analysis of variance indicated that 55 to 72% of the variation in leaf area was due to shade.Leaf area increasedand decreasedin a quadratic pattern with decreasingshade in red maple, northern red oak and black cherry, and in a cubic pattern in black oak (Figure 2). SHADE, LEAF GROWTH AND CROWN DEVELOPMENT 739

Table2. Analysis of varianceand polynomial orthogonalcontrasts for leaf area(cm1 of seedlingsof four tree speciesgrown in different shadetreatments for two years.

Species Sourceof R2 df P (F-ratio) P (polynomial orthogonal contrasts) variation Linear Quadratic Cubic Quartic Redmaple Block 2 Shadetreatment 0.607 7 0.065 0.174 0.064 0.168 0.984 Northernred oak Block 27 0.013 Shadetreatment 0.722 0.023 0.215 0.014 0.078 0.009 Blackoak Block 2 0.054 Shadetreatment 0.640 7 0.067 0.022 0.225 0.025 0.205 Blackcherry Block 2 0.058 Shadetreatment 0.551 7 0.262 0.154 0.096 0.989 0.245

800 A Red Maple 700

BOO "'~ ] 500

1400 -300 -'m 200 100

0 0 10 20 30 40 50 60 70 80 90 100 Shade (percent of flAl s~~t)

"'~ ~ .

~Gi

2. Mean leaf area(:t standard error of mean)of seedlings of four tree species grown in different shade treatments for two years. Significant orthogonal polynomial contrasts are plotted for the four speciesas follows: A. red maple-quadratic. B. northern red oak-quadratic. C. black oak-cubic. and D. black cherry-quadratic.

Leafsize Shading significantly affected the average area per leaf for red maple and black cherry (Table 3). In red maple, the leaves were significantly smaller in 8% of full sunlight than in the other treatments (Figure 3). Black cherry had significantly smaller leaves at both extremes of the range of shadetreatments (Figure 3). The

0.186Figure 740 GO1TSCHALK

Table3. Analysis of variance and polynomial orthogonal contrastsfor averageleaf size(cm1 of seedlings of four tree speciesgrown in different shade treatmentsfor two years.

R2 df (F-ratio) P (polynomialorthogonal contrasts) Quadratic Cubic Quartic

Red maple Block 2 0.658

Shade treatment 0.610 7 0.038 0.528 0.014 0.030 0.396

Northern red oak Block 2 0.574

Shade treatment 0.543 7 0.098 0.059 0.653 0.020 0.175

Black oak Block 2 0.049

Shade treatment 0.445 7 0.797 0.427 0.353 0.382 Black cherry Block 2 0.064

Shade treatment 0.827 7 0.000 0.013 O.OO} 0.002 0.295

25 A Red Ma~

~ 20 ) ~ !:! 16 CI) "m -"~ 10 3- I~ ; ~ 6

0, 0 10 20 30 40 50 60 70 60 90 100

Shade (percent of full Mig1t)

80 v. Black! D. B~ck Cherry

70 "'] 1- ~ ') CD 3020 ~ 60 ~ 10 (/) m "a; -1-i-- Q) CD -J 50 -' Q) co & i ~ \i < 40 -(

30 I , .., , , , , , 1u ...""" , 0 10 20 30 40 50 50 70 80 90100 0 10 20 30 40 50 60 70 60 gO 100

Shade (percent of flJl sun~ht) Shade (percent 01 llAl .OOight)

Figure 3. Mean leaf size (:1:standard error of mean)of seedlings of four tree speciesgrown in different shade treatments for two years. Significant orthogonal polynomial contrasts are plotted for the four species as follows: A. red maple-quadratic, B. northern red oak-cubic, C. black oak-linear (not significant), and D. black cherry-quadratic.

relationship between depth of shadeand leaf size was described by quadratic and cubic polynomials for all species exceptblack oak in which no clear pattern was evident (Figure 3). The analysis of varianlceexplained 44 to 83% of the variation in leaf size (Table 3).

~~~0.371PLinear SHADE, LEAF GROWTH AND CROWN DEYaOPMENT 741

Leaf weight per unit leafarea (W) Leaf weight per unit leaf area increasedsignificantly with decreasingshade for all four speciesafter both one and two years (Tables4 and 5). The increasecpnforms to a quadratic (or cubic) function (Figures 4 and 5). Within a speciesx shadetreatment combination, W values decreasedfrom the first year to the secondyear (Figures 4 and 5). The analysis of varianceexplained 59 to 84% and 77 to 97% of the variation in W for the fIrst and secondyears, respectively.

Leafarea ratio Leaf area ratio (LAR) decreasedsignificantly with decreasingshade for black oak and black cherry in the fIrst year (Table 6). In the second year, LAR decreased significantly with decreasingshade for all speciesexcept black oak (Table 7). The LAR decreasedin a linear or quadratic pattern with decreasedshade in the first year (Figure 6), and in linear, quadratic and quartic patternsin the secondyear (Figure 7). According to the analysis of variance shadeexplained 32 to 89% and 52 to 95% of the variation in LAR for the fIrst and secondyears, respectively (Tables 6 and 7).

Table4. Analysis of varianceand polynomial orthogonal contrastsfor leaf weight/leaf arearatio (W, mg cm-2) of seedlings of four tree speciesgrown in different shade treatmentsfor one year.

Species Source of df P (F-ratio) P (polynomial orthogonal contrasts) variation Quadratic Cubic Quartic

maple Block 2 0.330

Shade treatment 0.702 7 0.009 0.000 0.036 0.969

Northern red oak Block 2 0.018

Shade treatment 0.839 7 0.000 0.000 0.001 0.031 0.604

Black oak Block 2 0.817

Shade treatment 0.592 7 0.046 0.004 0.077 0.508 0.435

Black cherry Block 2 0.582

Shade treatment 0.843 7 0.000 0.000 0.004 0.361 0.456

Table 5. Analysis of variance and polynomial orthogonal contrastsfor leaf weight/leaf arearatio (W, mg cm -1 of seedlings of four tree speciesgrown in different shadetreatments for two years.

S~cies Sourceof R2 df P (F-ratio) P (polynomial orthogonal contrasts) variation Linear Quadratic Cubic Quartic

Red maple Block 2 0.561 Shade treatment 0.827 7 0.000 0.000 0.058 0.345 0.499 Northern red oak Block 2 0.077 Shade treatment 0.850 7 0.000 0.000 0.002 0.363 0.566 Black oak Block 2 0.307 Shade treatment 0.767 7 0.002 0.000 0.027 0.393 0.848 Black cherry Block 2 0.029 Shade treatment 0.968 7 0.000 0.000 0.000 0.049 0.144

0.701LinearR"Red 742 GO1TSCHALK

-9 -, 'I' 'i' D. Black Cherry ~ 8 ~ ~ E 8 E 6 ~ ~ .. ~ 7 Z 5 G ~ ~~ '! / I' :g,6 ~ ~ ~ y G .. -'G r .-' ~ ., , .., , ., 0 10 20 30 40 50 60 70 60 90 100 0 10 20 30 40 60 60 70 60 90 100 Shade (percent01 fLii Might) Shade (percent of ltAl M9'1t)

Figure 4. Mean leaf weight/leaf arearatio (:t standard error of mean)of leaves of seedlings of four tree speciesgrown in different shadetreatments for one year. Significant orthogonal polynomial contrastsare plotted for the four speciesas follows: A. red maple-quadratic. B. northernred oak-quadratic, C. black oak-quadratic. and D. black cherry-quadratic.

of LAR were stable for the oaks and only slightly lower for the other two speciesfrom Year 1 to Year 2 (Figures 6 and7).

Branchdevelopment During the fIrst year,branches were not presenton most of the seedlingsand the few that were presentwere consideredpart of the stem. In the secondyear, most stems produced somebranches and black cherry had 8-15 times as many branchesas the other three species.Black cherry developedfour to sevenindeterminate branches per seedlingin the secondyear. These indeterminate branches then developed secondary short branches resulting in large total numbers of branches. Shading only caused significant differenceswithin branchtypes for black cherry (indeterminatebranches,P = 0.045) and red maple (short branches, P = 0.001) (data not shown). Total numbers of branchesincreased with decreasingshade for red maple and northernred oak (Table 8). Branch numbers increased in a linear or quadratic pattern for all speciesexcept black oak in which there was no statistically significant relationship (Figure 8). Analysis of varianceexplained 42 to 78% of the variation in total branch numbers(Table 8).

Values SHADE, LEAF GROWTH AND CROWNDEVELOPMENT 743

Figure 5. Mean ratio of leaf weight/leaf area(.t standarderror of mean)of leaves of seedlingsof four tree speciesgrown in different shadetreatments for two years. Significant orthogonal polynomial contrasts areplotted for the four speciesas follows: A. red maple-quadratic, B. northernred oak-quadratic, C. black oak-quadratic, and D. black cherry-quadratic.

Table6. Analysis of variance and polynomial orthogonal contrastsfor leaf arearatio (LAR, cm2 g-l) of seedlings of four tree speciesgrown in different shade treatmentsfor one year.

Species Source of R2 df P (F-ratio) P (polynomial orthogonal contrasts) variation Linear Quadratic Cubic Quartic

Red maple Block 2 0.545

Shade treatment 0.319 7 0.633 O.~3 0.532 0.764 0.661

Northern red oak Block 2 0.749

Shade treatment 0.543 7 0.088 0.005 0.413 0.639

Black oak Block 2 0.443

Shade treatment 0.892 7 0.000 0.000 0.000 0.195 0.927

Black cherry Block 2 0.866

Shade treatment 0.776 7 0.001 0.000 0.004 0.260 0.936

Discussion Number of leaves and total seedlingleaf area increaseas shadedecreases in many species including Norway maple (Yakshina 1978), sessile oak (Jarvis 1964, Ig- boanugo 1990), beech (Loach 1970, Masarovicova and Minarcic 1984), several Eucalyptus species (Withers 1979), red maple, poplar and yellow-poplar (Loach

0.089 744 GOTTSCHALK

Table7. Analysis of varianceand polynomial orthogonal contrastsfor leaf arearatio (LAR, cm2 g-l) of seedlings of four tree speciesgrown in different shade treatmentsfor two years.

2 Species Source of R df P (F-ratio) P (polynomial orthogonal contrasts) variation Linear Quadratic Cubic Quartic

maple Block 2 0.376 Shade treatment 0.778 7 0.001 0.000 0.021 0.117 0.115 Northern red oak Block 2 0.261 Shade treatment 0.653 7 0.026 0.001 0.150 0.775 0.595 Black oak Block 2 0.944 Shade treatment 0.522 7 0.103 0.003 0.395 0.588 0.950 Black cherry Block 20.9510.140 Shade treatment 7 0.000 0.000 0.000 0.011

70 11 0 0 Black Oak N~~I a 100) ri -~"~-o;;;;-:D. Black Cherry - 70 60 "'g 5 90 ~ a: ~ 50 0( 80 '"Q a: 40 i70 ~ ~ 80 -( 30 0( ~ m .'1-l OJ 50 I j ~ I 'lO ~ 40 0 10 20 30 40 50 60 70 60 90 100 0 10 20 30 40 50 60 70 80 90 100

Shade (percent at full sunllghf) Shade (percent of full s.n~t)

Figure 6. Mean leaf area ratio (:1:standard error of mean) of seedlings of four tree species grown in different shadetreatments for one year. Significant orthogonal polynomial contrasts areplotted for the four species as follows: A. red maple-linear (not significant), B. northern red oak-linear, C. black oak-quadratic, and D. black cherry-quadratic.

1970), birch (Atkinson 1984), and northern red oak (McGee 1968, Loach 1970,Phares 1971, Farmer 1975). These increasesare usually closely related to height growth patterns (Shirley 1929, Phares 1971, Withers 1979, Blair 1982). Leaf area development largely follows height development for all the species in this study (Gottschalk 1985). Becauseshading had only a minor effect on height growth, it had only a minor effect on leaf area developmentin the shadetreatments above 8% of

1N0.011Red SHADE, LEAF GROWTH AND CROWN DEVELOPMENT 745

200 ~ A Red Maple'I 180 0 'co "'6 160 "'g ~ 140 ~-' 120 ~ 0 8'" ~ 100 II: '" e 80 ~ < -( ~ 80 ! 40

Shade (percent of full sun~t)

"'~ ~ d .Q OJ II: '" ~ < ':. j

Figure 7. Mean leaf area ratio (:1:standard error of mean) of seedlings of four tree species grown in different shadetreatments for two years. Significant orthogonal polynomial contrasts are plotted for the four speciesas follows: A. red maple-quadratic, B. northernred oak-linear, C. black oak-linear, and D. black cherry-quadratic.

Table 8. Analysis of variance and polynomial orthogonal contrasts for total number of branches of seedlings of four tree speciesgrown in different shade treatmentsfor two years.

Species Source of R2 df P (F-ratio) P (polynomial orthogonal contrasts) variation

Red maple Block 2 0.158 Shade treatment 0.775 7 0.002 0.000 0.151 0.069 0.124 Northern red oak Block 2 0.137 Shade treatment 0.705 7 0.012 0.006 0.002 0.653 0.116 Black oak Block 2 0.425 Shade treatment 0.419 7 0.372 0.149 0.468 0.082 0.609 Black cherry Block 2 0.034 Shade treatment 0.571 7 0.274 0.046 0.386 0.931 0.353

full sunlight. The averagesize of leaves increasesas shadeincreases (Blair 1982, Fails et at. 1982, Kappel and Flore 1983, Masarovicova and Minarcic 1984). Because the increasein averageleaf size is sometimesgreater in magnitude than the increasein number of leaves, there is a decreasein total leaf area as shadedecreases. In this

'0 746 GO1TSCHALK

5 ~ 4 .h -3 0 :. ~ 2 ~

:3 Black Oak

2 ~ 15 ~ ~ 1 r 2 i

0 10 20 30 40 50 60 70 60 gO 100 0 10 20 30 40 50 60 70 80 90 100

Shade (percent of fLJI su..ight) Shade (percent of full sunlight)

Figure 8. Mean total number of branches(:!: standard error of mean) for seedlings of four tree species grown in different shade treatments for two years. Significant orthogonal polynomial contrasts are plotted for the four speciesas follows: A. red maple-linear, B. northern red oak-

study,the large decreasein averageleaf size in the highestshade treatment counteredthis patternsomewhat, so leaf areadid not decreasemarkedly with decreasingshade. However, the quadratic relationship betweenshade and leaf area suggeststhat leaf areatends to decline with intensity of shade. Leaf weight/leaf arearatio (W) decreasesas shadeincreases for many tree species including birch (Nygren and Kellomaki 1983, Atkinson 1984), peach (Kappel andFlore 1983), sessileoak (Jarvis 1964, Igboanugo 1990), severalEucalyptus species (Withers 1979),dogwood and yaupon(Blair 1982),red oak, red maple,beech, poplar and yellow-poplar (Loach 1967, 1970), severalrain forest species(Langenheim et al. 1984),and severalCalifornia oaks (Callaway 1992).The changein dry weight per unit area is often accompanied by numerous morphological and physiological changes.In this study,there was a large increasein W with decreasingshade and in the same shadetreatment, the second-yearseedlings had lower values of W than flfSt"'yearseedlings. The mechanismunderlying this decreasewith age is unknown, but it may be relatedto increaseover time in self-shadingas crown size and total leaf area increase (cf. Marini and Marini 1983, Tooming and Tammets 1984). The two oak speciesexhibited smallerchanges in W with variation in shadeand age than black cherry and red maple.The greaterplasticity of black cherry andred maple gives them

~!0, SHADE, LEAF GROWTH AND CROWNDEVELOPMENT 747 the ability to respondto changingconditions betterthan the oaks.Similar differences have been reported betweennorthern red oak and red maple by Loach (1967), and also amongbirches (Nygren and Kellomaki 1983, Atkinson 1984). . Leaf area ratio (LAR) decreasedas shadingdecreased, as has beenobserved in severalspecies (Jarvis 1964, Loach 1970,Phares 1971, Farmer 1975,Withers 1979, Callaway 1992). This phenomenonhas been associatedwith plant size (i.e., the larger the plant, the lower the LAR, becausestem and roots occupy larger portions of the biomass) and should translate into an age-relateddecrease in LAR. In this study, there was also a year-to-year decreasein LAR, but the magnitude of the decreasewas less than that found for W. Although, the uniform shadecreated with shade-clothis a poor representationof actual light conditions in forests,due to the absenceof sunflecksand other irregular- ities in the quantity and quality of light reachingthe understory,the data indicate that seedlings of all four speciesgrow poorly and achieve low leaf areasin deepshade (8% of full sunlight) comparableto that below the canopy of unthinned mixed oak stands.Increasing light to more than 20% of full sunlight (similar to the below-can- opy irradiance after shelterwood cutting) increasedseedling growth and leaf area. I conclude from the results of this study and earlier studies (Gottschalk 1985, 1987) that establishmentand growth of northern red oak, black oak, red maple and black cherry seedlingscan be improved by increasingthe amount of light reaching the understory. Shelterwood cutting to increase irradiance from 5-10% of available sunlight in uncut stands(Chambers and Jenkins 1983)to 20% or more should result in increased growth of seedlings.Although the larger oak seedlingsproduced as a result of the increased light in shelterwoods have a better chance of competing successfullyonce released,the use of shelterwood cutting does not favor oaks over black cherry and red maple. Both red maple and black cherry exhibited more plastic leaf and crown responsesto shadethan the oaks indicating that they would be strong competitors under the shelterwood.

Acknowledgments This work was conducted while the author was at the NortheasternForest Experiment Station, Forestry Sciences Laboratory in Warren, Pennsylvania.The technical assistanceof Vonley Brown and Stephen Steele in collecting and processing the data is appreciated.Sandra Fosbroke,David Feicht and Regis Young assistedwith preparation of the graphics.

References Atkinson, C.J. 1984. Quantum flux density as a factor controlling the rate of growth, carbohydrate partitioning and wood structure of Betula pubescensseedlings. Ann. Bot. 54:397-411. Blair, R.M. 1982. Growth and nonstructural carbohydrate content of southern browse species as influenced by light intensity. J. RangeManage. 35:756-760. Callaway, R.M. 1992. Morphological and physiological responsesof three California oak speciesto shade.Int. J. Plant Sci. 153:434-441. Carvell, K.L. and E.H. Tryon. 1961. The effect of environmental factors on the abundance of oak regenerationbeneath mature oak stands.For. Sci. 7:98-105. Chambers,J .L. and M. W. Jenkins. 1983. Understory light intensity in bottomland hardwood stands.In Proc. Second Biennial SouthernSilvicultural Conf. Ed. E.P.Jones. USDA For. ServoGen. Tech. Rep. SE-24, Asheville, NC, pp 161-165. 748 GOTTSCHALK

Fails, B.S., A.J. Lewis and J.A. Barden. 1982. Anatomy and morphology of sun- and shade-grownFicus benjamina.J. Am. Soc. Hortic. Sci. 107:754-757. Farmer, R.E., Jr. 1975. Growth and assimilation rate of juvenile northern red oak: effects of light and temperature.For. Sci. 21:373-381. Gottschalk, K. W. 1984. Growth, biomass yield, crown development,and gas exchange of four inten- sively-cultured Populus clones in southernMichigan. Ph.D. Dissertation,Michigan State Univ., East Lansing, 156p. Gottschalk, K. W. 1985. Effects of shading on growth and developmentof northern red oak, black oak, black cherry,and red maple seedlings.I. Height, diameter,and root/shootratio. In Proc. Fifth Central Hardwood Forest Corn. Eds. J.O. Dawson and K.A. Majerus. SAF Publ. 85-05, Urbana, IL, pp 189-195. Gottschalk,K. W. 1987. Effects of shading on growth and developmentof northern red oak, black oak, black cherry, and red maple seedlings.II. Biomass partitioning and prediction. In Proc. Sixth Central Hardwood ForestCorn. Eds. R. Hay, F.W. Woods and H. DeSelm. Univ. Tennessee,Knoxville, TN, pp 99-110. Igboanugo, A.B.I. 1990. Effects of shading on shoot morphology, wood production and structure of Quercuspetraea seedlings.For. Ecol. Manage. 38:27-36. Jarvis, P.G. 1964. The adaptability to light intensity of seedlings of Quercus petraea (Matt.) Liebl. J. Ecol.52:545-571. Kappel, F. and J.A. Flore. 1983. Effect of shadeon photosynthesis,specific leaf weight, leaf chlorophyll content,and morphology of young peachtrees. J. Am. Soc. Hortic. Sci. 108:541-544. Langenheim, J.H., C.B. Osmond,A. Brooks and P.J. Ferrar. 1984. Photosyntheticresponses to light in seedlings of selectedAmazonian and Australian rain forest tree species.Oecologia 63:215-224. Loach, K. 1967.Shade tolerance in tree seedlings.I. Leaf photosynthesisand respiration in plants raised underartificial shade.New Phytol. 66:607-621. Loach, K. 1970. Shadetolerance in tree seedlings.ll. Growth analysis of plants raised under artificial shade. New Phytol. 69:273-286. Loftis, D.L. 1990. A shelterwoodmethod for regeneratingred oak in the southernAppalachians. For. Sci. 36:917-929. McGee, C.E. 1968. Northern red oak seedling growth varies by light intensity and seedsource. USDA For. ServoRes. Note SE-90, Asheville, NC, 4 p. Marini, R.P. and M.C. Marini. 1983. Seasonalchanges in specific leaf weight, net photosynthesis,and chlorophyll content of peachleaves as affected by light penetrationand canopyposition. J, Am. Soc. Hortic. Sci. 108:600-605. Masarovicova,E. and P. Minarcic. 1984. Photosyntheticresponse and adaptationof Fagus sylvatica L. trees to light conditions. 1. Growth of leaves, shoots,and trees.Biologia 39:867-876. Musselman,R.C. and G.E. Gatherum. 1969. Effects of light and moisture on red oak seedlings.Iowa StateJ. Sci. 43:273-284. Nygren, M. and S. Kellomaki. 1983. Effect of shading on leaf structure and photosynthesis in young birches, Betula pendula Roth. and Betula pubescensEhrh. For. Ecol. Manage. 7: 119-132. Phares,R.E. 1971. Growth of red oak (Quercus rubra L.) seedlings in relation to light and nutrients. Ecology 52:669-672. Sander,I.L. 1972. Size of oak advancereproduction: key to growthfollowing harvestcutting. USDA For. ServoRes. Pap. NC-79, St. Paul, MN, 6 p. Sander,I.L. 1977.Manager's handbookfor oaks in the North Central States.USDA For. ServoGen. Tech. Rep. NC-37, St. Paul, MN, 35 p. Sander,I.L., P.S. Johnsonand R.F. Watt. 1976. A guide for evaluating the adequacyof oak advance reproduction. USDA For. ServoGen. Tech. Rep. NC-23, St. Paul, MN, 7 p. Sander,I.L., P.S. Johnsonand R. Rogers. 1984. Evaluating oak advance reproduction in the Missouri Ozarks. USDA For. ServoRes. Pap. NC-251, St. Paul, MN, 16p. Shirley, H.L. 1929. The influence of light intensity and light quality upon the growth of plants. Am. J. Bot. 16:354-390. Tooming, K.G. and T.K. Tammets. 1984. Relationship of specific leaf weight to the irradiation density of adaptationand PAR regime in some plant species.Sov. Plant Physiol. 31:201-207. SHADE, LEAF GROWTH AND CROWNDEVELOPMENT 749

Tryon, E.H. and K.L. Carvell. 1958. Regenerationunder oak stands.West Virginia Agric. Exp. Sm. Bull. 424T, Morgantown, WV, 22 p. Withers, J.R. 1979. Studies on the status of unburnt Eucalyptuswoodland at OceanGrove, Victoria. N. The effect of shading on seedling establishment.Aust. J. Bot. 27:47-66. . Yakshina,A.M. 1978.Rate of respiration of the root systemin maple under different shadingconditions. Sov. Plant Physiol. 25:46-50.

.