Effect of shade on banana morphology, growth and production

SCIENTIA HORTlCULTuM Scientia Horticulturae62 ( 1995) 45-56 zyxwvutsrqponmlkjihgfedcbaZYXWVU Effect of shade on banana morphology, growth and production Y. Israeli”~“, Z. Plautb, A. Schwartzc zyxwvutsrqponmlkjihgfedc “Jordan Valley Banana Research Station, Zemach 15132, Israel bAgricultural Research Organization, The Volcani Center, Bet Dagan 50250, Israel “The Hebrew University of Jerusalem, Rehovot 76100, Israel Accepted9 December 1994 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSR Abstract The effect of three shade levels on morphology, growth and productivity of ‘Grand Nain’ (AAA) bananas during their first and second production cycles was studied in the Jordan Valley, Israel. In vitro propagated plants that were planted in the field in April 1990 were shaded with black Saran screens of different densities, installed above the canopy level. The resultant photosynthetic photon flux density (PPFD) was reduced to 80%, 60% or 30% of the unshaded control. Although only the heaviest shade affected plant vegetative growth in the first cycle, bunch weight was reduced by 7% and 32% under medium and heavy shade, respectively. A highly significant effect on vegetative growth and production was observed during the second cycle. Flowering date was delayed by 6 days, 9 days and 15 days, and bunch weight was reduced by 8%, 21% and 55% under light, medium and heavy shade, respectively. Yield was reduced by all levels of shade, owing to the combined effect of reduced bunch weight and a lower stand. Shading reduced the rate of leaf emergence, leaf and foliage area, plant height and pseudostem circumference. The leaves had thinner laminae, with a reduced number of stomata and higher chlorophyll content. Our observations indicate a significant effect of long-term shade on bananas, and the utilization of high levels of PPFD by the banana plant. Ke.vwords:Banana; Banana morphology;Banana yield; Chlorophyll: Irradiance; Shade; Stomata 1. Introduction Most wild banana species grow best in the open sun, as long as water is not a limiting factor (Simmonds, 1962). Under deep shade, growth is restricted and ultimately the plants die out. Most commercial banana production takes place in the tropics, where dense vege* Corresponding author. 0304-4238/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved SSDIO304-4238(95)00763-6 46 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Y. Israeli et al. /Scientia Horticulturae 62 (1995) 45- 56 tation and cloud cover affect the amount of incident radiation. However, the information available on the effect of light intensity on banana growth and production is scarce, although much has been published on responses to increased density, where mutual shading is involved. There is only one report available involving a controlled study of banana growth under different shade levels (Murray, 1961). In that experiment ‘Dwarf Cavendish’ bananas, interplanted with cacao plants at a wide spacing of 3 m X 6 m, were grown in Trinidad under shade which reduced incident light to 70%, 50% or 20% of full (100%) sunlight. Surprisingly, bunch weight was not affected, even at the heaviest shading. Rate of development, as indicated by planting to harvest time, was slower at 20% of full sunlight, but under 50% shade, neither production nor rate of development were affected. Stover ( 1984) observed that when the density of a commercial banana plantation is high and light transmitted to the understory is reduced to 10% of the above-canopy intensity, growth and production of the plants are severely affected. He also reported that in a tropical climate, rate of flowering declines significantly some 6 months after a period of low insolation (Stover and Simmonds, 1987), Robinson and Nel ( 1988,1989) observed a prolonged cycle time and a decrease in bunch mass under increased plantation density in cv. ‘Williams’ growing in a subtropical climate. They proposed that the reduction in incident light to the secondary canopy contributes to these effects. In this experiment, we investigated the effect of reducing the incident light to the primary canopy to 80%, 60% or 30% of full ( 100%) sunlight. Our main purpose was to simulate the effect of naturally reduced irradiance caused by long periods of cloudiness. However, shading could also be used as a management technique, e.g. to reduce frost damage, decrease water consumption or delay fruit maturation. The effect on plant morphology, growth and production during the first and second cycles is reported here. The effect on carbon assimilation and gas exchange characteristics will be reported separately. 2. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Materials and methods In vitro propagated ‘Grand Nain’ (AAA) banana plants were planted in April 1990 in the Jordan Valley, Israel (32”43’N, 35”36’E; 200 m below sea-level), in alluvial mineral soil. Mean daily minimum and maximum temperatures for the coldest month, January, are 9.1”C and 17.6”C, respectively, and for the hottest month, July, 22.6”C and 37.0°C, respectively. The main rainy season is December-March, with an average annual rainfall of 386 mm. Spacing was 3 m X 2.8 m, two bearing pseudostems per mat (a common commercially used spacing), resulting in a density of 2381 plants ha-‘. The plantation was subjected to standard agronomic practices, which included the incorporation of 200 m3 ha-’ cattle manure and 1200 kg ha-’ superphosphate into the soil before planting, and an annual application of 6 kg mat- ’ pelletized chicken manure. Both nitrogen (400 kg N ha- ’ year- ’ as ammonium nitrate) and phosphorus (50 kg P,O, ha- ’ year- ’ as phosphoric acid) were applied via the drip irrigation system. No special application of potassium was necessary, as the soil in the Jordan Valley is rich in this element. The plantation was irrigated daily during the dry period (from April to November) with a total application of about 2500 mm; Y. Israeli et al. /Scientia Horticulturae 62 (1995) 45- 56 300 2w I” E 240 f22a 200 b g % 70 - -, *,I- 2nd cycle - - -1 :- I b .’ L’ zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA iso % ,a i C 2nd cycle 2.0 ,.F----L.-----I F - zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 1.8 /’ /’ t 3 1.6 1.4 1.2 PI 40 lrradiice so 60 70 60 90100 (% of unshaded control) Fig. 1. Effect of reduced irradiance on (a) plant height (b) pseudostem circumference and (c) the area of the plant’s largest leaf at flowering. Plant height was measured from the pseudostem base to the bunch axis arch. Pseudostem circumference was measured at a height of 1 m from soil level. Leaf area was calculated from leaf length X leaf width X 0.83. The plant’s largest leaf was the third to fifth youngest leaf (spade leaf not included). Fig. 2. Effect of reduced irradiance on (a) zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFE leaf thickness (b) specific leaf weight and (c) stomatal density. Leaf thickness was measured with a micrometer at the central part of the lamina of the third to fifth youngest leaf, near the central vein. Leaves were measured near time of harvesting of second-cycle fruit. Samples from the same leaves were used to obtain leaf specific dry weight. Stomatal density was measured on the upper (adaxial) and lower (abaxial) leaf surface. The central part of the lamina of the third to fifth youngest leaf was sampled during the vegative stage of the first cycle. 48 Y. Israeli et al. /Scientia Horticulturae 62 (199.5) 45-56 class A pan evaporation for the same period was about 1500 mm. This resulted in low soilwater tension, ensuring that significant water stress was not a major limiting factor. Shading treatments were begun in July 1990, after an initial 3 month period of establishment. Black Saran screens of different densities were installed above the canopy at a height of 4.2 m, covering plots of 12 m X 20 m containing five banana rows per plot. Growth and production data were obtained from the internal three rows (30 plants per plot), whereas physiological observations were performed on plants of the central row only ( 10 plants per plot). The different densities of the Saran screens created different rates of shading. The actual rate of photosynthetic photon flux density (PPFD) under the screens was measured with a Li-190SA quantum sensor (Li-Cor, Lincoln, NE, USA) positioned horizontally at the level of the upper leaves. These measurements were also repeated routinely during gas-exchange measurements. The results indicated a transmission of 81%, 62% and 32% of available photosynthetic active radiation ( 100%) during the period of maximum solar radiation, with very small variation, for the light, moderate and heavy shades, respectively. Values of 80%, 60% and 30% are used for simplicity. The shade treatments were replicated three times in a randomized block design, except for the heaviest shade treatment, for which only one plot was established. Statistical analysis included analysis of variance and Duncan’s multiple range test to compare means of replicated treatments, and Student’s t-test to compare the mean of the population from the nonreplicated treatment with the means of the other treatments (Statistical Analysis Systems Institute, Inc., 1987). Plant height, pseudostem circumference and leaf emergence were recorded monthly. Leaf width and length were measured and leaf area was calculated using the formula length X width X 0.83 (Summerville, 1944). Leaf samples were taken from the central part of the lamina of the third to fifth youngest upper leaf from five plants per replication before fruit harvest. Ten 1 cm* discs representing the entire lamina width were punched, and used to obtain specific leaf weight (SLW) and chlorophyll content. Discs for chlorophyll extraction were kept in 98% ethanol, and the chlorophyll was extracted and its level determined according to Wintermans and De Mots ( 1965). Discs for SLW determination were wrapped with aluminum foil to minimize water loss, until fresh weight was determined. Dry weight was obtained after oven-drying for 48 h at 70°C. Stomata were counted on the upper and lower surfaces of the central part of the lamina of the third to fifth youngest leaf. A negative print of the epidermis was obtained by applying a thin layer of room temperature vulcanized silicone (RVT) adhesive. This was peeled off and kept for future reference. For microscopic observation, a positive print was obtained with a transparent polish. The number of stomata was counted under five microscopic observation fields on each side of the leaf surface. Bunch weight was recorded at harvest, and the central finger of the inner whorl of the third basal hand was sampled to record its fresh weight, length and circumference. Yield was calculated from the sum of gross weight of all bunches harvested from each experimental plot. In Israel, where most of the fruit is marketed as whole bunches, this gives a close approximation of commercial yield. Y. Israeli et al. /Scientia Horticulturae 62 (1995) 45-56 49 3. Results zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 3.1. Plant morphology and leaf chlorophy ll content Plant height, pseudostem circumference and leaf area (Fig. 1) were not affected by reducing growth irradiance to 80% or 60% of full sunlight, but decreased significantly when light was further reduced to 30% of full sunlight. A slight decrease in plant height was also noted in the plants grown under 60% of available sunlight in the second cycle (Fig. 1(a) ) . Of these characteristics, the strongest relative effect was recorded for plant height, which was reduced by 16% and 18% under the heaviest shade, during the first and second cycles, respectively, as compared with the unshaded control. Leaf thickness was reduced in an almost linear fashion under reduced irradiance, as was specific leaf weight (Figs. 2(a) and 2(b) ) . In both cases, the difference between 100% and 60% or 30% irradiance was statistically significant. Stomatal density was reduced under the heaviest shade, especially on the adaxial leaf surface (Fig. 2(c) ) . Leaf chlorophyll content increased with reduced h-radiance (Table 1) . 3.2. Rate of growth Date of flowering was not affected during the first cycle. It was delayed, however, by 6 days, 9 days or 15 days with growth irradiances of 80%, 60% and 30% of the control, respectively, during the second cycle (Fig. 3). Rate of growth in height was reduced during the second cycle by the heaviest shade (Fig. 4), but a major portion of this height difference was already evident at an early stage of sucker growth, suggesting a strong effect of the time of sucker initiation. Rate of leaf emergence was also affected. The number of leaves emerging during the 2 months between start of treatments and first-cycle flowering was 10.9 in full sunlight and only 9.4 in 30% of full sunlight, a statistically significant difference of 1.5 leaves. An accumulated difference of 3.8 leaves between the 30% treatment and the unshaded control was recorded during the whole period of sucker growth during the second cycle. A slower rate of leaf emergence was also recorded for the other shade levels, but the differences were less significant (data not shown). Table 1 Effect of reduced irradiance on leaf chlorophyll content (CIU, mg g -’ dry weight) during the first and second growth cycles Cm (mg g-l) b-radiance (96) of full sunlight) 100 First cycle Second cycle 6.0c 9.3c 60 30 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 80 6.6bc 10.P 7.4b 12.3” 9.7” 11.1* Samples were taken from the central part of the lamina of the third to fifth youngest leaf. Mean separation by Student’s t-test, P 2 0.05. Different letters in a row indicate statistically significant differences. 50 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Y. Israeli et al. /Scientia Horticulturae 62 (1995) 45- 56 zyxwvutsrqponmlkjihgfedcbaZY O fo c Tr II al Ii ‘c1 .rOlSEP $ g K 1st cycle l---.%._.I . - 2nd 30 40 50 lrradiance 60 .. -‘-.I.._ 70 80 (% of unshaded cycl&I 90 100 control) zyxwvutsrqponmlkjihgfedcbaZYXW Fig. 3. Flowering date, as affected by different levels of irradiance. The first cycle flowered in 1990, the second in 1991. 3.3. Fruit production Reduced irradiance resulted in a significant decrease in fruit production, especially during the second cycle. In the first cycle, bunch weight was not affected when irradiance was reduced to 80% of full sunlight. However, a reduction to 60% of full sunlight resulted in a zyxwvutsrqpo 30C I-- 250 l309 g E .o, 200l- P 150l- 100 03MAFf OBMAY 27JUN 16AUG Date (1991) Fig. 4. Growth in height of plants during the second cycle as affected by different levels of irradiance. Height of plants at the vegetative stage was measured from the base of the pseudostem to the intersection of the petioles of the two youngest leaves. Y. Israeli et al. /Scientia Horticulturae 62 (1995) 45- 56 51 40. 2nd cycle 8 G5 30 E .rs, ; 201 _-:::-1::ln*.. $ lo- m’ 30 40 50 60 70 80 90 100 lrradiance (% of unshaded control) Fig. 5. Bunch fresh weight as affected by different levels of irmdiance. 7% decrease in bunch weight, and a reduction to 30% of full sunlight resulted in a decrease of 32% in bunch weight (Fig. 5). During the second cycle, decreases of 8%, 21% and 55% in bunch weight were noted under it-radiances of 80%, 60% and 30% of the unshaded control, respectively. The effects of 60% and 30% available sunlight on bunch weight were statistically significant in both the first and second cycles. Reduced bunch weight was mainly the result of reduced finger weight, as the number of hands was only slightly affected (Figs. 6 and 7). Finger length was influenced mainly by the highest shade level. The combined effect of shade on finger weight and finger length 14.0v) 13.0IS c” 12.0. 2nd cycle __-. 1 _ -I - /.-.~----.-.~ I 9.07 1st cycle 8.07.0- I ’ II 30 40 1 ’ ,I, 50 60 1 I ’ ,a, 70 80 90 100 lrradiance (% of unshaded control) Fig. 6. Number of hands per bunch as affected by different levels of irradiance. 52 Y. Israeli et al. /Scientia Horticulturae 62 (1995) 45-56 151 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLK 90 40 50 60 70 80 90 lo o Im a dia nw (96 o f unsha de d c m tro l) Fig. 7. Finger fresh weight and length as affected by different levels of irradiance. The middle finger of the inner whorl of the third basal hand was measured at time of harvesting. resulted in lower fruit quality. Fruit grown under the heaviest shade had no commercial value. Plant density declined under the heaviest shade during the second cycle (Table 2). This was caused by poor sucker initiation and slow sucker growth; some of these suckers failed to flower before winter (winter flowering gives noncommercial bunches in Israel), In a few cases, choked bunches that failed to emerge were observed. The occurrence of these abnormalities was much more frequent in the third cycle, where they were also observed under moderate shade (60% of full sunlight). Decreased bunch weight and lower density resulted in a pronounced decrease in yield with reduced irradiance (Table 2). A significant difference in yield was even obtained when differences in bunch weight alone or density alone were not significant. Yield was reduced in the second cycle by about 12 t ha-’ ( 15%) under light shade, by 19 t ha-’ (24%) under moderate shade, and by 49 t ha-’ (61%) under the heaviest shade, as compared with the unshaded control. Harvest time was delayed, and the flowering-to-harvest interval increased with reduced irradiance (Figs. 8 and 9). This latter increase was already noticeable in the first cycle, but was more obvious in the second cycle. Y. Israeli et al. /Scientia Horticulturae 62 (1995) 45-56 53 Table 2 Effect of reduced b-radiance on plantation density and yield in plant crop (first cycle) and first ratoon (second cycle) Irradiance (% of full sunlight) 100 80 60 30’ 2280” 52.9a 2220” 52.0a 2130” 45.7b 2220 35.1 2350 80.68 2200” 68.9b 2280a 61.3b 2060 31.6 First cycle Density (plants ha-‘) Yield (t ha-‘) Second cycle Density (plants ha- ’ ) Yield (t ha- ’ ) Treatment means separation by Duncan’s multiple range test. Different lettersin a row indicate statistically significant differences. ‘The.30% light treatment was not analyzed statistically, as this treatment was not replicated. zyxwvutsrqponmlkjihgfedcb 4. Discussion A pronounced effect of reduced growth irradiance on banana plant morphology, anatomy, rate of growth and fruit production was found in this study. The much stronger effect during the second cycle was expected, as the shading treatments were begun only 2 months before flowering of the first-cycle plants. The second-cycle suckers emerged after the shade treatments had begun and were therefore fully influenced by these treatments. Thinner leaves (Fig. 2(a) ) , lower dry mass per unit leaf area (Fig. 2(b) ), fewer stomata, especially on the adaxial surface (Fig. 2(c) ) and a higher chlorophyll content (Table 1) are all wellknown plant responses to partial shading (Boardman, 1977; Bjorkman, 1981; Givnish, 1988). Some of these changes have been reported for bananas grown under the canopy of arecanut palms (Balasimha, 1989). Changes in single-leaf or foliage area and stem size in response to shading are more specific to the species involved. Increased leaf area in response to shade has been reported in a few tropical foliage plants (Fonteno and McWilliams, 1978) and in zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Ficus benjamina (Conover and Poole, 1977). No change in leaf area was induced by reduced irradiance in eggplant, lettuce, soybean or sweet potato (Wolff and Coltman, 1989). A decrease in leaf zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA area was noted in peanuts (Wolff and Coltman, 1989) and papaya (Buisson and Lee, 1993). In banana, we observed a decrease in leaf area, plant height and pseudostem circumference, especially under the heaviest shade (Fig. 1)) and a delay in the second-cycle flowering date at all shading levels, indicating a reduced rate of growth (Fig. 3). Reduced growth, shorter internodes and decreased stem diameter have also been recorded for papaya plants grown under artificial shade (Buisson and Lee, 1993). Murray ( 1961) reported shorter, compressed banana plants under heavy shade (only 20% of full sunlight). However, he did not observe a clear effect on growth rate. Our findings are in accordance with those of Stover ( 1984), who observed a suppression in sucker growth in the understory of a commercial plantation of very high density, and with those of Stover and Simmonds ( 1987)) who reported a marked effect of solar radiation on the rate of flowering. 54 Y, Israeli et al. /Scientia Horticulturae 62 (1995) 45-56 IOMAR ~ :wzl a IOFEB -0 ‘. t ‘. 5 2I ‘. ‘. j, ‘1 I” IOJAN ‘1 zyxwvutsrqponmlkjihgfedcbaZYXW ‘I 2nd cycle , lODEC30 I ’ 40 ’ I 60 50 I ’ 70 I ’ 80 I ’ 90 I 100 lrradiance (% of unshaded control) Fig. 8. Effect of different levels of irradiance on harvest date. First-cycle fruit was harvested in spring 1991, and second-cycle fruit at the beginning of 1992. 180- 1 i: \;;Ie, r” ‘.\, I .F 140- 6 h 1301 2nd cycle , 30 I’ 40 50 I ’ I ’ I ’ I ‘I 60 70 80 90 100 lrradiance (?A of unshaded control) Fig. 9. Effect of different levels of irradiance on the flowering-to-harvest interval. Robinson and Nel ( 1988) observed a significant decrease in leaf emergence rate and increase in cycle duration when density was increased from 1000 plants ha-’ to 2222 plants ha-’ and radiation available to the secondary canopy was reduced to 14% of full sunlight. A slower rate of leaf emergence was also indicated in our study. In Israel, bananas produce only one crop per year (and cycle). Date of flowering, however, is variable and highly indicative with regard to rate of growth and development. The delay in flowering date observed in our study is therefore comparable with the increased cycle interval under higher densities observed by Robinson and Nel ( 1988). Y. Israeli et al. /Scientia Horticulturae 62 (1995) 45-56 55 A most significant effect of irradiance on production, as indicated by total yield, bunch weight, and single fruit size, was observed in our experiment (Figs. 5,7 and Table 2). This is in contrast to the observations of Murray ( 1961), where extreme shade (20% of full sunlight available, as compared with 30% of available sunlight under the heaviest shade in our experiment) did not result in a statistically significant reduction in bunch weight. This disagreement could be explained, as already mentioned by the author himself, by the wide spacing used in his experiment, with practically no mutual shading, and by the fact that only first-cycle results were recorded. A decrease in bunch mass and finger size under increased density was reported by Robinson and Nel ( 1989). A response in bunch mass was also obtained under different spatial arrangements (Robinson et al., 1989)) indicating the sensitivity of this parameter to light conditions. A decrease in yield when available light was reduced to 70% of full sunlight has also been reported for eggplant, soybean, peanuts and sweet potato in Hawaii (Wolff and Coltman, 1989)) whereas lettuce production was not affected by this mild shading. It should be noted that the reduced bunch weight under reduced irradiance in our study was primarily the result of reduced carbon assimilation and/or allocation to the fruit or developing suckers. Morphogenetic changes, such as the number of hands per bunch, were relatively small (Fig. 6). An increase in the flowering-to-harvest interval under shade was noted in our experiment (Fig. 9) and has also been reported by Murray ( 1961) . This may also be the result of a poor supply of photosynthates to the developing fruit. Our research indicates that even a relatively small decrease in growth irradiance, to 80% of full sunlight, may result in some decrease in production and fruit size. An important question is whether these results are specific to the conditions under which our study was performed, or if they are valid for a wide range of environments. Seasonality is an important characteristic of banana growth and production in Israel (Oppenheimer, 1960; Ticho, 1970; Israeli and Blumenfeld, 1985). The summer is a period of rapid growth and development, whereas complete cessation of vegetative growth occurs during the winter. Solar radiation measurements at the Jordan Valley Banana Experiment Station indicate a peak of about 25 MJ m-* day-’ during June-July, a minimum 5 MJ m-* day-’ in January, and an annual average of about 16 MJ m-* day-’ (Y. Israeli, unpublished data, 1992). This later value is very close to those obtained in the tropics (Stover and Simmonds, 1987). Mutual shading in our experiment was high. The initial density of 2381 plants ha-’ was significantly higher than that of 1840-1900 plants ha-’ commonly used for ‘Grand Nain’ in the tropics. Also, the method of planting and growing two plants (doubles) per mat may have further contributed to mutual shading. These differences should be considered if the results of this experiment are to be extrapolated to different cultivation systems. Our study indicates the utilization of a high rate of solar radiation by banana plants, although the threshold value and the magnitude of the effect of reduced irradiance might change in accordance with local conditions. Limiting photosynthetic activity might be of crucial importance for banana in the subtropics, where the growth period is limited. However, other environmental factors, especially temperature, may also play an important role. In the tropics, light may be the main limiting factor when others, especially water and nutrient supply, are near optimal levels. 56 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Y. Israeli et al. /Scientia Horticulturae 62 (1995) 45-56 References Balasimha, D., 1989. Light penetration patterns through amcanut canopy and leaf physiological characteristics of intercrops. J. Plant. Crops, 16: 61-67. Bjorkman, 0.. 1981. Responses to different quantum flux densities. In: A. Pirson and M.H. Zimmerman (Editors), Encyclopedia of Plant Physiology, New Series, Vol. 12A. Springer, Berlin, pp. 57-108. Boardman, N.K., 1977. Comparative photosynthesis of sun and shade plants. Annu. Rev. Plant Physiol., 28: 355377. Buisson, D. and Lee, D.W., 1993. The developmental responses of papaya leaves to simulated canopy shade. Am. J. Bot., 80: 947-952. Conover, C.A. and Poole, R.T., 1977. Effects of cultural practices on acclimatization of Ficus benjamina L. J. Am. Sot. Hortic. Sci., 102: 529-531. Fonteno, W.C. and McWilliams, E.L., 1978. Light compensation points and acclimatization of zyxwvutsrqponmlkjihgf four tropical foliage plants. J. Am. Sot. Hortic. Sci., 103: 52-56. Givnish, T.J., 1988. Adaptation to sun and shade: a whole-plant perspective. In: J.R. Evans, S. von Caemmerer and W.W. Adams (Editors), Ecology of Photosynthesis in Sun and Shade. CSIRO, Melbourne, Australia, pp. 63-92. Israeli, Y. and Blumenfeld, A., 1985. Musa. In: A. Halevy (Editor), Handbook of Flowering, Vol. 3. CRC Press, Boca Raton, FL, pp. 390-409. Murray, D.B., 1961. Shade and fertilizer relations in the banana. Trop. Agric., 38: 123-132. Oppenheimer, C., 1960. The influence of climatic factors on banana growing in Israel. Publication, Series No. 350-B, National Univ. Inst. Agric., Rehovot, 8 pp. Robinson, J.C. and Nel, D.J., 1988. Plant density studies with banana (cv. Williams) in a subtropical climate. I. Vegetative morphology, phenology and plantation microclimate. J. Hortic. Sci., 63: 303-313. Robinson, J.C. and Nel, D.J., 1989. Plant density studies with banana (cv. Williams) in a subtropical climate. II. Components of yield and seasonal distribution. J. Hortic. Sci., 64: 21 l-222. Robinson, J.C., Nel, D.J. and Bower, J.P., 1989. Plant density studies with banana (cv. Williams) in a subtropical climate. III. The influence of spatial arrangement. J. Hortic. Sci., 64: 513-519. Statistical Analysis Systems Institute, 1987. SAS/STAT Guide for Personal Computers, Version 6 edition. Statistical Analysis Systems Institute, Inc., Cary, NC, 1028 pp. Simmonds, N.W., 1962. The Evolution of the Bananas. Longmans, London, 170 pp. Stover, R.H., 1984. Canopy management in Valery and Grand Nain using leaf area index and photosynthetically active radiation measurements. Fruits, 39: 89-93. Stover, R.H. and Simmonds, N.W., 1987. Bananas. Longman, Singapore, 468 pp. Summerville, W.A.T., 1944. Studies on nutrition as qualified by development in Muss cauendishii L. Queensl. J. Agric. Sci., 1: 1-127. Ticho, R.T., 1970. The banana industry in Israel. Trop. Sci., 13: 289-301. Wintermans, J.F.G.M. and de Mots, A., 1965. Spectrophotometric characteristics of chlorophylls a and b and their pheophytins in ethanol. Biochim. Biophys. Acta, 109: 448453. Wolff, X.Y. and Coltman, R.R., 1989. Productivity under shade in Hawaii of five crops grown as vegetables in the Tropics. J. Am. Sot. Hortic. Sci., 115: 175-181.