. . rill? 4.... . . r...r..mwn.t.. -..! n... .1. l. i. . L? Linn s ...... ... ...u‘t! ..i: 9:130:34: A”... 15.1.. I .3 ”Tuned“: . ' wit, a. .PIX ILA?” a} an. {VJ-'5' ' “I: 1 ‘59-»)..1.‘ I ...»! (In. )lvl‘ 4.343 Lamp-vhfi 1.1.. .99.:3. {pl}. 1 31.36“; : .. z ... €2.90“. auras-rot .. Sufi: 3......213. Ht: .... 2%.»! 33.1%...2... I ...! o. ¢\|i.\| hirui-hlnflrzl’fihl4 {col-l. lIJ. 2R \u.’\-.>vl". . .-.”- .»--.»---¢- Means of Dry BIomass From Prumngs (kg/m’) 7////////. / Pruning Height (m) Figure 3.9. Means of dry biomass from the second prunings. Bars topped with the same letter are not significantly different (LSD, p < 0.05). 78 3.3.1.5 Decomposition and Tissue Analysis Deco-position. In the initial clipping, only 38 percent of the total weight of samples decomposed over a period of six weeks, despite a fairly moderate amount of precipitation during these months (Figure 3.10 and Table 3.9). In the subsequent prunings, however, the accumulated weight loss of most pruning samples over six to eight weeks reached 48-60 percent (Figures 3.11 and 3.12). During high precipitation, when the prunings from the 1.5- and 2.0-m hedgerows were decomposing, the percent accumulated weight loss of samples after one week ranged from 40-46 percent for all the pruning heights. This percent accumulated weight loss is two to three times higher than the weight loss of the samples from the first pruning of hedgerows. Cumulative Wei ht Loss (Perceng 79 Figure 3.10. —d l I 3 Week N 43—1 (”—4 CD Decomposition of biomass from initial clipping over time. 80 I. d .. II Cumulative Wei ht Loss (Percenl?) e—e 2.0-m Pruning Height :'.- H 1.5—m Pruning Height ' H 1.0—m Pruning Height I 10 l l l l i l l I I I 11 1213 O No (A 4:. U" 07 \l a) 1.0 Week Figure 3.11. Decomposition of the first pruning biomass over time. 81 70— 3’, 60— o . _J .60 * a) 5 ~ 33 50~ 38 . 6v 4 3 .. g . L) ‘40-‘ ‘ H 2.0-m Pruning Height , ' e—e 1.5-m Pruning Height ‘ H 1.0—m Pruning Height 30 I I I T I I I I I 0 1 2 3 4 5 6 7 8 9 Week Figure 3.12. Decomposition of the second pruning biomass over time. 82 Table 3.9. Decomposition period of various prunings and the corresponding amount of precipitation during each period. Source of Start of Bad of Amount of No of biomass decompositi decompositi precipitati weeks to on period on period on in mm attain 50% decomposit ion Initial clipping 9/13/91 10/26/91 252.9 6' First pruning at 11/3/91 12/30/91 449.6 4 1.0-m high First pruning at 11/23/91 2/29/92 185.8 7 First pruning at 1/7/92 4/16/92 93.7 8 2.0-m high 2nd pruning at 2/29/92 5/1/92 69.6 7 2nd pruning at 4/16/92 6/11/92 225.9 8 1.5-m high 2nd pruning at 5/1/92 6/12/92 188.4 4 2.0-m high ' At week 6, the accumulated decomposition of biomass was only 38.6 percent. 83 Tissue Analysis. The tissue analysis (Table 3.10) was obtained to estimate nutrient contribution of the biomass. The data gathering was not designed. for statistical analysis with respect to the various treatments. The result, however, indicated that the average percent ash gradually declined with increasing pruning height (e.g. decrease in pruning intensity) in the first pruning; however, this pattern changed during the second pruning. Percent ash moderately increased with greater pruning heights in the second pruning. In the first and second prunings, CP exhibited a declining trend with increasing pruning height. This trend was more observable, however, in the second pruning. Percent CF, percent P, and percent K steadily declined during the first pruning; however, in the second pruning, percent CF and percent K gradually increased with increasing pruning height. Percent P declined during the second pruning. Percent N increased with pruning height during the first pruning, but‘ declined in the second pruning period. The general trend of N, CP, K, and P tended to decline from the initial clipping up to the second pruning. The tissue components from the first and second prunings and at various pruning heights were subjected to T-test to determine if there were significant differences in their means (p < 0.05) (Table 3.11). Only CF declined significantly from the first to the second pruning. The T-tests on the tissue components at various pruning heights were not significant (p < 0.05). 84 Table 3.10. Results of the tissue analysis'caf samples from the hedgerow prunings. =E= Prunings’ t Ash a as It or s x % p s 1: Initial 9.49 21.14 23.52 2.32 0.29 3.08 clippings 1st pruning at 10.89 16.41 35.31 2.06 0.34 3.65 1.0-m 2nd pruning at 7.65 20.83 18.66 2.81 0.27 1.91 1.0-m 1st pruning at 9.64 19.31 32.14 2.73 0.27 3.20 1.5-m 2nd pruning at 7.97 20.92 15.22 2.87 0.24 2.91 1.5-m 1st pruning at 7.39 17.73 20.16 3.34 0.27 2.47 2.0-m 2nd pruning at 9.61 21.93 20.44 2.70 0.25 2.91 2.0-m AVERAGE 8.94 19.75 23.63__ 2.69 .28 2.87 ' CP a crude protein; CF = crude fiber 2 Data on the initial clippings are the means of 12 samples, 4 randomly selected samples from each replicate; those of the subsequent prunings come from the average of three samples, one from each replicate. Table 3.11m T-test' for various tissue components from the first and second prunings. ' «we Er = Tissue Component in % Means of 1st Means of 2nd t-test pruning pruning value Ash 9.31 8.41 1.22 Crude protein 17.82 21.23 -2.08 Crude fiber 29.20 18.11 3.72*** N 2.71 2.89 -0.32 P 0.29 0.25 1.68 K 3.11 2.58 1.61 E Significant at one percent probability 85 3.3.2 Intercrops 3.3.2.1 Maise Grain and Stover First Maise Crop The effects of the number of hedges per contour line and within-row spacing on the yields (grain and dry stover) of maize were not significant in the first crop (Table 3.12). The average grain yield of the first maize crop was 144.6 g/m2 of grain. This could be roughly translated into about 1.4 tons per hectare. The control plots had an average yield of shelled maize of 173.6 g/mfl slightly higher than the 144.6 g/m2 of the treated plots. The average dry weight of stover from the control plots was 596 g/m’, higher than the 426 g/nfi from the treated plots. Table 3.12. F-values of analysis of variance of the air-dried weight of shelled maize and oven dry stover from the first crop. _ m Source of Variation Corn Stoverl Shelled Corn‘ Number of hedges/line 0.13 0.28 Within-row spacing * 0.61 0.26 Interaction 0.12 0.68 E g ’ All F-values are non-significant at p < 0.10. Second Maise Crop The F-tests (p < 0.10) for the effects of the number of hedges per contour line and the combination of pruning height and within-row spacing on the yields (air-dried grain and dry stover) of the second maize crop were not 86 significant (Table 3.13). The average yield of shelled maize in the second crop was 141.2 g/m2 or about 1.4 tons/ha, slightly lower than the average yield from the first maize crop. The control plots had average shelled maize yield of 135.6 g/nfi, lower than the yield of the first maize crap from the same plots. The average dry weight of stover was 691 g/m’, slightly higher compared with the first maize crop and with the treated plots (687 g/mfi. Table 3.13. iF-values of analysis of variance of the air-dried weight of shelled maize and oven dry stover from the second crop. Corn Stoverl Shelled Cornl i Number of hedges per line 0.19 0.22 I Pruning haight x within-row 1.39 1.53 - ll 0.76 n ' All F-tests are non-significant at p < 0.10. 0.59 i Interaction L—— 3.3.2.2 Mungbean In the first crop, the effect within-row spacing on mungbean yield were significant (p < 0.10) (Table 3.14). However, the effects of the number of hedges per contour line and the interaction effects were not significant. In the second harvest, the number of hedges per contour line, combination of pruning height and within-row spacing, and interaction did not significantly affect the air-dried yield of mungbean (p < 0.10) (Table 3.15). In this case, the number of hedges per contour line and the combinations of pruning height and within-row spacing were not found to influence the production of air-dried mungbean. 87 In the first mungbean crop, the highest yield came from the 20-cm within- row spacing, followed by those from the 40-cm within—row spacing. The least yield came from those S-cm within-row spacing (Table 3.16). There was significant difference in the means of 20-cm within-row spacing and those in the 5- and lO-cm within-row spacings. However, there was no significant difference in the means of 10-cm and 5-cm within-row spacings. Table 3.14. Analysis of variancel for the air-dried mungbean yield from the first crop. = _ == I Source of Variation F-Value I;Number of hedges per contour line 0.67 “ Within-row spacing 2.84* I Interaction 0.43 “E ' * Significant at 10 percent probability Table 3.15. Analysis of variance for the air-dried mungbean yield from the second crop. - ==================H F-Valuel Number of hedges per contour line 0.03 Pruning Height x Within-row spacing 1.69 Interaction 0.45 ' All values are non-significant at p < 0.10. The average mungbean yield from the treated and control plots from the first cropping were 52.4 g/m2 and 53.7 g/mz, respectively. These yields translate to about 0.52 t and 0.54 t per ha, respectively. In the second mungbean crop, the average yield from the treated and control plots declined to 40.1 g/m2 and 43.3 g/mz, respectively. 88 Table 3.16. Means of mungbean yield (g/m’) from the first crop. Treatments Means Ranked Order‘ . Within-row spacing of 20 cm 57.5 A Within-row spacing of 40 cm 52.2 AB i Within-row spacing of 10 cm 50.9 B Uta-m . ° _ ___ ,1 ___ 49-1 ' Values followed by the same alphabetical letter are not significantly different. 3.4 Discussion Contour hedgerows are the unique feature of alley cropping in uplands. The hedgerows serve as a vegetative barrier against soil erosion and as a ”factory” of organic matter for the intercrops. These intertwining service functions of hedgerows are the cornerstones of alley cropping. Without contour hedgerows, alley cropping in the uplands will not be sustainable. Contour hedgerows conserve top soil and restore soil fertility. Accordingly, the hedgerows' height and diameter growth, pruning biomass, mortality of hedgerows, and quality and decomposition rate of prunings are important variables in achieving the purposes of alley cropping. The present study on G. sepium hedgerows revealed that the above parameters can be directed to obtain optimum net gains while minimizing the unfavorable impacts of competition on the intercrop. Several controllable variables were highlighted which could be managed to improve alley cropping in the short- and long-term. The results also provided insights regarding the performance of G. sepium as a hedgerow species. The results of the height and diameter measurements, percent mortality of hedgerows, and, to a certain extent, the dry pruning biomass suggest that 89 G. sepium hedgerow plants compete for light and space in their early growth and development. This competition affects diameter and height growth, percent mortality over a period of time, and.production of pruning biomass. It was observed that diameter and height growth of hedgerows increased with increasing within-row spacing (less dense planting of hedgerows). Within the 12 to lB-month.observation.period, positive linear relationships still existed between height and diameter growth and‘within- row spacing of G. sepium hedgerows. The competition for light and space was more intense among hedgerow plants which were planted at S-cm within- row spacing, as evidenced by their highest mortality over the lB-month period and low height and diameter growth. As the densities of hedgerows increased, more weak and shaded plants died. Obviously, plants in the sparsely-planted.hedgerows (20- and.40-cm‘within-row spacing) did not have to compete for light and space early in their development stage. The above observation confirms Harper's (1977) statement that during the early growth of plants, yield is positively determined by density; but, this relationship changes as the plants reach the limits of the resource- supplying power of the environment. At this stage, yield becomes independent of planting densities. Then, plants begin to compete for limited and finite supply of resources (Ford, 1975). Within the 12 to 18- month period of measurement in the present study, however, the hedgerows did not appear to have come to the limits set by the available resources in the environment; dense hedgerows yielded the highest dry initial pruning biomass. Hedgerows in more dense within-row spacing were forced to maximize space and resource availability in their early growth; thus, the hedgerows yielded higher biomass despite their lowest attained height and diameter growth. Hedgerows in more dense within-row spacing could imediately contribute the most organic matter for the intercrops. Therefore, higher biomass yield is a net gain from dense planting of hedgerows even though the plants are subjected to early stress and 90 intraspecies competition. In addition, since hedgerows from more dense within-row spacing were short, they had the least threat of shading the crops compared with the taller, wide-spaced hedgerow plants. Also, at more dense planting, contour hedgerows are more effective in controlling soil erosion and slowing down the speed of water running down from the higher slopes (MBRLC Editorial Staff, 1988; Laquihon et al., 1991). Studies on leucaena hedgerow species, such as those of Lu and flu (1981); Guevarra et al. (1978); and Ella et al. (1989), concluded that biomass (prunings, leaves, or wood) increased with planting density. Desai et al. (1988), however, found that there was no difference in biomass production due to plant densities of leucaena over a three year period. Lu and flu (1981) also observed that leucaena plantings of more than 5,000 trees/ha had high mortality due to competition for light and nutrition. In G. sepium, Sumberg (1986) observed that yield increased with increasing density. He found that gliricidia produced the highest mulch when established at approximately 10 plants/m of hedgerow (i.e. lO-cm within- row spacing). Studies with other plants indicate that those in high densities are able to maximize yield per unit area during their early stage of growth and development (Harper, 1977) . This reasoning might also apply to hedgerows. Perhaps during the early stage of hedgerow development, the supply of resources for growth has not yet fallen below the combined demands of the individual plants (Donald, 1963); hence, higher pruning yields from dense planting of hedgerows. This may also explain why the number of hedges per contour line and the interaction effect did not produce significant difference in the height and diameter growth of hedgerows, percent mortality of hedgerows over time, and pruning from the initial clipping during the lZ-month period. Conceivably, the significant effect of the number of hedges per line on the lB-month diameter of hedgerows may be due 91 to the fact that the supply limit of available resources was nearing its critical point. Accordingly, between 6 and 18 months of hedgerow growth, an opportunity exists to harvest pruning biomass when shoot growth is still increasing but before it gets to an optimum point after which the yield declines (Cannell, 1983; and Huxley, 1985). In the short term, therefore, the benefits from densely planted G. sepium hedgerows in terms of biomass yield (organic matter) and erosion control (keeping fertile top soils in the alley) may yet outweigh the unfavorable impacts of intraspecies competitionmon the productivity of intercrops. It is suspected that during the early growth of hedgerows, when available resources are not yet limiting, intraspecies competition among hedgerows may not yet pose a threat to intercrops. Possibly, the hedgerows have not yet developed extensive lateral roots towards the alley and deprive the intercrops of moisture and nutrients. In the long term, the hedgerows will be regulated by the "law of constant final yield" (Kira et al., 1953 as cited by Harper, 1977). The density issue becomes invalid; the hedgerows will be limited by the resource- supplying power of the environment. Densely planted hedgerows would be forced to ”self-thin” in order to grow and survive (Harper, 1977). In densely established hedgerows, plants with large stem diameters would tend to obtain a greater proportion of available soil resources at the expense of the smaller hedgerow plants until an equilibrium of co-existense is achieved. When this happens, a temporal optimwm density for hedgerows would have been attained. 3.4.1 Key Variables in lanaging the Hedgerow: The present study highlighted key variables that could be manipulated to optimize benefits from the hedgerows for the alley crop. These variables 92 are plant density (number of hedges per contour line, within-row spacing) and pruning height. These variables can be managed to get an optimum pruning biomass, effect a certain level of hedgerow plants mortality, possibly influence the nutrient composition of prunings, check root growth to minimize the tree-crop interface, and minimize the hedgerows' shading effect. Initial pruning biomass is maximized from densely planted hedgerows, especially if they are initially cut to 30-40 cm above ground and maintained at that height (Briscoe, 1989; Watson, 1983; Garrity, 1991, pers. comm). In the present study, the average N, P, and K contribution of G. sepium biomass to the soil from the initial clipping (3.4 kg dry biomass/m2 alley or 17 t dry matter/ha) was estimated to be 144 kg, 14 kg, and 154 kg per ha, respectively. The average N, P, and K contribution from each subsequent pruning (0.7 kg dry biomass/m2 alley or 3.5 t dry matter/ha) was estimated to be 56 kg, 5 kg, and 60 kg per ha, respectively'. Considering the higher average density of 13,333 trees/ha in the study and the volcanic soil in the site, these estimates are within the range of those reported by Atta-Krah and Sumberg (1987) and Budelman (1986) as cited by Kang and Mulongoy (1987). Atta-Krah and Sumberg (1987), for instance, reported that pruning yields ranged from 2.85-3.06 t dry matter/ha with N contribution of 79-104 kg/ha. Budelman (1986) as cited by Rang and Mulongoy (1987) estimated an average of 15.2 t dry matter/ha from G. sepium at 10,000 trees/ha. Thus, planting hedgerows at high densities (5- and 1.0-cm within-row spacing) can be a strategy to ' These estimates were calculated based on the following formula: kg of element/ha :- kg dry biomass/m2 alley divided by 2 x percent of young twigs and leaves (30 and 60 t for the initial clipping and subsequent prunings, respectively) x amount in percent of the element from the tissue analysis x 10,000 m2 per ha. The dry biomass (kg/m2 alley) is assumed to be applied in two adjacent alleys. 93 maximize initial and subsequent pruning biomass and ultimately improve soil fertility. Combining within-row spacing (density) and pruning height forms another scheme to optimize pruning biomass for the benefit of the intercrops. The optimum pruning height in terms of yielding the highest biomass was 2.0 m for all the different within-row spacings. In the short term, however, the highest pruning biomass could be obtained from hedgerows with 5-cm within-row spacing. As the pruning heights were reduced (e.g. pruning frequencies increased), the biomass yields of G. sepium decreased. This is consistent with notion that there are great amounts of storage reserves (of carbohydrates and minerals) left on the plants when they were pruned high (Cannel, 1991; Erdmann et al., 1993). Thus, hedgerows should be cut back to 30 cm above ground when they reached the height of 2.0 m in order to maximize biomass yield. At the height of 2.0 m, however, the hedgerows may potentially shade the intercrops in the alleys. Previous work which supports the results of this study include those of Duguma, et al. (1988), Das and Dalvi (1981), Osman (1981); Guevarra et al. (1978); and Atta-Krah and Sumberg (1987). They maintained that with nitrogen-fixing species such as leucaena, G. sepium, and .Sesbania grandiflora, pruning biomass increased with pruning heights. Hedgerows attaining heights of more than 100 cm, however, should be pruned back to 30-40 cm above ground to minimize their shading the intercrops (Kang et al., 1985; Ong, 1989; and Briscoe, 1989). Das and Galvi (1981) even recommended pruning leucaena between 75-150 cm to obtain optimum biomass yield. Given the need to minimize the hedgerow's shading effect on the intercrop while maximizing pruning biomass, the question of optimum pruning height for G. sepium would then be a concern. Results of the first and second 94 pruning of hedgerows suggest that pruning at heights of 1.0 and 1.5 m will reduce biomass by about 30 to 75 percent per cutting based on the means of pruning at the height of 2.0 m. Will this reduction in biomass outweigh the benefit to the intercrop from reduced hedgerow shading? The yields of maize from the first and second cropping and mungbean in the second crop were not significantly affected by the various treatments. The maize crop yields were not affected by within-row spacing, pruning height, and number of hedges per contour line. However, within-row spacing affected the yield of the first mungbean crop. Other studies concluded that crops planted near the hedgerows had lower yield than those in the center of alley. This was observed in upland rice (Solera, 1992) with various nitrogen-fixing hedgerow species including G. sepium, and maize (Ong et al., 1992; Huxley et al., 1989; Salazar et al., 1993) with leucaena hedgerows. The lower yields of crops near the hedgerows were attributed to hedgerow shading and competition for moisture and nutrients by the roots of hedges and crops. Kang and Mulongoy (1987), in their studies of G. sepium loppings stressed that maximum benefit from green manure comes from the timely release and mineralization of its nutrients with regards to the requirements of the alley crops. The volume of pruning is important; but, the release of nutrients must be timed when the food crop needs them. Otherwise, most of the N will be lost through volatilization and leaching. Therefore, it is possible to prematurely apply a high volume of prunings in the alleys with minimal effect on the food crop. On the other hand, it is also conceivable that small amounts of prunings at regular intervals may provide more nutrient benefits for the alley crop. In this case, the lower pruning heights (shorter pruning intervals) would be more applicable. There would even be less hedgerow shading of the intercrops. 95 Accordingly, a relevant question with respect to pruning would be the level of nutrient concentration of biomass at every pruning activity and the corresponding rate of decomposition. In the study of Duguma et al. (1988), they reported that N concentration of prunings increased with decreasing pruning frequencies (e.g. higher pruning heights). In the present study, however, the result of the tissue analysis of G. sepium prunings indicated that N increased, though not statistically-tested, with higher pruning heights. However, the ligninzN ratios between repeated prunings and at higher pruning heights may have decreased because of the significant decrease in the means of crude fiber (CF) content between the first and second prunings and among the 1.0-, 1.5-, and 2.0-m pruning heights (Salazar and Palm, 1987). Thus, it was possible that the lower crude fiber content (or lower lignin concentration) in the second pruning facilitated the decomposition process. The materials that were left in the container after 8-13 weeks were mostly young twigs. Change in the CF content would probably have a minimal influence on decomposition rates because G. sepium leaves already have a low C:N ratio of 10:7 (Weeraratna, 1979) and lignin:N ratio of 2.1 (Salazar and Palm, 1987). The C:N ratio is below the upper limit of 30; thus, there is enough N to meet microbial needs (Foth and Ellis, 1988). .A determination of the 1ignin:N ratio of the G. sepium biomass at different pruning heights would have given a better indication of decomposition rates because lignin is the key in the breakdown of biomass (Palm and Sanchez, 1991; Melillo et al., 1982; Salazar and Palm, 1987). This becomes more important when young twigs and not only leaves are applied in the alleys, such as in this study. The result of the present study on decomposition suggests that the 50 percent loss of initial weight of the G. sepium prunings at the fourth to eighth week was more of an effect of moisture condition prevailing in the experimental site rather than the "decomposability" of the biomass from 96 different pruning heights. The slight differences in the decomposition pattern of the various prunings could be attributed to the initial drying up of green material before disintegration occurred, especially during the dry months when precipitation was quite low. Studies on G. sepium at International Institute of Tropical Agriculture (IITA) corroborate with the results of the study. For instance, Wilson et al. (1986) found that the number of weeks until 50 percent loss from G. sepium leaves ranged from 1.6 to 3.6, depending on the prevailing local rainfall pattern. Budelman (1987) used 20.3 days in his regression model as the time to lose half of the G. sepium mulch material. Yamoah, et al. (1986) observed that it only took 20 days for the G. sepium leaves to release 50 percent of their initial P content during the decomposition process. Differences in the decomposition pattern may also be attributed to the kind of biomass that was used in decomposition. Buldelman (1987), for instance, used G. sepium biomass with a ratio of leaves as a percent over total fresh weight with ranged from 8.1 to 12.4 percent. In another study, leaves of G. sepium was about 22 percent of the total biomass (Ghuman and Lal, 1990). In this study, young twigs and leaves ranged from 30-60 percent of the biomass; hence the decomposition period was longer. 3.4.2 gligigigig ggpigg as a Hedgerow Species In this study, G. sepium proved to be one of the more ideal hedgerow species. The species responded positively to various kinds of density plantings and repeated prunings. It produced the highest initial pruning biomass even at stressful high-density plantings. .Although, the G. sepium hedgerows suffered mortality in high density plantings and after intensive prunings, the species proved that it could still coppice vigorously from repeated prunings. There were some indications that the prunings, when placed in the alleys at the right time could mineralize enough nutrients (particularly N and P) for the alley crop (Kang et al., 1984; Kang and 97 Mulongoy, 1987). There were no significant reductions in N and P of biomass over repeated prunings and at different pruning heights. The N and P concentration (range of 1.13-4.85 percent and 0.05-0.32 percent, respectively) of G. sepium'are comparable with leucaena (range of 0.51-5.08 percent and 0.03-0.32 percent, respectively) (Lasco, 1991). It is not attacked by a major pest or disease, such as the jumping plant lice in leucaena (Anon, 1988). G. sepium belongs to a family of nitrogen- fixing' plants, nodulates profusely, has symbiotic relationship with rhizobia and mycorrhiza, and is a prolific seeder (Nanguiat et al., 1990; Rang and Mulongoy, 1987; Glover, 1986; ILCA, 1984). G. sepium, however, can only fix N at the rate of 13 kg/ha/yr compared with more than 100 kg/ha/yr for leucaena (Hanguiat et al., 1990; Young, 1989). Hedgerows of G. sepium'can be established from'direct seeding with high germination and survival (such as in this study) and also from cuttings (Solera, 1992). Laquihon (1988) and Laquihon et a1. (1991) have consistently ranked G. sepium as one of the few’promising hedgerow species for the uplands of the Philippines. It can tolerate a 4 to 6-month dry period and has the capacity to survive in marginal sites (Hensleigh and Hollaway, 1988). 3.4.3 Yields of Maise and Nungo None of the treatments affected the yields of maize (grain and stover) in either the first and second cropping season. The maize yields were within the average range of 1-2 t/ha in the Philippine upland areas (Tabinga and Gagni, 1985; Laquihon et al., 1991). In the first cropping, the hedgerows were not expected to compete withLmaize for incident light since both were planted at almost the same period. The first maize crop, however, was hit by a typhoon and rodents. In the second season, the maize benefitted from the initial prunings, which were roughly equivalent to a maximum of 34 g N and 3.5 g P per an2 alley (from 5-cm within-row spacing) and minimum of 98 22 g N and 2.2 g P per m2 alley, more than the requirements of maize in one growing season (Tabinga and Gagni, 1985). The timing of the maize crop in the present study, however, did not coincide with the nutrient availability from prunings; hence, most of them might have been volatilized, leached, or carried away by erosion (Mulongoy and van der Heersch, 1988; Tabinga and. Gagni, 1985). Also, the first of the subsequent prunings at the 1.0- and 1.5-m heights were applied in the alleys towards the end of the maize growing period, after tassling and fruiting. In the field, the second crop of maize did not suffer competition for light because all the hedgerows were cut at 30 cm above ground two weeks before planting. Mungbean is a nitrogen-fixing crop and as a crop it would compete with the hedgerows mainly for moisture, incident light, and P. In the first crop, the hedgerows probably competed for‘water and nutrients with the’mungbeans planted near the hedge. In addition, there might have also been a shading effect. Thus, the least mungbean yield was obtained from the S-cm'within- row spacing. In the second mungo crop, moisture stress probably caused the erratic yields from plot to plot because of the effect of uneven watering during the later part of the "El Nino" drought. In both the first and second mungbean crops, their average yields were below the 0.57 t/ha average yield of mungbean in Southern Tagalog, Philippines (Cagampang and Lantican, undated). Some studies, however, showed that higher yields with maize were obtained with G. sepium hedgerows as the source of green manure (Rang et al., 1984; Rang, 1987; Rang and Wilson, 1987). Other workers, however, reported a depression of crop yields with G. sepium hedgerows, especially in the vicinity of the hedges (Lal, 1989; Solera, 1992). These crops include yams, cassava, upland rice, and cowpea. They attributed the lower yields 99 to shading of the hedgerows, declining fertility of the soil, and competition for soil resources. 3.5 Conclusions and Recommendations 3.5.1 Conclusions Based on the results of this study on G. sepium as hedgerows for maize and f mungbean alley crops, the following conclusions are made: Optimum pruning biomass was obtained from hedgerows with 5-cm within-row spacing and pruning height of 2.0-m. Initial pruning and subsequent pruning biomass increased with decreasing (more dense) within-row spacing for a period of 22 months. Pruning yields also increased with pruning height in the two pruning periods. Height and diameter growth increased with increasing within-row spacing (decreasing planting density) for the 6- and 12-month measurements. After initial and subsequent prunings, mean stem diameter’was still highest in the least dense planting of hedgerows. Percent mortality of hedgerow plants was observed to be the highest after 18 months among hedgerows which were planted at 5-cm within- row spacing and pruned at 1.0-m. Mortality rates among hedgerows at all within-row spacing and number of hedges per contour line were not found to be significant within one year. The number of hedges per contour line did not significantly affect initial and subsequent pruning yields and crop yields. Hedgerow diameter growth after 18 months, however, was significantly affected 100 by the number of hedges per contour line. Stem diameter was lowest in the double hedgerows. Pruning did not significantly change the ash, CP, N, P, and R concentration of biomass in the first and second prunings. However, CF significantly declined between the first and second prunings. Except for CF and R, the other elements did not significantly change as the pruning heights increased. This study observed that G. sepium hedgerows at different planting densities and pruning heights did not significantly affect the yields of maize (grain and stover). The yield of the first mungbean crop was significantly lowest in the 5-cm within-row spacings. However, in the secOnd planting season, the number of hedges per contour line, pruning height x within-in row spacing, and interaction did not influence the yield of mungbean. In the long- term, however, crop yields are expected to increase, stabilize, and be sustained with contour hedgerows as organic matter accumulate in the alleys and soil erosion is minimized (Laquihon, et al., 1991; Lal, 1989; Tacio, 1993; Rang et al., 1990). 3.5.2 Recommendations Based on the above conclusions, this study recommends the following management practices to optimize net gain from the hedgerows and provide more benefits to the alley crops: Plant G. sepium in single hedgerows at 5- to 10-cm within-row spacing and.prune the plants after one year to reduce shading of the crop. For maize, pruning heights of 1.5-2.0 m may not cause depression of yields provided that the hedgerows were cut back to 30 101 cm during the planting. Other crops, however, like mungbean may need lower pruning heights to minimize the hedgerow's shading effect. Single hedgerows may be planted along the contour since there was not much significant increase in biomass from double hedgerows. Besides, with single hedgerows, only 10-12 percent of a hectare is used up for hedgerow establishment. Time the pruning of G. sepium hedgerows and application of biomass in the alleys so the crops will most likely use the nutrients from the green manure. With maize, application of prunings should be a week before planting and.*within four weeks after germination (Tabinga and Gagni, 1985). With mungbean, application of prunings should be done one week before and within 20 days after planting (Cagampang and Lantican, undated). G. sepium prunings will mineralize about 50 percent of their nutrient content within four to eight weeks, depending on the moisture condition of the area. Long-term trade-offs between higher and more frequent pruning biomass and shading effects of hedgerows on major alley crops should be investigated so that more definite recommendations on pruning heights and planting density can be made to the upland farmers. To obtain long-trends of G. sepium hedgerow pruning yields, mortality of hedgerow plants, and yields of maize and mungbean, the study needs to be continued for another two to three years. In the continuation of the study, the effects on crap yields of incident light before and after pruning and timing of application of prunings should also be evaluated. 102 Literature Cited Atta-Rrah AN and JE Sumberg (1987) Studies with Gliricidia sepium for crop/livestock. production systems in West Africa. In: Gliricidia sepium (Jacq.) Walp.: Management and Improvement. Special Publication 87-01 by NFTA and CATIE on the Proceedings of a Workshop, Turrialba, Costa Rica, 21-27 June Anon (1988) NET highlight: leucaena psyllids - a review of the problem and its solutions. Leucaena Research Reports 9:3-5 Briscoe CB (1989) Field Trials Manual for Multipurpose Tree Species. Manual No. 3. Winrock International-USAID, Washington, D.C. Budelman A ( 1987) Gliricidia sepium in the Southern Ivory Coast:production, composition, and decomposition of the leaf biomass . In: G1 iricidia sepium: Management and Improvement . Special Publication 87-01 by NFTA and CATIE on the Proceedings of a Workshop, Turrialba, Costa Rica, 21-27 June Cagampang IC and RM Lantican (Undated) Mungo Production in the Philippines. Department of Development Communication, College of Agriculture, University of the Philippines at Los Banos, College, Laguna, Philippines Cannell MGR (1983) Plant management in agroforestry: manipulation of trees, population densities and mixtures of trees and herbaceous crops. In: PA Huxley, ed, Plant Research in Agroforestry, pp 455-487. ICRAF, Nairobi, Renya Cannell MGR (1991) Plant management in agroforestry. In: MA Avery, MGR Cannell and C Ong, eds, Biophysical Research for Asian Agroforestry, pp 59-72. Winrock International and South Asia Books, USA Das RB and GS Dalvi (1981) Effect of internal and intensity of cutting L. leucocephala. Leucaena Research Reports 2:21 Desai SN, RB Rhot, SR Patil and JS Desale (1988) Evaluation of biomass production of eucalyptus hybrid and leucaena (R8) under various planting densities. Leucaena Research Reports 9:48 Donald CM (1963) Competition among crop and pasture plants. Advances in Agronomy. 15:1-114 Duguma 8, BT Rang and DUU Okali (1988). Effect of pruning intensities 6 three woody leguminous species grown in alley cropping with maize and cowpea on an alfisol. Agroforestry Systems 6:19-35 Ella A, C Jacobsen, W Stur and G Blair (1989) Effect of plant density and cutting frequency on the productivity of four tree legumes. Trop. Grassland 23:28-34 Erdmann TR, PRR Nair and BT Rang (1993) Effects of cutting frequency and cutting height on reserve carbohydrates in Gliricidia sepium (Jacq.) Walp. Forest Ecology and. Management 57:45-60 Ford. ED (1975) Competition and stand structure in some even-aged monocultures. J. Ecol. 63:311-333 103 Foth HD and BG Ellis (1988) Soil Fertility. John Wiley and Sons, New York, USA Garrity DP (1991) Personal communications Ghuman BS and R Lal (1990) Nutrient addition into soil by leaves of Cassia siamea and Gliricidia sepium grown on an ultisol in Southern Nigeria. Agroforestry Systems 10:131-133 Glover N (1986) Gliricidia sepium germplasm collection, conservation and evaluation. A MS thesis, University of Hawaii, USA Guevarra A, A Whitney and J Thompson (1978) Influence of intra-row spacing and cutting regimes on the growth and yield of leucaena. Agronomy Journal 70:1033-1037 Harper JL (1977) The influence of density on yield and mortality. In: Population and Biology of Plants, pp.151-194. Academic Press, Inc Hensleigh TE and BR Holaway, eds, (1988). Agroforestry Species for the Philippines. US Peace Corps/ Washington, D.C. Huxley PA (1985) The basis of selection, management and evaluation of multipurpose trees - an overview. Agroforestry Systems 3:251-266 Huxley P, T Darnhofer, A Pinney, E Akunda and D Gatama (1989) The tree/crop interface: a project designed to generate experimental methodology. Agroforestry Abstracts 2(4):127-145 ILCA (International Livestock Centre for Africa) (1984) Producing Seed of Gliricidia sepium. Addis Ababa, Ethiopa Rang BT, GF Wilson and TL Lawson (1984) Alley Cropping: A Stable Alternative to Shifting Cultivation. IITA, Ibadan, Nigeria Rang BT and B Duguma (1985) Nitrogen management in alley cropping systems. In: BT Rang and J van der Heide, eds, Nitrogen Management in Farming Systems in Humid and Sub-humid Tropics. Institute for Soil Fertility, Netherlands Rang BT and R Mulongoy (1987) Gliricidia sepium as a source of green manure in an alley cropping system. In: Gliricidia sepium (Jacq.) Walp.: Management and Improvement. Special Publication 87—01 by NFTA and CATIE on the Proceedings of a Workshop, Turrialba, Costa Rica, 21-27 June Rang BT and GF Wilson (1987) The development of alley cropping as a promising agroforestry technology. In: HA Steppler and PRR Nair, eds, Agro-forestry: A Decade of Development, pp 227-243. ICRAF, Nairobi, Renya Rang BT, L Reynolds and AN Atta-Rrah (1990) Alley farming. Advances in Agronomy 43:315-359. Lal R (1989) Agroforestry system and soil surface management of a tropical alfisol I. Soil moisture and crop yields. Agroforestry Systems 8:1-6 Laquihon W (1988) Small farm uses of multipurpose tree species in Mindanao. Mindanao Baptist Rural Life Center (MBRLC), Davao del Sur, Philippines 104 Laquihon W, JJ Palmer and H Watson (1991) Sloping agricultural land technology: a decade of experience on hillside sustainability. A paper presented at the 'Asian Farming Systems Symposium - 1992' at BMICH-Colombo, Sri Lanka, 2-5 November Lasco R (1991) Herbage decomposition of some agroforestry species and their effects as mulch on soil properties and crop yield. A PhD dissertation, UPLB Graduate School, College, Laguna, Philippines Lu C, and T Hu (1981) Biomass production of two-year old spacing trial plantation of leucaena in Taiwan. Leucaena Research Reports 2:53- 54 Manguiat IJ, VM Padilla, DM Mendoza and AM Perez (1990) Rhizobia- Mycorrhiza inoculation and N—P fertilization of gliricidia in a degraded. upland area” Nitrogen Fixing Tree Research Reports 8:140-144 MBRLC Editorial Staff (1988) Modern farming guide: a manual on how to farm your hilly land without losing your soil. ’ Mindanao Baptist Rural Life Center, Rinuskusan, Davao del Sur, Philippines Melillo JM, JD Aber and JF Muratone (1982) Nitrogen and lignin control of hardwood lead decomposition dynamics. Ecology 63:621-626 Mulongoy R and :0: van der Meersch (1988) Nitrogen contribution by leucaena prunings to maize in an alley cropping systems. Biol. Fertil. Soils 6:282-285 Ong CR (1989) Cropping Systems Approach to Agroforestry Research. ICRISAT, India Ong CR, MR Rao and M Mathuva (1992) Trees and crops: competition for resources above and below ground. Agroforestry Today 4(2):4-5 Osman AM (1981) The effects of cutting interval on the relative dry matter production of four cultivars of leucaena. Leucaena Research Reports 2:33-34 Palm CA and PA Sanchez (1991) Nitrogen release from the leaves of some tropical legumes as affected by their lignin and polyphenolic contents. Soil Biol. and Biochem 23:83-88 Salazar A and CA Palm (1987) Screening of leguminous trees for alley cropping on acid soils of the humid tropics. In: Gliricidia sepium (Jacq.) Walp.: Management and Improvement. Special Publication 87-01 by NFTA and CATIE on the Proceedings of a Workshop, Turrialba, Costa Rica, 21-27 June Salazar A, LT Szott and CA Palm (1993) Crop-tree interactions in alley cropping systems on alluvial soils of the Upper Amazon Basin. Agroforestry Systems 22:67-82 Solera CR (1992) Competition between upland rice and hedgerow species in an alley cropping system. A PhD dissertation, UPLB Graduate School, College, Laguna, Philippines Sumberg JE (1986) Alley farming with Gliricidia sepium: germplasm evaluation and planting density trial. Tropical Agriculture (Trinidad) 63(2):170-172 105 Tabinga GA and A0 Gagni (1985) Corn production in the Philippines. Department of Development Communication, UPLB, College, Laguna, Philippines Tacio, HD (1993) Sloping agricultural land technology (SALT): a sustainable agroforestry scheme for the uplands . Agroforestry Systems 22:145-152 Watson HR and W Laquihon (Undated) A manual on how to farm your hilly land without losing your soil. Mindanao Baptist Rural Life Center, Davao del Sur, Philippines Watson HR (1983) Developing a hillside farming techniques for Mindanao hill farmers. In: Proceedings of the Hilly Land Development Workshop, Cebu City, Philippines, 24-26 March Weeraratna CS (1979) Pattern of nitrogen release during decomposition of some green manures in a tropical alluvial soil. Plant and Soil 53:288-294 Wilson GF, BT Rang and R Mulongoy (1986) Alley cropping: trees as sources of green manure and mulch in the tropics. Biological Agriculture and Horticulture 3:251-267 Yamoah, CF, AA Agboola and R Mulongoy (1986) Decomposition, nitrogen release and weed control by prunings of selected alley cropping shrubs. Agroforestry Systems 4:239-246 Young A (1989) Agroforestry for Soil Conservation. Wallingford, UR:CABI International and ICRAF Chapter 4 ROOT GROWTH AND SOIL FERTILITY Abstract The root pattern and distribution of Gliricida sepium hedgerows at different number of hedges per contour line and combinations of pruning height and within-row spacing were evaluated in a 0.22-ha experimental site on Mt. Makiling, University of the Philippines at Los Banos, College, Laguna, Philippines. Periodic soil samples were analyzed to determine changes in soil pH, organic matter, N, P, and R over time. Maize (Zea mays) and.mungbean (Vigna radiata) were planted in rotation as alley crops during two cropping seasons. Root sampling with an auger and the trench profile technique were employed to determine root densities from the hedgerow base up to 50 cm towards the alley and at a fixed distance of 50 cm from the hedgerow base. Up to 90 percent of all roots were < l-mm diameter and more than 70 percent were located in the top 30 cm of the soil. The highest mean root densities (number of roots/dmfi) came from the single and double hedgerows at 5 and 10 cm within-row spacings. The root densities of the double hedgerows at 5 and 10 cm within-row spacings linearly decreased with soil depth and distance from the hedgerow base towards the alley. Root densities (mg/dm’) from auger sampling were not significantly affected by number of hedges per contour line, within-row spacing, and pruning height. The means of root densities in two sampling periods with auger, however, decreased*with.distance from the hedgerow>base towards the alley. Based on the rooting pattern, root densities, and distribution of G. sepium hedgerows, the alley crop would be partly deprived of nutrients and moisture as a result of intra- and interspecies competition. 106 107 Although the result of soil analysis was not subjected to statistical analysis, the average percent N, P, R, and pH declined after site clearing and the first maize crop. These elements, however, gradually increased after the hedgerows were established and after the incorporation of initial hedgerow clippings and subsequent prunings. Organic matter declined, then started to stabilize, after biomass application in the alley. 108 4.1 Introduction Early advocates of alley cropping tended to stress the importance and potential of the above-ground biomass production of hedges, yields of alley crops, control of soil erosion, improvement of soil properties, integration with livestock production, and above-ground hedgerow/crop manipulation (Rang, Wilson, and Lawson, 1984; Rang and Wilson, 1987; Wilson, Rang, and Mulongoy, 1986; Sumberg and Atta-krah, 1988; Watson and Laquihon, undated; Watson, 1983; Laquihon, 1988; Laquihon et al., 1991; and MBRLC Editorial Staff, 1988). In recent years, however, there has been an increasing realization that for alley cropping to be confidently promoted by technicians and accepted by farmers as a technology, there is a need to further understand below-ground conditions and determine how various growth processes function (Buck, 1986; Lal, 1989; Young, 1991). Specific questions on root competition, complementation, distribution, growth, and turnover require answers based on empirical data. There have been efforts to answer various hyphotheses on plant-soil processes in agroforestry; thus, some questions on erosion control, organic matter, soil physical properties, nitrogen fixation, nutrient cycling, soil toxicities, and soil water can now be answered based on direct or indirect evidence (Young, 1991). On the other hand, empirical studies that could answer specific questions on root growth, development, and competition have not been done or are lacking. There is little direct evidence on how tree roots in an agroforestry system grow, exploit soil resources, compete, and complement agricultural crops (Buck, 1986; Young, 1991; Solera, 1992). Conclusions and recommendations, which were drawn from a few studies on root system, such as those of Johnson et al. (1988), Torres (1983), Rang et al. (1981), Dhyani, Narain, and Singh (1990), Gillespie (1989), and Ong, Rao, and Mathuva (1992), appear to conflict with each other. Hence, a consensus to start focusing research on the 109 below-ground environment of alley cropping is emerging among scientists, researchers, and extension workers. There is now a greater awareness of the role of roots in various agroforestry systems. Is the focus on the below-ground environment of alley cropping and other agroforestry systems justified? Will further understanding of the rhizosphere help in designing and evaluating recommended cultural management practices for alley cropping? Will the focus give light on selecting hedgerow and crop ideotypes? Dickmann and Pregitzer (1992), in their review of the structure and dynamics of woody plant root systems, argued that understanding the morphology, ecology, and physiology of the aerial parts of the tree should go hand in hand with comprehension and knowledge of the root system. This reasoning, when applied to research in alley cropping, is highly commendable because the technology targets resource-limited upland farmers in the tropics. Upland farmers could not afford to invest their time, labor, and money in establishing hedgerows only to realize in later times that this vegetative structure will not be sustainable. Only a better understanding of the root system may nullify or confirm farmers' general apprehension that trees in association with crops will compete strongly with crops for nutrients and moisture (Dhyani, Narain, and Singh, 1990). Further understanding of hedgerow roots may direct future studies on the ”root silviculture” (Dickmann and Pregitzer, 1992) of alley cropping systems. A root silviculture that will minimize carbon investment on the root system so that more could be allocated to pruning biomass accumulation may be desirable. A clearer comprehension of roots will help in modifying or innovating above-ground cultural management practices so that hedges will reduce their subsidy for root production and increase investments on shoot growth (Caldwell, 1987). An enriched knowledge on 110 root systems could guide the screening and evaluation of potential hedgerow species that would meet the requirements of "hedgerow ideotypes" and pinpoint areas of complementarities and commensalism between hedgerows and crops (Gillespie, 1989; Dickmann, 1992; Young, 1991; van Noorwijk et al, 1988). This study attempted to understand the behavior and growth pattern of the roots of Gliricidia sepium; it is a response to the urgent need to examine roots of hedgerows in an alley cropping system. The research hypothesis is that root distribution from the hedgerow base towards the alleys at various depths decreases regardless of treatments applied. It is postulated that root densities and root distribution could provide an explanation of how the hedgerows react to various treatment combinations of pruning height, within-row spacing, and number of hedges per contour line. The study was based on the notion that the growth of hedgerow roots would affect the production of pruning biomass, crop yields, recycling of leached nutrients from the subsoil, and improvement of soil fertility. Intraspecific competition among hedgerow plants, as may be inferred from their pattern of root distribution, may constrain biomass production and indirectly favor or dampen the growth of alley crops. The research study proceeded with the understanding that roots are the major organs for nutrient absorption and movement of substances that are essential for plant growth. Lastly, the study hopes to contribute to the scarce, but increasing efforts on hedgerow root system. It provided an experience in examining the below-ground environment. Traditional methods were used, despite the fact that by themselves, they are considered inadequate, destructive, labor-intensive, and exacting (Boehm, 1979; Schuurman and Goedewaagen, 111 1965). The study did not have the luxury of using new and sophisticated methods of observing roots such as the minirhizotrons which require in- situ installation (Hendrick, 1992; Upchurch and Ritchie, 1983). It adopted Smucker's (1984) suggestion that the direct method of extracting root and soil samples, although laborious, could still provide excellent information if the roots are quantitatively separated from the soil. Thus, in the study, two methods were used in sampling roots. These are the sampling of roots by auger and the trench profile wall method. The latter was performed towards the end of the experiment and.was employed to confirm and check the result of the root sampling by auger. 4.2. Materials and Methods 4.2.1. Root Sampling with the Use of Auger 4.2.1.1 Design of the Anger and Sampling Procedures Figure 4.1 shows the design of the auger that was locally fabricated and used in sampling roots at 30-cm soil depth and at different distances from the hedgerows towards the alley. The design was adopted from the auger described by Schuurman and Goedewaagen (1965) and cited by Boehm (1979). The auger was made of stainless steel with inside diameter of 5.87 cm and length of 50 cm. Thus, at a sampling depth of 30 cm, the auger could bore soil with a total volume of 0.81 dmfi A week before sampling, the hedgerows and the alleys were weeded. The weeding was done to minimize the inclusion of roots of other species or weeds during the sampling. Sampling was performed at 0-, 25-, and 50-cm distances from the hedgerow base towards the alley. All borings in the lower and upper portions of the hedgerows were located in the mid-section 112 of each sub-plot. For each sub-plot, a total of six borings were done, three from each of the lower and upper hedgerows. Five boreholes is more than the five borings which are recommended by many researchers (Boehm, 1979). For the double hedges per contour line, sampling started at base of the hedgerow facing the alley. A two-person team was organized, taught, and supervised to do root sampling with the auger. The first person held the auger, while the second person hammered the auger into the ground with a mallet made of a heavy piece of wood. The auger was first driven into the ground at 15 cm, pulled out, and the soil with the roots inside the core was taken out. In the same hole, the auger was again driven down to 30 cm, pulled out, and the soil with the roots inside the core was extracted. Pre-sampling was conducted to determine the ease of driving the corer into the ground, extracting the soil from the core, ascertaining and familiarizing personnel with the color of the G. sepium roots, and washing the soil to separate the roots. Pre-sampling gave a rough estimate and benchmark for distribution of roots at various soil depths, i.e. 0-15 cm and 16-30 cm. After the pre-sampling, it was found that there were not that many roots in the soils at the depth of 16-30 cm. Thus, soils from the cores driven in the same spot at different soil depths (0-15 cm and l6-30 cm) were combined, and placed in coded plastic bags for washing. The samples were put in a jute sack for transport into the washing area. Two root samplings with auger were conducted. The first sampling was completed on January 10, 1992 after the second corn harvest and initial clipping of hedgerows. It was the end of the rainy season. Almost six months after, at the onset of the rainy season, on June 9, 1992, the second root sampling was started. This was after the harvest of the second mungbean crop and completion of the two subsequent prunings. The sequence of activities in both samplings were as follows: Root sampling within a selected.main.plot (12 sub-plots, 60 boreholes) during the first 113 I J 1.9 cm diueter and 15 cm long red welded in place 6.35 as outside dimter with ' / 0.48 cm sell steel tube I 0-3 em bevel to sin n ___} rp «In. Figure 4.1. Design of the auger for sampling roots at different distances from the hedgerow base and to a soil depth of 30 cm. 114 day, followed by washing of roots in the second day, and.weighing of roots in the third day. Thus, in each sampling period, the six main plots were completed in two weeks with an average of three main plots per week. 4.2.1.2 Processing the Root Samples In the wash area, each sample was placed on top of a fine-meshed aluminum screen and washed slowly with water from a faucet to separate the roots from the soil. Live roots were picked with a hair puller during the washing period. Visually, the roots of G. sepium were easy to determine. The color of the fine roots is light to almost dirty white while the larger diameter roots were light brownish in color. Efforts were exerted to isolate dead roots and roots of other species. After the roots were separated, they were allowed to drain dry of water, placed inside coded plastic bags, and weighed to the nearest mg at the University of the Philippines at Los Banos (UPLB) College of Engineering the following day. Three representative samples were randomly taken from each main plot (from each batch of root samples) for oven dry determination at 100 °C. The average moisture contents of the samples for each main plot (from each batch) were computed and used in determining the dry weights of root samples taken from the same main plot. The root dry weights were computed and translated into mg/dm’. The formula used for determining the volume of soil extracted by the auger (V) in dm3 was the following: V (.25) (3.14) (inside diameter of the auger in dm)2 (depth of sampling in dm). 115 With the auger's inside diameter of 0.587 dm and depth of sampling of 3 dm, the volume of soil taken by the corer was 0.81 de. Root density (dry mg/dm’) from sampling with auger was calculated by dividing the dry weight of roots per sample over the volume of the soil from the corer (0.81 awn. 4.2.2. Root Sampling with the Profile Wall Method 4.2.2.1 Design of the Trench Profile Sampling Figure 4.2 shows the rectangular counting frame that was used in the trenching technique. The frame was made of a chicken wire with an original mesh size of 2.5 cm. The rectangular frame had a width (depth) of 60.9 cm) and a length of 96.5 cm, with two sizes of grids within it. The design is an adaptation from what is described by Schuurman and Goedewaagen (1965) and Bohm (1979). From the ground surface down to a depth of 30.5 cm, the square grids used had a dimension of 5.0 cm x 5.0 cm). Then, from 30.5 cm down to 60.9 cm depth, larger grids with dimension of 5.0 cm x 10 cm‘were used because there were fewer roots found inside the grids. Thus, from the ground surface down to a depth of 60.9 cm, roots were counted in a total of nine grids per column. For the root profiling work, L-shaped trenches were dug. The relative locations of these trenches are listed in Table 4.1. Four trenches were dug on the lower portion of the alley and four on the upper portion. This precaution was taken to account for the possible influence of slope on the growth direction of roots because of a perceived soil fertility gradient from the upper and lower portions of the alleys. Along the contour hedgerows, each trench had an average length of 100 cm; but, the trench perpendicular to the contour hedgerows had a length of 50-60 cm. Both trenches formed an L-shape and had an average depth of 75-100 cm. A total 116 of eight trenches were dug, four trenches per main plot, each representing a within-row spacing of S-, 10-, 20-, and 40-cm. Table 4.1. Relative locations of trenches in the alleys. Treatmentl Location of the trench in the alley Double hedgerow -WRS of 5 cm Lower portion Double hedgerow -WRS of 10 cm Upper portion Double hedgerow -WRS of 20 cm Upper portion Double hedgerow -WRS of 40 cm Lower portion Single hedgerow -WRS of 5 cm Upper portion Single hedgerow -WRS of 10 cm Lower portion. Single hedgerow -WRS of 20 cm Lower portion Sin le hed erow Upper portion ' WRS s Within-row spacing For the trench profiling at a fixed distance of 50 cm from the hedgerow base, roots were counted in a total of 19 columns (total length of 96.5 cm) or a total of 171 grids per trench. In this case, the wall of the trench which paralleled the hedgerow was used in counting and profiling the roots. For the root profiling from the hedgerow base up to 50-cm distance towards the alley, roots were counted in a total of 90 grids per trench. The wall of the trench perpendicular to the hedgerow was used in the counting of roots. Figure 4.2. The rectangular counting frame that was used in counting roots within a profile. The frame had a depth of 60.9 cm and a length of 96.5 cm. 118 4.2.2.2 Preparing the Profile Wall After the L-shaped trenches were dug with bar and shovel, each working face of a profile was roughly prepared with spade. Smoothing the profile wall was done with a bolo which had a round and sharpened tip. After smoothing, the roots were exposed by lightly spraying the wall with water and slowly scraping the profile with a fork before counting the roots. The same procedure was followed for all the trenches that were dug. 4.2.2.3 Counting the Roots The rectangular frame*was placed against the profile in counting the roots per grid. Five root diameter classes were used: < 1 mm, 1-2 mm, 2-5 mm, 5-10 m, and > 10 am. To obtain visual familiarity of the various diameter classes, especially'during the initial root counts, bamboo sticks were prepared and calibrated for each class. Only the roots protruding within a grid were counted to minimize double counting of hanging roots. For each root diameter class, the root counts per grid were all converted into number of roots per dmfi. This was the root density figures that were used in the statistical analysis. 4 .2 .3 Soil Sampling Both the procedures of PCARRD (1985) and Shickluna (1983) were used as guides in taking composite soil samples from the experimental site. Three composite soil samples were collected from the whole area, one from each replicate, e.g. one from top, middle, and bottom sections of the experimental site. Each sample was a composite of soils from 10 holes in the alleys. A spade was used in digging 30-cm deep holes in the ground. From each hole, approximately 4-cm slice of soil from one side was taken and placed inside a plastic pail. The holes were approximately eight m 119 distance from each other and sampling followed a zigzag pattern (lower and upper portions of the alley). Each composite soil sample was thoroughly mixed and pulverized by hand before air drying. After air drying, four one—kg sub-samples were obtained from each composite sample for analysis and determination of pH, percent organic matter (OM), percent total N, P in ppm, and R in me/lOOg soil. The sub-samples were placed in a properly labelled plastic bags and sent to the UPLB Soils Laboratory, which uses a modified Rjeldahl method for N, molybdovanadate method for P, and flame photometer method for R. Soil. compositing was done four times during the duration of the experiment. The first sampling was conducted before planting the hedges and just after the site preparation (early September 1990). This sampling established more or less the baseline soil condition of the site before the start of the research. The second was performed in early January 1991, after the harvest of the first corn crop and before the planting of the first mungbean crop. The third sampling was done in early September 1991, imediately after the initial clipping of hedgerows and before planting the second corn crop. The last sampling was accomplished in early January 1992, after the harvest of second corn crop and before the planting of the second mungbean crop. Over the four sampling periods, similar procedures were adopted in taking individual soil samples from each of 10 holes and compositing the soils for further sub-sampling and soil analysis. 4.2.4 Statistical Analysis All the data from the root sampling with the auger and trench profiling were entered into the»Lotus 1-2-3 spreadsheet, converted into ASCII files, 120 and analyzed using MSTATC and/or SYSTAT software. Data from the sampling of roots with the auger were subjected to analysis of variance (ANOVA). To determine the relationship between root density (mg/de) and distances from the hedgerow base, the means of root density at different distances from the hedgerow were subjected to simple regression and correlation analysis. Regression and correlation ‘were also used for the root counts/dm2 at different distances from the hedgerow base towards the alley and at various soil depths for the data that were gathered in the trench profiling technique. The results of the soil sampling were tabulated and analyzed to determine the pattern and changes of each element over time. 4.3. Results 4.3.1 Root Densities from Sampling with the Auger In both the first and second root samplings with the auger, none of the treatments was significant with F-test (p < 0.05) (Tables 4.2 and 4.3). Average root densities from lower and upper hedgerow sampling positions and from all distances from the hedgerow base were not significantly affected by the number of hedges per contour line, combinations of within- row spacing and pruning heights, and interactions. The treatments did not affect root densities and their distribution from the hedgerow base up to 50 cm towards the alleys at a sampling depth of 30 cm. The means of root densities exhibited a decreasing trend from the base of the hedgerow towards the SO-cm distance for the lower and upper sampling positions (Table 4.4). The grand means of the second sampling were slightly lower compared with those from the first root sampling; however, the trend is more or less similar (Table 4.5). Table 4.2. (ms/dm’) 121 Analysis of variance for the average dry root densities from the lower and upper sampling positions from the first sampling with the auger. Sampling Position and Source of F-Value2 Distance from the Hedgerow Variation‘ Base 1. Averaged at 0-cm No of hedges/line 0.002 distance from the has; PH X WRS 0.761 Interaction 0.634 2. Averaged at 25-cm No of hedges/line 0.400 distance from the base PH X "RS 0.908 Interaction 0.644 3. Averaged at 50-cm No of hedges/line 1.297 distance from the bag; PH X WRS 0.557 Interaction 1.090 l-I-I-II Iri PH - Pruning height; WRS = Withinerow spacing None of the treatments was probability. significant at 10 percent 122 Table 4.3. Analysis of variance for the average dry root densities (mg/de) from the lower and upper sampling positions from the second root sampling with the auger. ,, ~ ______,__ — _,,, _, E l Sampling Position and Source of F-Value2 Distance from the Hedgerow Variation' Base ‘ 1. Averaged at 0-cm No. of hedges/line 0.187 distance from the base PH x WRS 0.517 Interaction 0.758 2. Averaged at 25-cm No. of hedges/line 0.045 distance from the base PH x WRS 0.912 Interaction 1.225 3. Averaged at 50-cm No. of hedges/line 0.234 distance from the base PH x WRS 0.651 Interaction 1.039 m: 3 = ' PH 8 Pruning height; WRS a Within-row spacing None of the treatments was significant at 10 percent probability. Table 4.4. Grand means (mg/dm’) and coefficient of variations of dry root densities from the first sampling with the auger. Sampling Position and Distance from Grand Means Coefficient the Hedgerow Base of Variation in S 1. Lower hedgerow at a 0-cm 0.71 182.5 distance from the base 2. Lower hedgerow at a 25-cm 0.52 255.8 distance from the base 3. Lower hedgerow at a 50-cm 0.46 296.9 distance from the base 4. Upper hedgerow at a 0-cm 0.63 231.7 distance from the base 5. Upper hedgerow at a 25-cm 0.58 281.6 distance from the base 6. Upper hedgerow at a 50-cm 0.28 308.5 distance from the base d 123 Table 4.5. Grand means (mg/dm’) and coefficient of variations of dry root densities from the second sampling with the auger. Sampling Position and Distance from Grand Means Coefficient of the Red erow Base Variation in % 1. Lower hedgerow at a 0-cm 0.58 223.4 distance from the base 2. Lower hedgerow at a 25-cm ‘ 0.41 395.8 distance from the base 3. Lower hedgerow at a 50-cm 0.15 420.6 distance from the base 4. Upper hedgerow at a 0-cm 0.35 250.4 distance from the base 5. Upper hedgerow at a 25-cm 0.53 339.3 distance from the base 6. Upper hedgerow at a 50-cm 0.17 465.2 E distance from the base Figure 4.3. shows the relationship between the grand means of root densities (across treatments, replications, and sampling positions) and distances from the hedgerow base towards the alley. The grand means of root densities from the first and second root samplings were used in running the regression and correlation analysis. The result of the t-test (p < 0.01), indicates that root density decreases from the hedgerow base up to 50 cm towards the alley within the 30-cm soil depth (r = - 0.73). Regression and correlation analysis was also used in determining the possible relationship between root density and within-row spacing. However, the r values for the first and second samplings were - 0.02 and - 0.13, respectively. Clearly, root densities did not show linear relationship with within-row spacings of hedgerows. The CVs for all the sampling positions and distances from the hedgerow base were extremely high (Tables 4.4 and 4.5). The average CV for the first root sampling with auger was 259.5 percent, slightly lower than the CV from the second sampling, which was 349.1 percent. 124 0.8 Y r 0.602 - 0.607 X —0.73 (p < 0.01) (I) .2 .3; C Q’A or) *gé a}, “53 (I) C 4 8 2 0.2“ ‘ .. A 0.1 A : 0'0 I I ' I I I ' I I I 0.0 0.1 0.2 0.3 0.4 0.5 Distance from the Hedgerow Bose (m) Figure 4.3. Relationship between the means of root densities (across treatments, replications, two sampling periods, and sampling positions) and distances from the hedgerow base towards the alley at 0-30 cm soil depth. The Gliricidia sepium hedgerows were between 16-month to 22-month- old when the samplings with auger were conducted. 125 4.3.2 Root Densities From the Trench Profile Method 4.3.2.1 Root Densities at a Fixed Distance of 50-cm from the Hedgerow Base At all soil depths at a fixed distance of 50 cm from the hedgerow base, the dominant root diameter class was < 1.0 mm, with an average of 89.9 percent (Table 4.6). The total aggregate percentage of the other root diameter classes was only 10.1 percent, with 1-2 mm class capturing 7.1 percent of the total number. Roots 2-5 mm diameter tended to occur at depths lower than 25 cm. Only a few roots fell under the 5 to 10 mm and > 10.0 mm diameter classes. Overall, at the depths of 15 cm and 30 cm, where most crops grow their roots, the corresponding cumulative percentages of hedgerow roots from all diameter were 48 percent and 83 percent, respectively (Figure 4.4). The rest of the roots occurred between the soil depth of 30-60 cm. Table 4.6. Percent distribution of root diameter classes at different soil depths and at a fixed distance of 50 cm from the hedgerow base. Percent of Root Diameter Classes I t of Roots —=—=—=-=_ based on < 1 mm 1-2 mm 2-5 mm 5-10 mm > 10 mm total count 90.6 8.3 0.9 0.2 0.0 16.2 89.8 8.3 1.9 0.0 0.0 15.2 92.8 5.0 1.9 0.3 0.0 16.7 90.4 6.2 2.7 0.5 0.2 14.6 86.7 7.9 4.2 0.9 0.3 11.9 88.8 7.1 3.7 0.0 0.4 8.7 91.6 5.7 2.4 0.3 0.0 6.7 89.2 6.8 3.7 0.3 0.0 5.4 89.3 8.2 1.2 0.9 0.4 4.6 89.9 7.1 = 2.5 0.4 = 0.1 100.0 126 1007 E om 80-: mm H0 a oCD ; 539 60— 0%» m"C3 .. w-m OI Na; 40- .C gut! 5% 3" 20- D o . O T I ' I ' l ' l ' I ' I 0 10 20 30 40 50 60 Soil Depth (cm) Figure 4.4. Cumulative percent of all roots at various soil depths. 127 In the regression and correlation analyses between density of roots belonging to the < 1.0-mm diameter class (number of roots/dmz) and soil depths (Table 4.7), both the trenches from the single and double hedgerows yielded significant t-test results for the 5- and lO-cm within-row spacing p1< 0.05 and p < 0.01, respectively. In the single hedgerows, the coefficient of correlations (r values) were -0.73 and -0.76 for the 5-cm and lO-cm within-row spacings, respectively. In the double hedgerow, the r values were much higher (-0.96 and -0.82 for the S-cm and 10-cm within- row spacing, respectively). The t-tests for the average number of roots with < 1.0-nun diameter/dm2 in the single and double hedgerows gave significant results (p < 0.01). The significant results of the t-tests Table 4.7. Results of regression and correlation analyses' between the density of roots with < 1.0-mm diameter and soil depths at a fixed distance of 50-cm from the hedgerow base. Dependent Mean r t-test Variable value2 Double hedgerow-WRS of 5 cm 14.45 -0.96 8.70*** Double hedgerow- WRS of 10 cm 10.67 -0.82 3.83*** Double hedgerow- WRS of 20 cm 3.54 -0.44 1.28 Double hedgerow- WRS of 40 cm 0.77 0.22 0.59 Double hedgerow-Average across WRS 5.67 -0.94 7.58*** Single hedgerow- WRS of 5 cm 6.12 -0.73 2.83** _Single hedgerow- WRS of 10 cm 10.18 -0.76 3.08** Single hedgerow- WRS of 20 cm 10.21 -0.65 2.28 Single hedgerow- WRS of 40 cm 7.33 -0.64 2.22 Single hedgerow-Average across WRS 8.46 -0.85 4.25*** ' WRS = Within-row spacing; r = Coefficient of correlation 2 *** Significant at one percent probability; ** Significant at five percent probability 128 for the dependent variables suggest that at a fixed distance of 50-cm from the hedgerow'base, soil depths may be a relevant and accurate predictor of density of roots belonging to the < 1.0-mm diameter class. The regression and correlation analysis between density of roots belonging to 1-2-mm diameter class and soil depths (Table 4.8) yielded significant results for the double hedgerow and within-row spacings of 5 cm (p < 0.01) and 10 cm (p < 0.05) with r values of -0.82 and -0.62, respectively. In the single hedgerow, only the trench with within-row spacing of 20 cm yielded a significant t-test (p < 0.01), with a r - -0.83. The results of the regression and correlation analyses between root density of roots from all diameter classes/dm2 and soil depths (Table 4.9) yielded significant t-tests on the double hedgerow with 5—cm and 10-cm within-row spacings (p < 0.01) and on the single hedgerow with S-cm, 10- cm, and 20-cm‘within-row spacings (pI< 0.05). In the double hedgerow, the r values were -0.96 and -0.82 for the 5-cm and lO-cm within-row spacing, respectively. The r values of the 5-cm, lO-cm, and 20-cm within-row spacings in the single hedgerow were 60.77, -0.67, and -0.69, respectively. The t-tests on the average of total roots in the double and single hedgerows yielded significant results (p < 0.01). The linear relationship between root density and soil depth at a distance of 50 cm from the hedgerow base was strongest in the 5- and 10-cm within- row spacings, especially for roots belonging to < 1.0 diameter class. Root density declined with increasing soil depth. Means of root densities were higher in more dense within-row spacing than in wider within-row spacing (20- and 40-cm) as shown in Figures 4.5 and 4.6. Comparing Tables 4.6, 4.7 and 4.9, it can be deduced that there were fewer roots belonging to larger diameter classes in the 5- and 10-cm within-row spacing than in the 20- and 40-cm spacings. 129 Table 4.8. Results of regression and correlation analyses' between density of roots with 1-2 mmIdiameter and soil depths at a fixed distance of 50 cm from the hedgerow base. Dependent Variable Mean r t-test value2 Double hedgerow-WRS of S cm 0.57 -0.82 3.76*** Double hedgerow- WRS of 10 cm 1.19 -0.62 2.07** Double hedgerow- WRS of 20 cm 0.17 -0.32 0.88 Double hedgerow- WRS of 40 cm 0.15 0.08 0.21 Double hedgerow-Average across WRS 0.52 -0.77 3.14** L§$291° hedgerow- WRS of 5 cm 0.05 -0.17 0.45 Single hedgerow- WRS of 10 cm 0.61 -0.29 0.79 Single hedgerow- WRS of 20 cm 0.89 -0.83 3.86*** Single hedgerow- WRS of 40 cm 0.77 -0.42 1.24 Single hedgerow-Average across WRS 0.58 -0.61 2.05 ' WRS I Within-row spacing; r 8 Coefficient of correlation 2 *** Significant at one percent probability; ** Significant at five percent probability 130 Table 4.9. Results of regression and correlation analyses' between the total number of roots in all diameter classes and soil depths at a fixed distance of 50 cm from the hedgerow base. nu- Dependent Variable Mean r t-test value2 Double hedgerow-WRS of 5 cm 8.46 -0.96 9.33*** Double hedgerow- WRS of 10 cm 12.28 -0.82 3.72*** Double hedgerow- WRS of 20 cm 3.82 -0.43 1.24 Double hedgerow- WRS of 40 cm 1.24 0.07 0.19 Double hedgerow-Average across 6.45 -0.94 7.41*** WRS _Single hedgerow- WRS of 5 cm 6.36 -0.77 3.23** Single hedgerow- WRS of 10 cm 11.16 -0.67 2.35** Single hedgerow- WRS of 20 cm 11.22 -0.69 2.48** Single hedgerow- WRS of 40 cm 8.158 -0.63 2.13 Single hedgerow-Average 9.23 -0.85 4.18*** across WRS _= ' WRS8 Within-row spacing; r = Coefficient of correlation 2 *** Significant at one percent probability; ** Significant at five percent probability 131 I'WMS=5cm1 Ewes-10cm . Ewes-20 cm --~tf EgWRS-40cm 101£Z-;f 2) mieere‘eeeeu‘ee‘I-esesseseeeeeoeeeee'se-lenoeeeeee 81.2 eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 61.2 \\\\\\\\\\\\\mm + ‘ 41.2 Root Densities from Single Hedgerows (No of roots/m 21.2 0-15 Soil Depth (cm) Figure 4.5. Means of root densities in single hedgerows at different within-row spacing. 132 . .VVRS-Scm ‘ Ewes-10cm . . Ewes-20 cm Ewes-40cm ..................................................................................... ...................................................................................... ...................................................................................... o . eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 16-30 . bee-mun..-eueneeenflhraanee-melube-emthae-hbhebLLuen ......................................................... ......................................................... ........................................................ eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee //////////// [m Ane\muoou no ozv msouoooom cannon sown moweemcoo Doom Soil Depth (cm) Means of root densities in double hedgerows at different within-row spacing. Figure 4.6. 133 4.3.2.2 Root Densities from the Hedgerow Base Towards the Alley More than 40 percent of all root counts were found within the depth of 0- 15 cm (Table 4.10). At least seventy percent of roots counted were located within the depth range of 0- 30 cm starting from the hedgerow base up to 50 cm towards the alley. Regression and correlation analyses between the percent distribution of all roots for all the three soil depths with the distance from the hedgerow base up to 50 cm towards the alley did not give significant t-tests (p < 0.05). The r values were 0.13, -0.61, and 0.38 for the percent distribution of roots at soil depths of 0-15, 16-30, and 31-60 cm, respectively. Based on the r values and the t-tests, there is sufficient evidence ‘to conclude that the percent distribution of all roots from the hedgerow base up to 50 cm towards the alley was not directly related in all the three soil depths because the percent distribution of roots at all depths towards the alley appeared to be uniform (Table 4.10). Roots belonging to the < 1.0-mm diameter class dominated in the 0-15, 16- 30, and 31-60 cm soil depths (Table 4.11). These fine roots comprised 88 percent of the total roots while, the rest belonged to the other diameter classes. A closer analysis of the roots belonging to < 1.0-mm diameter class revealed that the relationship between root density and the distance from the hedgerow base towards the alley was more consistent in the double hedgerows than in the single hedgerows (Tables 4.12 and 4.13). In the double hedgerow, there were strong relationships between root density and distance from the hedgerow’base for the‘within-row spacings of 5 cm (rs -0.71; p < 0.05) and 40 cm (r I -0.67; p < 0.05) at soil depth of 0-15 cm. In fact, for the within-row spacing of 5 cm, root densities at depths of 16-30 cm and 31-60 cm were highly correlated with the distance from the hedgerow base towards the alley (r values of -0.90 and -0.83, 134 Table 4.10. Percent distribution of roots from all diameter classes at different soil depths and from the hedgerow base up to 50-cm towards the alley. Distance from Percent Percent i the hedgerowr distribution distribution at distribution at . base towards at soil depth soil depth of soil depth of the alley in cm. of 0-15 cm 16-30 cm 31-60 cm 0-5 47 31 22 6-10 45 31 24 11-15 42 33 25 16-20 39 34 27 21-25 39 26 36 26-30 33 33 34 31-35 43 23 34 36-40 46 25 29 41-45 46 30 24 46-50 47 24 V 29 135 Table 4.11. Percent distribution of root diameter classes at three soil depths from eight trench profiles. Root Percent Percent Percent Average percent diameter distributio distributio distributio distribution class n at soil n at soil n at soil based on total depth of 0- depth of depth of root counts at 15 cm 16-30 cm 31-60 cm all depths I < 1.0 mm 89 89 87 88.2 1-2 mm 5 7 8 6.6 2-5 mm 4 3 4 3.6 5-10 mm 1 0 1 0.6 > 10 mm 1 1 1 1.0 respectively, and p < 0.01). Except for the within-row spacing of 40 cm and soil depth of 31-60 cm, root densities in the other within-row spacings at different soil depths did not yield significant t-test results. In ‘the single hedgerow' (Table 4.13), the ‘t-test for the coefficient of correlation between root densities at different within-row spacings and distance from the hedgerow'base towards the alley appeared to be unpredictable. There were no significant t-test results from the within-row spacings of 5 cm and 10 cm at depths of 0-15 and 16-30 cm. High correlations were only obtained from the*within-row spacings of 20 cm and 40 cm at soil depths of 0-15 and 16-30 cm (r values of -0.76, -0.89, - 0.88, and -0.77, respectively at p < 0.01). Surprisingly, even at the within-row spacing of 40 cm and soil depth of 31-60, root density was found to be correlated with distance from the hedgerow'base (r 8 -0.64 and p < 0.05). 136 Table 4.12. Results of regression and correlation analyses' between density of roots < 1.0-mm diameter and distance from the hedgerow base towards the alley in a double hedgerow and at three soil depths. Eu: Dependent Variable Mean r t-test valuez WRS of 5 cm - Soil depth of 0-15 cm 12.7 -0.71 2.83** WRS of 5 cm - Soil depth of 16-30 13.7 -0.90 5.73*** cm of 5 cm - Soil depth of 31-60 6.3 -0.83 4.23*** of 10 cm - Soil depth of 0-15 30.7 0.44 1.39 of 10 cm - Soil depth of 16-30 12.2 0.11 0.31 of 10 cm - Soil depth of 31-60 8.7 -0.38 1.15 of 20 cm - Soil depth of 0-15 14.2 -0.59 2.06 of 20 cm - Soil depth of 16-30 8.5 -0.67 2.58** of 20 cm - Soil depth of 31-60 4.5 -0.27 0.79 of 40 cm - Soil depth of 0-15 2.9 -0.66 2.51** of 40 cm - Soil depth of 16-30 1.7 -0.79 0.23 of 40 cm- Soil depth of 31-60 1.7 -0.72 2.96** I WRS 8 Within-row spacing; r 8 Coefficient of correlation *** Significant at one percent probability; ** Significant at five percent probability 137 Table 4.13. Results of regression and correlation analyses' between density of roots < 1.0-mm diameter and distance from the hedgerow base towards the alley in a single hedgerow and at three soil depths. ------== Dependent Variable Mean r t-test value2 WRS of 5 cm - Soil depth of 0-15 cm 11.1 -0.09 0.26 WRS of 5 cm - Soil depth of 16-30 cm 4.5 -0.29 0.84 WRS of 5 cm - Soil depth of 31-60 cm 2.5 -0.53 1.78 WRS of 10 cm - Soil depth of 0-15 cm 14.5 -0.32 0.955 WRS of 10 cm - Soil depth of 16-30 cm 5.7 0.11 0.31 WRS of 10 cm - Soil depth of 31-60 cm 1.9 0.71 2.83** WRS of 20 cm - Soil depth of 0-15 cm 9.8 -0.76 3.33*** WRS of 20 cm - Soil depth of 16-30 cm 15.1 -0.89 5.54*** WRS of 20 cm - Soil depth of 31-60 cm 6.7 0.45 1.41 WRS of 40 cm - Soil depth of 0-15 cm 11.5 -0.88 5.25*** WRS of 40 cm - Soil depth of 16-30 cm 13.4 -0.77 3.39*** “ WRS of 40 cm- Soil depth of 31-60 cm 10.1 -0.64 2.38** 2 WRS 8 Within-row spacing; r 8 *** Coefficient of correlation Significant at one percent probability; ** Significant at five percent probability 138 Since 88 percent of the total roots counted in the eight trenches belonged to < 1.0-mm diameter class, results of the simple regression and correlation analyses of the root densities of all diameter classes in both the single and double hedgerow (Tables 4.14 and 4.15) did not vary with those obtained in regressing densities of roots belonging to the < 1.0-mm diameter class. From the results, however, it can be summarized that root densities of all diameter classes were negatively correlated with the distance from the hedgerow base towards the alley in single hedgerows having within-row spacings of 20 cm and 40 cm. This was found to the soil depth of 30 cm. In the double hedgerows, negative correlational relationships existed. between root densities and distance from ‘the hedgerow base at within-row spacings of 5-, 20-, and 40-cm and a soil depth of 0-15 cm. Deeper in the soil, however, only a root density in the within-row spacing of 5 cm linearly declined from the hedgerow base towards the alley. Distance from the hedgerow base towards the alley appeared to be a reliable predictor of density of roots belonging to < 1e0-mm diameter class to a depth of 30 cm. The linearly declining root density with respect to soil depth, however, is strongest in the double hedgerows than in the single hedgerows of G. sepium. 139 Table} 4.14. Results of regression and correlation analyses' between density of roots in all diameter classes and distance from the hedgerow base towards the alley in a double hedgerow and at three soil depths. Dependent Variable Mean r t-test value2 WRS of 5 cm - Soil depth of 0-15 39.5 -0.66 2.48** cm WRS of 5 cm - Soil depth of 16-30 41.1 -0.90 5.731*** cm WRS of 5 cm - Soil depth of 31-60 20.3 -0.83 4.23*** cm WRS of 10 cm - Soil depth of 0-15 106.9 0.52 1.74 cm WRS of 10 cm - Soil depth of 16-30 41.9 0.08 0.22 cm WRS of 10 cm - Soil depth of 31-60 20.3 -0.50 1.64 cm WRS of 20 cm - Soil depth of 0-15 44.9 -0.59 2.04** cm WRS of 20 cm - Soil depth of 16-30 27.5 -0.65 2.42** cm WRS of 20 cm - Soil depth of 31-60 15.5 -0.29 0.86 cm WRS of 40 cm - Soil depth of 0-15 10.5 -0.70 2.75** cm WRS of 40 cm - Soil depth of 16-30 6.9 -0.29 0.86 cm I WRS of 40 cm- Soil depth of 31-60 6.2 -0.73 3.05** cm WRS 8 Within-row spacing; r 8 Coefficient of correlation *** ** Significant at one percent probability; Significant at five percent probability Table 44. 15. 140 Results of regression and correlation analyses' between density of roots in all diameter classes and distance from the hedgerow base towards the alley in a single hedgerow and at three soil depths. _ fl Dependent Variable Mean r t-test value2 WRS of 5 cm - Soil depth of 0-15 cm 36.8 -0.19 0.54 WRS of 5 cm - Soil depth of 16-30 20.2 -0.35 1.04 cm WRS of 5 cm - Soil depth of 31-60 8.7 -0.32 0.96 cm WRS of 10 cm - Soil depth of 0-15‘ 52.3 -0.42 1.32 " cm II WRS of 10 cm - Soil depth of 16-30 18.6 0.16 0.45 cm WRS of 10 cm - Soil depth of 31-60 6.9 0.62 2.22* cm WRS of 20 cm - Soil depth of 0-15 33.3 -0.83 4.14*** cm WRS of 20 cm - Soil depth of 16-30 50.4 -0.89 5.68*** cm WRS of 20 cm - Soil depth of 31-60 22.7 0.50 1.65 cm I I WRS of 40 cm - Soil depth of 0-15 39.5 -0.91 6.32*** cm WRS of 40 cm - Soil depth of 16-30 45.3 -0.63 2.31** cm 5 WRS of 40 cm- Soil depth of 31-60 18.4 -0.60 2.10 5 cm WRS 8 Within-row spacing; r 8 Coefficient of correlation *** Significant at one percent probability; ** Significant at five percent probability 141 4.3.3 Soil Sampling Over Time Although the results of soil analysis were not statistically analyzed, the average values over four sampling periods from September 1990 to January 1992 (Table 4.16) showed that pH declined after the first corn harvest which*was captured.during the second soil sampling period. However, after the harvest of the first mungo crop, four months of fallow with benefits from hedgerow litterfall, initial clipping, and subsequent prunings increased pH up to 6.2. OM steadily decreased at an average of 11-12 percent in the second and third soil sampling (from 3.6 to 2.8 percent). OM appeared to stabilize during the fourth sampling. Both P and R slightly decreased after the first corn crop, but these elements gradually increased after the four months fallow, initial clipping, and subsequent prunings. As expected, N declined from the first to the third sampling periods. It only picked up in the fourth sampling after the incorporation of the initial clippings and subsequent prunings into the soil. 142 Table 4.16. Average results of soil analysis' over four sampling periods from September 1990 to January 1992. Sampling Date pH OM in P in R in N in % 8 ppm me/lOOg Early September, 1990- 5.89 3.56 3.79 2.73 0.17 before first corn crop, after site preparation Early January, 1991 - after 5.66 3.13 3.69 2.59 0.16 first corn harvest, before first mungo crop Early September, 1991- 6.61 2.78 4.17 2.76 0.15 after first mungo harvest, four months fallow, and after initial clipping of hedgerow Early January, 1992- after 6.24 2.79 5.03 2.95 0.24 second corn harvest and before second mungo crop ' Values are means of three composite soil samples taken from the bottom, middle, and top portions of the experimental site. 143 4.4 Discussion The results of the study highlight basic concerns in sampling hedgerow roots, pruning biomass and nutrient management, intra- and inter-species competition, andIestablishment and treatment of hedgerows to enhance their delivery of benefits to the alley crops. In theory, higher root densities (mg/dm3 or number of roots/dm’) could be obtained from more dense within-row spacings and higher number of hedges per contour line because the degree of overlap and competition of roots among neighboring plants is more intense (Caldwell and Richards, 1983; Caldwell, 1987; Atkinson, 1976). Dense hedgerows would tend to deplete limited soil resources faster and producezmore branched roots resulting in higher root densities (Fitter, 1987). The results of the trench profile method for sampling hedgerow roots confirm the above theory. The highest root densities (number of rootsfimm) were obtained from the 5 and 10 cm within-row spacings in both the single and double hedgerows at a fixed distance of 50 cm from the hedgerow base. From the hedgerow base up to 50 cm towards the alley, the highest root densities also came from the 5 and 10 cm within-row spacings. At these within-row spacings, root densities were also negatively correlated with soil depth at a fixed distance of 50 cm from the hedgerowu Only the root densities in the double hedgerow 5 cm within-row spacing were highly correlated with distance from the hedgerow base towards the alley at all depths. On the other hand, root densities (mg/dm’) from the first and second samplings with an auger at various distances from the hedgerow towards the alley at a soil depth of 30 cm were not significantly affected by within- row spacing and the number of hedges per contour line. They appeared to be inconsistent with the theory and findings from the trench profile sampling of hedgerow roots. However, the grand means of root densities across treatments and sampling positions (lower and upper hedgerows) 144 linearly declined with the distance from the hedgerow base at a depth of 30 cm (r 8 -0.73; p < 0.05). The non-significant effect of number of hedges per contour line and within-row spacing on root densities (mg/dmfl ndght be attributed to the inability of augers to capture root growth plasticity and heterogeneity. Roots are known to change their growth orientation and turn downwards or sidewise when they detect intense competition for soil resources. This behavior may be triggered by a hormonal mechanism which controls the geotrophic curvature of roots (Russel, 1977; Taylor, Blake, and Pharis, 1982; van Noordwijk et al., 1988). Some roots may also invest more in developing deeper roots or forming clumped roots while others may grow shallower in the soil (Berendse, 1979; Passioura, 1988). This growth tendency of roots would render sampling with auger a bit problematic. Timing of sampling becomes crucial and is further complicated by an already heterogenous soil environment. The plasticity of root growth, soil heterogeneity, and dominance of fine roots in the top soil, as found in the trench profiles, might have contributed to the high CVs of root densities from the auger sampling method. The root's compensatory growth away from compacted (Smucker, 1990) or highly competitive areas and movement towards the areas of least resistance, more fertile soils or moist spots, or areas relatively free from competition (Lyr and Hoffman, 1967), could produce high variability of roots extracted from borehole to borehole with auger. The high CVs also might have been caused by the frictional resistance of the corer's inner wall, which may partly force soil with roots away during the sampling (Schuurman and Goedewaagen, 1965). Furthermore, in root sampling with an auger, there is high probability that in one of several boreholes, larger diameter roots would be included in the core. The weight of these roots would cause large variability in the results of soil sampling. 145 Thus, it is not surprising why Caldwell (1987) lamented that some results of root studies vary widely and the errors incurred are often sizeable. With an auger, it would even be possible to obtain false impression from the results (Sylvester-Bradley, 1979). Smucker (1990) pointed out that results of root measurements are a function of the method selected. Accordingly, other root sampling techniques could be more effective in determining hedgerow root densities and their distribution. Installation of minirhizotrons may provide a clearer graphic presentation of root competition in an alley cropping system. The method has been designed to observe roots' compensatory and plastic behavior in response to perceived stresses (Hendrick, 1992; Upchurch and Ritchie, 1983; Smucker, 1984). Indirect observations of the performance of alley crops may provide further inferential, but qualitative information on the behavior of roots (Solera, 1992; Ong, Rao, and Mathuva, 1992) as long as other effects are minimized or controlled (i.e. incident light, space). As found in the present study, the classic trench profile mapping technique (Boehm, 1979; Caldwell and Richards, 1983) may still be the most appropriate method of assessing root densities in developing countries, where equipment availability is limited but labor is abundant and inexpensive. The high root densities in the 5- and 10- cm‘within-row'spacings from both the single and double hedges explain why these hedgerows yielded the highest initial and subsequent pruning biomass. Thus, it is suspected that these hedgerows invested a large portion of their photosynthate on roots (Cannell, 1985). Given that root density is closely related to nutrient and water uptake of plants (Russell, 1977), the densely-planted hedgerows were able to maximize the use of above- and below-ground resources per unit area for their growth. These plants, however, suffered high mortality due to strong intraspecies competition. The densely- planted hedgerows attained the lowest total height and diameter growth in 146 one year. But, due to their extensive root systems, these hedgerows were able to yield the highest periodic pruning biomass when cut at 2.0-m high. They might even have contributed the largest amount of root litter in the soil because of their high mortality rates. High.biomass (fromjprunings and root turnover) and control of soil erosion in sloping areas during intense rainfall are the major benefits from densely-planted hedgerows (Laquihon et al., 1991). For instance, the hedgerows from the 5- and lO-cm within-row spacing produced an average of 1.24 dry kg/m2 alley area from young twigs and leaves in the initial clipping. Based on the tissue analysis of G. sepium,_ this would be equivalent to about 33.49 N, 3.49 P, and 35.69 R per'ufi alley area. On the average, the hedgerows from all within-row spacing in the single and double hedgerows produced a total dry biomass of 1.08 kg/m2 of alley area. If this amount of one-year biomass from the two parallel hedgerows is divided and applied in the two adjacent alleys, it would translate into a total of 144 kg N, 14kg P, and 154 kg R per ha. Thus, it is not surprising that despite the nutrient removal of the second corn and mungo crops, the result of the soil analysis showed that N, P, R gradually increased after the fallow period, initial clipping, and subsequent pruning of the hedgerows. The less than one-year-old hedgerows were still ineffective in minimizing sheet erosion during high rainfall; hence, the initial decline of OM. The general decline of OM in an alley cropping system‘was also observed by Lal (1989a) over a period of five years. Young (1991), however, argued that with leucaena, hedgerows were able to maintain soil carbon at a satisfactory level in the alleys over six years through added prunings and crop residues. Perhaps, in the present study the biomass from the initial clipping and. subsequent pruning stabilized ‘the» OM condition in the experimental site and prevented its loss during the rainy months (June, 147 1991-November 1991). Any firm conclusion from the present study on the pattern of OM content, however, is too early to make. Another two or three years of measurement will give more meaningful insights. The gradual decline of N up to the third sampling period (August 1991) may be attributed to sheet erosion which caused the loss of top soil during the first year, the consumption of N by the first corn crop that was not fertilized, the inadequacy of the crop residues (corn and mungo) and G. sepium litterfall to maintain. N at the original level, and the consumption of the hedgerows themselves. ZHowever, in the fourth sampling, N increased as soon as initial clippings and subsequent prunings decomposed in the alleys, even with the N removal of the second corn crop. The decline of soil pH, P, and R in the second sampling and their increases in the third sampling, reveal the contribution of hedgerow litterfall during the fallow period as well as biomass from the initial clippings and subsequent prunings, in improving soil fertility. The decrease of soil pH, however, in the fourth sampling reflected the depletion of cations from a cycle of leaching or erosion, uptake, and recycling of bases from deep subsoil to surface horizons (Lal, 1989a). The observation on P, however, contradicted those of Yamoah et al. (1986) and Garrity (1992) who reported that the G. sepium hedgerows were not able to replenish the removal of P after'a.maize cropu Garrity's work was done in acid upland soil, where pumping of P from.deeper soil layers is limited by aluminum toxic subsoils and low subsoil P reserves. On the other hand, Rang et a1. (1981) found that P and R were able to accumulate over a period of six years when leucaena hedgerows and maize crops were grown together. The pattern of horizontal and vertical distribution of G. sepium hedgerow roots was better defined from the eight trench profiles than from the 148 sampling with an auger. The dominance of roots < 1.0-mm diameter, the declining root densities with distance from the hedgerow base towards the alley, and the inverse relationshipnof root densities with increasing soil depth were detected from the trench profiles, especially with the 5 and 10-cm within-row spacings in double hedgerows. On the other hand, the auger sampling revealed the possible effect of repeated prunings on root production over time. These findings have implications on hedgerow-crop competition for nutrients and water and strategies to minimize competitive hedgerow-root systems. 'They confirm the notion that higher root densities at the hedgerow/crop interface (Huxley et al., 1989; Buck, 1986) would escalate inter-specific competition for limited below-ground resources. At a fixed lateral distance of 50 cm from the hedgerow base and from the hedgerow base toward the alleys at different soil depths, almost 90 percent of the G. sepium roots belonged to the < 1-mm diameter class. More than 70 percent of all roots occurred within the top 30 cm of the soil profile. ' Within this stratum, root densities tended to decline towards the alley and with greater depths, especially in more dense within-row spacing and in double hedgerows. This rooting pattern and distribution may be explained by the tendency of roots to occupy fertile tOp soil, provided.moisture is adequate, especially during the early stage of vigorous growth (St. John, Coleman, and Reid, 1983; Lyr and Hoffman, 1967; Rang et al., 1985). Dhyani, Narain, and Singh (1990) also observed this pattern of root distribution in five multipurpose species. In their case, they considered roots < 2-mm diameter as fine roots. In acid and volcanic soils, Solera (1992), who excavated roots of several hedgerow species, found that majority of the roots were located between 0-20 cm soil depth; but, the lateral distribution varied with species. Roots of G. sepium hedgerows in volcanic soils were observed to be within the top 20 cm of soil up to a lateral distance of 100 cm. 149 Under a condition where most roots were found in the plow layer and were < 1-mm diameter, the nitrogen-fixing G. sepium could potentially compete with alley crops for available soil resources, especially P, moisture, and R (Garrity, 1992). Clearly, the G. sepium rooting pattern and distribution in this study contradicts the hypothesis that hedges have a deeper rooting pattern than those of alley crops (Rang and Wilson, 1987; Young, 1991; Szott, Fernandez, and Sanchez, 1991) Therefore, the crop would be forced to invest more of its carbon resources in roots to survive the competition process (Berendse, 1979; Caldwell, 1987), and yield will decline. The crOp will reallocate photosynthates that would otherwise be used for biomass accumulation.or grain production (Dickmann and Pregitzer, 1992). Young (1991) and Garrity (1992) mentioned that this danger -- roots of perennial hedgerows robbing nutrients from the systems of annual crops -- would cause more problems when tree roots extend laterally beneath the area planted to crops. However, the extent of damage from this process and strategies of co-existense in an alley cropping system are still very much unknown. Each alley crop might have a ”threshold” level of competition. Again, the question of the "minimum" root density of a hedgerow that will be acceptable by a certain alley crop needs an answer 0 Other studies do not give much light on the unfavorable consequences of competition between roots of hedgerows and the alley crop. Using crop yield as the main indicator, Rang et al. (1981) and Torres (1983) contended that maize yields were not significantly affected by the roots of leucaena hedgerows at 0-20 cm depth and up to 100 cm lateral distance from the hedge. Solera (1992), who worked with upland rice, concluded that pruning the roots of G. sepium and Cassia spectabilis had no significant effect on the growth and development of upland rice compared with those plots that were not pruned. The MBRLC Editorial Staff (1993) and Laquihon et al. (1991) reported that after a decade of experience with 150 their sloping agricultural land technology (SALT), maize yields from their experimental farm continued to increase and were shown to provide higher incomes. Analysis of root patterns and distribution, however, were not available. Ong (1989) and Johnson et al. (1988) concluded that hedgerows of leucaena and other species would compete with maize and other crops for nutrients and water in areas that experience seasonality of rainfall. In acid soils, Garrity (1992) found that yields of maize and rice were reduced when intercropped with hedgerows of G. sepium or napier grass. The roots of the hedgerows spread laterally at a shallow depth of 20-30 cm into the alleys. Atta-krah (1983) and Ong, Rao, and Mathuva (1992) postulated that roots of hedgerows had minor influence on the tree-crop interaction. The latter installed root barriers (galvanized iron sheets which were buried between leucaena and maize to a depth of one m) which had only minor effects on the crop yield. The authors argued that the differences of crop yield in the alleys could be attributed to competition for light. The pattern of vertical and horizontal root distribution of G. sepium hedgerows in the top 30 cm of soil, the linearly decreasing root densities toward the alley and deeper soil layers, and the dominance of fine roots found in the present study pose a challenge on how to manipulate hedgerow roots so that they will first grow downwards below the topsoil, spread laterally, and function as ”safety net” that could intercept leached nutrients (Young, 1991; van Noordwijk et al., 1988). Root and shoot treatments may be performed to minimize competition and influence the allocation of carbon to increase yield or the harvest index (Dickmann and Pregitzer, 1992; Cannell, 1985). 151 Based on the observed pattern and distribution of G. sepium at a fixed distance of 50 cm from the hedgerow base, the roots could be pruned down to a soil depth of 15-30 cm especially in hedges at‘within-row spacings of 5 and 10 cm. In this way, the roots belonging to < 1.0 mm and 1-2 mm diameter classes will become part of root litter production and improve soil fertility (Szott, Fernandez, and Sanchez, 1991). Repeated root prunings may even condition the hedgerow roots to develop downwards because'of their plastic and compensatory response (Russel, 1977; Smucker, 1990; Bowen, 1985). Continuous and periodic root prunings may temporarily check shoot growth and, eventually, reduce the aggregate shading effect of hedgerows on the alley crop (Cannell, 1985). Root pruning, however, has a cumulative effect of reducing top pruning biomass production for recycling into the top soil. For example, periodic top pruning reduced the mean root densities of hedgerows based on the first and second sampling of roots with auger. It confirmed the notion that top pruning checks root growth (Cannell, 1985). Barriers (galvanized iron, plastic, trenches) between the hedges and the crops may be installed to keep the hedgerow roots from tapping soil resources in the alleys (Ong, Rao, and Mathuva, 1992; Solera, 1992). Putting these barriers, however, is laborious and impractical. Farmers might prefer plowing along and close to the hedgerows to reduce hedgerow/crop competition near the hedges. A study will have to examine whether or not regular plowing along and at least 50 cm away from the contour hedgerows will minimize hedgerow/crop competition. The linearly decreasing root densities from the hedgerow base up to 50 cm towards the alley within the top 30 cm indicates that plowing or hilling up along the contour will control lateral growth of dense G. sepium hedgerow roots. Plowing may even ”force” the hedgerows to develop roots in deeper soils. A no-tillage practice in the alley, on the other hand, will only encourage 152 lateral growth of hedgerows and will be unfavorable for crop growth. In the long term, evaluating and breeding promising hedgerow species such as G. sepium for their deep rooting characteristics may be a logical option (Dickmann, Gold, and Flore, 1994). Development of and application of a hedgerow-specific root growth inhibitor during the critical stage of growth of an alley crop may be another alternative. In any case, any treatment of hedgerow roots should be targeted towards minimizing the unfavorable effect of increased hedgerow-crop competition for limited soil resources . Lastly, hedgerows may be planted at 20- and 40-cm within-in row spacings to obtain low initial root densities at the tree/crop interface and less competition with the alley crops. These widely-spaced hedgerows had the lowest mean root densities compared with those spaced at 5 and 10 cm, but they did not optimize soil resources per unit of stem volume. At these within-row spacings, however, the hedgerows will not be as effective as the dense hedgerows in controlling soil erosion. Moreover, they had the highest diameter and height growth; hence, there is the possibility that they eventually would aggressively grow more roots vertically and horizontally at the expense of the alley crop to meet their demands for nutrients and moisture. 4.5 Conclusions 1. Root densities of G. sepium hedgerows were not significantly affected by number of hedges per contour-line, within-row spacing, and pruning height based on root samples taken with the auger equipment at 30 cm depth. However, root densities were highly correlated (e.g. declining) with distanCe from the hedgerow base 153 towards the alley. Based on the high CVs of the samples, there is doubt whether or not root sampling with an auger will be the appropriate method of determining root distribution and densities in an alley cropping experiment. Up to 90 percent of the roots belong to < 1.0-mm diameter class and more than 70 percent were located in the top 30 cm of the soil based on the trench profile method. Clearly, this pattern of root distribution requires proper management of contour hedgerows to minimize their unfavorable impact on the alley crops. Root densities were found to be negatively correlated with distance from the hedgerow base up to 50 cm towards the alley and with soil depths for single and double hedgerows at 5- and 10-cm within-row spacings. In more dense hedgerow plantings, root densities were more predictable with respect to soil depth and distance from the hedgerow base towards the alley. The root densities of less-dense hedgerows (20- and 40-cm within-in row spacings) were less predictable with respect to soil depth and distances from the hedgerow base. The negative correlations imply certain practices in hedgerow establishment and management (i.e. root treatments) to minimize tree-crop competition in the interface. Average N, P, R, and pH initially declined after site clearing and the first maize crop. ‘Values for these elements, however, gradually increased after the hedgerows were established and soil erosion declined, and after incorporation. of initial hedgerow pruning biomass in the alleys. OM declined and stabilized even after the initial hedgerow clipping and subsequent prunings. 154 Literature Cited Atta-Rrah AN (1983) The weedings of Leucaena leucocephala (Lam) De Wit and its control in leucaena-based agroforestry system. Seminar Paper, IITA, Ibadan, Nigeria, 17 June Atkinson D, D Naylor, GA Coldrick (1976) The effect of tree spacing on the apple root system. Horti. Res. 16:89-105 Berendse F (1979) Competition between plant populations with different rooting depths. Oecologia 43:19-26 Boehm W (1979) Methods of Studying Root Systems. Springer-Verlag, Heidelberg Bowen GD (1985) Roots as a component of tree productivity. In: MGR. Cannell and JE Jackson, eds, Attributes of Trees as Crop Plants, 592 p. Institute of Terrestrial Ecology, Huntington, UR Buck M (1986) Concepts of resource sharing in agroforestry systems. Agroforesty Systems 4:191-203 Caldwell MM and Richards J (1983) Competing root systems: morphology and models of absorption. In: T Givnish, ed, On the Economy of Plant Form and Function. Cambridge University Press, UR Caldwell MM (1987) Competition between root systems in natural conmunities. In: PJ Gregory, JV Lake, and DA Rose, eds, Root Development and Function. Cambridge University Press, UR Cannell MGR (1983) Plant management in agroforestry: manipulation of trees, population densities and mixtures of trees and herbaceous crops. In: PA Huxley, ed, Plant Research in Agroforestry. pp 455-487. ICRAF, Nairobi, Renya Cannell MGR (1985) Dry matter partitioning in tree crops. In: MGR Cannell and JE Jackson, eds, Attributes of Trees as Crop Plants, 592 p. Institute of Terrestial Ecology, Huntington, UR Dickmann DI and RS Pregitzer (1992) The structure and dynamics of woody plant root systems. In: CP Mitchell, JB Ford-Robertson, T Hinckley and L Sennerby-Forsse, eds, Ecophysiology of Short Rotation Forest Crops. Elsevier Applied Science, New York Dickmann DI (1992) Role of physiology in forest tree improvement. Silva Fennica 25(4):248-256 Dickmann DI, MA Gold and JA Flore (1994) The ideotype concept and the genetic improvement of trees. Plant Breeding Reviews (in press) Dhyani SR, P Narain, and RR Singh (1990) Studies on root distribution of five multipurpose tree species in Doon Valley, India. Agroforestry Systems 12:149-161 Fitter AH (1987) An architectural approach to the comparative ecology of plant root systems. New Phytol. 106:61-77 155 Garrity DP (1992) Conservation farming systems for the sloping lands: research and development imperatives. Background paper for the 'National Symposium of Conservation Farming', Soil Research and Development Center, Quezon City, Philippines, 18 July 18 Gillespie AR (1989) Modelling nutrient flux and interspecies root competition in agroforestry interplantings. Agroforestry Systems 8:257-265 Hendrick RL (1992) Fine root production, demography and turnover in Northern hardwood forests. A PhD dissertation, Michigan State University, East Lansing, MI, USA Huxley P, T Darnhofer, A Pinney, E Akunda and D Gatama (1989) The tree/crop interface: a project designed to generate experimental methodology. Agroforestry Abstracts 2(4):127-145 Johnson R, L Fidjeland, JA Maghembe and P Hogberg (1988) The vertical distribution of fine roots of five tree species and maize in Morogoro, Tanzania. Agroforestry Systems 6:63-69 Rang BT, GF Wilson and L Spikens (1981) Alley cropping maize (Zea mays L.) and leucaena (Leucaena leucocephala Lam.) in Southern Nigeria. Plant and Soil 63:165-179 Rang BT, GF Wilson and TL Lawson (1984) Alley Cropping: A Stable Alternative to Shifting Cultivation. IITA, Ibadan, Nigeria Rang BT and GF Wilson (1987) The development of alley cropping as a promising agroforestry technology. In: HA Steppler and PRR Nair, eds, Agra-forestry: A Decade of Development, pp 227- 243. ICRAF, Nairobi, Renya Lal R (1989) Potential of agroforestry as a sustainable alternative to shifting cultivation: concluding remarks. Agroforestry Systems 8:239-242 Lal R (1989a) Agroforestry systems and soil surface management of a tropical alfisol. III. Changes in soil chemical properties. Agroforestry Systems 8:113-132 Laquihon,W (1988) Small farm uses of multipurpose tree species in Mindanao. Mindanao Baptist Rural Life Center (MBRLC), Davao, Philippines Laquihon W, JJ Palmer and HR Watson (1991) Sloping agricultural land technology: a decade of experience on hillside sustainability. A paper presented at the 'Asian Farming Systems Symposium - 1992' at BMICH-Colombo, Sri Lanka, 2-5 November Lyr H and G Hoffman (1967) Growth rates and growth periodicity of tree roots. Int. Rev. For. Res. 2: 181-236 MBRLC Editorial Staff (1988) Modern farming guide: a manual on how to farm your hilly land without losing your soil. Mindanao Baptist Rural Life Center, Rinuskusan, Davao del Sur, Philippines Ong CR, MR Rao and M Mathuva (1992) Trees and crops: competition for resources above and below the ground. Agroforestry Today 4:4-5 156 Ong' CR (1989) Cropping systems approach ‘to agroforestry research. ICRISAT, India Parrota, JA and IP Singh (1992) Gliricidia sepium root system morphology and biomass allocation. Nitrogen Fixing Tree Research Reports 10:169-172 Passioura, J.B. 1988. Water transport in and to roots. Ann. Rev. Plant Physiol. Plant Mol. Biol. 39:245-265 PCARRD (Philippine Council for Agriculture and Resources Research) (1985) Research Techniques in Crops, 512 p. Los Banos, Laguna, Philippines Russell RS (1977) Plant Root Systems: Their Function and Interaction with the Soil. McGraw-Hill Book Co. (U.R) Ltd., London Upchurch DR and JT Ritchie (1983) Root observations using a video recording system in minirhizotrons. Agronomy Journal 75:1009-1015 Shickluna JC (1983) Sampling soils for fertilizer and lime recommendations. MSU Extension Bulletin E-498, Revised 1983. MSU, East Lansing, MI, USA Schuurman JJ and MAJ Goedewaagen (1965) Methods for the Examination of Root Systems and Roots. Institute for Soil Fertility. Center for Agricultural Publications and Documentation, Wageningen, Netherlands Smucker AJM (1984) Carbon utilization and losses by plant root systems. In: Roots, Nutrient and Water Influx, and Plant Growth. Soil Science Society of America, Crop Science Society of America, and American Society of Agronomy, 677 South Segoe Road, Madison, WI, USA Smucker AJM (1990) Isotopic and nonisotopic methods for measuring root dynamics and activities. A handout in Soil Biophysics course. MSU, East Lansing, MI, USA ' Solera CR (1992) Competition between upland rice and hedgerow species in an alley cropping system. A PhD dissertation, UPLB College of Forestry, College, Laguna, Philippines St. John TV, DC Coleman and CP Reid (1983) Growth and spatial distribution of nutrient-absorbing organs:selective exploitation of soil heterogeneity. Plant and Soil 71:487-493 Sumberg JE and AN Atta-Rrah (1988) The potential of alley farming in humid West Africa - a re-evaluation. Agroforestry Systems 6: 163-168 Sylvester-Bradley R (1979) Monitoring soils in agroforestry: soil biology and biochemistry. In: HO Mongi and PA Huxley, eds, Soils Research in Agroforestry: Proceedings of an Expert Consultation. ICRAF, Nairobi, Renya Szott LT, ECM Fernandez and PA Sanchez (1991) Soil-plant interactions in agroforestry systems. Forest Ecology and Management 45:127- 152Taylor JS, TJ Blake and RP Pharis (1982) The role of plant hormones and carbohydrates in the growth and survival of coppiced eucalyptus seedlings. Physiol. Plant. 55:421-430 Torres F 1983 Potential contribution of Leucaena hedgerows intercropped with maize to the production of organic 157 nitrogen and fuelwood in the lowland tropics. Agroforestry Systems 1:323-333 van Noordwijk M, R Hairiah and Syekhfani (1988)Pe1tophorum pterocarpa (DC) Back (Caesalpiniaceae), a ‘tree ‘with a root distribution suitable for alley cropping on acid soils in the humid tropics. In: Proceedings of the International Society of Root Research Symposium, Uppasala, Sumatra. Watson HR and W Laquihon (Undated) A manual on how to farm your hilly land without losing your soil. Mindanao Baptist Rural Life Center, Davao del Sur, Philippines Watson HR (1983) Developing a hillside farming techniques for Mindanao hill farmers. In: Proceedings of the Hilly Land Development Workshop, March 24-26, 1983, Cebu City, Philippines Wilson GF, BT Rang and R Mulongoy (1986) Alley cropping: trees as sources of green manure and mulch in the tropics. Biological Agriculture and Horticulture 3: 251-267 Young A (1991) Soil fertility. In: ME Avery, MGR Cannell and C Ong, eds, Biophysical Research for Asian Agroforestry. Winrock International and South Asia Books, USA Chapter 5 WATER POTENTIAL OF THE HEDGBROWB Abstract The day-time and pre-dawn'water potentials (t) of nine-month and 32-month- old Gliricidia sepium hedgerows were measured in a 0.22-ha alley cropping field experiment at Mt. Makiling, University of the Philippines at Los Banos, College, Laguna, Philippines. The treatments for the nine-month- old hedgerows were the number of hedges per contour line and within-row spacing. An additional treatment in the 32-month-old G. sepium hedgerows was pruning height. In the nine-month-old hedgerows, thetmost negative mean day-time and pre- dawn t was obtained from the S-cm within-in row spacing. The number of hedges per contour line did not significantly affect t of the hedgerows. Towards the end of the dry season, the nine-month-old G. sepium hedgerows were water stressed, indicated by pre-dawn.¢ more negative than - 0.5 MPa. The number of hedges per contour line, combinations of pruning height and within-row spacing, and their interaction significantly affected the day- time t of 32-month-old hedgerows. Although there was an indication that t was more negative with more dense within-row spacing and lower pruning height, Fisher's Protected LSD test (p.< 0.05) and regression analysis did not show a clear pattern of t response with respect to number of hedges per contour line, pruning height, and within-row spacing. It appeared that in fully-established G. sepium hedgerows, the demand for water to meet transpiration requirements was not singly influenced by any of the treatments e 158 159 The number of hedges per contour line, combination of pruning height and within-row spacing, and their interaction did not significantly influence the pre-dawn t of 32-month-old hedgerows. The measurements were performed in the middle of the 1993 dry season. The results of the study indicate that in newly-established G. sepium hedgerows, competition for water in the tree-crop interface will be high in densely-planted hedgerows. This competition will partly deprive intercrops of their needed water. In established hedgerows, competition for water occurred regardless of the number of hedges per contour line, pruning height, and within-row spacing. 160 5.1 Introduction In alley cropping, hedgerows compete for incident light and space above- ground, and nutrients, space and water below ground. This intraspecies competition directly affects the growth of hedgerows and indirectly influences the productivity of the alley crop because both share the same resource pools in the environment (Buck, 1986). Since hedgerows provide a service function in alley cropping, gains from intraspecies competition should be optimized and its unfavorable effects on intercrops minimized. Hedgerows serve as one of the means to improve upland productivity and enhance environmental stability. Intraspecies competition for water among hedgerow plants and how this affects intercrops are the least-studied aspect of alley cropping. Early work.focused on establishing hedgerows to control soil erosion and improve soil fertility (Rang, 1990; Laquihon, 1988; Watson, 1983), and, to a certain extent, crop yield. In the Philippines, alley cropping was developed in response to expanding deforestation, destructive slash-and- burn farming in logged-over areas, and degradation in open and grassland sites (Watson and Laquihon, undated; Granert and Sabueto, 1985; Sajise, 1985). Now that the role of hedgerows in controlling soil erosion and restoring soil fertility has been more or less established (Young, 1991; Tacio, 1993), inter- and intra-species interactions for water and nutrients have to be examined to test the validity of earlier assumptions on the servicing roles of hedgerows in an alley cropping (Huxley, 1983; Buck, 1986; Rang and Wilson, 1987; Rang et al., 1990; Ong, 1991). The phase of refining the technology of alley cropping has come. Hopefully, this refinement process will provide a deeper understanding of the hedgerows' service function to obtain sustainable intercrop yield. 161 Hedgerows have been documented to capture rainfall from canopy interception, increase soil water infiltration, and improve microclimatic condition (Rang and Wilson, 1987; Huxley et al., 1989). In MBRLC, for instance, they reported that SALT plots had infiltration rates seven times faster than the non-SALT ones. Moreover, by using moisture probes buried at 15-cm depth, the workers observed that the alley plots with contour hedgerows had higher soil moisture compared with non-SALT plots (Laquihon et al., 1991). These moisture-related improvements or increases in the alleys and the extent that they become self-serving to the hedgerows are least understood. Huxley et al. (1989) observed that the aggressive lateral roots of nearby Cassia siamea hedgerows intruded into the alleys and might have partly deprived castor beans of needed moisture for growth and transpiration. They also found that the soil near and below a hedge of Grevillea robusta dried almost to the wilting point but not at the deeper layers beneath the maize crap. Moisture improved with increasing lateral distance from the hedge. Huxley (1983) suggested the measurement of hedgerow water potential (V) to obtain a better understanding of the tree/crop interface in agroforestry systems. The leaf t of hedgerows under a given condition would indicate the rate of water uptake which reflects transpiration rate, size of the root system, and amount of available water in the soil (Rozinka, 1989). The t will show the extent of potential competition for water among hedgerows and between the hedges and the intercrop. By comparing the t of several species, their suitability as a hedgerow in alley cropping may also be evaluated. Accordingly, this study attempted to measure hedgerow t as affected by different number of hedges per contour line and various combinations of pruning height x within-row spacing; The study hypothesized that the t of hedgerows becomes more negative with increasing density (more hedges per 162 contour line and closer within-row spacing) and with higher pruning. The hedgerowe with very low (more negative) t'would cause intense intraspecies competition for water, and would deplete available soil moisture intended for the intercrops. 5.2 Materials and Methods 5.2.1 Sampling Procedure The first measurement of water potential occurred towards the end of the dry season in 1991 (May 25, June 1, and June 8), while the second was done at the middle of the dry season in 1993 (April 8-10). In both sampling periods, no major precipitation occurred at least two weeks before the measurement, except with the June 8 measurement when a slight rain fell in the area four days before the sampling. The hedgerows were about nine-months old during the first measurement. The measurement was performed before the initial clipping in August 1991. In the second measurement, the hedgerows were about 32-months old. Between the first and second measurements, the one-year old hedgerows were cut to 30 cm above ground in August 1991; periodically pruned back to 30 cm above ground upon reaching 1.0-, 1.5-, and 2.0-m heights between- November 1991 to May 15, 1992; allowed to grow until September 1992 after which they were cut back to 30 cm above ground; and periodically pruned back to 30 cm above ground between November, 1992 to March 1993 upon reaching 1.0-, 1.5-m, and 2.0-m heights. The treatments during the first sampling were only the number of hedges per contour line and within-row spacing. Accordingly, in each main plot the sub-plots which represented a within-row spacing treatment were selected for sampling. This was done for each replicate. Thus, for each 163 measurement, a total of 24 sub-plots were sampled per day. In each sub- plot, two hedgerow plants were randomly selected, one each from the lower and upper hedgerows. Three compound leaves were cut from each plant for water potential measurement. As suggested by Garrity (1991a), the leaves that were cut for measurement were those considered morphologically mature and active in photosynthesis. With G. sepium» these leaves normally start with the fifth to seventh leaf below the youngest leaf at the tip of the plant. The t was determined in a total of six leaves for each sub-plot, 24 leaves per main plot, 48 leaves per replicate, or 144 leaves per measurement. In the first measurement, the day-time leaf 8 of the hedgerows was completed three times, one full sampling per day. The day-time measurement started at 9:30 AM on May 25, 1991, 8:30 AM on June 1, 1991, and 10:15 AM on June 8, 1991. The measurements were over between noon to about 2:00 PM. The pre-dawn 1 measurement, however, was only done on June 1, 1991. The measurement started at 5:00 AM and ended at about 6:30 AM. Since there was only one- and a half-hour to do the measurements, only one hedgerow plant per sub-plot was selected. In each plant, three leaves were cut for t determination“ Thus, a total of 24 leaves were sampled for pre-dawn t measurement. The second i measurement was conducted on April 8-10, 1993. In this measurement, all 12 sub-plots per main plot were sampled. This was done to take into account the possible effects of number of hedges per contour line and the various combinations of pruning height and within-row spacing. Accordingly, only one replicate (composed of sub-plots from the single and double hedgerows) was completed per day. A total of 24 sub- plots, 48 hedgerow'plants, and 144 leaves were sampled each day during the day measurement. In the second measurement, the number of plants and 164 leaves, and location of the leaves for sampling, were the same as the first measurement. The pre-dawn measurement of t was conducted each day from April 8-10, 1993. Instead of sampling all sub-plots, however, only the number of hedges per contour line and within-row spacing were considered treatments in selecting sub-plots to be sampled. Only one hedgerow plant and three leaves per plant were measured at each sub-plot. Again, this procedure was adopted to maximize the one- and a half-hour sampling window at pre- dawn. During the three consecutive days, pre-dawn t measurements started at 5:00 AM and were completed on or before 7:00 AM. Thus, for the pre-' dawn measurements, only four hedgerow plants were sampled per main plot, a total of eight plants per replicate, or 24 leaves per measurement. 5.2.2 Equipment and Procedure Equipment. The pressure chamber method as described by Slavik (1974) was used in determining the water potential of the G. sepium hedgerow plants. A pressure chamber made by PMS Instruments, Corvallis, Oregon, USA was borrowed from the Agroecology Division of the International Rice Research Institute (IRRI), College, Laguna, Philippines. However, since the small nitrogen gas tank was not made available, a SO-kg nitrogen tank and required fittings were procured from a local supplier. Procedure. Before the actual t measurement, pre-sampling of G. sepium leaves with the use of the pressure chamber was performed to gain familiarity with the use of the equipment and to gain an estimate of the amount of time to complete measuring one leaf sample. The research aide was also taught and shown how and where to cut a leaf sample from the hedgerow plant. A day before the scheduled measurement, the nitrogen gas system was set and its locations established. The gas tank and the 165 pressure chamber were located in the middle of the main plots at each replicate. The pressure chamber was placed on top of a small table that was firmly put on the ground. A two to three-person team conducted the measurement of hedgerow t. The research aide cut the leaf with a blade from the hedgerow plant and took the leaf to the person operating the pressure chamber. The third person would hold the table or the chamber when it was being sealed or closed for reading. He also acted as an alternate runner person to get the leaf from the research aide. The whole sequence of activities took 3-5 minutes per sample, depending on the distance of the hedgerow plant to the pressure chamber. The pressure chamber operator gave the research aide a go signal to cut the leaf when he was ready for another sample measurement. I operated the pressure chamber throughout the t determination in both sampling periods. For the initial pressure chamber measurement, a hand lens was used in detecting the appearance of sap droplets on the cut surface of the leaf petiole. Over time, however, the determination of time when sap droplets appeared on the cut surface of the petiole was done with the naked eye. On a clear day, the use of a hand lens and a close examination by eye of the sap appearing on the cut surface of the leaf would almost give the same level of accuracy. In the pre-dawn measurements, a flashlight was used to see the appearance of sap droplets on the cut surface of the leaf petiole. 5.2.3 Statistical Analysis The data from the first and second measurements were converted from bars into megapascal units, MPa, using the conversion of one bar 8 0.1 MPa. Results from the three samplings in the first measurement were subjected 166 to analysis of variance and, when found significant, to the Fisher's Protected LSD multiple range test. The average of the six leaf samples was used in the analysis; each set of averages from the three sampling dates were analyzed separately. The values from the three sampling dates were also averaged and analyzed. In the second sampling, the daily results of the day-time and pre-dawn ¢ measurement were separately subjected to analysis of variance and LSD multiple range test, when applicable. Accordingly, the r values from the hedgerow plots in top, middle, and bottom of the slope were individually analyzed. In each case, the average water potential values from each hedgerow'plant were used as a replicate. Ultimately, however, the average data from the top, middle, and bottom plots were combined for analysis. Hedgerow 8 values from the top, middle, and bottom plots were also pooled in running the regression analyses to determine the linear relationship between t, within-row spacing, and pruning height. All the data.were entered into the LOTUS 1-2-3 spreadsheet, converted into ASCII files, and analyzed with the use of MSTATC/SYSTAT software. The derived figures were all generated with PLOTIT software. 5.3 Results 5.3.1 Water Potential in the First Measurement Means of three leaves per replicate were used in the analysis of variance in all the three sampling dates. Based on the result, hedgerow t on May 25 was not significantly affected by the treatments and their interaction (Table 5.1). However, on June 1, t of the hedgerows was significantly affected by the interaction of number of hedges per contour line and the within-row spacing (p < 0.01). On June 8, within-row spacing 167 significantly affected 1 of the hedgerows (p < 0.01). The F-test on the average t from the three sampling dates yielded significant effects from within-row spacing and the interaction (p < 0.01). In all the three sampling dates and the average, the hedgerow t was not significantly influenced by the number of hedges per contour line. The number of hedges per contour line and the interaction did not significantly affect pre—dawn y at p < 0.01 (Table 5.2). However, the within-row spacing treatments significantly affected pre-dawn hedgerow w (p < 0.01). In both the June 8 and average from the three sampling dates, the day-time t means from the 5-cm within-row spacing were significantly lower (more negative) than those from the 10-, 20-, and 40-cm within-row spacing (Figure 5.1). The mean i from the S-cm within-row spacing was found to be significantly lower than those in the 10-, 20-, and 40-cm within-row spacings (Figure 5.2). The average pre-dawn t of the hedgerows on June 1 was - 0.51 MPa, below the average water potential of the soil at field capacity (- 0.1 MPa) (Kramer, 1969). 168 Table 5.1. F-values of analysis of variance'csf day-time hedgerow water potential on May 25, June 1, and June 8, 1991. Source of Variation May 25 June 1 June 8 Average ‘ Number of 0.07 0.06 0.40 0.00 hedges/line Within-row spacing 0.45 1.73 6.07*** 5.09*** I Interaction 0.82 6.15*** 1.54 4.03*** ll ' *** Significant at one percent probability Table 5.2. Analysis of variance'oi pre-dawn hedgerow water potential on June 1, 1991. Source of Variation F-Value Number of hedges/line 1.49 Within-row spacing 7,1o*** 2.85 Interaction ' *** Significant at one percent probability 169 1.20- d 1.18— -! 1.16— LSD=0.07 ‘ 1.14- d 1.12- d .10— B . B B 1.08— d 1.06- q 1.04- d 1.02- - 100 I I I I I I II 0 510152025303540 Mean Water Potential (—MPO) Within—row Spacing (cm) Figure 5.1. Day-time water potential (mean of three sampling dates) of hedgerows from different within-row spacing. Bars topped with the same letter are not significantly different (LSD, p < 0.05). Mean Water Potential (—MPa) Figure 5.2. within-row spacing on June 1, 1993. 0.70; 065—: 0.605 0.555 0.50—j 045—: 0.40% 0.35—j 030—: 0.25g 020—: 015—: 010—: 0.055 170 L30 = 0.11 0.00 i 0 i 5 Within—raw Spacing (cm) I I I I I I I 10152025303540 Means of pre-dawn water potential of hedgerows at different Bars topped with the same letter are not significantly different (LSD, p < 0.05). 171 5.3.2 Water Potential in the Second Measurement The effect of pruning height x within-row spacing on day-time t of hedgerows at the top of the slope was significant at p < 0.01 (Table 5.3). The interaction effect was also significant at p < 0.05. At the middle of the slope, the number of hedges per line, pruning height x within-row spacing, and the interaction significantly affected the day-time t of the hedgerows (p < 0.01). At the bottom of the slope, the number of hedges per contour line and pruning height x within-row spacing significantly influenced hedgerow t at p < 0.05 and p < 0.01, respectively. There was no significant interaction effect. When the average i values from the three sampling dates were further averaged, the treatments and the interaction showed no significant effects on the day-time t of hedgerows. The t from the different slope locations of the hedgerows (top, middle, and bottom) did not greatly vary (-1.18 MPa from the middle; -1.18 MPa from the bottom; and -l.13 from the top). It should be noted that the plots at the top of the slope are shaded starting at about 2:30-3:00 PM. At the middle and bottom of the slope, the hedgerow plots were significantly affected by the number of hedges per contour line. The single hedgerows had more negative mean t (-1.22 MPa) than the double hedgerows (-1.13 MPa) because the plots are more exposed to sunlight. In theimiddle plots, however, the double hedgerows had.more negative t (-1.27 MPa) than those in the single hedgerow plots (-1.09 MPa). The number of hedges per contour line, pruning height x within-row spacing, and interaction did not significantly affect pre-dawn t of the hedgerows in any slope location of the plots (Table 5.4). The mean pre- 172 dawn t of hedgerows at the middle of the slope was -0.32 MPa, followed by those in the top (-0.29 MPa), and bottom of slope (-0.26 MPa). Some treatment means from the top, middle, and bottom of the slope had significant differences from each other; however, these differences did not show a definite pattern or trend. Perhaps, the strong interaction effect of number of hedges per contour line and pruning height x within- row spacing complicated the pattern of mean differences. Nevertheless, a matrix was prepared to determine the frequency of occurrence of the most and least mean t at different pruning height and within-row spacing (Table 5.6). The most negative mean 8 occurred the highest at 5- and 10-cm within-row spacings at 1.0- and 1.5-m pruning heights. The least negative t occurred the greatest in the 20- and 40-cm within-row spacings at 1.0- and 1.5-m pruning heights. In the top plots, where hedgerows get shaded from the afternoon sunlight as early as 2:30 PM, the hedgerows with 1.5-m PH x 5-cm*within-row spacing had the least negative t (-1.03 MPa); the most negative was obtained from, the 1.5-m pruning height x 10-cm within-row spacing plots (-1.31 MPa). In the middle plots, where hedgerows get most of the afternoon sunlight, the 1.0-m pruning height x 40-cm‘within-row spacing had the most negative V (- 1.29 MPa) and the least was from the 1.5-m pruning height x 20-cm within- row spacing (-0.99 MPa). Most treatment 173 Table 5.3. F-values of analysis of variance'cmfday-time water potential of hedgerows from April 8-10, 1993. Source of Top Middle Bottom Average Variation2 Number of 2.97 256.76*** 47.81** 0.34 hedges/line PH x WRS 5.11*** 10.69*** 5.91*** 0.87 Interaction 2.84** 11.41*** 1.41 0.78 Table 5.4. *** ** Significant at one percent probability; Significant at five percent probability PH 8 Pruning height; WRS8 Within-row spacing; Top8 Main plots at the top of the slope; Middle= Main plots at the middle of the slope; Bottom-8 Main plots at the bottom of the slope; Average8 Average water potential from the top, middle, and bottom plots. F-values of analysis of variance' of pre-dawn hedgerow water potential on April 8-10, 1993. [ Source of Variation’ Top Middle Bottom fl Number of hedges[line 3.77 3.19 0.36 I Pruning height x within- 2.78 0.34 3.03 row spacing Interaction 1.76 1.88 0.71 None of the F-values are significant at P < 0.05. Top 8 Main plots at the top of the slope; Middle8 Main plots at the middle of the slope; Bottom8 Main plots at the bottom of the slope. Table 5.5. Fisher's Protected LSD test' potential of hedgerows in - MPa on April 8-10, 1993. for the means of day water Treat- Tap-Means Treat- Middle- Treat- Bottom- ment2 Ranked ment2 Means ment2 Means Order Ranked Ranked Order Order 1.5-PH 1.31 1.0-PH 1.29 1.0-PH x 1.32 lO-WRS A 40-WRS A S-WRS A 40-WRS AB S-WRS AB 20-WRS A lO-WRS BC lO-WRS B S-WRS A 1.5-PH 1.16 2.0-PH 1.22 1.5-PH x 1.24 20-WRS BCD 20-WRS B S-WRS AB 2.0-PH 1.12 2.0-PH 1.21 1.0-PH x 1.22 40-WRS BCDE 40-WRS BC 40-WRS ABC 1.0-PH 1.11 1.0-PH 1.21 1.0-PH x 1.20 40-WRS BCDE 5-WRS BC 10-WRS ABCD 2.0-PH 1.11 2.0-PH 1.18 2.0-PH x 1.15 S-WRS BCDE lO-WRS BC lO-WRS BCDE lO-WRS CDE lO-WRS BC lO-WRS BCDEF 20-WRS CDE 5-WRS BC 20-WRS CDEF 20-WRS DE 40-WRS C 20-WRS DEF 1.0-PH 1.06 1.0-PH 1.08 2.0-PH x 1.04 S-WRS DE 20-WRS D 40-WRS EF 1.5-PH 1.03 1.5-PH 0.99 1.5-PH x 1.03 S-WRS E 20-WRS 40-WRS F _ — letter are not Values significantly different. followed by the same alphabetic 2 PH 8 Pruning height; WRS8 Within-row spacing 175 Table 5.6. Occurrence of the significant most and least negative mean water potential in different pruning height and within-row spacing. 53;! 1.0-m PH 1.5-m PH 2.0-m PB Total "“3 139 IN LN NN LN NN’ LN 5 cm 1 1 1 1 1 3 2 10 cm 1 1 2 0 i 20 cm 2 1 1 1 3 I 40 mm 1 1 1 1 2 I Total 3 3 3 3 1 1 7 7 ‘ PH8 Pruning height; WRS = within-in row spacing 2 MN8 Most negative t LN8 Least negative t means in the rest of the hedgerow plots were found to be not significantly different from each other. In the bottom plots, the most negative mean W was taken from 1.0-m pruning height x 5-cm‘within-row spacing (-1.32 MPa); the least negative came from 1.5-m pruning height x 40-cm within-row spacing (-1.03 MPa). To obtain possible linear relationships between t and pruning height, and between.t andwwithin-row spacing, regression analyses were conducted.using the pooled values from all the plots. The T-test gave a non-significant result; thus, there is insufficient evidence to say that there is a linear relationship between t and pruning height (p < 0.10; r 8 0.10). The same result was obtained from the T-test between t and within-row spacing (p < 0.10; r - 0.08). 176 5.4 Discussion In the first measurement of day-time t of nine-month old hedgerows, the effect of within-row spacing was significant. There was no significant influence of the number of hedges per contour line on the hedgerow t or an interaction. In the measurement of 32-month old hedgerows, the number of hedges per contour line, pruning height x within-row spacing, and the interaction influenced the t of hedgerows at different locations of the slope. The effect of the number of hedges per contour line, pruning height x within-row spacing, and interaction on hedgerow t existed; however, how the treatments account for that effect could not be explained by the multiple range test and the regression analyses. There were significant differences in the pre-dawn t of the nine-month old hedgerows but this was not observed in the 32-month old hedgerows. The first measurement was taken towards the end of the dry season; hence, the plants experienced more stress and had more negative pre-dawn 8 (average of - 0.51 MPa). The nine-month old hedgerows were already water stressed at the beginning of the day before the transpiration process began. The pre-dawn t during the second measurement, which was done in the middle of the dry season, had an average of - 0.29 MPa, slightly below the water potential of soil at field capacity (-0.1 MPa), but still within the optimum range of soil matric potential (-0.1 to -0.5 MPa). At this level of pre-dawn t, the hedgerows in the second measurements were not as water stressed as in the first measurements. None of the mean water potential values from the first and second measurements exceeded the limit of the wilting coefficient (-1.5 MPa). The i values of the nine-month old hedgerows became more negative with more dense plantings because at closer within-row spacing, there are more plants per unit area and, therefore more transpiring leaf surface area 177 (Rramer and Rozlowski, 1979). In fact, the hedgerows in 5-cm within-row spacing had the most negative pre-dawn t values. They suffered water stress which resulted in high mortality and reduced height and diameter growth. These hedgerows were not able to acquire enough soil water to fully rehydrate at night and equilibrate to a V level similar with hedgerows from less dense within-row spacing (Huck, 1984), despite the fact that these hedgerows had high root densities. In fact, the hedgerows at closer within-in row spacing consumed more water and depleted soil water faster compared with those in the less dense within-in row spacing Ibecause they had more roots/db? (Tables 4.7 and 4.14, Chapter 4). The root density of the hedgerows from the 5-cm within-row spacing was also negatively correlated with soil depth ranging from 0 cm (surface layer) down to 60 cm. The dense hedgerows invested more carbon in developing an extensive root system to supply the high demand for water during their transpiration period. In similar studies, yields of maize and upland rice in rows near hedgerows were reduced despite regular pruning of above-ground biomass to reduce competition for light and some root pruning to reduce competition for nutrients and.water (Ong, Rao, and Mathuva, 1992; Solera, 1992; Lal, 1989; Laquihon et al., 1991). My results confirm the observation of Huxley et al. (1989) that soil moisture near and below hedgerows up to a lateral distance of 0.5-1.5-m and depth of 200 cm could be reduced by the hedgerows. My findings also provide empirical evidence to support Garrity's (1991) suspicion that the lateral spread of hedgerows of (L sepium and napier towards the alley and beneath the plow layer will rob much of the soil moisture intended for the intercrops. The high water demand of densely-planted hedgerows was probably the reason why mungbean yield from the first crop was the lowest in S-cm within-row spacing. There was intense competition for water in the tree/crop 178 interface. Thus, at closer within-row spacing of hedgerows, the intercrop's net gain from intraspecies competition for water in the hedgerows was negative. The major positive gains were production of a high volume of biomass to be used as soil amendment and reduction of water evaporation from the soil surface because of shading and cover. The results of the t measurement in the 32-month-old hedgerows give an insight into how hedgerows make alley cropping productive. In the medium- term, after the hedgerows were initially cut and subsequently pruned several times, G. sepium hedgerows still intensely competed for water among themselves and gradually depleted available soil water for the intercrop. This statement is clearly reflected in the mean t values obtained at different number of hedges per contour line, within-row spacing, and pruning height. The vague response pattern and weak linear relationship of t and within-row spacing and pruning height treatments imply that intraspecies competition for water became a factor in the hedgerows. Established hedgerows appeared not to be singly affected by within-row spacing, number of hedges per contour line, and pruning height. The effect of these variables on V of hedgerows existed; but their strong interaction renders accounting of the effect on # difficult. There was indication, however, that t was still significantly influenced by within- row spacing in established hedgerows. It is obvious that the hedgerows were controlling soil erosion, enhancing microclimatic condition, and providing organic matter for the intercrops, while also depriving the crops of water. This is the trade-off in adopting alley cropping system in the uplands. Based on the t pattern of the 32-month G. sepium, fully-established hedgerows appeared to have developed roots in both the lateral and vertical direction to meet the demand for water by the transpiring leaves. At this age, hedgerow planting density ceased to be a factor and the limit of the resource-supplying power of the environment (Harper, 1977) took 179 over. Available water and growing space for expansion in the soil became the limiting factors. Therefore, the question of how to manipulate the hedgerows to reduce their demand for water to a level that will not jeopardize the intercrop becomes the issue. Will more regular and frequent prunings check root growth (Cannell, 1985) and result in a temporary reduction of water consumption by the hedgerows? Will periodic root prunings minimize interspecies competition for water between the hedgerows and the intercrops (Solera, 1992; Ong, Rao, and Mathuva, 1992)? Will it be feasible to select hedgerow species based on a definition of the idealized rooting pattern and transpiration rate as part of the "hedgerow ideotype” (Dickmann, 1992; Dickmann, Gold and Flore, 1994; Young, 1991)? Will some G. sepium provenances and strains adapt better in seasonally dry areas with deep rooting patterns (Glover, 1986)? Definitely, these questions need answers that this study on water potential cannot provide. Given the reality of intraspecies competition for water among hedgerow plants which threatens alley crops, but the need for the benefits of hedgerows, the results of this t study suggest that crops must be evaluated in terms of their productivity, threshold for interspecies competition for water, complementary rooting pattern, and response to improving the benefits of hedgerows (i.e. pruning). .Alley crops, however, should be evaluated not in the context of strictly agricultural intercropping, such as corn/bean/squash system (Amador and Gliessman, 1991), but rather on their performance as crops or perennials growing together with contour hedgerows. Through this process, a system may evolve where synergism and complementarity between the hedgerows and intercrops are the norm. 180 Conclusions Based on the results and analysis of t at different ages and hedgerow culture, the following can be concluded: 1. In ‘the short-term, newly-established. G. sepium lhedgerows will compete for water with the intercrops the most if they are planted at very close within-row spacing. Dense hedgerows have high demand for water owing to their large transpiring leaf surface areas. Therefore, intense intraspecies competition for water in the hedgerows will deprive the intercrops of their needed soil moisture. Fully-established G. sepium hedgerows will compete for water with intercrops to a similar degree regardless of the number of hedges per contour line, within-row spacing, and pruning height. Effects of these variables on hedgerow t were sporadically significant; but, they could hardly be predicted. A clear pattern of response in hedgerow t did not emerge as a result of increasing or decreasing within-row spacing and pruning height. It is speculated that the strong interaction effect clouded possible linear relationships between water potential, within-in row spacing, and pruning height. Hedgerows do not only provide benefits that enhance intercrops' productivity, conserve top soil from further erosion, and restore soil fertility. In the process of providing these benefits, the hedgerows' strong intraspecies competition for water partly deprived the intercrops of neededwwater. 'Therefore, I suggest that hedgerows must be dealt with as a competitor, not only as a provider of benefits in an alley cropping system. The hedgerow's high demand for water should be considered in promoting alley cropping with contour hedgerows in seasonally dry areas. 181 Literature Cited Amador MF and SR Gliesmann (1991) An ecological approach to reducing external inputs through the use of intercropping. In: SR Gliessman, ed, Researching the Ecological Basis for Sustainable Agriculture. New York: Springer-Verlag Buck M (1986) Concepts of resource sharing in agroforestry systems. Agroforestry Systems 4:191-203 Cannell MGR (1985) Dry matter partitioning in tree crops. In: MGR Cannel and JE Jackson, eds, Attributes of Trees as Crop Plants. Institute of Terrestrial Ecology, Huntington, UR Dickmann DI (1992) Role of physiology in forest tree improvement. Silva Fennica 25(4):248-256 Dickmann DI, MA Gold and JA Flore (1994) The ideotype concept and the genetic improvement of trees. Plant Breeding Reviews (in press) Glover N (1986) Gliricidia sepium (Jacq) Steud germplasm collection, conservation and evaluation. MS thesis, University of Hawaii, USA Garrity DP (1992) Conservation farming systems for the sloping lands: research and development imperatives. Background paper for the National Symposium of Conservation Farming at Soil Research and Development Center, Quezon City, Philippines, 18 July Garrity DP (1991) Personal communication on the use of pressure chamber and sampling leaves of G. sepium Granert W and W Sabueto (1985) Watershed protection and amelioration. A paper presented at the Workshop on Site Protection and Amelioration Aspects of Agroforestry, University of the Philippines at Los Banos, College, Laguna, Philippines, 4-11 September Harper JL (1977) The influence of density on yield and mortality. In: Population and Biology of Plants. Academic Press, Orlando, Florida Hillel D (1980) Applications of Soil Physics. Academic Press, New York Huck.MG (1984) Water flux in the soil-root continuum. In: Roots, Nutrient and Water Influx, and Plant Growth. Soil Science Society of America, Crop Science Society of America and American Society of Agronomy, 677 South Segoe Road, Madison, WI, USA Huxley PA, T Darnhofer, A Pinney, E Akunda and D Gatama (1989) The tree/crop interface: a project designed to generate experimental methodology. Agroforestry Abstracts 2:127-145 Huxley PA (1983) The tree/crop interface or simplifying the biological/environmental study of mixed cropping agroforestry systems. ICRAF, Nairobi, Renya Rang BT, L Reynolds and AN Atta-krah (1990) Alley farming. Advances in Agronomy 43:315-359 182 Rang BT and GF Wilson (1987) The development of alley cropping as a promising agroforestry technology. In: HA Steppler and PRR Nair, eds, Agroforestry: A Decade of Development, pp 227-243. ICRAF, Nairobi, Renya Rozinka V ( 1989) Uptake and transport of water. In: J Rolek and V Rozinka, eds, Physiology of the Plant Root System. Rluwer Academic Publishers Rramer PJ (1969) Plant and Soil Water Relationships: A Modern Synthesis. McGraw-Hill Book Company Rramer PJ and TT Rozlowski (1979) Physiology of Woody Plants. Academic Press, Orlando, Florida Lal R (1989) Agroforestry systems and soil surface management of a tropical alfisol: I. soil moisture and crop yields. Agroforestry Systems 8:7-29 Laquihon W, J Palmer and H Watson (1991) Sloping agricultural land technology (SALT): a decade of experience on hillside sustainability. A paper presented at the Asian Farming Systems Symposium - 1992, BMICH, Sri Lanka, 2-5 November Laquihon W (1988) Small farm uses of multipurpose tree species in Mindanao. Mindanao Baptist Rural Life Center, Davao del Sur, Philippines Ong CR, MR Rao and M Mathuva (1992) Trees and crops: competition for resources above and below the ground. Agroforestry Today 4:4-5 Ong CR (1991) Interactions of light, water, and nutrients in agroforestry systems. In: ME Avery, MGR Cannell and C Ong, eds, Biophysical Research for Asian Agroforestry. Winrock International and South Asia Books, USA Sajise PE (1985) Some perspective in upland development. A paper presented to the Philippine .Forest Research Society, College, Laguna, Philippines, 21 August Slavik B (1974) Methods of Studying Plant Water Relations. Academia, Prague Solera CR (1992) Competition between upland rice and hedgerow species in an alley cropping system. A PhD thesis, UPLB Graduate School, College, Laguna, Philippines Tacio HD (1993) Sloping agricultural land technology (SALT): a sustainable agroforestry scheme for the uplands . Agroforestry Systems 22:145-152 Watson H (1983) Developing a hillside farming techniques for Mindanao hill farmers. In: Proceedings of the Hilly Land Development Workshop, Cebu City, Philippines, 24-26 March Watson H and W Laquihon (Undated) A manual on how to farm your hilly land without losing your soil. Mindanao Baptist Rural Life Center, Davao del Sur, Philippines Young A (1991) Soil fertility. In: ME Avery, MGR Cannell and C Ong, eds, Biophysical Research in Agroforestry. Winrock International and South Asia Books, USA Chapter 6 Contour hedgerows hold the key to the future of alley cropping in the Philippine uplands. If properly managed, they can fulfill at least four major functions: improve and sustain crop jproductivity through the restoration and improvement of soil fertility via the application of top prunings and turnover of roots; produce forage for livestock production; control soil erosion; and increaseIwater infiltration that would gradually recharge ground acquifers. 'The first two functions would directly benefit resource-limited upland farmers. The last two will profit on-site the upland farmers who cultivate alley farms and off-site the lowland farmers/residents who are the beneficiaries of soil and water conservation practices. These function may be used as a justification by the government to subsidize training of farmers and provide inputs during the initial establishment of contour hedgerows. In this study, I evaluated tree-crop interactions in an alley cropping with Gliricidia sepium hedgerows as the source of pruning biomass for the alley crop. The study was anchored on the rationale that contour hedgerows are established in alley farms to serve the intercrops - simultaneously supplying high volume of green manure and causing minimum competition in the tree/crop interface. Accordingly, hedgerows were established at different number of hedges per contour line and within-row spacings and were subjected to three pruning heights. The growth, pruning biomass, decomposition rate, mortality, root densities, and water potential of the hedgerows were determined. In addition, the yields from the two croppings of maize and mungbean were ascertained. 183 184 The hedgerows from the 5- and 10-cm within-row spacings produced the highest initial clipping and subsequent pruning biomass when out at 30 cm and 2-m, respectively. These hedgerows were the shortest and had the smallest stem diameter; hence, they produce less shade but they suffered the highest mortality, and probably contributed the largest volume of dead roots in the soil. Thus, from the point of view of crop production, densely-planted hedgerows have the greatest potential to provide nutrients and enrich the soil in the alleys. This conclusion can be partly inferred frem the results of periodic soil analysis. The pruning biomass and, possibly, root turnover increased and sustained soil fertility (pH, N, P, R, OM) after two cropping seasons, even in the face of soil erosion after clearing and establishment of the research plot. In short, the densely- planted hedgerows were able to optimize the use of incident light, space, and below-ground resources, produced the highest biomass, and had the least threat of shading the alley crop. The densely-planted hedgerows had the highest root density. In general, root densities were negatively correlated with distance from the hedgerow base up to 50 cm towards the alley and.with soil depth. Ninety percent of roots counted were < l-mm diameter and 70 percent of these were found in the top 30 cm of the soil. This explains why the newly-established hedgerows in the 5- and lO-cm within-row spacings produced the highest volume of biomass and had the most negative water potential. Their intense intraspecies competition forced them to optimize and explore nearby available soil resources for nutrients and water. It was possible that the hedgerows that were producing biomass to enrich the soil were also consuming most of the nutrients that they have contributed in the alleys. Thus, densely-planted hedgerows‘will probably deprive alley crops of soil resources, especially in crop rows planted near the hedges. Unfortunately, densely-planted. hedgerows are also» more effective in controlling soil erosion. 185 While the maize crop yields fromItwo croppings and.mungbean yield from the second cropping were not significantly affected by number of hedges per contour line, within-row spacing, and pruning height, there is suspicion that in established hedgerows, competition will clearly deny alley crops of needed nutrients and moisture, especially in the tree/crop interface. This was partly shown by the lowest mungbean yield obtained from the 5-cm within-row spacing during the first crop. The interaction of number of hedges per contour line, within-row spacing, and pruning height significantly affected the water potential of the 32-month old hedgerows. At this time, however, there were no significant differences in water potential of hedgerows from different within-row spacing, number of hedges per contour line, and pruning height. Nanetheless, there was strong intraspecies competition for moisture in the hedgerows, posing a definite threat to the alley crops. From this, it can be deduced that strong intraspecies competition within hedgerows will increase intercrop competition for limited resources in the alleys. Therefore, reducing intra- and inter-species competition is the ultimate action to optimize the net gain of the alley crops from the hedgerows in addition to benefits from control of erosion. This can only be done by developing a set of "root silvicultural prescriptions" that are geared towards minimizing unfavorable competitive effects of hedgerows on the crop. These prescriptions may include the following: periodic plowing along and at a distance of 50 cm from the hedgerow, application of root growth inhibitors during the critical stage of growth of a high value alley crop, heavy top prunings to check root growth before and during the growth of the crop, and crop fertilization only after root prunings. In the long term, however, the answer to alley farming will be to screen and develop hedgerow and crop ideotypes and use them in breeding new hedgerow cultivars that would fit the requirements and specifications of the system. ‘These ideotypes would co-exist with and complement crop varieties 186 in a highly competitive environment while allowing the hedgerows to perform other environmental services. APPENDIX 187 2 Appendix Table 3.1. Treatments and respective area of sub-plots (m alley). u=-------=---=---=---==: Treatment and respective code ‘ Area (m2 alley) TOP 1 - Two hedges per contour line B3C1- PH of 2-m x 5-cm WRS 11.3 B2C3- PH of 1.5-m x 20-cm WRS 8.8 3104- PH of 1-m x 40-cm was 12.3 ll B3C2- PH of 2-m x 10-cm WRS 12.6 H B3C3- PH of 2-m x 20-cm WRS 13.0 “ 82C4- PH of 1.5-m x 40-cm WRS 12.3 “ BlCl- PH of 1-m x 5-cm WRS 12.4 83C4- PH of 2-m x 40-cm WRS 12.3 BlC3- PH of 1-m x 20-cm WRS 13.6 82C2- PH of 1.5-m x 10-cm WRS 12.5 82Cl- PH of 1.5-m x 5-cm WRS 12.9 BlC2- PH of l-m x 10-cm WRs 12.2 TOP 2 - One hedge per contour line B3C1- PH of 2-m x 5-cm WRS 8.0 BlCl- PH of 1-m x 5-cm WRS 8.3 83C2- PH of 2-m x lO-cm WRS 7.9 83C4- PM of 2-m x 40-cm WRS 8.8 82C4- PH of 1.5-m x 40-cm WRS 10.3 82C3- PM of 1.5-m x 20-cm WRS 10.3 B3C3- PH of 2-m x 20-cm WRS 10.0 ” BZCl- PH of 1.5-m x S-cm WRS 10.0 82C2- PH of 1.5-m x lO-cm WRS 10.0 " BlC4- PH of 1-m x 40-cm WRS 9.8 BlC3- PH of 1-m x 20-cm WRS 8.5 BlC2- PH of 1-m x lO-cm WRS 9.0 n ‘ Sub-plot Control Number 1 4. 13.5 n ' PH 8 Pruning height; WRS 8 Within-row spacing; 188 Appendix Table 3.1. (cont'd). ==-=-=====-==, _ 7 Treatment and respective codel Area (I? alley) MIDDLE 1- One hedge per contour line 83C2- PH of 2-m x 10-cm WRS 10.4 BZC2- PH of 1.5-m x 10-cm WRS 10.0 8303- PH of 2-m x 20-cm WRS 11.0 BBCl- PH of 2-m x 5-cm WRS 8.4 83C4- PH of 2-m x 40-cm WRS 11.4 BlC4- PH of l-m x 40-cm WRS 9.0 BlCl- PH of 1-m x S-cm WRS 12.0 BlC2- PH of 1-m x 10-cm WRS 11.5 82C1- PH of 1.5-m x 5-cm WRS 11.3 82C4- PH of 1.5-m x 40-cm WRS 8.8 BlC3- PH of l-m x 20-cm WRS 10.9 82C3- PH of 1.5-m x 20-cm WRS 10.9 MIDDLE 2- Two hedges per contour line B2C3- PH of 1.5-m x 20-cm WRS . B3C2- PH of 2-m x 10-cm WRS . B2Cl- PH of 1.5-m x 5-cm WRS . BlC4- PH of 1-m x 40-cm WRS 8.1 BlC3- PH of 1-m x 20-cm WRS 5.3 83C1- PH of 2-m x 5-cm WRS 5.2 83C4- PH of 2-m x 40-cm WRS 5.5 B3C3- PM of 2-m x 20-cm WRS 6.8 BZC4- PH of 1.5-m x 40-cm WRS 7.4 BlCl- PH of 1-m x 5-cm WRS 6.9 8202- PH Of 1.5-m x 10-cm WRS 7.0 BlC2- PH of 1-m x 10-cm WRS 5.3 ._ 9b'219t19°“tF9£ “Faber % 12-3 Appendix Table 3.1. (cont'd). 189 H Treatment and respective code‘ Area (m2 alley) I Barton 1- Two hedges per contour line 8101- PH of 1-m x 5-cm WRS 12.3 B2C3- PH of 1.5-m x 20-cm WRS 11.8 B2C1- PH of 1.5-m x 5-cm WRS 10.5 I BlC2- PH of 1-m x 10-cm WRS 13.9 H BlC4- PH of 1-m x 40-cm WRS 11.7 I B3C1- PH of 2-m x S-cm WRS 11.0 B3C4- PH of 2-m x.40-cm WRS 12.0 I B2C4- PH of 1.5-m x 40-cm WRS 11.0 B2C2- PH of 1.5-m x 10-cm WRS 10.0 BlC3- PH of l-m x 20-cm WRS 10.0 H B3C2- PH of 1.5-m x 10-cm WRS 10.0 H B3C3- PH of 1.5-m x 20-cm WRS 8.8 BOTTOM 2- One hedge per contour line BlC3- PH of 1-m x 20-cm WRS 10.0 83C4- PH of 2-m x 40-cm WRS 14.4 83C2- PH of 2-m x lO-cm WRS 14.3 B2C4- PH of 1.5-m x 40-cm WRS 12.2 B3C3- PH of 2-m x 20-cm WRS 10.0 H BlC2- PH of 1-m x 10-cm WRS 10.7 H B3C1- PH of 1.5-m x 5-cm WRS 13.0 H BlC4- PH of 1-m x 40-cm WRS 12.6 " B2C2- PH of 1.5-m x 10-cm WRS 10.3 BlCl- PH of l-m x 5-cm WRS 9.5 82C3- PM of 1.5-m x 20-cm WRS 11.0 BZCl- PH of 1.5-m x 5-cm WRS 10.4 ' PH 8 Pruning height; WRS 8 Within-row spacing 190 Appendix Table 3.2. Coefficient of variations (CVs) in the analyses of variance of different above-ground parameters. Parameter Coefficient of Variation (8) Six-mo height of hedgerows (m) 4.76 I 12-mo height of hedgerows (m) 5.70 HSix-mo diameter of hedgerows (cm) 3.65 12-mo diameter of hedgerows (cm) 7.46 H 18-mo diameter of hedgerows (cm) 10.10 H Six to 12-mo % mortality 57.37 12 to 18-mo % mortality?) 105.18 Initial clipping (dry kg/mFIalley) 19.42 First pruning (dry kg/mF alley) 27.54 Second pruning (dry kgjnP alley) 26.05 H Shelled maize from first crop (g/mfi 20.25 Stover of maize from first crop (g/mfi 16.40 lShelled maize from second crop (g/mfl 21.86 Stover of maize from second crop (g/mfi 18.37 HMungbean yield from first crop (g/m’) 10.02 H Mungbean yield from second cropEr (g/m’) 29.73