Printed in Sri Lanka USE OF CHLOROPHYLL FLUORESCENCE PARAMETERS TO ASSESS DROUGHT TOLERANCE OF COCONUT VARIETIES

Chlorophyll fluorescence analysis as a measure of photosynthesis is increasingly used in eco-physiological studies. It is particularly useful in investigations on the photosynthetic performance of plants in stressful environments. Chlorophyll fluorescence, which can be measured rapidly and conveniently, provides insights into a plant’s ability to tolerate environmental stresses such as those of a drought. Damage to the photosynthetic apparatus due to moisture or heat stress, which would reduce photosynthetic assimilation and eventually the yield, can be detected at a very early stage of the drought. This technique is widely used to identify stress tolerant plants and crop varieties. However, this technique has not been utilized to identify putative drought tolerant coconut varieties. Hence, the objective of this study was to set out the basic principles of the technique, discuss its applications in eco-physiological studies, and evaluate its potential to screen coconut varieties for drought tolerance. San Ramon, a tall form of coconut introduced to Sri Lanka few decades ago and a few other local tall accessions in the germplasm conservation blocks in the Coconut Research Institute of Sri Lanka appeared to possess a photosynthetic apparatus which was clearly more tolerant to moisture stress conditions than the other accessions. SRR, SRG, SA, CL PI & PW were the most drought tolerant of the 40 genotypes tested. The results generated by using Fv/Fm, were consistent with results of other important parameters such as the rate of photosynthesis and water use efficiency. Even the basic parameters of chlorophyll fluorescence, for instance the maximum quantum yield of photosystem II (Fv/Fm) measured on dark-adapted coconut leaves showed promise as a selection criterion in future screening programmes due to its quick, non-invasive and reliable nature. Similarities of fluorescence parameters observed between seedlings and adult palms enables fairly confident predictions on the performance of adult palms, based on studies with seedlings. In general, chlorophyll fluorescence appears as a promising tool for screening coconut palms for their drought tolerance.


INTRODUCTION
In recent times the chlorophyll fluorescence technique has been increasingly used in ecophysiological studies with important crops and plant species.Investigations on the photosynthetic performance of plants in stressful environments are now deemed incomplete without information on chlorophyll fluorescence (Maxwell and Johnson, 2000) especially, since the introduction of a variety of sophisticated, highly user-friendly, portable chlorophyll fluorometers.Despite the simplicity of taking measurements, the theoretical aspects of the technique and the interpretation of data are complex and controversial.Although the potential for using the chlorophyll fluorescence technique to screen plant species for putative drought tolerance has been established, it has not been tested for coconut.Hence, this paper discusses the basic principles of the chlorophyll fluorescence technique, its eco-physiological applications and limitations, and proceeds to evaluate the potential for using the technique to screen coconut varieties for drought tolerance.The results of preliminary studies with coconut seedlings under plant house conditions and adult palms under field conditions are presented.

Chlorophyll fluorescence -the underlying principles
Each quantum of light absorbed by a chlorophyll molecule raises an electron from ground state to an excited state.Part of this light energy, absorbed by chlorophyll molecules in the leaf, is used to drive the photosynthetic process (photochemistry); the excess energy is either dissipated as heat, or re-emitted as red light, which is called chlorophyll fluorescence.These three processes occur simultaneously; an increase in the efficiency of one process will reduce the yield of the other two.It is therefore possible to derive information on the efficiency of photochemistry and heat dissipation by measuring the yield of chlorophyll fluorescence.
The chlorophyll fluorescence spectrum is clearly different from that of absorbed light; the fluorescence peak has a longer wavelength than that of absorbed light.Measuring chlorophyll fluorescence is therefore relatively easy, even though total chlorophyll fluorescence is only 1-2% of total light absorbed.The fluorescence yield can be quantified by exposing a leaf to a defined wavelength and measuring the amount of light re-emitted at a longer wavelength.The advent of a 'modulated' measuring system (a modification to the basic measuring device), was pivotal in revolutionizing the application of chlorophyll fluorescence (Quick and Horton, 1984).Here, the light source used to induce fluorescence is 'modulated' (switched on and off at a high frequency) and the detector is adjusted to detect only the fluorescence excited by the measuring light.This enabled measuring the relative yield of fluorescence in the presence of background light, even under full sunlight in the field.

Kautsky effect
In 1960, Kautsky and co-workers first observed that transferring photosynthetic material from dark into the light increased the yield of chlorophyll fluorescence for about one second (Kautsky et al, 1960).This increase was explained later as being due to the reduction of electron acceptors in the photosynthetic pathway, down stream of photosystem II (PSII), more precisely plastoquinone, Q A .When PSIl absorbs light and Q A accepts an electron, it cannot accept another electron until it passes the electron to the subsequent electron carrier Q B .Until then, the reaction centre is said to be 'closed'.The closing of reaction centres leads to an overall reduction in the efficiency of photochemistry and a simultaneous increase in the yield of fluorescence.The magnitude of these changes, at any given time, is related to the proportion of closed centres.
When a dark-adapted leaf is transferred to light, PSII reaction centres are progressively closed giving rise to an increase of chlorophyll fluorescence for a very brief period.However, the fluorescence level drops over a period of few minutes and this phenomenon is termed fluorescence quenching.There are two processes that contribute to fluorescence quenching.Firstly, the rate at which electrons are transported away from PSII is increased due to the light-induced activation of enzymes involved in carbon metabolism and opening of stomata.This is 'photochemical quenching'.Secondly, the efficiency of energy conversion to heat is increased simultaneously and this is termed 'non-photochemical quenching (NPQ).These two processes attain a more or less steadystate condition within about 15-20 minutes, depending on the plant species (Maxwell and Johnson, 2000).
The clear separation of the photochemical and non-photochemical components of quenching is important to extract information on the photosynthetic performance of a plant by measuring chlorophyll fluorescence.The fluorescence yield is estimated with one component process operating and the other 'switched off.The common practice is to switch off the photochemical component.Although addition of the chemical Diuron (DCMU) inhibits PSII and reduces photochemistry to zero, it is not an appropriate method from a physiological and practical point of view.On the other hand, the 'Light doubling technique 1 i.e. giving a high intensity, short duration flash of light, which momentarily reduces photochemical quenching to zero, is non-destructive and more convenient (Bradbury and Baker, 1981).In the absence of any photochemical quenching the fluorescence yield reaches the maximum attainable (maximum fluorescence FJ.Comparing this value (FJ with the steady-state yield of fluorescence in the light (F) and the yield of fluorescence in the absence of photosynthetic light (FJ gives information on the efficiency of photochemical quenching and also the performance of PSII (Maxwell and Johnson, 2000).Although the levels of F m are subjected to change with changes in heat dissipation (non-photochemical quenching), it is not possible to totally inhibit heat dissipation and measure the yield of chlorophyll fluorescence in the absence of nonphotochemical quenching.Therefore, all estimates of non-photochemical quenching have to be taken relative to some dark-adapted point termed as F' m .Therefore, it is important to take a dark-adapted, non-stressed reference point, which has to be a pre dawn value in field experiments.Some parameters are referred to by different terminology in the literature and this may cause some confusion.However, the general calculation of fluorescence parameters can be explained by reference to a typical experimental trace as shown in Fig. 1, reproduced from Maxwell and Johnson (2000).The measurement is initiated by switching on the measuring light, giving a measure of the minimal level of fluorescence (F a ).A saturating pulse of light is then applied which allows the measurement of F m in the dark-adapted state (F' m ).Then, an actinic (photosynthetic) light is applied with further saturating flashes at appropriate intervals in which the maximum fluorescence in the light is measured (F m ).The fluorescence immediately prior to the flash is termed steady-state fluorescence (r 7 ,) after which actinic light is removed with simultaneous application of far-red light (wavelength >680 nm) which completely nullifies the effect of actinic light and allows measuring F a .This far-red light is important to ensure that all PSII reaction centres open rapidly after the end of illumination.

Photochemical processes
One of the important parameters that can be measured is the efficiency (or the quantum yield) of Photosystem II photochemistry (<t> PSII ) which, measures the proportion of the light absorbed by chlorophyll associated with PSH and used in photochemistry as shown below (Genty etal, 1989).This provides a measure of the rate of linear electron transport which is an indication of the overall photosynthesis.Strong linear relationship has been observed between 0 PS!1 and the efficiency of carbon fixation under laboratory conditions but some discrepancies may occur under stress conditions due to changes in the rate of photorespiration (Fryer et and Johnson, 2000).

Non-photochemicalprocesses
The dark-adaptation of a full night and 24 hour period is generally used under laboratory conditions to measure the dark-adapted values of F m which is important in quantifying non-photochemical quenching.Moreover, it is assumed that the plant material is not exposed to any stress prior to the start of the experiment.This ensures the obtaining of the reference level of F m in which photochemical efficiency is at its maximum and heat dissipation is at a minimum.The pre-dawn values of F m are used as a reference in field experiments.The ratio of the change in F M to the final value of F m is the frequently used method for quantifying non-photochemical quenching as given below (Bilger and Bjorkman, 1990).Although there were some brief showers during the 77-day 'drought' period the rainfall was no greater than 5 mm on any given day.The 77-day drought was broken by natural rainfall.

Chlorophyll fluorescence
In seedlings, the youngest, fully-expanded leaf was selected for measuring chlorophyll fluorescence.In mature palms, it was the ninth leaf (counting from the top) which is representative of the physiologically most active middle whorl (Braconnier and Bonneau, 1998).In tall, adult trees, conducting in situ leaf measurements was impractical.Hence, detaching a complete leaflet from the ninth leaf, and dropping it to the ground for rapid measurements was adopted as the standard practice.Previous tests in 3-year old palms showed that determining stable leaf gas exchange values and other physiological parameters on detached leaves was acceptable provided that the measurements are completed within two minutes of detachment (Nainanayake, 2004).The maximum quantum yield of photosystem II, given by the dark-adapted FJF m (the ratio of the variable to the maximum chlorophyll fluorescence), was measured using a fluorometer (PEA Mark 2, Hansatech Instruments Ltd., Kings Lynn, UK).
A preliminary experiment was conducted on AS seedlings to determine the minimum period required for dark adaptation, in preparation for measuring the dark-adapted FJF m .
A portion of the youngest fully expanded leaf was covered by a clip provided with the instrument, and allowed to remain for 2 minutes before recording the FJF m value.Recordings were repeated, increasing the time allowed for dark adaptation by 2 minutes, up to 10 minutes dark adaptation; and then increasing by 5 minutes up to 30 minutes.Measurements were made on both well-watered and droughtcd seedlings; six of each category.

Physiological parameters
The rate of photosynthesis (A, umol m 2 s"'), stomatal conductance (g,, mol m"V) and the rate of transpiration [E, mmol m" 2 s"') were measured using gas exchange technique in a closed-system portable photosynthesis system (Li-6200, LI-COR Inc., USA) and the instantaneous water use efficiency (co inil , umol C0 2 mol"' H 2 0) was calculated as the ratio ofA/E.The rate of photosynthesis observed at the end of the drought is termed minimum rate of photosynthesis (A mill ).The leaves selected for these measurements were the same as for chlorophyll fluorescence.

Soil moisture content
In the plant house experiment, mean soil moisture content (9 %) was determined gravimetrically on approximately 5 g samples of soil taken at a depth of 45 cm, through a small access tube in each pot.Soil samples were drawn at the time of fluorescence measurements.
In the field experiment, the mean gravimetric soil moisture content of the top 150 cm (9nwD %)»where the highest concentration of coconut roots are found (Vidhana Arachchi, 1998), was determined at 14-day intervals, during the drought period.Soil core samples of approximately 200 cm 3 were drawn from two points in each form/acc plot, 1 m away from the base of two palms, and at 50 cm, 100 cm, and 150 cm depths.0 values at different depths viz.50 cm, 100 cm, and 150 cm (0 jO , 0 lO0 , 0 liO ) were determined gravimetrically.The average of the means at the 3 depths, at both sampling points, was taken as the plot mean 6 (0 mtM ), and as a measure of its soil moisture status.The 6^ was assumed to be applicable to the other palms in the form/acc plot, as all six palms were adjacent to each other and in the middle of the form/acc block.

Data analyses
The software package, Statistical Analysis System (SAS Release 8.2, SAS Institute Inc., USA) was used in statistical analyses mainly for Analysis of Variance (ANOVA).

RESULTS
Although the plant house was well ventilated with 3 large exhaust fans, the air temperature ranged from 29°C to 45°C during the day and 23°C to 29°C during the night.
Relative humidity ranged from 65% to 75% during the day and 75% to 85% at night.Light penetration through the roof was appreciable, and the incident photosynthetic PPFD was up to 950 umol m' 2 s" 1 .At the time leaf measurements were taken, between 09:30 and 12:00 hrs, PPFD was between 600 and 950 umol m" 2 s 1 .According to Madurapperuma et al, 2002, the diurnal time course of leaf gas exchange in coconut is more or less constant between 10 and 13h.

Soil moisture content
In the plant house experiment, the mean soil moisture content (averaged over form/acc) in the well-watered pots was steady over the trial period, ranging from 37.4% to 38.6% (Fig. 2a).In the draughted pots however, 8 declined to about 4.5% at the end of the imposed 63-day drought.There were substantial differences in the moisture depletion patterns of the form/acc (Fig. 2b).Soil moisture depletion in the DTand SR pots was more rapid than in the DG and AS pots.This resulted in low 0 in pots with Diana* SR at the end of dry period (3.3% and 2.5% respectively) compared to DG and AS (6.6% and 5.6% respectively).

(Control -No plant, only soil, DG -Dwarf Green, AS -Ambakelle Special, DT-Dwarfx Tall, SR -San Ramon; SE of the mean is indicated by vertical bars).
In the field trial, all coconut palms were exposed to a 77-day natural drought from January to March 2003.At the onset of the drought, the 0 mean averaged 33.4% over all form/acc; the difference between form/acc was small, only about 2.0%.Although the initial rates of decline were similar (barring few exceptions) there were substantial differences in 0 mcan among form/acc towards the latter part of the drought (Fig. 3).Differences in the rate of decline were reflected in the minimum soil moisture (0 min ) recorded for form/acc (Fig. 4) where SRR, SRG, DTand DYTrecorded < 8.0% at the end of the drought.DBY, KS, CRD and KC took up less water and recorded a 0 min as high as 16.0% while the rest ranged between 11.0 and 14.0%.Generally, across all form/acc, the reduction of 0 mcan was about a 60% reflecting the severity of the drought.
The experiment was sited on a block of land which was quite uniform and with a flat terrain.Two bare plots, with no palms, were also included to serve as 'controls'.The 0 mean of these plots were 11.2 and 12.9%.It appears that soil water loss by evaporation in the block is quite uniform, and therefore substantial differences in 0 nican at the end of the drought could be ascribed to differences in the quantum of water absorbed and/or lost by transpiration by the form/acc concerned

Maximum quantum yield of photo system II (FJFJ
In preliminary investigations, FJF m reached a constant maximum value after a darkadaptation period of 20 minutes, in both well-watered and droughted coconut seedlings (Fig. 5a and 5b).Hence, a minimum of 20 minutes dark-adaptation was used for subsequent measurements.In adult palms, FJF m was measured on detached leaflets of the 9 th leaf and the dark-adaptation was achieved by placing the leaflets in a black polythene bag for 20 minutes before measuring.In the plant house experiment, well-watered seedlings recorded slightly higher (4%) dark-adapted FJF m values than the other three form/acc which, were not significantly different (Fig. 6a).In the droughted treatment, in all form/acc, the reduction of FJF m was marginal up to day 35 of the drought, but declined sharply after day 49 (Fig. 6b).During this sharp decline phase, SR maintained a marginally higher value of F/F m \ no difference between other form/acc was observed at the end of the drought.The recovery with rewatering was generally slow with SR slightly ahead of the others (Fig. 6b).
When F/F m was plotted against corresponding 0 the patterns were similar but the differences more conspicuous; SR maintained markedly higher FJF m ratios at lower levelsof 9 (Fig. 6c).AtaOof 10%, SR recorded the highestF/F m of0.71 while DT,AStmd DGrecorded0.66,0.59and0.52 for respectively.In the case of adult palms, rain fed but with no moisture stress, there were no significant differences in dark adapted FJF m values (averaging 0.756) between form/acc (data not shown).With the onset of the drought, although there were no major changes in the FJF m values up to day 14, most form/acc showed a slight decline from day 28 followed by a sharp decline after day 42 (Fig. 7 shows four selected form/acc with contrasting behaviour).
The minimum FJF m at the end of the drought was significantly higher (> 0.5) in eight form/acc (SRR, SRG, SA, DBY, KS, CL, PI&PW) the rest; the lowest values (< 0.4) were recorded in BD, HK, KK and KD (Fig. 8).The high F/F m group of form/acc appears to have been able to protect the integrity of the thylakoid membranes of the photosystem II and thereby maintain the rate of photosynthesis with little or no change.This indicates relatively greater drought tolerance.In fact, A min (rate of photosynthesis at the end of the drought) of these form/acc more or less matched the ranking based on the maintenance of high photosynthetic rates.For example, SRR, SRG, CL, SA and PI grouped together had a significantly higher A mw than the rest (Fig. 9).
The only exceptions were DB YandKS which maintained a high F/F m at the end of the drought but with substantially lower rates of photosynthesis than the rest.Grouping based on the water use efficiency was more or less similar to that of A mm (data not shown).Most form/acc were slow to recover and their FJF m ratios were 10-25% less than the initial values even 3 weeks into the wet period.However, SRR, SRG, DBY and KS recovered well; their F/F m were only about 8% less than the initial F/F m .Therefore, in general, the screening for drought tolerance based on yield related physiological parameters such as the rate of photosynthesis and/or water use efficiency matches well with screening based on chlorophyll fluorescence.

Seedlings versus adult palms
The general pattern of decline in the F/F m values during the course of the drought was similar in seedlings and adult palms (Fig. 10).However, in the case of seedlings the decline was somewhat steeper than in adult palms.This may be ascribed to the moisture in the relatively small mass of soil in the pots being rapidly depleted in contrast to the much larger mass of soil and soil moisture availability to adult palms in the field.

Chlorophyll fluorescence versus yield as a parameter for screening
Plant breeders are interested in testing morphological and physiological traits, as reference parameters to screen crop species for putative drought tolerance, instead of the most integrative, comprehensive and heavily used parameter, yield.The compelling factors here are the desire to save time and the difficulty in evaluating the effect of a particular dry spell on the yield of a perennial tree crop such as coconut.Coconut has a long generation period of about 15 years from seed nut to a bearing palm with a stable yield.It takes a year from the opening of the inflorescence to produce mature nuts.This visible cycle is preceded by a 32-month non-visible cycle of development from flower primodia up to the opening of the inflorescence -a development cycle of 44 months in all.This complexity and lengthy generation period makes it very difficult to separate the effects of drought on coconut yields.Button nut setting, their growth, development of nut water and kernel occurs over different periods during the year-long visible maturation cycle.The extent of adverse drought effects such as immature nut fall, reduced nut size or poorly developed kernel will therefore depend on the stage of nut development exposed to the drought.To add to the complexity: nut development stages are sequential in a single inflorescence, but they overlap in the palm and in the field.Droughts could also impact on different stages of the non-visible development cycle.And drought tolerance evaluation based on yield is a very long-term exercise.Therefore, the development of a rapid, reliable and less invasive technique/s such as chlorophyll fluorescence appears important and advantageous in the context of screening for drought tolerance.
There is a long standing controversy in regard to using parameters, other than yield, to screen for drought tolerance.Lawlor (1995) argued that there is no direct relationship between physiological parameters such as rate of photosynthesis and crop yield.However, evidence to the contrary has emerged in. the recent past.For instance, Mann (1999) reported a good correlation of increasing rate of photosynthesis with higher crop yields while Mitchell et al, (1999)

Figure 1 ;
Figure 1; Sequence of a typical fluorescence trace.A measuring light is switched on (MB) and the zero fluorescence level is measured (F a ).Application of saturating flash of light (SP) allows the measurement of maximum fluorescence level (F°m).A light to drive photosynthesis (AL) is then applied.After a period of time, another saturating light flash (SP) allows the maximum fluorescence in the light (F' m ) to be measured.The level of fluorescence immediately before the saturating flash is termed F r Turning off the actinic light (AL) typically in the presence of far-red light allows zero level fluorescence in the light to be estimated (From Maxwell and Johnson, 2000).

While O ps1i relates
to achieved efficiency, qP and FjF m provide information about the underlying processes which have altered efficiency.While changes in qP indicates the closure of reaction centres, resulting from a saturation of photosynthesis by light, a change in F/F m shows a change in the efficiency of non-photochemical quenching.The potential quantum efficiency of PSII represented by the dark-adapted values of F/F m , is a sensitive indicator of plant photosynthetic performance.Optimal values of FJF m are around 0.83 for most plant species (Maxwell and Johnson, 2000; Bjorkman and Demmig, 1987).When plants are exposed to stress, FJF m values would be lower than optimal, indicating the phenomenon of photoinhibition (Maxwell mNPQ is linearly related to heat dissipation and could range from zero to infinity.Typical values for a plant are in the range 0.5 to 3.5 at saturating light intensities(Maxwell and Johnson, 2000).Chlorophyll fluorescence -applications in eco-physiologyIn general, chlorophyll fluorescence provides information on how the light energy absorbed by chlorophyll impacts on PSII -the extent to which PSII uses the energy and the extent to which PSII is damaged by excess light(Maxwell and Johnson, 2000).The flow of electrons through PSII is indicative of the overall rate of photosynthesis, and this enables a rapid estimation of photosynthetic performance, under conditions where other methods would fail.Moreover, PSII is the most vulnerable part of the photosynthetic apparatus to light-induced damage and this damage is the first manifestation of stress in a leaf.Although these experiments of chlorophyll fluorescence requires very careful designing and are generally coupled with other techniques such as gas exchange measurements, it provides quick but precise estimates of photosynthetic performance even under initial stress conditions.An advantage of chlorophyll fluorescence is that it provides a measure of photosynthesis.Chlorophyll fluorescence can be converted into a relative rate of linear electron transport.Photosynthesis is the gross rate of carbon fixation.PSII electron transport and C0 2 fixation correlate well under laboratory conditions(Genty et al, 1989;Edwards and Baker, 1993).However, the relationship is obscure under field conditions due to competing processes such as photorespiration, nitrogen metabolism and electron donation to oxygen (Mehler reaction).Furthermore, the heterogeneity between leaves, especially when grown in different microclimates, makes comparisons less precise.Hence, gas exchange measurements still remain as the core parameter in eco-physiology.Despite above limitations dJpg,, provides useful information on photosynthetic performance in the field with rapid and minimally invasive measurements.Hence, the method has been used successfully in varietal comparisons in very different plant species such as rice(Murchie et al, 1999) and loblolly pines trees(Hymus et al, 1999).The linear relationship observed between quantum yield of C0 2 fixation and O^,, in experiments with simultaneous measurements of C0 2 assimilation and O ps " under nonphotorespiratory conditions (elevated CO, or 1-2% 0 2 ) at different light intensities provide a way to estimate the extent of photorespiration in vivo(Epron et al., 1995).This protocol has been used in different plant species to evaluate the significance of photorespiration as a photoprotective mechanism during drought stress (Cornic and Briantais, 1991; Valentini et al., 1995).Meyer and Genty (1998) using high resolution images of <t> psll , proved that stomatal closure is the primary effect of drought stress which in turn leads to the reduction in intercellular C0 2 concentration (C,) and thus limits carboxylation.Moreover, Fryer et al, (1998) used fluorescence analysis to understand the effect of low and high temperatures on the photosynthetic process.Most importantly, chlorophyll fluorescence gives insights into the ability of a plant to tolerate environmental stresses and also the extent to which those stresses damaged the photosynthetic apparatus (Maxwell and Johnson, 2000).Diurnal measurements of chlorophyll fluorescence in different plant species, using portable fluorometers developed recently, have provided a wealth of information on NPQ, electron transport rates, quantum efficiency and the extent of photorespiration in response to light, temperature and other environmental stresses (Bilger et al, 1995).Gamon and Pearcy (1989) observed a decrease in FJF m and an increase in F" which indicated photo inhibitory damage in response to high temperature.Similar effects were also observed for low temperature (Groom and Baker, 1992), high PFD (Ogren and Sjostrom, 1990) and water stress (Epron et al, 1992).With the recent improvements in the technology, changes in FJF m and F 0 appear to be valid, reliable diagnostic indicators of photo inhibition and are being widely used (Valladares and Pearcy, 1997; Hakam et al, 2000; Percival and Sheriffs, 2002).For example, a study conducted by Baker and Adams (1997) on a cactus species revealed that cladodes that received high PFD contain a large xanthophyll cycle pool with de-epoxidation of violaxanthin to anthcraxanthin and zeaxanthin which indicated a high NPQ.Chlorophyll fluorescence -as a tool for screening coconut varieties for drought tolerance MATERIALS AND METHODS Two experiments, one in the plant house and the other in the field, were carried out to evaluate the potential for using chlorophyll fluorescence to identify putative drought tolerant coconut genotypes.The plant house experiment was conducted with seedlings of four genotypes grown in pots under controlled environment.The field experiment evaluated adult palms from 40 forms and accessions (form/acc).Plant house experiment The four coconut forms (in parenthesis) selected for the plant house experiment represented the main 'varieties' of coconut: var.typica, or 'talis' (Ambakelle Special, AS)\ var.nana or 'dwarfs' (Dwarf Green, DG); hybrids between dwarfs and tails (Dwarf x Tall, DT); and a tall form of Pacific region origin, acclimatized to Sri Lanka San Ramon, SR.).Mature nuts selected from well-grown palms, were germinated and raised in a soil and coir dust mix under well-watered conditions, for seven months, as per standard nursery practice at CRISL.Seedlings of uniform size in terms of number of leaves, seedling height and girth, were transferred to 40 litre capacity plastic pots containing sandy loam soil and placed in a plexiglass roofed, open sided, plant house.The soil used was from a Madampe Series site (Somasiri et al, 1994), similar to that of the field experiment.Seedlings were uniformly fertilized one month after planting with 10 g of dolomite and 50 g of the CRI Young Palm Fertilizer Mixture (YPM).A Randomized Complete Block Design (RCBD), with 3 replicates, was used for the experiment.There were 8 treatments (4 varieties and two watering regimes viz.well-watered and 'droughted' i.e. withholding water for 63 days before watering to field capacity), with 3 plants per plot.AH seedlings were allowed to establish under the well-watered regime for 3 months prior to applying the 'draughted' treatment -a 63-day drought broken by watering to field capacity.Field experiment Forty forms or accessions (form/acc) including local and exotic dwarfs, tall forms and hybrids (Table 1), approximately 15 years old, from the germplasm collection at Pottukkulama Research Station, Pallama, Sri Lanka were used for the field experiment.All form/acc were planted in single random blocks of at least 30 palms each, using a 7.8 x 7.8 x 7.8 m triangular planting system.The entire plantation has been fertilized regularly with 3 kg per palm per year of the CRI Adult Palm Fertilizer Mixture (APM) and managed as per practices recommended by CRISL (Liyanage, 1999), from the time of planting.The site was excellent for coconut with a flat terrain and a uniform deep sandy loam Madampe Series soil in Land Suitability Class S 1 (Somasiri et al., 1994).Six adjacent palms from the centre of each block were monitored over 14 consecutive weeks during January to April 2003; the first 11 weeks were the natural drought period.

Figure 2 :
Figure 2: Mean soil moisture content (0, %) in pots (a) averaged over the four form/acc in the watered and draughted treatments, and (b) of each form/acc over the 63-day drought period.(Control -No plant, only soil, DG -Dwarf Green, AS -Ambakelle Special, DT-Dwarfx Tall, SR -San Ramon; SE of the mean is indicated by vertical bars).

Figure 3 :Figure 4 :
Figure 3: Time course of soil moisture content (0 mean %) in adult palms of eight selected form/acc during the 77-day drought period.

Figure 5 :
Figure 5: Effect of duration of dark adaptation on the mean FJF m of the youngest fully expanded leaf of 9-month old (a) well-watered and (b) droughted coconut seedlings grown in pots in a plant house (n = 6)

Figure 6 :
Figure 6: Mean ratio of F/F m of coconut seedlings in the well-watered and draughted treatments (a and b respectively).F/F m plotted against the corresponding soil moisture content (0, %) in the drought phase is at c.

Figure 7 :Figure 8 :
Figure 7: Mean FJF m of four selected form/acc of adult coconut palms over the 77-day drought period and the period followed by a 21-day rainy period.

Figure 9 :
Figure 9: The rate of photosynthesis in adult palms at the end of the drought (A min , umol m 2 s' 1 ) for different form/acc of coconut.

Figure 10 :
Figure 10: Mean FJF m of coconut seedlings (S) and adult palms (A), of two selected form/acc (DG and SR), over the 63-day and 77-day drought periods, respectively, and the 21 -day wet periods (SandA indicate seedlings and adult palms respectively).
found a 50% average increase of photosynthesis in a single genotype of spring wheat grown at seven sites in Europe (Germany, UK, Ireland, Belgium and The Netherlands) which increased grain yield by an average of 35%.Ainsworth et al, (2002) reported comparable increases in photosynthesis and yield in soybean, under elevated C0 2 .Thus the use of parameters, other than yield, to screen crops is becoming more and more acceptable with evidence on different crops from different parts of the world.In this context, chlorophyll fluorescence emerged as a quick, non invasive tool to investigate the activity of photosystem II of the photosynthetic machinery in the plant leaf.This method has been heavily used in the recent past to evaluate the impact of stress conditions on the photosynthetic machinery and consequently crop productivity under given conditions.Applicability of chlorophyll fluoresence in coconut and other cropsChlorophyll fluorescence works on the principle that photosynthesis is one of the core functions in the physiology of plants.The functional state of photosynthesis has been considered an ideal physiological activity to monitor the health and vitality of plants(Clark et al, 2000).The health or the integrity of the internal apparatus (thylakoid membrane) drives the photosynthetic process within a leaf.This integrity which can be assessed by chlorophyll fluorescence provides a rapid and accurate means of detecting and quantifying the plant's tolerance to stress.Moreover, previous researches indicate that chlorophyll fluorescence values of detached leaves strongly correlate with plant mortality due to environmental stresses such as freezing, heat chilling and salinity(Smillie and Hetherington, 1983;Greaves and Wilson, 1987;Brennan and Jefferies, 1990;Yamada et al, 1996;Hakam et al, 2000).It permits quantification of species or varietal tolerance with limited whole plant experiments.Using FJF m values Percival and Sheriffs (2002) selected 30 woody plant species with the ability to maintain photochemical efficiency even after a 70-day drought.The ranking by visual observation was similar to that based on the rate of photosynthesis.As photosynthesis is the primary physiological function that contributes to the productivity of a plant, any species identified as drought tolerant based on FJF m may reflect this tolerance in their yield or productivity parameters.Adopting a similar approach to evaluate adult coconut form/acc in the present experiment, it was possible to group 6 form/acc (SRR, SRG, SA, CL PI <£ PW) as drought tolerant based on high readings for both F/F", and the rate of A observed at the end of the drought.The exceptions observed were DBY and KS which maintained a high FJF m at the end of the drought but with relatively low rates of photosynthesis.Stomatal conductance (g,) of these two form/acc revealed a probable cause for this discrepancy.Their g, decreased markedly at a fairly early stage of the drought, when 0^ was around 25%, indicative of an early response to conserve water.Restriction of gaseous exchange at stomatal level appears to be the primary cause for the rapid reduction of A observed in KS and DBY.These results point to the possibility of using F/F m as a promising tool to screen coconut palms for their drought tolerance.Concurrent monitoring of the gas exchange parameter would be useful not only to provide confirmation but also to identify the exceptions.Consistency of past selections with present findingsKSand DBY had been originally selected for inclusion in the germplasm conservation pool because their parent populations had withstood severe droughts.Hence they were also expected to possess a degree of drought tolerance.Closure of stotnata from the very early stages of the drought is a likely adaptation to survive long and frequent dry spells.SRR, SRG and CL were of common origin from the Pacific region, probably from the Philippines and introduced to Sri Lanka few decades ago.Frequent changes in climate in the Pacific region may have contributed to their adaptation to withstand a wide range of stressful environments.Some inherent characteristics of these form/acc such as a thick kernel may have contributed to the survival of developing seedlings under adverse weather conditions.The common origin of these three form/acc and the CL being an open-pollinated population of the variety San Ramon was recently been confirmed with evidence from molecular studies (Perera, 2005).BD, HK, KK and KD which appeared to be sensitive to drought may have evolved in good, deep soils with higher water holding capacity and/or in the wetter parts of the intermediate zones of the coconut growing areas.Therefore, most of the form/acc selected for the CRI germplasm collection, based on different morphological or other general characteristics, show some degree of consistency with the present evaluation based on the chlorophyll fluorescence.Consistency of grouping patterns based on F/F lliy the rate of photosynthesis and water use efficiency provides a firm basis for using chlorophyll fluorescence as a selection criterion.Therefore, identifying putative drought tolerant coconut form/acc based on the rapid, non invasive tool of chlorophyll fluorescence is a distinct possibilty.The similarity observed in the performance of seedlings and adult palms is a great advantage.This enables the identification of inherent capacities and potentials at the seedling stage thus making a huge saving in the time requirement for evaluating coconut varieties.The preliminary findings clearly demonstrate the potential of chlorophyll fluorescence.Nevertheless, the fluorometer used for these measurements was a basic model that was not equipped with a modulated light system that would have enabled a full range of measurements, as discussed in the introductory sections.A more comprehensive evaluation of the coconut germplasm collections should therefore be conducted with the full spectrum of fluorescence parameters under natural light conditions.It is also important to re-evaluate and confirm the above findings using long-term yield data so as to fully validate the use of this rapid and convenient selection technique.CONCLUSIONSChlorophyll fluorescence which has been widely used for screening many crops and plant species for drought tolerance appears to be also applicable to coconut.The maximum quantum yield or the efficiency of PSII, as indicated by the dark-adapted F/F", is a promising, non-invasive parameter for screening coconut germplasm.Using this approach, SRR, SRG, SA, CL PI & PW v/cre identified as the most drought tolerant of the 40 genotypes tested.The results generated by using F/F mt and other important, yieldrelated physiological parameters such as the rate of photosynthesis and water use efficiency, were consistent.This provides a firm basis for using leaf chlorophyll fluorescence as a selection criterion.The great similarities observed between seedlings and adult palms with respect to fluorescence parameters is a key finding that enables fairly confident predictions on the performance of adult palms, based on studies with seedlings.A detailed evaluation of the CRI germplasm collections using the full spectrum of chlorophyll fluorescence parameters, and relating the results to long-term yield data, is important to make a strong case for using the technique in coconut.

al. t 1998). The rate of linear electron transport (J) can be calculated using <E> reil as shown below (Genty et al., 1989): y=O PSII xPFDax(0.5) where, PFDa is absorbed light (umol photon m 2 s" 1 ) and 0.5 is a factor that accounts for the partitioning of energy between PSII and PSI. Provided that absorption of light is constant, the relative changes of J between two leaves can be compared simply by multiplying O^,, by incident light. Photochemical quenching (qP) is calculated as follows: qP=(F' m -F l
)/(K-F