Effects of crown-pruning on seed and pollen cone production in two lodgepole pine seed orchards in British Columbia

The effects of top pruning on subsequent seed and pollen cone production in lodgepole pine (Pinus contorta Dougl.) grafted clones in two seed orchards were evaluated. In the older orchard (14 years old), three pruning regimes were applied: an untreated control, a moderate pruning where 25% of the crown was removed and a severe pruning where 40% of the crown was removed. In the younger orchard (ten years old), only a control pruning and a severe pruning (50% of crown removed) were applied. Five-year seed cone production was lower in treated ramets compared to controls in both orchards. However, when cumulative seed cone yields were adjusted for reduced tree heights after pruning, differences were no longer significant in the older orchard. Strong clonal variation was observed in seed cone yields in both orchards. Pollen cone development was less variable among treatments in the older orchard. In the younger orchard, controls consistently produced heavier pollen crops. Pruning recommendations for similar orchards are made and results discussed for reducing the impacts of losses in seed cone yields.Deceased     

Effect of local stand density on reproductive processes of the sub-boreal conifer Picea jezoensis Carr. (Pinaceae)

The effect of local pollen donor density (conspecific) and local total tree density (all species) on pollen pool diversity, outcrossing rate, and filled seed ratio were examined in a natural stand of the sub-boreal conifer Picea jezoensis Carr. Seed cones were collected from 17 mother trees, and the percentage of filled seed were calculated for five cones per tree. In addition, local stand density of pollen donors and total trees were investigated within r<30 m from the focal mother trees. Using four polymorphic microsatellite markers, the average pairwise kinship coefficient F_S and multilocus outcrossing rate t_m were estimated on germinated seedlings sampled from 14 mother trees. F_S showed significant positive correlation to local total tree density (r=0.541, p=0.046). Since F_S is inversely proportional to the effective number of pollen donors N_ep, genetic diversity of pollen pool was decreased in mother trees with higher local total tree density. Applying a generalized linear mixed model to filled seed ratio and local stand density of pollen donors and total trees, a negative impact of local pollen donors on filled seed ratio was found. We conclude that both local stand density as well as global stand density are important in the management of a natural stand of P. jezoensis.     

Rejuvenating indigenous trees in agroforestry parkland systems for better fruit production using crown pruning

The effects of crown pruning of mature indigenous fruit trees of Vitellaria paradoxa C. F Gaertn (commonly known as karité) and Parkia biglobosa (Jacq.) R. Br. ex G. Don (commonly known as néré) on recovery of crown size and fruit yield were assessed during 6 years in an agroforestry parkland system in Burkina Faso. Three treatments of crown pruning (total-pruning, half-pruning and a control of no-pruning) were applied to karité and néré. Each treatment comprised ten individuals of each species or a total of 60 trees of both species. Six years after pruning, higher recovery (81%) of crown diameter was achieved in total-pruned trees of néré as opposed to karité which recovered by only 73%. On the contrary, fruit production in total-pruned trees of karité recovered by 83% 5 years after pruning and fully (100%) 6 years after pruning as opposed to néré which recovered by only 57% 5 years after pruning but declined to 16% on the sixth year probably due to interannual variability. Fruit yields did not differ significantly between unpruned and half-pruned trees of both species throughout the experiment period. Total pruning may, therefore, be recommended to farmers to rejuvenate old trees of karité in parklands on the basis of fast recovery of fruit and slow recovery of crown in the species. Slow recovery of crown in pruned trees is the most desirable characteristic in parklands in order to avoid the negative effect of tree shade on adjacent crop.     

Gene flow pattern and mating system in a small population of Quercus semiserrata Roxb. (Fagaceae)

Pollen flow from external sources is important for the conservation of tree species in fragmented forests or small populations, because it can be sufficient to prevent differentiation among them, and appears to be able to prevent the loss of their genetic diversity through genetic drift. In this study, we examined the genetic heterogeneity of pollen pools accepted by each Quercus semiserrata seed parent at the Khun Wang Royal Agriculture Research Center, Thailand, both within and among two mast fruiting years (2005 and 2007), using paternity analysis and analysis of molecular variance (AMOVA). The mating systems of the trees were also examined using the multilocus mating system model (MLTR), after determining the genotypes at eight microsatellite loci of 26 seed-trees and 435 seeds from 8 seed-trees in the 2 mast fruiting years. The average distance of effective pollen flow within the plot was estimated to be 52.4 m, and 95% of effective pollen was dispersed within 200 m, indicating that effective pollen flow is highly localized and that most effective pollen is contributed by near-neighbor trees. The proportion of effective pollen that immigrated from external sources was estimated to be 26.2%. The AMOVA analysis based on the pollen haplotypes showed that the pollen pools, both total and for each reproductive year, significantly genetically differed among the seed parents. Using a mixed mating model, the estimate of biparental inbreeding for the total population (t_m − t_s) was 0.013, indicating that a low proportion of mating occurred among close relatives. The effective number of pollen donors (N_ep) was estimated to be 9.987 using the TwoGener model, or 10.989 using the mixed mating model. The effective number of pollen donors of seeds was higher in the mast fruiting year 2005 than in the other examined year, 2007. Consequently the allelic richness and genetic diversity of seeds produced in 2005 were higher than those produced in 2007. Overall, the results show that high outcrossing rates, high levels of gene flow from other populations and heterogeneity in the pollen received by an individual may enhance the ability of populations to maintain effective population sizes. Therefore, these processes may be sufficient to prevent loss of genetic diversity through genetic drift of Q. semiserrata at this study site.     

Pollen dispersal in a hinoki (Chamaecyparis obtusa) seed orchard detected using a chloroplast DNA marker

Pollen dispersal was estimated in two test plots in a hinoki (Chamaecyparis obtusa) seed orchard using a chloroplast DNA marker, the spacer region between thetrnD andtrnY genes, and SSCP (single strand conformation polymorphism). In Plot 1, 2,020 seeds from 40 trees within 30 m of the marker tree were analyzed using the PCR-SSCP method. In Plot 2, 1,850 seeds from 37 trees were analyzed in the same manner. The results revealed that the maximum pollen dispersal distance in the two plots exceeded 25 m. Pollen dispersal appeared to be inversely proportional to the distance from the marker tree. The effective pollen dispersal was suggested to be less than about 20 m in a mature hinoki seed orchard. Adjacent trees had an excessive influence when the pollen density was increased by artificial flower stimulation. Therefore, it was suggested that seed production better resembles ideal random mating when carried out as naturally as possible. In conclusion, the SSCP chloroplast DNA marker was a useful tool for amassing basic information on pollen management in seed orchards of coniferous species.     

Productivity of annual cropping and agroforestry systems on a shallow Alfisol in semi-arid India

An experiment was conducted at ICRISAT Center, Patancheru, India from June 1984 to April 1988 on a shallow Alfisol to determine whether the productivity of annual crop systems can be improved by adding perennial species such as Leucaena leucocephala managed as hedgerows. Except in the first year, crop yields were suppressed by Leucaena due to competition for moisture. The severity of competition was high in years of low rainfall and on long-duration crops such as castor and pigeonpea. Based on total biomass, sole Leucaena was most productive; even on the basis of land productivity requiring both Leucaena fodder and annual crops, alley cropping had little or no advantage over block planting of both components. Application of hedge prunings as green manure or mulch on top of 60 kg N and 30 kg P ₂ O ₅ha⁻¹ to annual crops did not show any benefit during the experimental period, characterized by below average rainfall. Indications are that (i) alley cropping was beneficial in terms of soil and water conservation with less runoff and soil loss with 3 m alleys than with 5.4 m alleys, and (ii) root pruning or deep ploughing might be effective in reducing moisture competition.     

Productivity and nutrient cycling in young Acacia senegal farming systems on Vertisol in the Blue Nile region, Sudan

Acacia senegal, the gum arabic-producing tree, is the most important component of traditional dryland agroforestry systems in the␣Sudan. The spatial arrangement of trees and the type of agricultural crop used influence the interaction between trees and crops. Tree and crop growth, gum and crop yields and nutrient cycling were investigated over a period of 4 years. Trees were grown at 5 × 5 m and 10 × 10 m spacing alone or in mixtures with sorghum or sesame. No statistically significant differences in sorghum or sesame yields between the intercropping and control treatments were observed (mean values were 1.54 and 1.54 t ha⁻¹ for sorghum grain and 0.36 and 0.42 t ha⁻¹ for sesame seed in the mixed and mono-crop plots, respectively). At an early stage of agroforestry system management, A. senegal had no detrimental effect on crop yield; however, the pattern of resource capture by trees and crops may change as the system matures. A significant positive relationship existed between the second gum picking and the total gum yield. The second gum picking seems to be a decisive factor in gum production and could be used as an indicator for the prediction of the total gum yield. Soil organic carbon, N, P and K contents were not increased by agroforestry as compared to the initial levels. Soil OC was not increased by agroforestry as compared to sole cropping. There was no evidence that P increased in the topsoil as the agroforestry plantations aged. At a stocking density of 400 trees ha⁻¹ (5 × 5 m spacing), A. senegal accumulated in its biomass a total of 18.0, 1.21, 7.8 and 972 kg ha⁻¹ of N, P, K and OC, respectively. Agroforestry contributed ca. 217 and 1500 kg ha⁻¹ of K and OC, respectively, to the top 25-cm of soil during the first four years of intercropping.     

Inhibition of flower bud formation inBetula platyphylla andBetula pendula by gibberellin (GA3) application

Birch (Betula) trees produce a large amount of pollen, which is a cause of serious pollinosis. To control the pollen scattering, we examined the inhibitory effect of gibberellin (GA₃) on flower bud formation of birch. Field-grown trees ofBetula pendula, nine years of age, were treated with a foliar spray of GA₃ (1, 10, and 100 mg L⁻¹),Betula platyphylla trees, 17–23 years of age, were treated with a stem injection of GA₃ (5, 50, and 500 mg per plant) in June or July, 1998. Male flower buds were not observed at the top of new shoots in June, but were observed in July. The number of inflorescence that emerged from randomly sampled branches was counted the next spring. The foliar spray of GA₃ in June at the dose of 10 or 100 mg L⁻¹ decreased the number of inflorescence. Stem injection in June at a dose higher than 50 mg per plant completely inhibited flower bud formation, although it increased bud mortality in some trees. The most effective dose of GA₃ for stem injection to inhibit flower bud formation was between 0.04 and 0.44 mg per cm² of stem cross-section area at breast height. These results indicate that GA₃ applied during the period of floral differentiation markedly inhibits flower bud formation in birch.     

Integration of pines and pastures for hay and grazing

When combining pine and cattle production on improved pastures, grazing may have to be delayed for several years until trees are large enough to resist injury. During this period forage would be lost unless harvested for hay. This study in south Georgia, USA, examined hay production during the first 3 years, cattle production during the next 3 years, and effects of this management system on survival and growth of slash pine(Pinus elliottii) planted in widely-spaced rows within the pastures. Pines were planted at 225 trees/ha in configurations of 3.0 x 14.6 or 4.9 x 9.1 m in pastures recently sprigged or seeded with Coastal bermudagrass (Cynodon dactylon) or Pensacola bahiagrass(Paspalum notatum). Pastures were fertilized annually and cut for hay four or five times each year. Wide-row spacings permitted haying operations, but tree rows and turning areas removed 5 to 8 percent of the area from production. However, hay yields and liveweight gains by yearling cattle were near normal for this locale. Slash pine planted in these fertilized pastures survived well, grew rapidly, and only a few trees were killed during haying operations. After 6 years, 83 percent of the trees survived and averaged 6.5 m in height and 13.2 cm in diameter. Cattle killed a few trees by rubbing during the 4th year. Wide-row spacing of pines permits harvesting forage for hay while waiting for the trees to become large enough to permit grazing. This approach to agroforestry produces annual returns to the landowner while awaiting maturity of the pines.     

Genetic variation in fruitfulness in a hinoki (Chamaecyparis obtusa Endl.) seed orchard and its impact on the maintenance of genetic diversity in seedlots

Genetic variation in seed/cone production among clones was studied in a hinoki (Chamaecyparis obtusa Endl.) seed orchard containing 25 plus-trees by analyzing the number of cones, the yield of cones and seeds of individual ramets for 5 successive years (1982 to 1986). There was significant variation among clones each year and parental contribution in the seed orchard. Specifically, in the years 1982, 1983, 1984, 1985, and 1986, 20% of the clones produced 37.2, 60.6, 36.0, 44.3, and 44.8% of the total cones, respectively. The size of the crop greatly influenced the parental balance in the resulting seed/cone crops. The product moment correlation coefficients and Spearman’s coefficients of rank correlation were small and insignificant between consecutive years, but large and highly significant between alternate years, suggesting the presence of carry-over effects in seed/cone production. The broad-sense heritability on a clone mean basis was 0.74 ± 0.15 for the number of cones, 0.72 ± 0.14 for the yield of cones, and 0.68 ± 0.13 for the yield of seeds. The corresponding heritabilities from analyses combined over all years were 0.24, 0.558, and 0.724, respectively. These results indicate that seed/cone production in hinoki is under strong genetic control. Several managerial measures are discussed that maintain the genetic diversity in seedlots used for reforestation, by reducing the variation in seed/cone production among clones and producing seed crops with equal contributions from all parents.     

In situ forcing of pollen maturation in Jack pine and Japanese larch male cones

Paper bags equipped with a polyethylene window were placed on branchlets 6 and 9 weeks before the time of pollen dispersal of Japanese larch and Jack pine, respectively. This technique produced a localized greenhouse effect around the branchlets, accelerating male cone development. For Jack pine trees, buds in bags with the window oriented upward had the shortest maturation time. In fact, it was possible to collect these pollen cones about 11 days before the unbagged control without decreasing the production of pollen. The mean germination rate of the pollen from the bags with the window pointed upward, however, was significantly lower than the control, by about 15%. Buds in bags with the window oriented downward had an accelerated development of about 9 days (compared to the control) but did not experience a decrease in pollen production or quality. This same technique caused the abortion of male cones of the Japanese larch trees. Thermal probes in the bags enabled us to follow the progression of temperature inside and outside the bags during the development of the male cone in each species. They revealed that the bags increased mean temperature, daily temperature range, and degree-day accumulation, and that this effect was most pronounced for Jack pine, when the polyethylene window was oriented upward.     

Photosynthetic response to green crown pruning in young plantation-grown Eucalyptus pilularis and E. cloeziana

The loss of foliage through pruning of live branches may reduce tree growth or it may be compensated by photosynthetic up-regulation of the remaining crown. Here, the changes in light-saturated photosynthesis following pruning to remove 50% of green crown length were examined in 4-year-old Eucalyptus pilularis Sm. and Eucalyptus cloeziana F. Muell. trees. The objectives of the study were to: (1) compare leaf-level physiological (light-saturated photosynthesis (A_max), stomatal conductance (g), transpiration (T), dark respiration (R_d), quantum yield (Φ), light compensation point (Γ), water-use efficiency (WUE), nitrogen-use efficiency (NUE)) traits in species with contrasting crown dynamics and structure, (2) examine the effect of crown position on these traits, and (3) examine the effect of pruning on A_max, g, T, WUE, NUE, leaf N and P concentrations and specific leaf area (SLA). Prior to pruning there were no differences in R_d, Γ and Φ between E. pilularis and E. cloeziana but differences in A_max, T, g, leaf N, leaf P, WUE, NUE and SLA. Whereas the rate of physiological processes (A_max, T, and g) and leaf N and P concentrations increased with crown height, R_d, Γ, Φ and SLA declined along this vertical gradient, except in the upper crown of E. cloeziana where A_max, T and g were not different to the lower crown. No up-regulation of photosynthesis or changes in leaf physiology occurred between 6 and 13 months after pruning in either species. The results provide an important basis for modelling pruning effects in process-based tree growth models.     

Pruning effects on root distribution and nutrient dynamics in an acacia hedgerow planting in northern Kenya

Tree pruning is a common management practice in agroforestry for mulching and reducing competition between the annual and perennial crop. The below-ground effects of pruning, however, are poorly understood. Therefore, nutrient dynamics and root distribution were assessed in hedgerow plantings of Acacia saligna (Labill.) H.L. Wendl. after tree pruning. Pruning to a height of 1.5 m was carried out in March and September 1996. In July and October 1996, the fine root distribution (< 2 mm) and their carbohydrate contents were determined at three distances to the tree row by soil coring. At the same time, foliar nutrient contents were assessed, whereas nutrient leaching was measured continuously. The highest root length density (RLD) was always found in the topsoil (0–0.15 m) directly under the hedgerow (0–0.25 m distance to trees). Pruning diminished the RLD in the acacia plots at all depths and positions. The relative vertical distribution of total roots did not differ between trees with or without pruning, but live root abundance in the subsoil was comparatively lower when trees were pruned than without pruning. In the dry season, the proportion of dead roots of pruned acacias was higher than of unpruned ones, while the fine roots of unpruned trees contained more glucose than those of pruned trees. Pruning effectively reduced root development and may decrease potential below-ground competition with intercropped plants, but the reduction in subsoil roots also increased the danger of nutrient losses by leaching. Leaching losses of such mobile nutrients as NO₃^– were likely to occur especially in the alley between pruned hedgerows and tended to be higher after pruning. The reduced size of the root system of pruned acacias negatively affected their P and Mn nutrition. Pruning also reduced the function of the trees as a safety net against the leaching of nutrients for both NO₃^– and Mn, though not for other studied elements. If nutrient capture is an important aim of an agroforestry system, the concept of alley cropping with pruning should be revised for a more efficient nutrient recycling in the system described here.This revised version was published online in November 2005 with corrections to the Cover Date.     

Tree/crop complementarity in an arid zone runoff agroforestry system in northern Kenya

We tested the hypothesis that shallow-rooted crops and deep-rooted trees will share the available water in a complementary manner, when grown together, in a field trail in the Turkana district of northern Kenya during 1994 to 1996. Such studies have been few in dryland agroforestry. The effects of two different Acacia saligna (Labill.) H. Wendl. tree planting densities (2500 and 833 trees per ha), tree pruning (no pruning vs. pruning) and annual intercrops (no intercrop vs. intercrop) on total biomass production and their interactions were tested. In 1996 Sorghum bicolor (L.) Moench was used during the first vegetation period and Vigna unguiculata (L.) Walp. during the second. We used naturally generated runoff water for irrigation to supplement low rainfall amounts typical for the area. High biomass production (> 13 t ha^–1 over a two year period) was observed irrespective of intercropping of pruned trees or sole tree stands. Although the pruning treatment reduced total tree biomass yields by a quarter, the introduction of annual intercrops after the pruning of trees outweighed this loss. The yields of the intercrops in the pruned tree treatments were similar to their yields when grown as monocrops. The calculation of land equivalent ratios showed overyielding for intercropped, pruned systems. The high values for LER (1.36 at low and 1.47 at high density of trees) indicate that there is complementarity in resource use between the different species.This revised version was published online in November 2005 with corrections to the Cover Date.     

Changes in arboreal ant populations following pruning of coffee shade-treesin Chiapas,Mexico

Reducing or eliminating shade cover in coffee (Coffea arabica L.) agroforestry systems affects fungal disease and pest outbreaks, coffee yields, and can result in biodiversity loss of important predators, such as ants. Less dramatic changes in shade structure or composition may also affect ants. Shade tree pruning, a common management practice in shaded coffee systems, has unknown consequences for ant communities. The effects of pruning on arboreal ant communities were investigated by measuring ant abundance, distribution, and species richness in the short (1 week) and long-term (6 months) after shade tree pruning in one 25×50 m plot. Shade tree pruning significantly affected the distribution and abundance of two of the most common ant species (Azteca instabilis F. Smith and Camponotus senex textor Forel), and in general did not affect other ants. After pruning, C. senex textor ants were 80% more abundant on coffee plants and shade trees, whereas A. instabilis abundance dropped by 40% on coffee plants and 73% on shade trees after pruning. Additionally, C. senex textor were significantly more widespread, whereas A. instabilis distributions were more restricted. The effects of pruning were strong over the short-term, but were not evident over the long-term. Shade tree pruning did not affect ant diversity. Thus shade tree pruning largely affected certain aspects of arboreal ant communities in one coffee agroforestry system, with important implications for biological control.     

The effects of between-row (alley widths) and within-row spacings ofGliricidia sepium on alley-cropped maize in Sierra Leone

The effects of spacings between hedgerows (alley widths) and the spacings of trees within hedgerows ofGliricidia sepium on growth and grain yield of maize were investigated at Senehun in southern Sierra Leone. Four between-row spacings (2, 4, 6 and 8 m) were combined with three within-row spacings (0.25, 0.50 and 1.00 m) in a split block design. Maize, at densities of 20,000, 40,000 and 53,333 plants ha^–1, was established in the alleys and also as pure crops. N, P and K fertilizers were applied to all plots before pruning of the trees began. When pruning started, only the pure maize plots received fertilizer; prunings from the hedgerows were returned to the appropriate alleys in the other plots.Plots with the highest maize populations consistently gave the best yields before pruning started, but lower populations gave improved yields after pruning. Yields of maize increased with increasing alley widths before the start of pruning, after which the narrower alleys of 2 and 4 m outyielded the wider ones by almost double, probably because of the large amount of nutrients applied in prunings. Lack of light limited grain yields before the start of pruning, when there was some shading by the hedgerows. Alleys of 2–4 m wide, planted no closer than 0.50 m within rows, resulted in more than twice the yields of maize than in the 8-m alleys planted at 0.25 m within rows, once the hedgerows were well established and were being managed.     

Water use efficiency and uptake patterns in a runoff agroforestry system in an arid environment

Water is the most limiting factor for plant production in arid to semiarid regions. In order to overcome this limitation surface runoff water can be used to supplement seasonal rainfall. During 1996 we conducted a runoff irrigated agroforestry field trial in the Turkana district of Northern Kenya. The effects of two different Acacia saligna (Labill.) H. Wendl. tree planting densities (2500 and 833 trees per ha), tree pruning (no pruning vs. pruning) and annual intercrops (no intercrop vs. intercrop: Sorghum bicolor (L.) Moench during the first season and Vigna unguiculata (L.) Walp. during the second season) on water use were investigated. The annual crops were also grown as monocrops. Water consumption ranged from 585 to 840 mm during the first season (only treatments including trees). During the second season, which was shorter and the plants relied solely on stored water in the soil profile, water consumption was less than half of that during the first season. Highest water consumptions were found for non-pruned trees at high density and the lowest were found for the annual crops grown as monocrops. Tree pruning decreased water uptake compared to non-pruned trees but soil moisture depletion pattern showed complementarity in water uptake between pruned trees and annual intercrops. The highest values of water use efficiency for an individual treatment were achieved when the pruned trees at high density were intercropped with sorghum (1.59 kg m^–3) and cowpea (1.21 kg m^–3). Intercropping and high tree density increased water use efficiency in our runoff agroforestry trial. We ascribe the observed improvement in water use efficiency to the reduction of unproductive water loss from the bare soil.This revised version was published online in November 2005 with corrections to the Cover Date.     

Reduced fertilization capacity of SO2-fumigated Picea omorika-pollen

Dry pollen of Picea omorika was fumigated with 820 μg/m³ SO₂ for 3 , 4 , or 5 days. Few or no adverse effects occurred during subsequent in vitro germination: germination percentage was not affected at all, and reduction of the pollen tube length was only 10% in pollen exposed for 5 days. Nevertheless pollination with fumigated pollen resulted in reduced seed production. 41% fewer seeds per cone were produced, and only 33% were filled seeds, compared to 70% filled seeds of the control. Seed-weight was reduced from 3.6 g per thousand in the control to 3.0 g per thousand, when cones were pollinated with pollen fumigated for 5 days. Similar but less striking effects were found with pollen fumigated for 3 or 4 days.     

Millet production under pruned tree crowns in a parkland system in Burkina Faso

As a tree management tool, three treatments of crown pruning (total-pruning, half-pruning and no-pruning) were applied to Vitellaria paradoxa (karité) and Parkia biglobosa (néré) in agroforestry parkland systems in Burkina Faso. The area under each tree was divided into four concentric tree influence zones (Zones A: up to 2 m from the tree trunk, B: up to half of the radius of the tree crown, C: up to the edge of the tree crown and D: up to 2 m away from the edge of the tree crown). Millet production under these zones and outside was assessed during two cropping seasons over the study period of three years and the results showed that tree crown pruning had significant effect on millet production and the highest millet grain yield and total dry matter were produced under total-pruned trees (507 ± 49 and 2033 ± 236 kg ha⁻¹ year⁻¹, respectively). Light transmission, transpiration and soil nutrient status under the trees were also analysed in relation to millet production. The results of the analysis showed that total-pruned trees gave the highest millet production due to the reduction by crown pruning of the effects of large tree crowns on PAR transmission below crowns and rates of transpiration by trees. Soil was more fertile closer to the tree trunks than outside tree crowns. This may also be one of the reasons why millet overall performed better under Zone B than outside tree crowns. The higher production of millet under Zone B than under Zone A, the zone closer to the tree trunk, may be due to lower light intensity and more intense competition for water between trees and crops under Zone A. It was concluded that at least in the short term millet production could be improved by crown pruning of both karité and néré, but long term effects may depend on the ability of the trees to maintain the amelioration of soil fertility and on how quickly the trees recover from pruning. This revised version was published online in June 2006 with corrections to the Cover Date.     

Effects of pruning on radial growth and biomass increment of trees growing in homegardens of Kerala, India

To evaluate the effects of pruning on stem radial growth increment and leaf and twig biomass production, an experiment with four pruning intensities (0, 50, 75 and 90%) on ten locally important tree species (Ailanthus triphysa, Albizia odoratissima, Artocarpus hirsutus, Bombax malabarica, Bridelia crenulata, Erythrina indica, Grewia tiliifolia, Macaranga peltata, Terminalia paniculata and Xylia xylocarpa ), was carried out. The results did not support the contention that a certain level of pruning promotes stem growth in trees. Instead, all species have a level of pruning that reduces annual increment in stem diameter. In Ailanthus triphysa and Artocarpus hirsutus trees subjected to different pruning intensities showed a decline in the annual increment in stem diameter while in other species diameter increment reduced when the pruning intensity was 75% and 90%. Response to pruning in terms of biomass production also varied from species to species. In Erythrina indica, Macaranga peltata and Terminalia paniculata annual foliage and branch production in pruned trees was significantly more than that of the un-pruned trees. However, in Ailanthus triphysa, Albizia odoratissima, Artocarpus hirsutus, Bridelia crenulata, Grewia tiliifolia and Xylia xylocarpa pruned trees produced comparatively more amount of foliage and branches produced annually than that by the un-pruned trees when the pruning was carried out once in 2 years. Based on these observations it is recommended that trees of Erythrina indica, Macaranga peltata and Terminalia paniculata may be pruned at 50% level annually while the trees of Ailanthus triphysa, Albizia odoratissima, Artocarpus hirsutus, Bridelia crenulata, Grewia tiliifolia and Xylia xylocarpa may be pruned at the same pruning intensity once in 2 years.