163
SURVIVAL, MORTALITY AND COPPICE SHOOT PRODUCTION AND GROWTH PATTERNS OF COLOPHOSPERMUM MOPANE KIRK EX J.
LÉONARD (KIRK EX BENTH) AND ANDROSTACHYS JOHNSONII PRAIN AFTER HARVESTING
N.R. Rathogwa1,2, J.J. Midgley1 and W.J. Bond1
1. Botany Department, University of Cape Town, Private Bag, Rondebosch, 7700 2. Division of Water, Environment and Forestry Technology, CSIR,
P.O. Box 395, Pretoria, 0001
Abstract
Survival, shoot production and growth patterns after harvesting Colophospermum mopane and Androstachys johnsonii were investigated. Trees were divided into two size classes: trees <15 cm and >15 cm in basal diameter. Ten trees of each species in each size class were harvested ≤30 cm above ground to the approximate equivalents of 100% and 25-75% harvesting intensity.
A 100% and 25-75% harvesting intensity resulted in 100% survival rate of the cut stumps in the two size classes of C. mopane trees. Full harvesting of A. johnsonii resulted in 100% and 70% mortality for the
>15 cm and <15 cm size classes respectively. Partial harvesting of A. johnsonii resulted in 60% and 40%
mortality for the >15 cm and <15 cm size classes respectively. For C. mopane, there was no significant relationship between the initial tree size and number of shoots produced as well as total length of shoots. A significant positive relationship was observed between initial tree size and equivalent diameter of shoots.
Similarly, a significant positive relationship was observed between light regime available to cut stumps and equivalent diameter of shoots. However, a significant negative relationship was observed between the sum of ratios of nearest neighbour diameters to distances and equivalent diameters of shoots.
Thus, the response of trees to similar harvesting regimes is species-specific, depending among other things on the population biology of the target species.
1. Introduction
Access to fuelwood and construction timber in indigenous woodlands is one of the primary basic needs of rural communities throughout the developing world, southern Africa included. Most rural populations are still reliant upon natural fuelwood and indigenous timber for most of their energy needs, building houses, roofing, making kraals and fencing (Conroy, 1996; Shackleton, 1997) as a result of low educational levels, low family income and high unemployment levels (Rathogwa, 1997 - unpublished data). The negative impacts of wood collection from woodlands include a decline in the availability of plant resources and hence deforestation (Chidumayo, 1987; Goodman, 1987; Griffin et al., 1993; Banks et al., 1996).
Deforestation in developing countries is a problem that increasingly evokes worldwide concern due to the need to balance conservation development and socioeconomic aspects of the rural communities. The demand for fuelwood and construction timber will obviously increase with an increase in human population size in rural areas. The collection of plant resources in rural areas is already outstripping the vegetation around them (Conroy, 1996). Consequently, there is a resource crisis because available supplies are insufficient to meet demand, particularly as the wood-using populations are increasing (Banks et al., 1996).
In arid savannas, the proposed solutions to deforestation, such as the establishment of woodlots and social forestry programmes (Goodman, 1987; Arnold, 1987; Shackleton, 1993, 1994) are more likely to fail than succeed, due to slow growth rates of trees (Helliwell, 1987) and unpredictable weather conditions such as drought (Goodman, 1987). A complementary approach is to provide access to sustainable harvesting in neighbouring areas presently unavailable to rural communities, including conservation areas and private land (Shackleton, 1993; Cunningham, 1996; Wild & Mutebi, 1996). The success and sustainability of this new approach requires information on the effects of tree damage, such as felling and lopping, on stump survival, shoot production and growth, and the effects of coppice browsing. All these are a basic requirement for assessment and management of trees for wood production (Milton, 1988).
While measures such as the provision of access to sustainable harvesting in neighbouring areas presently unavailable to rural communities (i.e. protected or conservation areas) are advocated (Cunningham, 1996; Wild & Mutebi, 1996), very little is known about the population biology of the target species, including their responses to harvesting and subsequent coppice shoot growth patterns. A preliminary survey on resource use patterns by the rural community (Rathogwa, 1997 - unpublished data) identified C. mopane and A. johnsonii as the most sought-after species for energy and construction respectively. It is against this background that the two species were selected for this study.
This study was subsequently undertaken as a preliminary investigation into the effect of tree size, intensity of stem harvesting, height of cutting and season of harvesting on stump survival, coppice production, and coppice shoots growth patterns. The main aim of the study is to investigate methods of plant resource
164
harvesting and management interventions that are likely to be sustainable in the long term. The objectives of this study were:
! To determine the influence of tree size and intensity of stem harvesting on survival and mortality patterns of cut trees.
! To determine the effects of stump size on coppice shoot production and growth.
! To determine the effects of nearest neighbour plants and light availability on coppice shoot production and growth.
2. Study area
The study was conducted at Makuya Nature Reserve in the north-eastern part of the Northern Province, about 60 kilometres from Thohoyandou, Venda. The study area lies between 300 50'E and 310 05'E, and 220 25'S and 220 35'S. The area occurs on loamy sand and clayey soils in the undulating granitic landscape of the northern Kruger National Park. Annual summer rainfall is from 250 to over 500 mm. Temperatures vary between extremes of 1.5 and 42.5 0C, with an average of 22 0C (Low and Rebelo, 1996). The vegetation is classified as Mopane Bushveld (Low and Rebelo, 1996) and is characterised by a fairly dense growth of Mopane (Colophospermum mopane) and mixtures of Combretum apiculatum, associated with Acacia nigrescens, Adansonia digitata, Commiphora spp, Boscia albitrunca, Terminalia prunioides, Kirkia acuminata and, on rocky outcrops, Androstachys johnsonii. The sandy-loam soils, low rainfall, high temperatures and lack of frost influence the distribution of this vegetation type. Cattle, goat and game farming, ecotourism and mining are the most important economic land uses within the study area.
3. Methods
3.1
Stump survival and mortality patterns after harvestingDuring December 1997, two size classes (i.e. trees <15 cm and trees >15 cm in basal diameter) were identified within homogeneous pure stands of Colophospermum mopane and Androstachys johnsonii species.
Within each size class, 10 trees of variable basal diameter sizes were selected for total stem harvesting (i.e.
100% of stems of an individual tree were harvested) and another 10 trees selected for partial harvesting (i.e.
25 -75% of stems harvested) for both species. Thus, a total of 20 trees was selected for both full and partial- stem harvesting for each species.
Before harvesting, the equivalent stem basal diameters (i.e. basal diameter of a multi-stemmed tree) were measured ≤30 cm above ground level, or above the basal swelling, using a diameter tape and recorded.
Each tree was then fully or partially harvested using a chainsaw ≤30 cm above ground. The number of stems removed from each tree were noted and recorded. The trees were left unattended for a period of one year.
One year later, during December 1998, the sites were revisited. Within each site all harvested trees were assessed to see if they had died or not, and to see if they had produced new shoots (coppice).
3.2 Effect of stump size on Colophospermum mopane coppice shoot production and growth
One year after harvesting, the number of shoots produced, as well the diameter and height of each shoot, were measured and recorded from stumps that were still alive. The relationships between the initial stem size and i) number of sprouts produced, ii) total length of a sprouts and iii) equivalent diameter of sprouts were investigated.
3.3 Effect of nearest neighbours and light availability on Colophospermum mopane coppice shoot production and growth
One year after harvesting, the basal diameters of the three nearest neighbours in the first three 900 quadrats and their distances to the cut stems were also measured and recorded. The percentage light available during the day to the cut stems was estimated visually and recorded. The relationships between the sum of nearest neighbour diameter to distance ratios and the percentage light available to stumps and i) number of sprouts produced, ii) total length of sprouts and iii) equivalent diameters of sprouts were investigated.
4. Results
4.1
Stump survival and mortality patterns after harvestingWhen C. mopane was cut at ≤30 cm above ground level during summer, a 100% survival rate was observed for all the cut stumps in the two size classes one year after harvesting, with the exception of three trees that were removed during road construction (Table 1).
Table 1. Responses of C. mopane and A. johnsonii to varying harvesting intensities
Species Size class 100% harvest level 25 - 75% harvest level
Dead Live n Dead Live n
C. mopane <15 cm class 0 (0%) 10 (100%) 10 0 (0%) 10 (100%) 10
≥15 cm class 0 (0%) 10 (100%) 10 0 (0%) 7 (100%) 7
A. johnsonii <15 cm class 7 (70%) 3 (30%) 10 4 (40%) 6 (60%) 10
≥15 cm class 10 (100%) 0 (0%) 10 6 (60%) 4 (40%) 10 Many stumps of A. johnsonii died when the stems were cut at ≤30 cm above ground level in summer. This pattern was common when all stems of an individual tree are harvested in the two size classes. Seventy percent of A. johnsonii stumps from the <15 cm size class died when all the stems of an individual were removed, while 100% mortality was also observed for the ≥15 cm size class (Table 1). Forty and 56 percent respectively of A. johnsonii died in the <15 cm and ≥15 cm size classes when stems of an individual tree were partially harvested (25-75% harvesting regime).
4.2 Effect of stump size on Colophospermum mopane coppice shoot production and growth
The initial size of the cut tree did not significantly influence the number of coppice shoots produced and total length of coppice shoots. However, there was a significant positive relationship between the initial size of the cut tree and equivalent basal diameter of coppice C. mopane when all the stems of an individual tree were harvested (Figure 1). The mean number of shoots produced per year, the mean total length of shoots and mean equivalent basal diameters of coppices as well as the correlation results between stump size and number of shoots produced, total shoot length and equivalent diameter of shoots are given in Table 2.
Initial tree diameter (cm)
Equivalent diameter of shoots (cm)
0.6 1.0 1.4 1.8 2.2 2.6 3.0 3.4 3.8
2 6 10 14 18 22 26 30 34
Figure 1. Relationship between the initial tree (stump) size and equivalent diameter of shoots.
Sample size, r2 and p values are given in Table 2. Shoot diameter = 1.38 + 0.06 8 x initial tree size.
165
Table 2. Relationship between initial tree basal diameter and number of shoots produced, total length of shoots produced and diameter of shoots produced. N/A means that there are results.
Means and standard deviations are given in brackets.
Species 100% har vest level 25-75% harvest level
C. mopane Basal diameter vs number of shoots Basal diameter vs number of shoots
r2 p-value n r2 p- value n
0.00 0.935 20 (12.3 ± 10.3) 0.00 0.97 16 (5.8 ± 5.4) Basal diameter vs total length of shoots Basal diameter vs total length of shoots
r2 p-value n r2 p- value n
0.042 0.397 20 (665.6 ± 367.8) 0.025 0.56 16 (285.2 ± 276.9) Basal diameter vs diameter of shoots Basal diameter vs diameter of shoots
r2 p-value n r2 p- value n
0.36 0.005* 20 (2.3 ± 0.67) 0.00 0.95 16 (1.1 ± 0.6) A. johnsonii Basal diameter vs number of shoots Basal diameter vs number of shoots
r2 p-value n r2 p- value n
N/A N/A N/A 0.22 0.147 12 (7.8 ± 8.1)
Basal diameter vs total length of shoots Basal diameter vs total length of shoots
r2 p-value n r2 p- value n
N/A N/A N/A 0.103 0.336 12 (82.0 ± 75.7)
Basal diameter vs diameter of shoots Basal diameter vs diameter of shoots
r2 p-value n r2 p- value n
N/A N/A N/A 0.007 0.803 12 (0.5 ± 0.3)
4.3 Effect of nearest neighbours and light availability on Colophospermum mopane coppice shoot production and growth
The size and proximity of nearest neighbour C. mopane trees did not influence significantly the number of coppice shoot produced and the total length of coppice shoots (p>0.05). Again, there was a significant negative relationship between the sum of nearest neighbour size to distance ratio (n = 20, r2 = 0.20; p = 0.05;
Figure 2). The amount of light available to the cut stumps did not significantly influence shoot production and shoot growth (p > 0.05) of C. mopane sprouts. However, there is a significant positive linear relationship between the amount of light available to cut stumps and the equivalent diameter of sprouts for C. mopane when all stems of an individual tree are harvested (n = 20, r2 = 0.27, p < 0.05).
5. Discussion
5.1
Stump survival and mortality patterns after harvestingResults indicate that C. mopane and A. johnsonii species respond differently to harvesting in summer and to the same height of cutting as well as the intensity of stem harvesting. The survival rate of both the big and small size class C. mopane trees is very high, with no mortality observed over the one-year period irrespective of the harvesting regime employed. On the other hand, mortality rate of A. johnsonii is generally higher than the survival rate irrespective of the harvesting intensity, except for the small trees under partial harvesting (Table 1).
166
Sum of neighbours' diameter divided by distance (cm m Equivalent diameter of shoots (cm) 0.6
1.0 1.4 1.8 2.2 2.6 3.0 3.4 3.8
6 12 18 24 30 36
-2)
Figure 2. The relationship between equivalent basal diameter of shoots and sum of nearest neighbour diameter divided by distance ration for C. mopane after 100% harvest regime.
Shoot basal diameter = 2.952 - 0.039 x sum of nearest neighbour diameter divided by distance.
5.1.1 Colophospermum mopane
Stump size does not influence survival pattern in C. mopane after harvesting. Also, harvesting stems ≤30 cm above ground level during the summer does not cause mortality to the stumps. Johansson (1992b), while studying the resprouting of 10 to 50-year-old Betula pubescens, found a mean stump survival rate of 84% one year after felling. In this study, almost 100% of cut stumps are still living one year after cutting. Johansson also noted that large stumps of B. pubescens do not survive as well as smaller stumps but this study has shown that survival patterns of both big and small trees after harvesting are similar. Thus, the season of harvesting and the height of cutting employed in this study favour the survival of the cut stumps of C. mopane.
The high survival rate of C. mopane may be due to the fact that the root system is decisively shallow rooted (Smit, 1994; Smit & Rethman, 1998) and this possibly allows the species to exploit soil moisture efficiently throughout the year. It is generally accepted that the pre-existing or pre-disturbance root system plays a very important role in recovery (Castell et al., 1994) and C. mopane could have survived so well because of its root system. Coppice production seems to have been successful due to the buds, which are located over the whole tree (Potgieter et al., 1997), in which case stumps seem to have overcome the difficulties with which any individual bud will have in developing independently of the cluster.
Physiologically, the explanation for this high survival rate of C. mopane trees after harvesting may be that trees were felled during the summer season but before leaf flush, hence during the inactive growth period, when the carbohydrate reserves are high (Johansson, 1992b; Vila & Lloret, 1996). Observations from stumps that were either burnt or cut a long time before indicated that many stumps had coppice shoots growing from the ground around the cut or burnt stumps, suggesting that lignotubers (Zammit, 1988; Canadell & Lopez- Soria, 1998), i.e. woody underground stems that support dormant buds and are thought to store carbohydrates and nutrients, are likely to be some characteristic features of C. mopane. It is known that the prime function of the lignotubers is to assure survival of the tree after disturbances rather than to provide storage tissue (Zohar et al., 1978). The distribution of buds along the whole tree argument is also supported by results of a recent study in Mopani veld (Potgieter et al., 1997).
Therefore, it can be concluded that the size of a tree does not influence the survival and shoot production respectively of this species after cutting. Also, harvesting stems partially or fully during the summer will not induce mortality of the cut stumps.
The period that can be used for cutting C. mopane trees with the aim of yielding many living stumps is not known, but this study has demonstrated that harvesting in summer will result in 100% survival and the production of many coppice shoots per stumps (12±10.3 shoots per stumps). The emerging sprouts usually form the bulk of the future stock and, due to their fast growth rates, they can be managed with the aim of biomass production for local people. In communal land, harvesting trees close to the ground, as was done in this study, is undesirable due to the presence of browsers that feed on coppice shoots during the dry months.
A possible solution will be to cut trees higher up the stem. This seems to be the best possible solution because the growth rates of resprouts from trees cut higher up the stems is reportedly higher than when cut close to the ground (Potgieter et al., 1997). However, it must be noted that much of the biomass will be left standing and this may lead to uncontrolled harvesting of these stems close to the villages.
5.1.2 A. johnsonii
Small and big trees are more sensitive to both the height of cutting as well as the harvesting intensity and season employed in this study (Table 1). Initial tree (or stump) size influences the survival of cut A. johnsonii
167
stumps. Thus, a range of size classes of this species cannot be harvested, since some will not persist under such harvesting pressures. In this study almost 100% of the big 100%-harvest-intensity-level stumps died one year after harvesting, suggesting that it is the bigger trees that will suffer the highest mortality after harvesting.
Johansson (1992b) also noted that large stumps do not survive as well as smaller stumps and this study proved that the two size classes indeed do not survive in a similar fashion after they are cut. The findings suggest that the most important causes of mortality in this species after harvesting are the size of the tree and the intensity of stem harvesting, but either the height or season of cutting or a combination of these factors might also be responsible. As a result, we cannot conclusively rule out the importance of the height and season of cutting as important determinants of survival patterns of the species after harvesting.
The low number of coppice shoots produced by A. johnsonii may be due to the old age of buds on the stems. It is known that buds on old and large trees lose viability and die (Chidumayo, 1993). Stem swellings on which buds are concentrated were observed on this species, located close to the ground on young trees and up along the stems on older trees (Figure 3).
Figure 3: The swellings of stems in A. johnsonii due to the concentration of buds along the stem.
Black areas indicate swellings (=bud clusters). A to C are different sizes of a tree from the small to large individuals. The arrow indicates the movement of swellings as the tree grows.
For old trees that were cut close to the ground, it is possible that no bud clusters were left and hence there were no more sources of buds. It is possible that these clusters are not only sources of buds but also storage structures. The fact that partial harvesting resulted in a 60% survival rate may indicate that the contribution of current photosynthesis to survival of A. johnsonii is large (Richards & Caldwell, 1983). The location of buds in A. johnsonii as clusters above the stem base seems to be causing difficulties as to which individual bud must develop independently of the cluster.
The period as well as the height of cutting with the aim of producing many live stumps for the production of many coppice shoots has not been investigated. If the emerging sprouts are to be used for biomass production, then more focused research into the response of this species to height as well as season of harvesting is warranted.
5.2 Effect of stump size on Colophospermum mopane coppice shoot production and growth
The fact that the initial size of a tree did not influence resprouting and height growth of sprouts in C. mopane (Table 2) suggests that size of a tree is not an important factor in the production of coppice shoots and coppice shoot growth patterns. This also suggests that carbohydrate reserves are likely to be less important relative to concurrent photosynthesis to regrowth (Richards & Caldwell, 1983). Assuming that big trees have larger food reserves, we expected large trees to produce many shoots and to have shoots of large height sizes. A positive linear relationship between the initial size of a tree and the equivalent diameter of sprouts for C. mopane (Figure 1; Table 2) when all stems of an individual tree are harvested, suggests that the size of the cut tree is an important factor with regard to diameter growth of coppice shoots, as well as the importance of inter-shoot competition. Thus, the production of shoots that can grow quickly in diameter can be achieved by cutting bigger trees. This finding is of significant importance since this species is used mainly for fire and roofing and fencing by local communities around the study area. Rathogwa’s (1997) unpublished data indicate that people in the communal area prefer wood of small dimensions over big ones for firewood, roofing
168
169
and fencing. Preference for young stems over large unsuitable ones is a well-known phenomenon in rural communities (McGregor, 1994). The production of these desirable materials can therefore be achieved within a very short time after harvesting from coppice regrowth. If the areas currently not used by the community are envisaged for the production of wood of smaller dimensions, then a coppice system should be adopted for the management of the system.
5.3 Effect of nearest neighbours and light availability on Colophospermum mopane coppice shoot production and growth
The significant negative relationship between equivalent basal diameters of sprouts and sum of nearest neighbour diameters to distance ratio suggests that the farther apart the cut stumps are to their neighbours, the more quickly the sprouts can grow in basal diameter (Figure 2). This is likely so if the number of shoots produced is fewer and, as a result, the resources are then located to stem circumference increments rather than height growth. Also, this suggests that competition at the early stages of sprout growth is intense, manifesting itself through the increment in basal diameter and height growth. Another way to test this assumption would be to prune coppice shoots or to monitor coppice shoots growth patterns over a number of time periods.
Similarly, the observed significant relationship between the percentage light available to sprouts and equivalent diameter of sprouts is an indication that production is limited by competition for this resource.
Harvesting of large trees will create this light environment and ultimately, the combined effects of initial tree size, available light regime and reduced competition from harvesting will lead to increased production of the desired woody material within a very short period.
However, the amount of light available to a stump during the day did not influence shoot production and shoot height growth of C. mopane sprouts, suggesting that competition is not limiting shoot production and height growth but only diameter growth. Thus, competition may be more intense below ground than above ground. The absence of detectable influence of competition for C. mopane shoot production and shoot height growth is contrary to the findings of most previous competition studies in savannas (Yeaton et al., 1977;
Smith & Walker, 1983; Smith & Grant, 1986; Smith & Goodman, 1986; Grundy et al., 1994). The results, however, support the findings of a recent study in savannas (Shackleton, 1997) that competition is not limiting woody productivity.
Kauppi et al. (1988) and Johansson (1992a,b) noted poor sprouting results from old stumps. Kauppi et al. (1988) concluded that the locations of buds in a stem as cluster are a reason for the poor sprouting capacity due to the difficulties that any individual bud had in developing independently of the clusters. C.
mopane buds are located both in large clusters as well as single buds, but this does not seem to cause bud development problems, most probably because buds develop during the growing period when there is enough moisture on the stems and stem bases. The number of sprouts, equivalent diameter and total length reported in this study are also very high (Table 2) but it is still too early to predict how long it will take the sprouts to reach the preferred size classes, since this depends also on the environmental conditions.
6. Conclusions and recommendations
A complementary approach to woodlots and social forestry entails the provision of access to sustainable harvesting in neighbouring areas presently unavailable to rural communities, including conservation and private land (Shackleton, 1993; Cunningham, 1996; Wild & Mutebi, 1996). Information on the effects of harvesting disturbance on yield components of trees indicates that C. mopane and A. johnsonii respond differently to harvesting and browsing regimes. These findings can be used as a basis towards sustainable resource utilisation in protected areas. In such cases, the responses of these two different species to harvesting regimes should be used as guidelines towards harvesting regimes that will promote natural regeneration of the desired species.
It is recommended that detailed research on the responses of these two species to season and height of cutting as well as the effects of coppice shoot pruning be conducted. The impacts of browsers on coppice shoot should also be investigated.
Acknowledgments
Financial assistance from the Division of Water, Environment and Forestry, CSIR, is gratefully acknowledged.
We would like to thank the Department of Agriculture, Tourism and Environmental Affairs, Northern Province for permission to conduct research at Makuya Nature Reserve. Mr C. Tshisikule and Mr E. Mainganye of the Makuya Nature Reserve deserve special thanks for providing the research team with accommodation, as does Mr Sivhugwana and Mr Phungo Rathogwa for their assistance while cutting trees and collecting coppice growth data respectively.
170
References
ARNOLD, J.E.M. (1987). Community forestry. Ambio 16(2-3): 122-128.
BANKS, D.I., GRIFFIN, N.J., SHACKLETON, C.M., SHACKLETON, S.E. & MAVRANDONIS,J.M. (1996).
Wood supply and demand around two rural settlements in semi-arid savanna, South Africa. Biomass and Bioenergy 11(4): 319-331.
CANADELL, J. & LOPEZ-SORIA, L. (1998). Lignotuber reserves support regrowth following clipping of two Mediterranean shrubs. Ecology 12: 31-38.
CASTELL, C., TERRADAS,J. & TENHUNEN, J.D. (1994). Water relations, gas exchange, and growth of resprouts and mature plant shoots of Arbutus unedo L. and Quercus ilex L. Oecologia 98: 201-211.
CHIDUMAYO, E.N. (1987). A survey of wood stocks for charcoal production in the miombo woodlands of Zambia. For. Ecol. Manage. 20: 105-115.
CHIDUMAYO, E.N. (1993). Responses of Miombo to Harvesting: Ecology and Management. Energy, Environment and Development Series No. 24, SEI/SIDA.
CONROY, C. (1996). Socio-economic aspects of Mopane use in Omusati region, Namibia. Forestry Publication No. 6. ODA, UK.
CUNNINGHAM, A.B. (1996). People, park and plant uses: recommendations for multiple-use zones and development alternatives around Bwindi Impenetrable National Park, Uganda. People and Plants Working Paper 4, UNESCO, France.
GOODMAN, G.T. (1987). Biomass energy in developing countries: problems and challenges. Ambio 16: 111- 140.
GRIFFIN, N.J., BANKS, D.I., MAVRANDONIS, J.M., SHACKLETON, C.M. & SHACKLETON, S.E. (1993).
Fuel use in six rural settlements in Gazankulu. Journal of Energy in Southern Africa 4: 68-73.
GRUNDY, I.M., CAMPBELL, B.M. & FROST, P.G. (1994). Spatial pattern, regeneration and growth rates of Brachystegia spiciformis and Julbernardia globifera. Vegetatio 115: 101-107.
HELLIWELL, D.R. (1987). Multiple-use forestry in the United Kingdom. Ambio 16(2-3): 129-133.
JOHANSSON, T. (1992a). Sprouting of 2- to 5-year-old birches in (Betula pubescens Ehrh. and Betula pendula Roth) in relation to stump height and felling time. For. Ecol. Manage. 53: 263-281.
JOHANSSON. T. (1992b). Sprouting of 5- to 10-year-old Betula pubescens in relation to felling time. For.
Ecol. Manage. 53: 283-296.
KAUPPI, A., RINNE, P. & FERM, A. (1988). Sprouting ability and significance for coppicing of dormant buds on Betula pubescens Ehrh. stump. Scand. J. For. Res. 3: 343-354.
LOW, A.B. & REBELO, A.G. (1996). Vegetation of South Africa, Lesotho and Swaziland. Department of Environmental Affairs and Tourism, Pretoria.
MCGREGOR, J. (1994). Woodland pattern and structure in a peasant farming area of Zimbabwe: ecological determinants and present and past use. For. Ecol. Manage. 63: 97-133.
MILTON, S.J. (1988). The effects of pruning on shoot production and basal increment of Acacia tortilis. S.
Afr. J. Bot. 54(2): 109-117.
POTGIETER, M., WESSELS, D. & NEL, C. (1997). Optimising mopane production. Farmers Weekly, January 3. pp. 27- 29.
RATHOGWA, N.R. (1997). Unpublished household survey data for Mukomawabani rural village in Venda, Northern Province, South Africa.
RICHARDS, J.H. & CALDWELL, M.M. (1985). Soluble carbohydrates, concurrent photosynthesis and efficiency in regrowth following defoliation: a case study with Agropyron species. Journal of Applied Ecology 22(3): 907-920.
SHACKLETON, C.M. (1993). Demography and dynamics of the dominant woody species in a communal and protected area of the eastern Transvaal Lowveld. S.Afr. J. Bot. 59(6): 569-574.
SHACKLETON, C.M. (1994). Growing more trees for fuelwood in the Northern Transvaal or redistribution after sustainable harvesting? Development Southern Africa 11(4): 587-598.
SHACKLETON, C.M. (1997). The Prediction of Woody Productivity in the Savanna Biome, South Africa.
Unpublished PhD Thesis, Science Faculty, University of the Witwatersrand, Johannesburg, South Africa.
SMIT, G.N. (1994). The Influence of Intensity of Tree Thinning on Mopani Veld. PhD-thesis, University of Pretoria, Pretoria.
SMIT, G.N. & RETHMAN, N.F.G. (1998). Root biomass, depth distribution and relations with leaf biomass of Colophospermum mopane. S. Afr. J. Bot. 64(1): 38-43.
SMITH, T.M. & GRANT, K. (1986). The role of competition in the spacing of trees in a Burkea africana- Terminalia sericea savanna. Biotropica 18(3): 219-223.
SMITH, T.M. & GOODMAN, P.S. (1986). The effect of competition on the structure and dynamics of Acacia savannas in southern Africa. J. Ecol. 74: 1031-1044.
SMITH, T.M. & WALKER, B.H. (1983). The role of competition in the spacing of savanna trees. Proc. Grassl.
Soc. Sth Afr. 18: 159-164.
VILA, M. & LLORET, F. (1996). Herbivory and neighbour effects on the sprout demography of the Mediterranean shrub Erica multiflora (Ericaceae). Acta Oecologica 17: 127-138.
WILD, R.G. & MUTEBI, J. (1996). Conservation through community use of plant resources: establishing
171
collaborative management at Bwindi Impenetrable and Mgahinga Gorilla National Parks, Uganda. People and Plants Working Paper 5. UNESCO, France.
YEATON, R.I., TRAVIS, J. & GILINSKY, E. (1977). Competition and spacing in plant communities: the Arizona upland association. J. Ecol. 65: 587-595.
ZAMMIT, C. (1988). Dynamics of resprouting in the lignotuberous shrub Banksia oblongifolia. Australian Journal of Ecology 13: 311-320.
ZOHAR, Y., WAISEL, Y. & KASCHON, R. (1978). Regeneration after cutting and the role of lignotubers in the regeneration of Eucalyptus occidentalis Endl. Aust. For. Res. 8: 115-123.
