CHANGE IN GROWTH AND YIELD CHARACTERISTICS OF MUSA CV MYSORE ( " EMBUL " BANANA ) WITH INCREASING AGE

Banana (Musa spp.) is a major semi-perennial fruit crop in Sri Lanka. A study was conducted at the Ftegional Agricultural Research and Development Centre, Angunakolapelessa to investigate the effect of age on growth and yield of banana cv Mysore ("Embul", AAB group). Thirty-seven "Embul" banana accessions collected from various ecological regions of Sri Lanka were studied in a Randomized Complete Block Design with three replicates. The observations were recorded in three consecutive years. Number of leaves per clump, LA1 (leaf area index) and penetration of photosynthetically active radiation (PAR) to the crop were recorded a t 4,9,20 and 32 months after planting. Pseudostem height and girth of the crop were investigated a t different ages. Number of combs and fruits per comb, fruit maturity period, bunch weight, total fruit weight, average fruit weight and peduncle weight were also measured. Comb weight, fruit circumference, fruit length and diameter were recorded in the second and last combs. Leaf number and W increased, while penetration of PAR decreased with the age. Except weight of second comb, fruit length in second comb and fruit diameter in last comb, all other characteristics varied significantly with age of the crop. Fruit maturity period increased with the age of the crop to second year and then decreased, while total fruit weight decreased up to second year and then increased. Number of fruits per comb and bunch weight significantly increased after the second year. Stem height and girth, number of fruits per bunch, weight of the peduncle and length of the male axis consistently increased with the age of the crop investigated up to the third year.


INTRODUCTION
Banana (Musa spp.) is a major semi-perennial fruit crop in Sri Lanka.There are about 27-28 cultivars grown in the c0untry.lNational annual production of 731,000 m.tons was recorded from a n estimated 44,578 ha of banana cultivated throughout the c o ~n t r y .~ Cultivar Mysore ("Embul", AAB group) is most popular, yields well, has good keeping qualities, and can be cultivated in most agro-ecological zones.The life-span of banana crop mainly depends upon the level of crop management.Some growers in ~r i Lanka harvest only the first bunch of the crop and no suckers are allowed to grow for subsequent crops.
However, majority of the growers in Sri Lanka maintain their crop for many years depending upon management conditions.The gross yields of banana has been found highest in the plant crop and decreased in ratoon crops to a constant level.l Thirty-five percent yield reduction in first ratoon over the plant crop was observed in '.'Dwarf Cavendish" banana in French Guinea.ls3It is not possible to maintain a crop for long periods, if the harvest price is dependant on the grade of the fruits since this declines with the age of the crop.'However, the effect of the crop's age on the yield and quality of banana has not been experimentally assessed and causes for such effects have also not been identified.There are no reports available on the yield parameters affected by the age of the crop in local cvs of Sri Lanka.Therefore, the main objective ofthe present study was to carry out investigations on the effect of crop's age on yield parameters of "Mysore" banana in Sri Lanka.

METHODS AND MATERIALS
Crop Management: The experiment was conducted a t the Regional Agricultural Research and Development Centre, Angunakolapelssa, Sri Lanka (1200 mm average annual total rainfall and 30 m elevation) during the period 1992 to 1995.Thirty-seven accessions of Musa cv Mysore with diverse morphological characters were selected and multiplied in the field using naturally emerging suckers.The accessions collected were grown in a Randomized Complete Block Design with three replicates.Sword suckers with the height of 1.5 -2.0 m, uniform in size and age were used as planting materials.Blades of broad leaves of suckers and all the dead portions of the rhizomes were removed.The rhizomes were treated with fungicide "Dithane" mixed with dry wood-ash.Treated suckers were kept under shade for about 24 h before planting.The suckers were planted in July 1992 in planting holes (45 cm x 45 cm x 45 cm) with distance of 3 m both within and between rows.Each planting hole was filled with well decomposed cow-dung mixed with top soil.600 g of dolomite were applied to each planting hole as recommended by the Department of Agriculture, Sri Lanka, as calcium and magnesium source.Bana;na cultivar Seenikesel (Pisang awak) plants were established around the experimental area to avoid border effects and wind damages.15 g of carbofuran (3% G.) was applied to each planting hole at the time of planting to control banana nematode Radopholus sirnilis.Irrigation was done once in ten days satisfying the field capacity of the soil depth a t 75 cm when dry climatic conditionprevailedmore than 10 d.Fertilizer mixture 12: 8: 34: (N: P:K) recommended by the Department of Agriculture was applied at the rate of 450 g per clump, two months after planting.This was repeated at four months intervals.Fertilizer was applied, 45 cm radius around the clump a t 3-5 cm depth of the soil.First daughter sucker was allowed to emerge a t four months of planting and the second at flowering of the plant crop (8-9 months).The third one was allowed to emerge at the time when the bunch matured (10-12 months).Subsequently, one additional sucker was allowed to emerge whenever the mother plant was removed from the clump because of harvesting the bunch.Therefore the clump consisted of one plant until 4 months of planting, two plants from 4-9 months and three plants at 12 months onwards.Thereafter, three plants in a clump with an age difference of 3-4 months were maintained.
Growth Cl~aracters: The time of planting, time of flowering and fruit maturity in each plant of the clump were recorded with the growth of the crop.The number of leaves per clump was recorded a t 4,9,20 and 32 months of planting of the crop.The number of plants available for leaf measurements was one at four months, two at nine months, three at the end ofthe first year and thereafter.The leaf area was calculated from the length of leaf blade using the method described previously.'14 Therefore, leaf blade lengths were recorded in their order of insertion to estimate leaf areain eachindividual leaf.The total leafarea was estimated for the clump and then for each block of banana separately.The total leaf area in each block was divided by block area to compute leaf areaindex (LAI).The leaf-blade length was measured from distal end to base of the leaf blade at 4,9,20 and 32 months of planting, with a light wooden stick calibrated in cm.The pseudostem height from ground level to base of the leaf petioles and the pseudostem girth a t 10 cm above ground level were measured, at the time of bunch harvesting.
Fruit Measurements: Harvesting was done a t the time when one fruit ofthe first or second comb turned green to yellow.Bunches with male axis and male bud were cut at 60 cm ofpeduncle before the first comb.After harvesting, combs were counted and removed from peduncle and tagged to identify their sequence in the bunch.Fruit clusters consisting of more than four filled fruits were considered as combs.The weight of the second comb and the last comb in each bunch was also recorded.The total fruit weight was taken by weighing the combs.The bunch weight was measured excluding male axis and male bud.Number of fruits per bunch was also counted.The weight ofthe peduncle was measuredincluding the male axis.Fruit circumference was measured a t the middle of the fruit by wrapping a measuring tape around it.The fruit diameter was measured in the cross section taken at the middle of the fruit.This measurement was .taken in three different directions of the cross section ofthe fruit using a Vernier calliper and the mean value was calculated.Fruit length was measured after removing fruit stalk and the bottle-neck part in distal end of the fruit.All the fruit measurements were taken for three fruits in a comb.Fruit measurements were taken in second and last combs in each bunch.AI1 .thegrowth and fruit measurements were taken for the fruit bunches harvested during 1993 to 1995 and average value for each replicate was presented for the statistical analyses.Data were analysed using the F test at significance range of 0.1-5%.

Physiological Measurements:
The daily measurements of sunshine hourswere taken during the experimental period from the Meteorological Unit, Regional Agricultural Research and Development Centre, Angunakolapelessa and presented as daily mean per month (Table 1).These measurements were taken on sunshine scales (Casella,'London) with the recorder model MI8068 (Negretti, Zambra, London).Photosynthetically active radiation (PAR) was also measured at 4,9,20 and 32 months after planting in conjunction with LAI measurements taken.PAR was measured with a PAR meter model SKP-200 (Sky Instruments, Llandriolod Wells, UK).Measurements were taken at the bases of the plants, towards five cardinal points Erom 0900 to 1500 h at 3h intervals for one week and averaged for daily variation for different age of the crop mentioned (Table .2).

RESULTS
Average number of leaves per clump increased as 4 f 0.5,12 + 1,18 f 1.5 and 1 8 f 2 during the measurements taken a t 4, 9, 20 and 32 months of age respectively.Similarly, the LA1 ofthe crop also increased as 0.61 + 0.11,3 + 0.21, 4.2 f 0.28 and 4.5 f 0.34 measured a t 4,9,20 and 32 months ofage respectively.The light penetration to the crop measured as PAR was reduced from 1046 to 503 pm~l.m-~.s-l(mean measurement from 0900 -1500 h) during the period from 4 to 32 months of age (Table 2).
The growth and yield parameters that were significantly different between different ages of the crop and the level of significance are indicated in Table 3.The pseudostem height, pseudostem girth, number of fruits per bunch and length of male axis increased significantly (p < 0.001) from first year to third year of age (Table 4).Similarly, number of combs per bunch and weight of peduncle increased significantly (p < 0.001) during this period (Table 4).The average fruit weight of the bunch and fruit length of the last comb decreased significantly (p < 0.001) with the age of the crop (Table 4).The fruit length in the second comb, weight of the second comb, fruit diameter and fruit circumference in last comb did not vary significantly (p > 0.05) with the age.The fruit maturity period was significantly (p < 0.01) increased in second and third year over the first year, but there was no significant difference (p > 0.05) between second and third year (Table 4).Bunch weight, total fruit weight and number of fruits per comb were not significantly increased (p > 0.05) up to the second year, but there was a significant increase in these characters (p c 0.01) during the third year (Table 4).The fruit diameter in second comb did not signifiicantly vary (p > 0.05) from first year to second year but a significant decrease (p <0.01) was observed in the third year (Table 4).The weight of the last comb increased significantly (p < 0.001) from first year to second year but no significant difference (p > 0.05) was observed in this character between second and third year (Table 4).

DISCUSSION
It was revealed that plant height and girth increased from first year to third year in this experiment.A positive correlation between pseudostem height and the girth has been previously reported for "Dwarf Cavendish" banana in Queensland,5 Musa AAB group in Africa6 and East African highland bananas.'Therefore, it is evident that positive correlation between pseudostem height and the girth in cv Mysore evaluated in this study exists.The pseudostem height has also been increased when plant density increased in "Cavendish" banana (AAA group) in Indiae and in the second and third cycles of William" (AAA group) banana in A f r i ~a .~ Similarly, in this experiment plant density increased during the first year because of the increase in plant number in the clump.Therefore, increase in pseudostem height in ratoon crops in this experiment may be due to increase in plant density resulting from increased clump size.Light penetration in the form of PAR was reduced with increasing age of the crop to a maximum of 50% measured from 4 months to 32 months of age in this experiment.Seventy percent reduction of light penetration was reported with the increasing canopy for "William" banana in third cycle grown with the density of 1000 clumpdha in A f r i ~a .~ Therefore, the elongation of pseudostem height in ratoon crop of this experiment could be due to reduced light penetration resulting from increased foliage.
LA1 increased in ratoons with increasing age of the crop up to a maximum of 4.5.LA1 also increased up to 6.3 inUWilliam" banana in the third cycle grown a t the density of 2222 clumpdha in A f r i ~a .~ Further, the leaf area has also been positively correlated with pseudostem height of "Cavendish" banana in Ivory Coast.1°The peduncle weight and the male axis length also increased with the age of the crop in this study probably be due to increased leaf area and leaf number.
Bunch weight, total fruit weight and number of fruits per comb increased in the third year over the first and second year.The number offruits per bunch and the number of combs per bunch progressively increased from the first year to the third year of the crop.Higher yield has been correlated with higher number of leaves, greater LAI, greater pseudostem height and girth in highland bananas in Uganda.7However, bunch yield has been better correlated with pseudostem girth than the pseudostem height in Musa AAB cv Agbagba in Nigeria.llThe bunch yield has also been correlated with the leafnumber in the first and second cycles of the crop in a mulching experiment in South Nigeria.12In contrast, there were some reports on the reduction of gross yield in ratoons of "Gross Michel" (AAA group) banana in commercial fields.ls3However, these authors have not assessed the yield related vegetative characters or yield components.Therefore, increase in fruit yield, fruit number, comb number, fruit weight and comb weight in ratoon crops in this study could be due to increase in LA1 in ratoon crops over the plant crop.Increase in yield related vegetative characters such as pseudostem height and girth in ratoon crop strongly support the yield increase ih this experiment from year one to three.
The average fruit weight, fruit circumference and fruit length in the last comb decreased from year one to three in this study.Fruit weight has been negatively correlated with density of the plants in "Cavendish" banana.sPoor light penetration into increased foliage in ratoon crops may be the reason for poor grade of the fruits in ratoon crops investigated up to three years in this experiment.

Table 4 :
Variation of growth and fruit characteristcs in Musa cv Mysore ("Embul" banana) at different ages of the crop.GrowtWFruit Variation a t different ages S.E.D (220 dQ" Year 1 Year 2 Year 3 Pseudostem height (cm) Pseudostem girth (cm) F. maturity period (days) Bunch wt.(kg)Total fruit wt.(kg) No. of combs per bunch No. of fruits per bunch Av. no. of F. per comb Av. wt. of the fruit (g) F. diameter i n second comb (cm) Wt. of last comb (g) F. length in last comb (cm) Wt. of peduncle (kg) Male axis length,(cm) * = Standard error difference at 220 degrees of freedom.

Table 2 : Daily mean variation of light penetration to banana crop at different ages in comparison to direct light penetration.
" = mean value of the four measurements (taken on sunny days)