[From the U.S. Government Printing Office, www.gpo.gov]


                                                                       cm- @1331


            FDNR/FMRI-Fi1e Code: FC010-90-A'










              CATASTROPHIC MORTALITY OF THE SEAGRASS Thalassia testudinum


                                     IN FLORIDA BAY












           ANNUAL COMPLETION REPORT TO OFFICE OF COASTAL ZONE MANAGEMENT



           PERIODCOVERED: OCTOBER 1, 1990 TO DECEMBER 31, 1991



          .SUBMITTED: JANUARY 21, 1992














                                            NT Of C-0



            BY:      Michael J. Durako,  Ph.D., Paul R. Carlson, Jr., Ph.D.,
                      Timothy R. Barber, Laura  A. Yarbro, PhD, Katie Kuss,
                      and Scott Brinton

                     Florida Department of Natural Resources
                     Florida Marine Research Institute
                     100 Eighth Avenue, SE
                     St. Petersburg, Florida 33701






     X






                                       SCOPE OF WORK


                 In a previous CZM-funded study, our research focused on
             Thalassia morphometrics, productivity, and stress physiology as
             part of a collaborative die-back research group including
             investigators from FMRI, Everglades Park, University of Georgia,
             University of Virginia, and FIU. In this study we continued some
             of the previously described sampling and initiated new studies
             which centered on seagrass community dynamics at the basin level
             and initiated more extensive experimental studies on causal
             mechanisms.   The data from these studies have already been
             valuable to researchers who are now studing effects of the die-
             back on invertebrates and fish species in Florida Bay.            We
             continued with our use of Sunset Cove, Rankin Lake, Johnson Key
             Basin, and Rabbit Key Basin as primary sampling sites based on
             our previous study results.


             TASK 1 CHARACTERIZATION OF DIE-BACK SPREAD AND RECOVERY RATES.

             SUBTASK 1.1.  Dieback patch mapping.

                                           METHODS

                 Die-back  mapping patches were    established in Sunset Cove,
             Rankin Lake,  and Johnson and Rabbit Key Basins.        A permanent
             central stake or anchor was established in the approximate center
             of either the open die-back area or where die-back was extensive
             (e.g., Rankin Lake and Sunset Cove) , within a small Thalassia
             patch. Distances from the reference stake to the ecotone of the
             patch were measured at 300 intervals during each bimonthly
             sampling.  Patch area was estimated by calculating the area of
             the triangle formed by the individual 300            segments and
             calculating the sum of the 12 segments.

                 During the initial establishment of the Johnson Key basin
             and Sunset Sove map sites, PVC stakes were placed at the ecotone
             at the cardinal compass points and these stakes were used as
             reference points for paired fixed-frame photoquad sampling (focal
             length four feet, sampling area ca. 6.25 cm2). These sites were
             photographed approximately bimonthly from March 1989 to October
             1991.  The resulting slides were projected onto a 10 row x 15
             column grid and the number of grid points contacting Thalassia
             were counted as an estimate of coverage.

                                  RESULTS AND DISCUSSION


                  The amount of regrowth and resulting 'fingers' of Thalassia
             in the original die-back map patch in Johnson Key (Figure 1) made
             determination of the ecotone impossible.    Therefore, two new map
             patches were established in this basin.      New map patches were
             also established in Rabbit Key basin and in Rankin Lake.
           4Unfortunately, the floats marking the two    patches in Rankin Lake
  C*Y)    q-
       ps







            disappeared after the second set of measurements were taken and
            the sites could not be relocated. The two-new Johnson Key basin
            die-back patches and the new Rabbit Key basin die-back patch
            exhibited an overall recovery trend during this study (Figure 2).
            The JKA patch did show a slight increase in area between the June
            and August sampling periods which corresponds to the period in
            which a major die-back event was observed in this basin.            All
            three patches exhibited relatively rapid regrowth between August
            and October.     In contrast, the Sunset Cove Thalassia patch
            exhibited rapid spread between June and August and a possible
            loss in area between August and October (Figure 3). The reasons
            for this difference in growth patterns is unknown, but it may
            reflect so-me basic distinctions between the seasonality of
            Thalassia growth at western and eastern Florida Bay sites.

                  Photoquad data for the original Johnson Key and Sunset Cove
            map patches also indicates differences in seasonal regrowth
            patterns although both sites exhibited an overall recovery trend
            over the 32 month study period (Figure 4). The Johnson Key basin
            patch exhibited loss of Thalassia between December 1989 and
            February 1990, and between June 1990 and August 1990.            Sunset
            Cove exhibited little or no regrowth of Thalassia between April
            and June of 1989 and a slight loss between September 1990 and
            February 1991..    This latter period corresponds to the time of
            year that the die-back was first observed in Sunset Cove (i.e.,
            January and February of 1989).


            SUBTASK 1.2. Measure spread and recovery of die-back patches in
            permanent quadrats.

                                           METHODS


                  Because of the initiation of a standardized vegetational
            monitoring program (> 50 - 25 x 25 m permanent quadrats
            throughout Florida Bay) by ENP research personnel, the focus of
            this subtask was altered to assess, on a basin level, the
            frequency, abundance and relative density of the major plant
            species/groups in Rankin Lake, Johnson Key Basin and Rabbit Key
            Basin.    This 'was accomplished. using the Braun-Blanquet cover-
            abundance scale (Mueller-Dombois and Ellenberg, 1974).             This
            scale" uses absolute values to provide species- or plant group-
            specific quantitative information requiring little time per
            sample area, thus allowing a larger area to be effectively
            sampled.

                  In each basin, 10 sample sites were randomly selected from a
            0.5 km sample-point grid. At each sample site, four 0.25 m2
            quadrats (thrown north, east, south and west of a reference
            point) were sampled for species occurrence and quantity (40
            quads/basin) using the following scale: 5-any number of
            individuals with > 75% cover, 4-any number of individuals with
            50-75% cover, 3-any number of individuals with 25-50% cover, 2-
            any number of individuals with 5-25% cover, 1-plentiful, but less
            than 5% cover, 0.5-sparse, with small cover.        Using this scale,


                                          2







            the frequency of occurence         of quads where a species was
            observed/total' # of quads) , abundance (sum of cover-abundance
            scale values/#.of occupied quadrats) and relative density (sum of
            cover-abundance scale values/total # of quadrats) of each
            seagrass species and the dominant algal species or groups were
            calculated.   Braun-Blanquet sampling was initiated in February
            and continued on a bimonthly basis to October.


                                  RESULTS AND DISCUSSION.

                 The results of each bimonthly sampling are presented in
            Figures 5-9. It is readily apparent from these figures that each
            basin has distinctive f loral characteristics with Rankin Lake
            being especially distinct from Johnson and Rabbit Key basins.
            Throughout the year, Rankin Lake was dominated by Halodule and
            Batophora, while Thalassia was dominant in Johnson Key and Rabbit
            Key basins. The latter two basins also consistently had a higher
            algal species diversities than Rankin Lake. .       Syringodium was
            consistently observed in Johnson Key quadrats throughout the year
            at a lower frequency than Halodule, but it had higher abundances
            than Halodule during the February, April and June samplings.
            This pattern reflects the patchy distribution of Syringodium in
            Johnson Key Basin, but indicates that it grows to relatively high
            densities where it occurs.    Syringodium was only detected during
            the April and August samplings in Rabbit Key Basin. This was due
            to its patchy occurrence in this basin (we've only observed
            Syringodiu in the extreme western portion of the basin) rather
            than to seasonal variability.     Syringodium was not observed in
            Rankin Lake.    In contrast, Rupipia was only observed in Rankin
            Lake and was only detected in the sampling quadrats during June
            and August.   The low frequency, abundance and density of Ruppia
            reflect its patchy and sparse growth habit.         Ruppia exhibits
            pronounced seasonality in its growth, producing abundant (and
            easily observable) festoon-like stems from March through August.
            These reproductive stems break off during the fall and winter and
            Ruppia becomes virtually in distinguishable from Halodule, where
            they grow intermixed.

                 Figure 10 summarizes the changes in frequency, abundance and
            density of the two dominant seagrass species (Thalassia and
            Halodule) within the three basins sampled during this study. In
            Rankin Lake, which has experienced the most severe die-back,
            Thalass-ia frequency exhibited a decline while its abundance and
            density were relatively unchanged. This indicates that Thalassia
            distribution is becoming more patchy within this basin and may
            also suggest that Thalassia is continuing to decline through the
            loss or reduction in size of surviving patches rather than
            through a gradual reduction in short-shoot densities (i.e. stand
            thinning). In contrast, the frequency of occurrence of Halodule
            in Rankin Lake increased from 78% during reproductive sampling in
            September 1990 to consistently approaching or equaling 100%,
            indicating a uniform basin wide distribution of this species.
            Both abundance and density increased during the study period
            reflecting continuing colonization and growth of Halodule within


                                        3









             this basin.

                  In Johnson Key Basin, Thalassia exhibited a        decline in
             frequency of occurrence from June to October (Figure 10) .      This
             corresponded to the period in which a major die-back event was
             observed by ENP research personnel.   This die-back was initiated
             during mid-June.  Abundance and density of Thalassia in Johnson
             Key Basin declined throughout the study, but the most dramatic
             decline occurred between August and October.      The frequency of
             Halodule also declined in Johnson Key Basin during the study
             indicating the distribution of this species became more patchy
             and may have been affected by the die-back event. Both Thalassia
             and Halodule populations remained relatively stable throughout
             the study within Rabbit Key Basin. If anything, Thalassia
             abundance and density seemed to increase over time with little
             change in these parameters for Halodule. Frequency of occurrence
             of Halodule in Rabbit 'Key Basin increased from February to
             October.



             SUBTASK 1.3 Determination of flowering short-shoot and seedling
             densities.


                                          METHODS


                  During the bimonthly Braun-Blanquet frequency/abundance
             survey in April, the numbers and sex of-flowering short-shoots of
             Thalassia occurring within sample quadrats were recorded.       Four
             replicate 0.25 m2 quadrats in each of ten randomly selected
             sample sites in Rankin Lake, Johnson Key and Rabbit Key Basins
             were surveyed.     During the June and August surveys, the
             occurrence and numbers of fruits and/or seedlings were recorded.

                                  RESULTS AND DISCUSSION

                  Similar to what was observed during the previous year
             (Carlson et al., 1990), the distribution of reproductive short-
             shoots was very patchy.    Again, no flowering short-shoots were
             observed in Rankin Lake.    Flowering was observe.-I in 28% of the
             quadrats in Johnson Key Basin and in 15% of the quadrats in
             Rabbit Key Basin. Observed reproductive densities were higher in
             Johnson Key Basin compared to Rabbit @Key Basin with 0.6 female
                 0.8 male short-shoots m 2 in the former basin and 0.3 female
             and 0.5 male short-shoots m-2 in the latter.       Both basins had
             male-biased sex ratios - male:female=1.3 in Johnson Key Basin and
             1.7 in Rabbit Key Basin.

                  During the June sampling, Thalassia fruits were observed
             only in Rabbit Key Basin and they occurred in 10% of the
             quadrats. Fruit density was estimated to be 0.9 fruits m_2 based
             on quadrat observations.   The lack of observed fruit production
             in Johnson Key Basin may have been due to the high turbidity of
             the water in this basin during June, or could have reflected the
             occurrence of a widespread, recent die-off event during this
             period.


                                         4








                  Seedlings of Thalassia were observed during the August
            sampling only in Rabbit Key Basin.     This concurs with our fruit
            observations and suggests local recruitment rather than import
            from other basins.    The absence of reproductive short-shoots in
            Rankin Lake for the past two years suggests that if recovery of
            Thalassia within this basin is to occur, it will have to be based
            on vegetative regrowth.


                                     OVERALL CONCLUSIONS

                  The Braun-Blanquet data indicated that, at the basin level,
            there was relatively little or no net recolonization by Thalassia
            within Rankin Lake.     Because no flowering short-shoots, fruits,
            or seedlings of Thalassia have been observed in either Rankin
            Lake or the adjacent north portion of Whipray Basin for the past
            two years,   the only mechanism for recovery available to this
            species is recolonization by vegetative growth.              However,
            regrowth of  Thalassia has been outstripped by the rapid spread of
            Halodule as indicated by the increase in Halodule's frequency,
            abundance and density during the past year.       The net ef f ect of
            these species-specific differences in regrowth patterns is that
            Rankin Lake has been changed from a Thalassia-dominated system to
            a Halodule and Batophora-dominated system.          This change in
            species dominance can be expected affect habitat function within
            the Lake (Zieman, 1982).      Recolonization of previously barren
            die-back patches by Halodule seems to have had some beneficial
            effects - water clarity seemed much better than last year (not
            quantified) and blue-green algal mats were less common.       We also
            observed, for the first timel the consistent presence of fishing
            activity within the Lake and we observed tarpon and dolphins
            feeding in the shallow flats adjacent to Rankin Key. While this
            information is anecdotal, it does suggest that the Rankin Lake
            system has become more ecologically stable.

                  In contrast, Johnson Key Basin experienced an overall loss
            of Thalassia. and Halodule over the past year.            This loss
            corresponds to the occurrence of an extensive die-back event
            which initiated during June. This die-back seemed to affect not
            only Thalassia, but may have led to a reduction (or at least an
            increase in patchiness) in     Halodule as well.      Map patch and
            photoquad data demonstrated    that on a smaller scale, surviving
            plants were producing new short-shoots and that there was
            regrowth into existing die-back patches, so the basin-level
            patterns nay be due to an increase in the number of new die-back
            patches. The difference in patterns determined using basin level
            versus die-back patch level sampling point out the value of a
            multi-level sampling strategy in assessing a phenomenon of this
            magnitude.

                  Rabbit Key Basin exhibited the most stability in seagrass
            species abundance patterns.        The map patch data indicated
            regrowth of Thalassia into existing die-back patches.               No
            additional basin wide decline of Thalassia was indicated during







            the past year, but there was an increase in the frequency of
            occurrence of Halodule (>50% by October) indicating possible
            recolonization of previously unvegetated sites by this species.

                 The initiation of a major die-back event within Johnson Key
            Basin during a summer with relatively high rainfall (salinities
            were about 10 o/ 0,0 lower during the summer of 1991 than during
            the same periods in 1989 and 1990, see Figure 11) and during the
            period with longest day lengths raises questions regarding some
            of the previously suggested environmental/seasonal conditions
            (i.e. the warm water temperatures and decreasing daylength of
            fall) responsible for causing this die-back. Over the course of
            our studies, active seagrass die-back has been observed during
            summer and fall (western Florida Bay sites) and winter (Sunset
            Cove), and during during both relatively dry and wet years.
            These observations suggest the possibility that die-back may
            linked to characteristics of the particular seagrass population
            (i.e. the age or successional stage of the bed) rather than, or
            coupled with, external environmental factors.


            TASK 2. ROLE OF SYNERGISTIC STRESS IN DIE-BACK AND RECOVERY OF
            Thalassia testudinum IN FLORIDA BAY.

                 The objective of this task was to determine the capacity of
            healthy and diseased Thalassia to avoid hypoxic stress and
            sulfide toxicity. Because previous CZM-funded research indicated
            that hypoxic stress and sulfide toxicity of Th-alassia roots and
            rhizomes play important roles in the die-back process, we
            reasoned that any phenomenon which interfered with the ability of
            seagrasses to maintain aerobic conditions in their roots and
            rhizomes could contribute to die-back. Because previous studies
            also indicated that the pathogenic slime mold Labyrinthula plays
            a major role in the die-back process (Porter, 1989)               we
            hypothesized that Labyrinthula may induce rhizome hypoxia       and
            sulfide toxicity in Thalassia by 1) physically disrupting       the
            oxygen-conducting system of the plant and/or 2) reducing        the
            amount of photo synthet i ca 1 ly-pr oduc ed oxygen available   for
            belowground tissue.      Subtask 2.1 comprised field and        lab
            experiments to determine the ability of healthy and diseased
            plants to supply oxygen to their belowground tissues.        Subtask
            2.2 involved a field experiment to test the effect of       sediment
            sulfide and sedinent-borne pathogens on seedling and rhizome
            transplant survival in die-back patches.

            SUBTASK 2.1. Oxygen transport rates through healthy   and diseased
            Thalassia shoots.


                                          METHODS


                 Laboratory measurements of potential diffusive oxygen flux
            rates through intact plants were made using methane and ethane as
            conservative tracers.   Thalassia plants collected from Tampa Bay
            and Florida Bay were incubated in gas-filled chambers to estimate
            diffusive resistances of different parts of the plant (ie.


                                        6







             leaves, stems, rhizomes, roots).         Thalassia rhizomes with a
             healthy apex and at least three lateral (short) shoots were
             collected from Fort Desoto Park and Lassing Park in Tampa Bay.
             Healthy and diseased rhizomes from Florida Bay were collected
             from two of our previous study areas: Johnson Key Basin and
             Sunset Cove.


                  To estimate the   flux of gas through the internal airspaces
             (aerenchyma) of Thalassia testudinum, we used the distal 20-30 cin
             rhizome segment with at least one short-shoot. We first cut each
             rhizome 1 cm behind the apex and again approximately 10 cm behind
             the first fully-expanded short shoot. Swagelok connectors and/or
             Tygon sleeves were used to connect Mylar bags to the cut ends of
             the rhizome.   A cylindrical polyethylene chamber with a flexible
             gas reservoir was slipped over the short shoot and sealed at the
             base.   Tracer (methane or ethane) was added to the short-shoot
             chamber, and accumulation of tracer was monitored in the Mylar
             bags enclosing the cut rhizome ends.        The rhizome was wrapped
             with non-adhesive Teflon tape to impede radial loss of gases, but
             a gap in the tape was left at each rhizome end to prevent gas
             channeling.

                  Tracer concentrations in the source reservoir and Mylar bags
             enclosing the cut rhizome ends were measured on a Carle AGC-100
             gas chromatograph. C1 and C    2 hydrocarbon gases were separated on
             a 51 x 1/811 stainless steel column packed with Porapak Q and a 11
             silica gel pre-column.      The flame ionization detector response
             was calibrated with a standard hydrocarbon mixture (Scott
             Specialty Gases), and concentrations were estimated from peak
             areas measured by a SpectraPhysics SP4270 integrator.

                  Tracer concentrations in source and rhizome bags were
             determined every 10-30 minutes. source bags were sampled using a
             50 ul gas-tight syringe, while rhizome bags were sampled using a
             500 ul gas-tight syringe.     Over the course of a 3 to 6 hour flux
             experiment, a negligible volume of gas was removed from each bag.

                  After 5-8 data points were collected from the "whole-shoot"
             experiment, the upper portion of the short shoot was removed, and
             the resistance of the shoot base and rhizome were determined by
             attaching a new tracer and reservoir to the cut shoot base.
             Tracer flux through the shoot base and rhizome was measured for
             2-4 hours before the shoot base was removed. Fluxes through the
             "mature". (proximal) and apical (distal) segments of the, rhizome
             were then measured.       Typical results for a series of flux
             measurements on two shoots are shown in Figure 12.

                  At the end of each experiment, leaf area, rhizome porosity,
             diffusion path length, and rhizome cross-sectional area were
             determined.    Image analysis of aerenchyma area was attempted
             (Figure 13), but proved unsuccessful.          Aerenchyma volume was
             determined gravimetrically by weighing a rhizome segment before
             and after flooding the internal air spaces with distilled water.




                                           7









                                     MODEL DEVELOPMENT

                 To describe gas transport in Thalassia rhizomes, we used a
             generalized version of Fick's first law:

                 F= KL  * (C Hi_CLo)
             where F    flux in mol/cm2/s  ' KL = the exchange coefficirt in
             Cm/s, CHi = the'source concentration of tracer in mol/cm , and
             CLQ@ = tracer concentration at the cut rhizome ends (Boudreau and
             Guinasso, 1978).   This equation applies to unidirectional flow
             and assumes KLI CHil and CLo are constant.
                 The exchange coefficient, KLI in turn, is defined as the
             effective diffusion coefficient (Ds) divided by the diffusion
             path length (1) measured in cm:

                 KL       Ds
             Ds is defined as follows:
                 Ds       Do 02

             where D is the diffusion coefficient of the gas in air, and 0 is
             the tor@uosity (Berner, 1980).    Tortuosity, in this case, would
             be the actual distance travelled by a tracer molecule through a
             rhizome segment with length 1.       The effective rhizome cross-
             sectional area (aerenchyma area) was calculated by multiplying
             the total cross-sectional area by the rhizome porosity.

                                  RESULTS AND DISCUSSION

                 A frequency plot of rhizome porosity is shown in Figure 14.
             Modal porosity values for all rhizomes used in our experiments
             are .27-30 %. These data indicate that aerenchyma comprise 21-40
             percent of the total rhizome cross sectional area, providing a
             ventilation system capable of transporting photosynthetically-
             produced oxygen from the shoots to the rhizomes and roots.

                 The effective diffusion coefficient (D ), the exchange
             coefficient (KL), and tortuosity (0) have been Sefined above. We
             have also separated the total tortuosity of the diffusion path in
             component tortuosities of the upper shoot and leaves, shoot
             bases, and rhizomes.

                 Plant source, or population, had statistically significant
             effects on all of the diffusive flux model parameters used as
             dependent variables in one-way analysis of variance (Table 1).
             The strength of the statistical relationship between plant source
             population and most model parameters was greater for shoot base
             and rhizome fluxes than for whole-shoot fluxes but lower than for
             rhizome-only fluxes. No significant differences between apical
             and mature rhizome segments were observed for any model
             parameters.    Significant statis   'tical effects of plant vigor
             (disease status) were noted only rhizome-only flux measurements,


                                         8







             and only for two model parameters- effective diffusion
             coefficient and the component tortuosity.

                   In whole-shoot flux measurements, Thalassia from Lassing
             Park in Tampa Bay had generally lower values         for DS and KL, as
             well as higher tortuosity values, than Florida Bay Thalassia
             (Table 2) .    In shoot base and rhizome-only f lux measurements,
             model parameters for Lassing Park and Sunset Cove plants were
             similar, while Johnson Key plants had significantly high
             component tortuosity and significantly lower D           and K     values.
             Because Johnson Key rhizomes have higher porosHies tha Lassing
             Park and Sunset Cove plants, we would expect lower tortuosity and
             higher conductance in Johnson Key plants.              Apparently, the
             conductance of rhizomes and shoots depends on more than porosity
             alone. Anatomical comparisons of rhizomes might show differences
             in. the abundance and resistance of aerenchyma cross-walls or
             septae.

                   Plant vigor had statistically significant effects on all
             flux model parameters, but only for rhizome-only flux
             measurements (Table 3).       Tortuosity and conductance values for
             healthy Thalassia and Thalassia with lesions were very similar
             for whole shoot measurements and shoot base flux measurements.
             As noted for comparisons of source population characteristics,
             Thalassia with lesions had higher rhizome porosity than healthy
             plants, but healthy plants had higher conductance (Ds and KL)
             values. Unlike the results of the source population comparisons,
             sick Thalassia rhizomes with high porosity had lower tortuosity
             values than healthy rhizomes with lower porosity.

                                      CONCLUSIONS-Task 2.1 .

                   The negative, and curvilinear relationship of whole-plant
             tortuosity to two-sided leaf surface area (Figure 15) suggests
             -that shoot maturity affects the ability of Thalassia shoots to
             provide oxygen to rhizome tissue.        For shoots with more than 75
             CM2 leaf area, changes in whole-plant tortuosity are negligible,
             perhaps as the result of rapid,lysigenous development of
             aerenchyma.     Internal resistance of larger shoots is probably
             offset by the effect of a greater surface area for exchange with
             the overlying   water.

                   We have  continued developing the oxygen flux model given the
             tortuosity and conductance parameters determined empirically in
             these experiments (Figure 16). These models provide confirmation
             of field data gathered in 1988 and 1989 which showed a
             significant relationship between late-night water column
             dissolved      oxygen     concentrations       and    rhizome       oxygen
             concentrations.     Using the data from Figure 15 and a            rhizome
             respiration rate of 0.01 unoles/g dry weight/hr at 25             OC, we
             calculate that a water column oxygen concentration of 6 ppm will
             support rhizome respiration rates up to 4 cm from any shoot. At
             2.0 ppm dissolved oxygen concentration, diffusion of water column
             oxygen'may provide approximately 30% of respiratory oxygen needs
             in the same rhizomes.



                                            9








                 These experiments demonstrate that the greatest barrier to
            gas exchange between above-ground and below-ground Thalassla
            tissue is the stem base. While component tortuosities for upper
            shoots range from 4 to 6, shoot base values range from 20 to 40.
            However, the high tortuosity of shoot bases is offset by their
            short length.  Rhizomes other than those from Johnson Key Basin,
            on the other hand, typically have tortuosities close to 1.
            Higher rhizome and shoot base tortuosities in Johnson Key Basin
            seem to be particularly maladaptive, given the high porewater
            sulfide concentrations in Johnson Key Basin sediments.

                 In summary, we have empirically determined the rates of
            diffusive gas flux through Thalassia testudinum plants and have
            estimated parameters for inclusion in oxygen models for healthy
            and diseased seagrass beds.    Preliminary comparison of healthy
            and sick Thalassia indicates that higher porosity in sick
            Thalassia rhizomes is offset by lower conductance values.     As a
            result, sick Thalassia may be less able to supply oxygen to their
            belowground tissue, and this deficiency may contribute to' the
            die-back phenomenon.



































                                        10








             SUBTASK 2.2. FIELD MEASUREMENT OF THE SUSCEPTIBILITY OF Thalassia
             AND Halodule TO SULFIDE TOXICITY.

                  In this task, we have measured survival and growth of
             Thalassia seedlings and mature shoots in die-back patches using
             natural and amended sediments. Thalassia seedlings collected in
             Florida Bay and along Atlantic beaches during Augu-st 1990 were
             grown in a culture system at the Florida Institute of
             Oceanography field station at Long Key.             In March 1991,
             we,transplanted Thalassia seedlings into pots containing
             builder's sand or natural sediment at four previous ly-studied
             sites:  Whipray Basin, Rankin Lake, Northeast Johnson Key Basin,
             and Johnson Key Basin near Johnson Key.          In April 1991, we
             transplanted apical rhizome segments with at least three emergent
             shoots into trays containing builder's sand or natural sediments
             at the same four sites.         Donor material for the rhizome
             transplants came from the northwest side of Man O'War Key in
             Johnson Key Basin.    The proximal, cut ends of the rhizomes were
             capped with rubber serum caps to reduce transplant shock.

                                            RESULTS


                  Despite our attempts to stabilize our transplant pots and
             trays, large fish (presumably toadfish, Opsanus spp.) repeatedly
             knocked over our plants in Johnson Key Basin.         While physical
             disturbance was less of a problem in Whipray Basin and Rankin
             Lake, sediment resuspension rapidly added natural sediment to all
             of our pots and trays.       As a result, some of our transplant
             mortality should probably be attributed to physical disturbance.

                  Of 78 seedlings and 52 rhizomes planted in March and April,
             respectively, 6 seedlings (7.6%) and 20 rhizomes (38%) survived
             through the summer (Table 4). While the low numbers of survivors
             preclude statistical examination of treatment effects, a few
             generalizations can be made. Survival was lowest in Rankin Lake-
             0% seedlings and 8% rhizomes.      Survival was highest in Johnson
             Key Basin, and intermediate in Whipray Basin.

                                          CONCLUSIONS


                  No conclusions regarding sediment characteristics can be
             made from the low numbers of surviving individuals.            We can
             generally conclude that, despite the passage of time since the
             major die-back episodes in Rankin Lake, the area is still
             extremely stressful for transplanted Thalassia.      While seedlings
             showed poor survival in Whipray Basin, rhizomes      did quite well.
             overall poor survival of transplant material          indicates that
             Thalassia does not have the vigor of a pioneer species like
             Halodule.   Rather, as a climax species, it it -much more easily
             disrupted.










                                       REFERENCES

           Berner, R. A. 1980.    Early Diagenesis.   Princeton Univ. Press,
           Princeton, NJ. 241 pp.

           Boudreau, B. P. and N. L. Guinasso.     1982.  The inf luence of a
           sublayer on the accretion, dissolution, and diagenesis at the sea
           floor. pp. 115-145 in K. A. Fanning and F. T. Manheim, eds. The
           Dynamic Environment of the ocean Floor. Lexington Books,
           Lexington, KY.

           Carlson, P. R., M. J. Durako, T. R. Barber, L. A. Yarbro, Y.
           deLama, and B. Hedin. 1990.       Catastrophic mortality of the
           seagrass Thalassia testudinum in Florida Bay. Annual
           Completion report to the Florida Department of Environmental
           Regulation, office of Coastal Zone Management, 51 pp.

           Mueller-Dombois, D. and H. Ellenberg. 1974. Aims and Methods of
           Vegetation Ecology. John Wiley and Sons, New York, 547 pp.

           Zieman, J. C. 1982.     The ecology of the seagrasses of south
           Florida: A community profile. U. S. Fish Wildl. Serv., Office
           Biol Servc. FWS/OBS-82/25, Washington, D.C., 123 pp.

































                                       12













                                       LIST OF FIGURES



            FIGURE 1.       Seasonal changes in area for original die-back map
            patch in Johnson Key basin.

            FIGURE 2.       Seasonal changes in area f or two die-back patches in
            Johnson Key basin (JKA and JKB) and one die-back patch in Rabbit
            Key basin.

            FIGURE 3.       Seasonal changes in areal extent of a Thalassia map
            patch within a die-back area in Sunset Cove.

            FIGURE 4.       Seasonal changes in Thalassia cover in photoquads
            located along the ecotone of a die-back patch in Johnson Key basin
            and along the ecotone of a Thalassia patch in Sunset Cove.

            FIGURE 5.       Braun Blanquet frequency, abundance, and density
            data f or Rankin Lake and Johnson and Rabbit Key basins during
            February 1991.

            FIGURE 6.       Braun Blanquet frequency, abundance, and density
            data for Rankin Lake and Johnson and Rabbit Key basins during April
            1991.

            FIGURE 7.       Braun Blanquet frequency, abundance, and density
            data for Rankin Lake and Johnson and Rabbit Key basins during June
            1991.

            FIGURE 8.       Braun Blanquet frequency, abundance, and density
            data for Rankin Lake and Johnson and Rabbit Key basins during
            August 1991.

            FIGURE 9.       Braun Blanquet frequency, abundance, and density
            data for Rankin Lake and Johnson and Rabbit Key basins during
            October 1991.

            FIGURE 10.      A summary of the seasonal changes in Braun Blanquet
            frequency, abundance, and density data for Thalassia testudinum and
            Halodule wrightii in Rankin Lake and Johnson and Rabbit Key basins
            during 1991.

            FIGURE 11.      Water temperature and salinity data for Rankin Lake
            and Johnson and Rabbit Key basins.

            FIGURE 12.      Time course of tracer flux through healthy and
            diseased Thalassia testudinum from Florida Bay.        A.'Wh.ole shoot
            flux, B. Fluxes through shoot base and rhizome, and C. Fluxes
            through rhizome only.

            FIGURE 13.      A. Cross section of Thalassia testudinum rhizome
            showing aerenchyma as open spaces. B. Computer enhanced image.








          FIGURE 14.     Frequency distribution of rhizome porosity.

          FIGURE 15.     Rhizome tortuosity versus two-sided leaf surface
          area.

          FIGURE 16.     output from a whole shoot flux model. Oxygen flux
          as a function of rhizome length.

















                                DIE-BACK PATCH AREA
                                          Johnson Key

                    70-



                    65-



                    60-



                    55-



                    50-


                    45-  A@9    O@9    D8'9   F@O    A60     AO     A60
                                            Month/Year



           FIGURE 1. Seasonal changes in area for original die-back map patch
           in Johnson Key basin.














                             DIE-BACK PATCH AREA
                              Johnson and Rabbit Keys 19 9 1
                  35-



                  30-



                  25-



                  20-



                  15-



                  10-                                           T_
                          APR          JUN         A6G         OCT
                                            Month


                                     JK-A   JK-B    RK



           FIGURE 2.   Seasonal changes in area for two die-back patches in
           Johnson Key basin (JKA and JKB) and one die-back patch in Rabbit
           Key basin.

















                                THALASSIA PATCH AREA
                                         Sunset Cove 19 9 1

                     250-




                     200-



                   C14 150-
                   E

                   t3
                   (D
                     100-




                       50-




                        0-
                             FEB        APR        J6N        A6(;       OCT
                                                  Month



            FIGURE 3.     Seasonal changes in areal extent of a Thalassia map
            patch within a die-back area in Sunset Cove.
















                            PHOTOQUAD DATA
                                   Florida Bay

                   100%_




                   75%-
                 0



                   50%-

                 0


                   25%-


                CL

                    0%     . . . . . . . .
                     M-8 9   S-89    M-9     @-q 0   M-9 1    S-91
                                        Monf h/year


                                Johnson Key1Sunset Cove



          FIGURE 4.    Seasonal changes in Thalassia cover in photoquads
          located along the ecotone of a die-back patch in Johnson Key basin
          and along the ecotone of a Thalassia patch in Sunset Cove.









                1.0 -                                               THALASSIA
                                                                    HALODULE
                0.8 -                                               SYRINGODIUM
                                                                    RUPPIA
                                                                    BATOPHORA
            >-                                                 EEM  CAULERPA
                CO). 6
            z                                                       GREEN ALGAE
            Li                                                      DRIFT RED ALGAE
            M                                                  OM   OTHER ALGAE
            Cy  0.4
            Ld
            w
            LL-
                                           N

                0,2



                0.0



                   4


            U.j
            0
                   3



                         N
                   2     N
                         x
            m            X
                         N
                         X'
                         N
                         N
                         X
                         N
                         N

                   0



                   4



                                                             7
                   3

            Ld

                   2
                         N
                         N
                         N
                         N
                   1     X


                   0
                                                             71











                                                              N
                                                              N



















                          RANKIN            JOHNSON            RABBIT
                      FEeRUARY 1991

            FI GURE 5.   Braun Blanquet frequency, abundance, and density data
            for Rankin Lake and Johnson and Rabbit Key basins during February
            1991.







              1.0                    7                       THALASSIA
                                                             HALODULE
                                                             SYRINGODIUM
              0.8                                            RUPPIA
                                                             BATOPHORA
          >_                                            EHE  CAULERPA
          U   0.6                                            GREEN ALGAE
          z                           X
                                      X
          Uj                                                 DRIFT RED ALGAE
          Z>                                                 OTHER ALGAE
          C31 0.4

          U_

              0.2

                                                      N

              0.0



                4



           Cj
           z    3


                2
           cn         N






                0



                4



           (n   3

           Ld
           C)
                2


                      N





                0
                        RANKIN        JOHNSON           RABBIT
                  APRIL. 199 1

           FIGURE 6.  Braun Blanquet frequency, abundance, and density data
           f or Rankin Lake and Johnson and Rabbit  Key basins during April
                                                     71
                     77

















                                     X
















           1991.









               1.0                     7                       THALASSIA
                                                               HALODULE
                                                               SYRINGODIUM
               0.8
                                                               RUPPIA
                                                               BATOPHORA
                       IN                                      CAULERPA
               0.6.-
                                                               GREEN ALGAE
                                                               DRIFT RED ALGAE
                                                       /X      OTHER ALGAE
           CY                                          /N
           Lij 0.4                                      X
                       N  z
                      7\
           U-
                                                        X
               0.2



               0.0



                 4


           Uj


           z



           z
           D     2
           Co


                 1



                 0



                 4




           z
           Uj          N
                       N
                 2
                                       X






                 0
                         RANKIN        JOHNSON            RABBIT

                   @JUNE 1991

           FIGURE 7.   Braun Blanquet  frequency, abundance, and density data
           for Rankin.Lake and Johnson and Rabbit Key basins during June 1991.
                                         DQ

                                       XM
                                       ',DQ
                                       'Ism
                                       IIDQ
                                       XDQ














                       N

                       N











             1.0    NZ
                    N                                  THALASSIA
                    N
                                                       HALODULE
             0.8                                       SYRINGODIUM
                                                       RUPPIA
                                                       BATOPHORA
                                                       CAULERPA
             0.6
          z                                            GREEN ALGAE
          W
                                                       DRIFT RED ALGAE
          CY                                           OTHER ALGAE
          LLJ


          U-

             0.2



             0.0



               .4



          z    3


               2
                    N
                    N
          <         \11
                    N




               0



               4




          z
          Ld

               2    X
                    N
                    N
                    l< -                        X
                    x
                    N
                    N"
                    N
                    N
                    X,                p
               0                          IN
                      RANKIN       JOHNSON         RABBIT

                 AUGUST 1991

          FIGURE 8. Braun Blanquet frequency, abundance, and density data
          for Rankin Lake and Johnson and Rabbit Key basins during August
          1991.
                                  71
                                                X
                    N
                    N
                    N
                    N
                    N
                    N
                    N
                    N

                    N
                    N
                    N
                    N
                    N
                    N
                    N
                    N
                    N
                    N
                    N'

                    X,
                    X
                    x



















                                 X
                                                X








             1.0                                       THALASSIA
                                                       HALODULE
                    N
                    x                             M    SYRINGODIUM
             0.8    'N
                                                       RUPPIA
                                                       BATOPHORA
                                                       CAULERPA
             0.6
         z                                             GREEN ALGAE
         Uj                                    /N      DRIFT RED ALGAE
                                               / XIN
                                               /N      OTHER ALGAE
         W   0.4                  N
                                  N
         U-

             0.2



             0.0                                        rrn



              4                                7

          Uj
                    N
          z   3     N
                    x
          <         X
                    N
                    N
          z
          ::D 2     X
                    X
                    N
                    Is'
          <
                    N
                    X
                    N
                    N
                    I
                    N






              4



              3
          z
          LLJ

              .2


                                     EL,         \1 r-1 rrn

                     RANKIN       JOHNSON          RABBIT

                 OCToeER 1991

          FIGURE 9. Braun Blanquet frequency, abundance, and density data
          for Rankin Lake and Johnson and Rabbit Key basins during October
          1991.
                                  71













                1.00 -
                                                                    0  Rankfn Tt
                                                                    0  Rankin Hw
           >-   0.75 -                                              13 Johnson Tt
           U                                                        N  Johnson Hw
           77                                                       &  'Rabbit Tt
           LLJ                                                      A  Rabbit Hw
           ZD   0.50 -
           CY
           W

           LL-  0.25 -







            LLJ
            U      4


            z      2

            00
            <



                   5






            V)     3

            Ld
                   2







                   0
                           FEB     APR      JUN     AUG     OCT
                                         MONTH

            FIGURE 10.   A summary of the seasonal changes in Braun Blanquet
            frequency, abundance, and density data for Thalassia testudinum and
            Halodule wrightii in Rankin Lake and Johnson and Rabbit Key basins
            during 1991.




































                             0
                               %-111      32    -

                               Ld
                               11@        28    -
                               D


                                          24    -
                               LLJ
                               CL
                               :2         20    -
                               Lij


                                          16
                                                    J S N J M M d S N J M M J S N J M M J S 0 N
                                          70    -
                                          @65   -                                                                                             0 RANKIN LAKE
                                                                                                                                              v JOHNSON KEY
                                    0     60
                                    0                                                                                                         N RABBIT KEY
                             '--          55
                             0
                                          50

                                          45
                               Z          40
                                    I
                               <          35
                               (n         .30

                                          25
                                                          S N J M M J S N J M M J S N J M M J S 0 N
                                                        1988                          1989                               1990                                1991
                                                                                            MONTH/YEAR







                             FIGURE 11.                          Water temperature and salinity data for Rankin Lake
                             and Johnson and Rabbit Key basins.





                                                                                                                  PLAFLEX # 1 OA - 12/11/91
                                                                                                                      WHOLE SHOOT FLUX

                                                     A
                                                                                                                                                                                         X
                                                                                               C3
                                                                                               APX-S"
                                                                                    wp-
                                                                                               X                                                                         X
                                                                                               "K41111                                                                                 Cm

                                                                                    sw         UATAIV


                                                                                                                                                                                  M
                                                                             T      too-                                                               CEO
                                                                             U                                                                    so                C3
                                                                                                                                            00
                                                                                                                                         w M
                                                                                                                                         C3
                                                                                                                  Cal
                                                                                                              "ZI


                                                                                                                                      TIME (min)



                                                                                                                  PLAFLEX #1 OA - 12/11/91
                                                                                                                          NO SHOOT FLUX

                                                       B


                                                                                               C3
                                                                                    800@       APX@=K
                                                                                               1%ICK
                                                                                               X                                                                            X
                                                                                               APXALT
                                                                                                 T-                                                    >C
                                                                                                                                                                                   C3
                                                                                                                                                                       C:I

                                                                                                                                                       CH)
                                                                                                                               X                 E=1






                                                                                         a                iD          ;Q              ;0               16           I@Q            16             110
                                                                                                                                      TIME (min)



                                                                                                                  P LAFLEX #I OA - 12/11/91
                                                                                                                            AHIZOME FLUX

                                                        C                           3-

                                                                                               C3
                                                                                    30w                                                                                          X

                                                                                               &Ar-&ICK
                                                                                    25OD-      X
                                                                                               Ap"M

                                                                                    2000.      OAAT41LT

                                                                                                                                         X
                                                                                    1800-                                                                                     C:3
                                                                                                                                            120

                                                                                    to0a.                                                        C3
                                                                                                                                      C3



                                                                                                          C3
                                                                                             C=I
                                                                                                          20          Z)                                            1;0            In              46
                                                                                                                                      TIME (min)


                                FIGURE 12.                                          Time                  course of tracer flux                                                             through healthy and
                                diseased Thalassia testudinum from Florida                                                                                                               Bay.                   A. Whole shoot
                                f lux, B. Fluxes through shoot base and rhizome, and C. Fluxes
                                through rhizome only.

























                                            ......... .


                                                Rig
                                                :4M,








                                jx@












































          FIGURE 13.    A. Cross section of Thalassia testudinum rhizome
          showing aerenchyma as open spaces. B. Computer enhanced image.

















           ci
           w
           t3j


                             20-

                             18-

                             16-

                             14-
                         z
                         0-  12-

           En
                             lo-

                         w
                         cf)  8-
                         0    6-

                              4

                              2-
           0

                              01
           0                          13-15 15-18 18-21 21-24 24-27 27-30 30-33 33-36 36-39
           0                                             % POROSITY

           rt-
                                  LASSING PARK    FORT DESOTO     SUNSET COVE     JOHNSON KEY















                 m  a
                 ru cl






                                           15-

                                                          WE                                                                JKB-HLT-APX

                    N                                                                                                       JKB-HLT-MAT
                    0
                                                                                                                            x
                                                                                                                            JKB-SIC-APX
                    rt                                                                                                      CK
                    0                                +                                                                      JKB-SIC-MAT
                    Fl                     10-
                    rt                                             x                                                        4_
                    rl                                                                                                             -
                    0                                X                                                                      LAS-HLT APX
                    En
                    P_
                    rt,               0                                                                                     LAS-HLT-MAT
                                      F-             +
                    M                 cl:
                                      0
                    En                F-
                    En                       57
                    rt                                                                                                 x
                    c                                                               >&
                    0








                    FJ
                                            0
                    In                        0          25          @o         i5         16o        125         150        175         200
                                                                    TWO-SIDED LEAF SURFACE AREA (cm2)

                    0
                    m

















            En H



            0 Ch                                                PLANT FLUX MODEL
            rt.
            p-                                      02=6,p=0.4,SB=2(l.5),NS=d/2(l 0),R= (1.2)
            0
            @:l                      3-
            0
            @-h 0
               r-
            @i ft                    --                               X  02=2ppm 0 6ppm           loppm

            N rt                   2.5-
            0                               +
               0                                +
                                                   +
            z P)                     2-                +
            Q
                              75
                                                           +
               0              X                               +
               FJ
               (D                                                 +
                                                                     +
                              LL  1.5    0                               +
                                                                             +
               0              z                                                 +
               0              w                                                     + +
               rt                                                                          +
                                                              0                                       +
                                                                     0                                   +
                              0
               x                                                                0
               5
               0                                                                           0 0    0,0 0
                                  0.5" X x x x
                                                                                        x  x x    x x    x

                                     0-1     1
                                     0       1      2      3      4             6       7                    lb
                                                              RHIZOME LENGTH (cm)

               x













             TABLE 1:       EFFECTS OF PLANT SOURCE, VIGOR, AND RHIZOME TISSUE TYPE ON RHIZOME POROSITY AND GAS
             CONDUCTANCE. Data are F-Ratios of one-way analyses of variance. See methods for definitions of gas
             conductance dependent variables.




                                                                                     Dependent Variables


                                                               Rhizome                            Aggregate    Component
                        Independent Variables         df       Porosity        D s        KL      Tortuosity Tortuosity




             A. Whole-Shoot Gas Fluxes


                        Plant Source                  2          ---         4.29*       5.04*        8.97***     4.00*


                        Plant Vigor                   I          ---         0.34        0.15         0.46        0.85


                        Rhizome Tissue                1          ---         0.05        0.08         0.03        0.03



             B. Shoot Base and Rhizome Fluxes (No     Shoot)


                        Plant Source                  2          ---        11.75***    12.44***      6.66**      6.55**


                        Plant Vigor                   1                      1.57        0.09         2.49        2.00

                        Rhizome Tissue                1                      1.61        3.10         0.31        2.78



             C. Rhizome Fluxes (Shoots and Shoot Bases Removed)


                        Plant Source                  2         32.12***    18.67***     6.66**       ---       19.59***


                        Plant Vigor                   1          1.62        6.84*       1.50         ---         9.45**

                        Rhizome Tissue                1          0.27        0.24        0.08         ---         0.54



                          P= 0.05; ** P= 0.01; ***    P=0.001













             TABLE 2:      EFFECTS OF PLANT SOURCE ON RHIZOME POROSITY AND GAS CONDUCTANCE. Data are mean values
             of dependent variables.       Values of the same variable with the same letter subscript are not
             signficantly different according to Duncan's multiple range test.           See methods for definitions of
             gas conductance model parameters. K     L values are multiplied by 1000 for presentation.




                                                                                     Flux Model Parameters


                                                            Rhizome                             Aggregate   Component
                        Plant Source                         Porosity       Ds         KL      Tortuosity Tortuosity




             A. Whole-Shoot Gas Fluxes


                   Lassing Park- Tampa Bay                    ---        0.011 b      0.78 b      8.00 a     11.40 a


                   Sunset Cove- Florida Bay                   ---        0.033 a      2.22 a      2.90 b      3.80 b


                   Johnson Key- Florida Bay                   ---        0.020 ab     1.29 b      3.86 b      4.04 b



             B. Shoot Base and Rhizome Fluxes (No Shoot)


                   Lassing Park- Tampa Bay                    ---        0.064 a     10.5 a       1.77 b     16.6 b


                   Sunset Cove- Florida Bay                   ---        0.064 a     10.2 a       1.76 b     21.2 ab


                   Johnson Key- Florida Bay                   ---        0.034 b      4.5 b       2.47 a     33.3 a



             C. Rhizome Fluxes (Shoots and Shoot Bases Removed)


                   Lassing Park- Tampa Bay                    0.19 b     0.180 a     31.1 a        ---        1.13 b


                   Sunset Cove- Florida Bay                   0.20 b     0.170 a     28.3 a        ---        1.17 b


                   Johnson Key- Florida Bay                   0.32 a     0.120 b     18.5          ---        1.37 a












          TABLE 3:       EFFECTS OF PLANT VIGOR ON RHIZOME POROSITY AND GAS CONDUCTANCE.        Data are mean values
          of dependent variables.       Values of the same variable with the same letter subscript            .are not
          signficantly  different according to Duncan's multiple range test.         See methods for definitions of
          gas conductance model parameters. K    L  values are multiplied by 1000 for presentation.




                                                                                 Flux Model Parameters


                                                        Rhizome                           Aggregate    Component
                     Plant Vigor                        Porosity        Ds         KL     Tortuosity Tortuosity



          A. Whole-Shoot Gas Fluxes


                     Healthy Thalassia                   ---        0.017 a      1.21 a     5.21   a    5.17 a


                     Thalasssia with lesions             ---        0.024 a      1.38 a     4.04   a    4.91 a



          B. Shoot Base and Rhizome Fluxes    (No Shoot)


                     Healthy Thalassia                   ---        0.043 a      6.54 a     2.31 a     29.7 a


                     Thalassia with lesions              ---        0.043 a      5.58 a     2.20 a     28.5 a



          C. Rhizome Fluxes (Shoots and Shoot Bases Removed)


                     Healthy Thalassia                   0.27 b     0.148 a    24.1 a         ---       1.25 a


                     Thalassla with lesions              0.31 a     0.116 b    18.5 b         ---       1.39 b





















              TABLE 4:   SURVIVAL OF SEEDLING AND RHIZOME TRANSPLANTS IN NATURAL AND
              ARTIFICIAL SEDIMENTS AT FOUR SITES IN FLORIDA BAY.








                                                 Seedlings               Rhizomes


                  Site/Sediment           Planted   Survivors        Planted   Survivors




                    Whipray Basin


                      Natural  Sediment      12           2               6            2


                      Sand                   12           1               6            4




                    Rankin Lake


                      Natural Sediment       12           0               6            0


                      Sand                   12           0               6            1




                    Northeast Johnson


                      Natural Sediment       12           0               6            3


                      Sand                   12           1               8            5



                    Johnson Key


                      Natural Sediment       --           --              6            1


                      Sand                     6          2               8            4









                                                                                                                       I    NOAA COASTAL SERVICES CTR LIBRARY



                                                                                                                            3 6668 14111714 5


































                                                      i