| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2 Department of Biology, The University of Mississippi, University, Mississippi 38677 USA; and 3 Department of Physics and Astronomy, The University of Mississippi, University, Mississippi 38677 USA
Received for publication January 12, 1999. Accepted for publication December 16, 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: ballistic seed dispersal • ballistic seed projection • Croton capitatus • Euphorbiaceae • explosive dehiscence • Fabaceae • functional morphology • Vicia sativa ssp. nigra.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Niklas, 1992
; and Vogel, 1994
). For example, the terminal velocity of various windborne diaspores of different shapes has been measured (Augspurger, 1986
) and its effect on dispersal distance has been modeled (Greene and Johnson, 1989, 1992
). Ballistic seed projection is an ideal system in which to link form and function, because seed projection angle, seed projection height, seed drag, and initial seed velocity are the consequences of a combination of physical features of the plant. Here we model seed projection distance (Beer and Swaine, 1977
; Swaine, Dakubu, and Beer, 1979
) using measured parameters. We will use the expression "ballistic seed projection" to refer to explosive dehiscence or splitting of a fruit with the forceful ejection of its contents due to the elastic contraction of the fruit. The term "projection distance" as used here refers to the possible distance a seed could be projected in the absence of other factors such as wind, rain, surrounding vegetation, or secondary dispersal, by ants, for instance. We will use the term velocity to refer to a vector comprising angle and speed.
Most ecological studies of ballistic seed dispersal have emphasized statistical correlations and descriptive characterization of selected physical characteristics and seed shadow, the two-dimensional distribution of seeds around the parent plant. For example, Lee (1984)
found no relationship between the number of seeds per pod and either the seed shadow or the mean dispersal distance in Cassia fasciculata (Fabaceae). Pod length was positively correlated with increased seed dispersal distance in Impatiens capensis (Balsaminaceae) (Schmitt, Ehrhardt, and Swartz, 1985
). Thiede and Augspurger (1996)
, analyzing seed shadows, found that pod length was correlated with decreased mean dispersal distance in Lepidium campestre (Brassicaceae), while it increased the kurtosis of the seed shadow. Projection height had the opposite effect (Thiede and Augspurger, 1996
). Beattie and Lyons (1975)
concluded that, while fruit capsules of Viola species with ballistic dispersal are held vertically on strong stems and reinforced with thick-walled cells, sclereids, fruits of species relying on ant dispersal were procumbent on weak stems and generally had larger seeds (Beattie and Lyons, 1975).
A limited number of high-speed photographic studies employing stroboscopic techniques and one camera angle have elucidated various aspects of seed and spore projection mechanisms. Hinds, Hawksworth, and McGinnies (1963)
found that Arceuthobium (Loranthaceae) seeds tumble in flight and that the viscin sheath, composed of gelatinous material surrounding the seeds, is separated from the seeds in a matter of microseconds after they are ejected. Hinds and Hawksworth (1965)
measured initial velocities of seeds of four dwarf mistletoe (Arceuthobium) species. Beer and Swaine (1977)
, Swaine and Beer (1977)
, and Swaine, Dakubu, and Beer (1979)
were the first to use the laws of projectile motion to interpret mathematically the physical relationship between seed projection angle and height and the projection distance of a seed. They coupled high-speed stroboscopic photography of ballistic seed projection in Hura crepitans (Euphorbiaceae) with theoretical models, which are the inspiration for the present study. Projection angle of seeds of Blepharis ciliaris and Ruellia brittoniana (Acanthaceae) were estimated from high-speed video by Witztum and Schulgasser (1995a, b)
. Page (1964)
microscopically photographed spore projection in the fungus Pilobolus kleinii, confirming that the spore-containing structures are expelled with a jet of fluid.
In this study we used improved techniques to determine whether plants with ballistic seed projection will project their seeds in a manner that enhances seed projection distance from the parent plant. We apply physical principles to test the following predictions: (1) seeds will be projected at angles that yield a greater distance than the average of seeds projected at random for a given initial speed; and (2) because seed projection distance increases with increasing height, more seeds will be projected from the upper rather than the lower half of the plant.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. According to USDA, NRCS (1999)
synonyms for Vicia sativa L. ssp. nigra (L.) Ehrh. include V. angustifolia L., V. angustifolia L. var. segetalis (Thuill.) W.D.J. Koch, V. angustifolia L. var. uncinata (Desv.) Rouy, V. sativa L. var. angustifolia (L.) Ser., V. sativa L. var. nigra L., and V. sativa L. var. segetalis (Thuill.) Ser. Specimens also keyed out to V. angustifolia Reichard in Radford, Ahles, and Bell (1968)
. The name Vicia angustifolia Reichard is not found in USDA, NRCS (1999)
but is listed by Steyermark (1981)
as a synonym for Vicia sativa var. nigra (L.) Ehrh. The many synonyms reflect the fact that it is difficult to distinguish among a number of species or subspecies in the Vicia genus (Gil and Cubero, 1993
). Hereafter, the two species will be referred to as Vicia sativa and Croton capitatus. Voucher specimens of both species have been deposited at the Thomas M. Pullen Herbarium at the University of Mississippi (MISS). Accession numbers for V. sativa specimens are 61460, 61461, 61462, and 61463. Those for C. capitatus are 61464, 61465, and 61466.
Vicia sativa is a climbing annual vine, typical of disturbed areas, with flowers in the upper leaf axils (Radford, Ahles, and Bell, 1968
). One of us (WJG) found that ballistic seed projection is accomplished by sudden separation of the two halves or valves of the pod along both sutures, accompanied by twisting; that submerging the twisted valves in water causes them to straighten; and that the number of seeds per pod is variable, ranging from six to 12.
Croton capitatus is an herbaceous annual plant typical of disturbed areas, especially abandoned pastures. The fruit is a tricarpellate capsule, with each carpel containing one seed. Each seed has an eliasome (Radford, Ahles, and Bell, 1968
) at its distal end. The fruit is septicidal and loculicidal, i.e., it splits along the septa between carpels, and also along the locules between the septa (Gray, 1880). After ballistic seed projection, the two parts of the valves usually remain attached to each other at their proximal end, but WJG notes that it is also not uncommon for them to separate and that as the fruit ripens the calyx becomes reflexed.
The ballistic model
We used a computer model to calculate how far from the parent plant a seed will be projected in the absence of other factors. Our model was modified from Schmid, Spitz, and Losch (1988)
and hereafter is called the ballistic model. The ballistic model requires inputs of six values. Four were mean values for each species: seed mass, seed volume, seed projection height, and seed drag. The fifth and sixth, initial seed speed and seed projection angle, were experimentally determined for individual seeds (N = 78 seeds from ten fruits of Vicia sativa, and N = 27 seeds from nine fruits of Croton capitatus). The model calculated the projection distance for each individual seed. The model also determined greatest possible projection distance and an average projection distance for seeds projected at random angles, for any given initial velocity. The ballistic model employs an iterative process whereby changing seed velocity is constantly assessed as a function of negative acceleration of drag and gravity at discrete time intervals, and the changing seed position is in turn assessed as a function of velocity at the same time intervals. We have included an analytical treatment of seed dispersal in the Appendix. The model was tested before use in these experiments with data obtained with manually projected KixTM cereal pieces and black eyed peas, recorded by stroboscopic photography.
Seed mass
Seeds of both species were weighed on an electrobalance in the laboratory.
Seed dimensions and volume
For both species, seed length, width, and height were measured for N = 54 seeds of V. sativa and N = 18 seeds of C. capitatus, using a stereoscope equipped with an ocular micrometer. The seed volume was considered to be (1/6)(
/wh).
Seed projection height
In order to test the prediction that seeds were projected from the upper rather than the lower half of the plant and to obtain a value for average projection height, in 1994 the heights of all fruits on ten plants of each species were recorded in the field as was the height of each plant (V. sativa, N = 137 fruits; C. capitatus, N = 407 fruits). No particular method was used to select the ten plants. To determine the relative projection position of the seeds we converted the continuous variable of height of fruit to a categorical variable because we were interested in the relative position of the fruit, upper or lower half of the plant, rather than the absolute height of the fruit.
Seed drag
Drag was determined by measuring the horizontal deflection, in response to different wind velocities, of a seed supported on a wire, acting as a vertically oriented cantilever spring, and then calculating the weight (mass x gravity) that it would take to deflect the wire vertically an equal amount. This method was used because the forces to be measured proved experimentally to be too small to register on a standard force transducer.
Drag is independent of the frame of reference (Vogel, 1994
), so drag on a seed exposed to moving air, as in a wind tunnel, is the same as drag on a moving seed in still air, as would be experienced by a ballistically projected seed. We used an 18-m long wind tunnel at the National Center for Physical Acoustics at the University of Mississippi, after preliminary experiments indicated that a bench top fan created unacceptable amounts of turbulence. The wind tunnel had several features that minimized turbulence, including a flow-straightening grid at the air intake and a motor located at the far end of the tunnel rather than near the intake. The wind tunnel contained a transparent Plexiglas section equipped with ports for the attachment and manipulation of an apparatus inside the tunnel. Each seed was attached to a 0.20-mm diameter wire, a high-e guitar string, via a small hole drilled into the seed. One end of the wire was clamped in a vice grip that was attached to an arm extending through a side port. Preliminary experiments showed that 8 cm of wire was best for obtaining a measurable amount of deflection and for detecting a difference between the seed on the wire and the wire alone. The wire was bent at a 45° angle just behind the seed, and the side of the seed without the wire was oriented toward the wind. Croton capitatus seeds were oriented with the flat side down and the eliasome facing into the wind. Because V. sativa seeds were more nearly spherical, they were oriented at random with respect to the wind direction. The seed was then exposed to wind speeds between 1 and 9.5 m/s for 30 s at 0.5 m/s intervals. These speeds were used because they included the ranges of the initial velocities of ballistically projected seeds measured in other parts of this study.
We recorded the horizontal deflection of the seed in response to different wind speeds by videotaping it through the Plexiglas section of the wind tunnel with a macro lens. Scale was calibrated by videotaping a ruler placed immediately below the seed. To minimize interference with wind flow, the ruler was removed before the experiments began, and the distances that the seed moved were measured later from a video monitor using the videotaped ruler to convert screen distance to real distance. The deflection of the wire alone was also filmed at the same wind velocities.
Using the same wire and the same camera distance and focus that had been used in recording seed deflection, we hung small weights on the wire and recorded its vertical deflection in the range of 0–20 mm, a range that included that of the seed's horizontal deflection in response to various wind velocities as described above. We then obtained the mass of the weights needed to deflect the wire those distances and prepared a standard curve of mass as a function of deflection distance. The slope of this line was obtained using a least squares regression and forcing the intercept through the origin (Chapra and Canale, 1988
). Because the relationship between mass and deflection distance proved to be linear, we were able to use the slope of the line thus obtained to calculate the seed drag at different velocities as
Drag was then plotted as a function of the square of wind velocity for the wire alone and for each of the means of seeds for both C. capitatus (N = 7) and for V. sativa (N = 5). These relationships were linear. The Reynolds numbers for our seeds (
850 for Vicia sativa and 1300 for Croton capitatus) indicate that they are in a range where the drag coefficient of a sphere remains fairly constant, leading to a v2 dependency (Vogel, 1994
). The difference between the drag of the wire alone and that of the mean of the seed attached to the wire for each species was assumed to be the drag of the seed alone.
Seed projection speed and angle
After preliminary experiments with high-speed video at 2000 frames per second proved to yield images that were blurred, we used high-speed stroboscopic techniques to photograph seeds as they were ballistically projected from their fruits. A custom-made wooden box of interior dimensions 141 x 141 x 141 cm served as an environmental and photographic chamber (Fig. 1). The interior was painted black to minimize light reflection during photography. An 18 x 25 cm Plexiglas window was inlaid in the removable lid so that the box interior could be viewed without disturbance of the contents. Two hinged doors allowed access to the box for manipulation of the apparatus inside. We affixed a grid to two adjoining walls of the box in order to track the positions of seeds in photographs from two cameras, as described below. We constructed the grid of red yarn because it minimized reflected light and provided contrast to the seed images, which appeared white in the developed prints.
|
After the fruit was positioned, the protocol summarized in Fig. 2 was followed. Positioning of the fruit, positioning of the motion detector, and centering and focusing the cameras took between 45 min and 1 h. Explosive dehiscence took place anywhere from 1 min to 4 h from the time the experiment started, with a typical time elapsed between turning on the heater in the box and ballistic seed projection of 15–20 min. The 19 usable sets of photographs, ten for V. sativa and nine for C. capitatus, were the result of 155 repetitions of the experiment, using 98 rolls of film.
|
We used C. capitatus fruits collected the same day as the experiment. We found that ballistic seed projection was more likely and more rapid when the fruits were attached to stems kept alive in water, so the stem was placed in a plastic film can of water and adjusted on the apparatus so that the pedicel was at 90° to the horizontal. This angle was used because it was one commonly measured in the field. Each C. capitatus fruit invariably had three seeds.
Two photographs, one from each camera, resulted from each successful experiment. Each photograph showed five images of each seed. The five images were formed by consecutive flashes of the strobe, tracing each seed's trajectory. Because some V. sativa pods had ten seeds, there were up to 50 images on each photograph that needed to be interpreted. It was not necessarily possible to select a seed in the photograph taken by camera 1 and then to identify the same seed in the photograph taken by camera 2 by visually comparing the photographs. This was done by computer by assuming that each seed had to be somewhere along a line extending from the camera, whose exact position was known, and the seed's appearance against the grid in the photograph. For each seed, the line described above for photograph 1 was compared to such a line in photograph 2. We then mathematically found the minimum distance between the two lines, and this was declared to be the seed position. The program in FORTRAN is available upon request. Only those sets of photographs with a minimum distance of less than 2 cm for every seed image were used in the analyses, because a larger distance indicated a poor match. Half the difference between these two lines was taken as the seed position.
After this procedure, the position of each seed in space was known at each strobe flash. A least squares regression on time vs. seed position in the x, y, and z directions was performed, and the slope of that line was considered to be the velocity in each direction. The initial speed was calculated from
Comparison of seed projection distances
Using the ballistic model, a maximum possible seed projection distance for the experimental range of velocities was determined. In addition, an average seed projection distance, had seeds been projected at a random combination of angles, was also calculated by the ballistic model. The distance calculated for the experimental seeds was then plotted along with the maximum and average distances, visually inspected, and the median of fractional distance error (maximum distance - seed distance)/(maximum distance) was determined for seeds of each species.
Seed projection distance indoors
We measured seed projection distances to compare later to distances predicted by the model, to see whether the predictions of the model were reasonable. During the summer of 1993, we measured projection distances of seeds of V. sativa (N = 87 seeds from seven fruits on five plants). The stem containing the dried fruits was clamped to a ringstand at a heights of 0.5–0.8 m above the floor. A 150-W incandescent bulb suspended above the plant was turned on to trigger ballistic seed projection, which usually occurred in 15 min to 1 h. Drop cloths were placed on the floor of the Old Gymnasium on the University of Mississippi campus to facilitate seed collection and measurement of seed projection distances. We drew concentric rings at 1-m intervals to 9 m on the drop cloths, thus covering a 254-m2 area.
Field fruit angles
We were curious about field fruit angles because they would ultimately affect the projection angle. For V. sativa, the angle was considered to be the intersection of the horizontal and a line extending from the point of attachment of the fruit and its pedicel to the distal end of the fruit. In order to measure these angles, we established four 9-m transects through a population. The angle of every ripe pod on every stem within 10 cm to either side of the tape was measured with a handheld protractor. In this manner, we recorded the angles of 100 fruits on 90 V. sativa plants. On C. capitatus the angle measured was considered to be defined as the intersection of the horizontal and the pedicel. We measured such angles only on calyces of fruits that had already dehisced because it appeared that as fruits matured the angle of the fruit became closer to 90°, although we did not document this. We measured each such calyx (N = 255) on ten individuals from one population. No particular method was used to select the ten individuals.
Seed projection position
A chi-square test was used to test the null hypothesis that fruits were equally likely to be found in the top and lower half of the plant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Seed dimensions and volume
Vicia sativa seeds have a mean length of 2.8 mm (SD 0.15); mean width, 2.6 mm (SD 0.12); and a mean height of 2.6 mm (SD 0.19). Those of C. capitatus have a mean length of 4.3 mm (SD 0.16); mean width of 3.4 mm (SD 0.11); and mean height of 2.7 mm (SD 0.17). Seed volumes were 9.9 x 10 -9 m3 and 2.1 x 10-8 m3 for V. sativa and C. capitatus, respectively.
Seed projection height
The mean projection height for seeds was 0.31 m (SD 0.17) and 0.68 m (SD 0.12) for V. sativa and C. capitatus, respectively.
Seed drag
The regression line of mass as a function of vertical deflection had a slope of 10.89 mg/mm when the intercept was forced through the origin. Virtually all of the variation in the measured deflection was accounted for by the regression line (r2 = 0.997). The regression line of drag of the wire on the square of velocity with a slope of 0.737 accounted for nearly all of the variation in drag (r2 = 0.998). This relationship was true for the drag measured on the seeds of both species attached to the wire. For V. sativa, the average slope of drag on the seed attached to the wire on the square of velocity was 9.39 x 10-6 N s2/m2, and for C. capitatus, 1.017 x 10-5 N s2/m2. When drag on the wire was subtracted, the slopes of the regression lines were 2.02 x 10-6 N s2/m2 and 2.87 x 10-6 N s2/m2 for V. sativa and C. capitatus respectively.
Seed projection speed
The range, distribution, mean, and variance of speeds of projected seeds of the species were similar. The range of speeds of projected seeds of V. sativa was 0.83–9.03 m/s (N = 78 seeds from ten fruits), while for C. capitatus it was 1.43–8.53 m/s (N = 27 seeds from nine fruits). The means were 4.64 m/s (SD 2.17), and 4.71 m/s (SD 1.81) for V. sativa and C. capitatus, respectively.
Seed projection angle
The species had a similar range, distribution, mean, and variance in projection angles as well. Of a possible 180° range, from -90° to 90°, the range of projection angles for V. sativa was -50° to 69°, with a mean of 25° (SD = 24°) (N = 78 seeds from ten fruits). For C. capitatus the range was -51° to 75°, with a mean of 31° (SD 25°) for N = 27 seeds from nine fruits.
Comparison of seed projection distances
Figure 3 shows the projection distance calculated by the ballistic model for each individual seed of V. sativa and C. capitatus as a function of its initial speed. Also shown for comparison are lines indicating a maximum (solid line) and average distance yielded by seeds projected at random angles (dashed line) as a function of initial seed projection speed. The median of fractional distance error between the maximum possible and calculated seed projection distances were 0.11 and 0.04 for V. sativa and C. capitatus, respectively.
|
Field fruit angles
Of 100 V. sativa pods on 90 plants, most (67) were situated in the field at angles between 20° and 60° (mean 34°, SD 23°). None were encountered at angles less than -40° (Fig. 4A). Of 255 C. capitatus fruits on ten plants, the greatest number (80) were in the range from 80° to 90° (Fig. 4B). The mean angle was 56° (SD 29°). Only six were found at angles below -10°.
|
2 = 12.7; P < 0.001). For C. capitatus, of N = 407 fruits on ten plants, all were located on the upper half of the stem. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
In our analysis, we regarded each seed as an individual observation. Other possibilities were to consider each fruit an individual observation, or even to consider each plant an individual observation. We chose to consider each seed as an individual observation because, while the pericarp of the fruit is maternal tissue, each sexually reproduced seed is a genetically unique individual. The projection of each seed is part of the parent plant's reproductive strategy, and the effect, the seed projection angle, was of interest to us independent of whether each seed was projected alone, or whether ten seeds were projected from a single fruit. It seems likely that the projection angle of V. sativa seeds would be affected by their position in the pod, but we did not analyze this. Because the fruit of C. capitatus was more uniform, seeds were projected at more uniform angles. It was not possible to ascertain whether seed projection angles varied predictably with seed mass, because multiple seeds were projected at once, and differences in mass could not be obtained from the photographs.
The results obtained with the dual camera method used are a considerable improvement over those that would be obtained using only one camera. This is because the apparent position of a seed in a photograph depends on the angle of the viewer or camera. With two camera angles, the seed's position in space can be determined rather than its apparent position from a two-dimensional photograph.
Although literature documentation of seed projection angles is rare, similar results to this study have been reported in other species. A high-speed video of Blepharis ciliaris (Acanthaceae) showed a seed projection angle of 60° to the horizontal (Witztum and Schulgasser, 1995a
). The authors proposed that in the field the average angle of the pod would lead to a seed projection angle in nature of 49°, well within the range leading to greatest seed projection distance. In Hura crepitans, one-dimensional stroboscopic photography indicated seed projection angles ranging from 20° to 48° (Swaine and Beer, 1977
). Stamp and Lucas (1983)
roughly estimated seed projection angles for Geranium maculatum, G. carolinianum, G. molle, Phlox drummondii, and Viola eriocarpa, and suggested that the mean angles ranged from a low of 45.5° for G. carolinianum to a high of 75.8° in P. drummondii. Witztum and Schulgasser (1995b)
undertook a unique and detailed engineering analysis of ballistic seed projection in Ruellia brittoniana and found that fruit structures, jaculators, direct the seeds at an upward angle of 40° as they are projected. These data strongly suggest selection on seed projections angle.
Measured seed projection distances obtained in the gym for V. sativa were generally consistent with results from the model, although distance was slightly greater than that predicted by the model for those initial velocities and angles experimentally determined, probably because the seeds rolled after hitting the ground. Similar studies have used seed traps covered with sticky material to prevent seeds from rolling (e.g., Thiede and Augspurger, 1996
). Such traps are placed only at specified locations and do not cover the entire area.
Most other reported seed projection distances are in a range similar to those of V. sativa and C. capitatus, up to 5 m, with the exception of Ceanothus cuneatus and C. leucodermis (Rhamnaceae), which dispersed their seeds up to 9 m (Evans, Biswell, and Palmquist, 1987
). In the field, the final distance a seed travels might differ considerably from the calculated ideal distance because of the influence of the surrounding vegetation or environmental factors. Vegetation or other obstacles may reduce the travel distance, and wind speed at the time of ballistic seed projection may increase or decrease the distance. For example, the distribution pattern of ballistically projected seeds of Lepidium campestre (Brassicaceae) was altered by the presence of vegetation, which decreased the mean and standard deviation of dispersal distance presumably because seeds hit nearby plants and were prevented from completing their trajectory (Thiede and Augspurger, 1996
). Winds blowing toward quadrants southwest and northwest of Tsuga heterophylla infected with Arceuthobium tsugense (Loranthaceae) were believed to be responsible for the increased number of dwarf mistletoe seeds found in those quadrants as compared to a random distribution (Smith, 1973
). This study does not address such possible variations on projection distance. However, seeds of C. capitatus, which have eliasomes, were removed from the lid of a jar by ants (W. Garrison, personal observation), suggesting that final dispersal distance may be farther than projection distance.
Most field angles of C. capitatus calyces of fruits that had already dehisced were between 80° and 90° to the horizontal. We observed that as they developed, the angle of orientation seemed to change, and that as the fruit ripened, the sepals opened. In the genus Dalechampia of the Euphorbiaceae, pre-explosion changes in the calyces were also noted. Of 16 species studied, all with ballistic seed dispersal, the calyx or bracts surrounding the fruit in all but one species dropped or withered prior to fruit ballistic seed projection (Armbruster, 1982
). Changes in the angle of Blepharis ciliaris fruits also occur prior to ballistic seed projection (Witztum and Schulgasser, 1995a
.). Fruit orientation angles in the field in both species studied here converged around particular angles, with modes at 80°– 90° for C. capitatus and 30°– 40° in V. sativa. Similar findings were reported for three Geranium species, two violets (Viola), and one species of Phlox by Stamp and Lucas (1983)
, where field fruit angles ranged from 45° to 75° with standard deviations ranging from 13° to 24°. The exact relationship between fruit orientation angles and seed projection angles has not been determined in the present study or other studies, but fruit angles necessarily affect the angles at which seeds are projected, and a convergence among fruit orientation angles suggests an adaptation for seed projection distance.
An increase in projection height always results in an increase in projection distance, all other parameters remaining equal. Whether projection height is an adaptation to seed projection is not known. Certainly there is a structural cost to supporting fruits on the upper part of the plant, just as there is a cost to leaf height on herbaceous plants (Givnish, 1982
). To be sure, the height of fruits may have any number of selective advantages, functions, or constraints. For example, a given flower height may enhance pollination. But such multiple functions are not mutually exclusive. The question of whether seeds are projected from a relative height enhancing distance can be tested independently of whether the trait is adaptive for other reasons.
Thus this study supports a hypothesis that seeds are projected at angles and from positions that enhance projection distance. In so doing, it is the first to analyze seed projection angles and velocities with a dual-angle method and is the first to accurately measure drag on small seeds by using a low-turbulence wind tunnel.
|
FOOTNOTES |
|---|
4 Author for correspondence (phone: 662-915-1089; e-mail: bywjg{at}olemiss.edu
; FAX: 662-915-5144).
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Augspurger, C. K. 1986 Morphology and dispersal potential of wind-dispersed diaspores of neotropical trees. American Journal of Botany 73: 353–363[CrossRef][ISI]
Beattie, A. J., and N. Lyons. 1975 Seed dispersal in Viola (Violaceae); adaptations and strategies. American Journal of Botany 62: 714–722[CrossRef][ISI]
Beer, T., and M. D. Swaine. 1977 On the theory of explosively dispersed seeds. New Phytologist 78: 681–694[CrossRef][ISI]
Chapra, S. C., and R. P. Canale. 1988 Numerical methods for engineers, 2nd ed. McGraw Hill, New York, New York, USA
Denny, M. W. 1988 Biology and the mechanics of the wave-swept environment. Princeton University Press, Princeton, New Jersey, USA
Evans, R. A., H. H. Biswell, and D. E. Palmquist. 1987 Seed dispersal in Ceanothus cuneatus and C. leucodermis in a Sierran oak-woodland savanna. Madroño 34: 283–293
Gil, J., and J. I. Cubero. 1993 Multivariate analysis of the Vicia sativa L. aggregate. Botanical Journal of the Linnean Society 113: 389–400[CrossRef]
Givnish, T. J. 1982 On the adaptive significance of leaf height in forest herbs. American Naturalist 120: 353–381[CrossRef][ISI]
Gray, A. 1880 The botanical text book, 6th ed., part 1, Structural botany. Ivison, Blakeman, Taylor and Company, New York, NewYork, USA
Greene, D. F., and E. A. Johnson. 1989 A model of wind dispersal of winged or plumed seeds. Ecology 70: 339–347[CrossRef][ISI]
———, and ———. 1992 Can the variation in samara mass and terminal velocity on an individual plant affect the distribution of dispersal distances? American Naturalist 139: 825–838[CrossRef][ISI]
Hinds, T. E., and F. G. Hawksworth. 1965 Seed dispersal velocity in four dwarfmistletoes. Science 148: 517–519
———, F. G. Hawksworth, and W. J. McGinnies. 1963 Seed discharge in Arceuthobium: a photographic study. Science 140: 1236–1238
Lee, T. D. 1984 Effects of seed number per fruit on seed dispersal in Cassia fasciculata (Caesalpiniaceae). Botanical Gazette 145: 136–139[CrossRef][ISI]
Niklas, K. J. 1992 Plant biomechanics. University of Chicago Press, Chicago, Illinois, USA
Page, R. M. 1964 Sporangium discharge in Pilobolus: a photographic study. Science 146: 925–927
Radford, A. E., H. E. Ahles, and C. R. Bell. 1968 Manual of the vascular flora of the Carolinas. University of North Carolina Press, Chapel Hill, North Carolina, USA
Schmid, E. W., G. Spitz, and W. Losch. 1988 Theoretical physics on the personal computer. Springer-Verlag, Heidelberg, Germany
Schmitt, J., D. Ehrhardt, and D. Swartz. 1985 Differential dispersal of self-fertilized and outcrossed progeny in jewelweed (Impatiens capensis). American Naturalist 126: 570–575[CrossRef][ISI]
Smith, R. B. 1973 Factors affecting dispersal of dwarf mistletoe seeds from an overstory western hemlock tree. Northwest Science 47: 9–19
Stamp, N., and J. R. Lucas. 1983 Ecological correlates of explosive seed dispersal. Oecologia 59: 272–278[CrossRef][ISI]
Steyermark, J. A. 1981 Flora of Missouri. Iowa State University Press, Ames, Iowa, USA
Swaine, M. D., and T. Beer. 1977 Explosive seed dispersal in Hura crepitans L. (Euphorbiaceae). New Phytologist 78: 695–708[CrossRef][ISI]
———, T. Dakubu, and T. Beer. 1979 On the theory of explosively dispersed seeds: a correction. New Phytologist 82: 777–781[CrossRef][ISI]
Thiede, D. A., and C. K. Augspurger. 1996 Intraspecific variation in seed dispersion of Lepidium campestre (Brassicaceae). American Journal of Botany 83: 856–866[CrossRef][ISI]
USDA, NRCS 1999 The PLANTS database. (http://plants.usda.gov/plants). National Plant Data Center, Baton Rouge, LA 70874-4490 USA. (Visited 10 December, 1999)
Vogel, S. 1994 Life in moving fluids, the physical biology of flow, second edition, revised and expanded. Princeton University Press, Princeton, New Jersey, USA
Witztum, A., and K. Schulgasser. 1995a Seed dispersal ballistics in Blepharis ciliaris. Israel Journal of Plant Sciences 43: 147–150[ISI]
———, and ———. 1995b The mechanics of seed expulsion in Acanthaceae. Journal of Theoretical Biology 176: 531–542
This article has been cited by other articles:
|
|
 |
|
  E. Narbona, M. Arista, and P. L. Ortiz Explosive seed dispersal in two perennial Mediterranean Euphorbia species (Euphorbiaceae) Am. J. Botany, March 1, 2005; 92(3): 510 - 516. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
