The Effect of Plant Location and Leaf Position on the Adult Leaf Form of Monstera tenuis in Costa Rica Heather J. Allen Department of Biological Aspects of Conservation, University of Madison, Wisconsin _____________________________________________________________________________________ ABSTRACT Leaf morphology of Monstera tenuis a tropical forest vine differs greatly between growth stages. The mature leaf growth form has open areas presumably to preve nt shading of the lower leaves on the same plant or to prevent overheating. Therefore, leaves highest on a plant, those exposed to most sun should show the most open area. Plants located on the edge of forest are exposed to more evenly distributed light; t herefore, they are expected to have more open area per leaf on average. These ideas were tested in the forest and on the forest edge near the EstaciÃ³n BiolÃ³gica Monteverde. One hundred and forty leaves were collected for analysis from twelve plants at each location. The edge plants had a significantly higher mean of open area per total leaf surface area F1, 264 = 6.770, P= 0.0011. The forest site showed a significant correlation between leaf height and open area per total surface area Rho = 0.0273, P =0. 0011. These results suggest that the location of a plant and the placement of a leaf may explain leaf morphology, specifically the proportion of open area per leaf. Leaf height and leaf area has a significantly positive correlation, indicating that leaf s ize is likely a result of developmental changes. Leaf area appears to be independent of environmental conditions as determined by this study. RESUMEN Las formas morfolÃ³gicas de las hojas de Monstera tenuis una parra de bosques tropicales tienen diferencias entre los niveles de edades. Las hojas maduras tienen espacios abiertos para no obscurar hojas bajas en la misma planta, y para emitir calificaciÃ³n. Las hojas que se encuentran mas altas tienen mas expo siciÃ³n al sol, deberÃa tener lo mas espacios abiertos. Las plantas que estÃ¡n en la orilla del bosque reciben el sol igualmente en todos las alturas, ellos deberÃa tener mas espacios abiertos en media. Estas ideas estaban examinando en el bosque y la orilla del bosque cerca de la EstaciÃ³n BiolÃ³gica Monteverde. ColeccionÃ© 144 hojas de 12 plantas en cada lugar. Las plantas de la orilla tienen un medio mÃ¡s alto de espacios abiertos. Las plantas en el bosque tienen una relaciÃ³n significativa entre las alturas de hojas y espacios abiertos. Los resultados sugieren que la posiciÃ³n de cada hoja y lugar de cada planta podrÃa explicar las formas morfolÃ³gicas. Las alturas y las Ã¡reas de hojas tienen una relaciÃ³n significativa. El resultado indica que el Ã¡rea de hojas es decidido de cambios de desarrollo. El Ã¡rea es independiente de la naturaleza.
INTRODUCTION Monstera tenuis Araceae is a vine native to the tropics with unique leaf morp h ology. Vines are an important and largely neglected component of tropical communit ies despite their large species contributions to flora, forest above ground cover, biomass, and litter. They have a great effect on community productivity and nutrient cycling Castellanos et al. 1986. Unlike woody species, vines invest little in self sup porting tissue; rather they concentrate on rapid growth and elongation. Competition for light may represent one of the strongest adaptive pressures for vines, demanding particular growth morphologies Castellanos et al. 1986. Juveniles of M. tenuis are usually found in the forest understory where light levels are very low and diffuse. Oberbauer and Noudali 1998. The plant produces leaves of differing shapes and sizes depending on the growth stage at which each leaf is generated. They begin as roun d leaves flattened against the host trunk, whereas leaves higher on the plant are large, pinnatifid, and horizontal. The plant produces leaves of differing shapes and sizes depending on the growth stage of the plant. The progression to mature leaves is a g radual heteroblastic change that is, it will occur without environmental cues Ray 1983. However, vines on trees exposed to higher light levels produce leaves of 25 30 cm more quickly, at which point the morphology changes. Thus, the transition can occu r lower on the host trunk due to heightened growth rates induced by increased light exposure Ray 1983. Many vines display marked change in leaf size and shape as they grow taller or older and presumably have well developed physiological plasticity. Yet, few species may have the capacity to produce leaves that function efficiently in the understory and flourish in the upper canopy Lowman and Nadkarni 1995. The differing ecological requirements of the understory and canopy necessitate specific growth for ms in vines. Three distinctive growth forms are observed in M. tenuis : the leafless seedling, the climbing stem, and the descending stem ground runner. The climbing stem produces the juvenile form with small round leaves held against the trunk. Once they a ttain the size of 25 30 cm the leaves develop deep clefts and are pinnatifid; this is the mature leaf form. These leaves are held away from the trunk and can increase to 125 cm in length Ray 1983. Mature leaf forms on a single plant can vary dramaticall y in shape. Surprisingly, a mature pinnatifid leaf may be highly dissected, while the leaf located on a subsequent node will be entire, lacking visible clefts. The gradation may be a consequence of environmental cues. The relationship between open areas pe r total leaf surface area will be referred to as Â€open areaÂ. There are several ecological explanations for the phenomenon of Â€open areaÂ in mature leaves. Â€Open areaÂ in leaves may be a method for M. tenuis to prevent self shading, a need that understory plants often share Rundell and Gibson 1996. Self shading is especially problematic in an understory environment in which leaves must successfully intercept and use diffuse, rather than direct light Oberbauer and Noudali 1998. Highly lobed leaves allow light to pass through the new growth to leaves below, thereby maximizing use of the solar energy available to the entire plant.
In addition to preventing self shading, Â€open areaÂ may assist in alleviating the stresses associated with accumulated heat. L eaves found higher on a vine are often exposed to more sun and heat; and may require more rapid dissipation of accumulated heat. An increase in the number of lobes provides a higher surface area to volume ratio, therefore a higher rate of heat dissipation Taiz and Zeiger 1991. I expected to find an increase in Â€open areaÂ on leaves located higher on a plant, because the associated advantages such as heat dissipation and non self shading increase as with leaf height. I also expect to find that plants exp osed to more even distributions of light, such as those in a forest will exhibit more Â€open areaÂ for the same ecological benefits. Light in the forest filters through several layers of leaves, therefore light is more diffuse in general, but exhibits a stronger gradient of light levels because of the obstructions. MATERIALS AND METHODS I performed the study in the lower montane wet forest Holdrige Life Zone near the EstaciÃ³n BiolÃ³gica Monteverde, from the dates of October 20 through November 13 of 2000. Two transects were used as study sites, and 12 vines were sampled at each site. The first transect was located on the edge of an undisturbed climax forest. I sampled vines located one meter or less into the forest away from the pasture edge. This area is exposed to more sunlight, than th e continuous forest. The forest transect was located at an area farther in the continuous forest least four meters from the pasture edge within the forest interior. I collected 12 leaves from each vine at least two meters tall, starting at the first leaf nearest the ground and obtaining every other one. I used an extensible leaf clipper, with a tape measure attached to measure and record vine height and the height of each leaf. Leaves were numbered consecutively beginning with the first leaf closest to th e ground, such that leaf number corresponds with height. I determined that percentage of Â€open areaÂ by using a one centimeter square grid transparency Fig. 1. All data were tested for normality using the Chi squared test. I used a Spearman Rank Test to test for correlation between the height of the leaf on each plant and Â€open areaÂ. Spearman Rank was repeated to test for a correlation between leaf height and actual leaf area. I used a Two Way Analysis of Variance to test the effects of leaf number cor responding to increasing leaf height, and plant location on mean Â€open areaÂ. To test the effects of leaf number and plant location on mean leaf area, I used a Two Â‚ Way Â‚ Analysis of Variance. RESULTS I collected 12 leaves from each of 12 plants on the ed ge and in continuous forest sites, and quantified the Â€open areaÂ of 288 leaves total. The distributions of leaf height and leaf area and Â€open areaÂ were normal in both the edge and forest sites Table 1. There was no significant correlation between leaf height and Â€open areaÂ in the edge site Spearman Rank; Rho corrected for ties = 0.043, Tied P Value = 0.6077. However, the continuous forest site had a significant correlation between leaf height and Â€open areaÂ
Fig. 2. Leaf area shows a strong positi ve correlation with leaf height in both the edge and forest sites Fig. 3, 4. Leaf number had a significant effect on mean Â€open areaÂ as did site location Fig. 5, 6. FisherÂƒs PLSD showed significant differences between pairwise comparisons of leaves in regards to actual leaf area Table 2. FisherÂƒs PLSD revealed significant differences in average Â€open areaÂ Table 3. Often leaves located furthest away from one another other on the plant, show the greatest differences in actual area and Â€open areaÂ. Finally, though site and leaf number affects mean Â€open areaÂ, there is no significant interaction effect Fig. 7. Leaf area in both sites increases with leaf number, but site does not have a significant effect on leaf area Fig. 8. DISCUSSION Forest Monstera tenues located in the forest site have a significant correlation between leaf height and leaf area. As one move towards the canopy the leaves on M. tenuis plants in the forest, become larger. The forest edge also shows this same correlation. Presu mably, because both sites share the trend, environment is not an essential factor. Larger leaves are produced at higher heights, due to developmental progression. This result supports the conclusion that increased leaf sizes, as well as the basic morpholog ical successional changes, are heteroblastic Ray 1983. Light availability under the forest canopy is limited to sunflecks that filter through the treetops. The highest leaves are exposed to the most sun and heat; and they may shade lower leaves on the p lant. The leaves of the forest vines exhibit a significant correlation between Â€open areas Â with increasing leaf height, perhaps to prevent over heating or self shading. Leaf number, which corresponds to leaf height on the plant, also correlates with Â€open areaÂ in the forest. This result lends support to the hypothesis that leaves in the forest benefit most ecologically, if higher leaves have more Â€open areasÂ. In the understory, M. tenuis plants have a lower mean of Â€open areaÂ than those plants loc ated on the forest edge. Light is diffuse and less available in the forest understory than on the forest edge. Monstera tenuis is exposed to less sun and therefore is likely to accumulate heat. Therefore prevention of self shading may be more relevant to the understory environment, because light availability is limited and light is vertically stratified. Edge Leaf height and leaf area are positively correlated in plants located on the forest edge, as was observed in the continuous forest. It is apparent t hat site has no effect on leaf area and therefore the gradients of leaf area observed in M. tenuis are the results of growth patterns which are not influenced by plant location. Leaves on vines located in the edge site did not show a significant correlation between leaf height and Â€open areaÂ. This result can be explained by the abiotic
conditions of edges. Edge effects can arise from habitat fragmentation, and forests can experie nce an altered microclimate along their edges; conditions may be hotter, more desiccating and windier at the edge and as much as 100 meters into the forest Wheelwright 2000. Plants found in edges receive more sunlight from oblique angles because they bor der unobstructed areas such as pastures. Leaves at varying heights are exposed to similar levels of light and heat; therefore, they have similar heat dissipation requirements; subsequently, amounts of Â€open areaÂ do not correlate with height. In addition, the lateral light availability should exclude possible self shading as a cause of increasing leaf Â€open areaÂ with increasing leaf height. Leaf number positively correlated with Â€open areaÂ in both study sites. This correlation is surprising in the edge s ite because, leaf height does not correlate with Â€open areaÂ on the forest edge. Leaf number increases consecutively up the vine towards the canopy, as doe s leaf height. The discrepancy may be explained by variation in internode length. If leaves were not consistently distant from one another then tests using the two variables leaf height and number may show different results. Alternately, there may be a general trend of increasing Â€open areaÂ with height on the edge. This trend is not statistically signifi cant, yet it is helpful when explaining the correlation between leaf height and Â€open areaÂ. The photosynthetic capacity of a collection of leaves will be greatest if all leaves receive an intermediate, non saturating level of illumination such that the r ate of change of assimilation with light intensity is maximized for each leaf. Within an individual plant, this can be achieved by reducing the light incident on surfaces near the top of the crown and thus increasing light arriving at leaves farther down Lowman and Nadkarni 1995. This alternative to the self shading theory may explain the even distribution Â€open areaÂ of M. tenuis in the edge site. The edge plants have a higher mean Â€open areaÂ. The edge site receives more light on average at all stra ta. Therefore it is beneficial to have more Â€open areasÂ to prevent over heating. Amount of Â€open areaÂ most likely increases in the edge due to the increased availability of sunlight. Environmental factors appear to be responsible for the gradation in ope n area that corresponds with increased Â€open areaÂ on the forest edge. These same factors may help explain increasing Â€open areaÂ with height found in forest vines. While intuitive, the higher mean Â€open areaÂ on the forest edge may result from sampling m ethods. Edge plants were sampled more completely from bottom to top, because they are on average shorter than those sampled in the forest. One must reach the top of the canopy to obtain a representative sample of the average Â€open areaÂ for a vine located in the forest. I was unable to reach the uppermost leaves of the forest vines the leaves expected to have the most Â€open areaÂ. If these leaves had been included in the statistics they may have increased the average of Â€open areaÂ for all the forest leaves . Further investigation of the causes of Â€open areaÂ of M. tenuis leaves would be beneficial. A study to determine the effect of light, temperature and self shading at each
leaf location would help to tease apart the two possible causes for the amount of open area found in each leaf. ACKNOWLEDGEMENTS I would like to thank the effervescent Karen Masters for making a daunting project pleasantly challenging. I would like to thank Andrew Rodstrom for tireless academic and life support; he is truly a Renaissance man. Thanks to Tim Kuhman for his calming influence. I give many thanks to my host family who embraced me with warmth and affection. Finally thank you to the many friends I have made here, I will never forget you. I would also li ke to thank Arnoldo Beche and the EstaciÃ³n BiolÃ³gica Monteverde for the generous permission to study on their property, without whom this study would have been impossible. LITERATURE CITED Castellanos, A., Mooney, H., Bullock, S., Jones, C., and Robichaux, R. 1986. Leaf, Stem and Metamer Characteristics of Vines in a Tropical Deciduous Forest in Jalisco, Mexico. Biotropica. 21: 41 45. Lowman, M., and Nadkarni, N. 1995. Forest Canopies. Academic Press. San Diego. 312, 412 419. Oberbauer, S., and N. 1998. Potential Carbon Gain of Shingle Leaves in Juveniles of the Vine Monstera tenuis Araceae in Costa Rica. American Journal of Botany. 856: 850 854. Ray, T. 1983. Monstera tenuis In: Costa Rican Natural History. D.H. Janzen, ed. The University of Ch icago Press, Chicago. 278 280. Rundel, P. and Gibson, A. 1996. Adaptive Strategies of Growth Form and Physiological Ecology in the Neotropical Lowland Rain Forest Plants. In: Neotropical Biodiversity and Conservation. Gibson, A. C. ed. University of Califo rnia, Los Angeles. 33 71. Taiz, L. and Zeiger, E. 1991 . Plant physiology. The Benjamin/Cummings Publishing Company, Inc, Redwood. 256. Wheelwright, Nathaniel T. 2000. Conservation Biology. In: Monteverde Ecology and Conservation of a Tropical Cloud Forest. Nadkarni, Nalini M. and Wheelwright, Nathaniel, T. eds. Oxford University Press, New York 422.
_____________________________________________________________________________________ TABLE 1. The distributions for leaf height and Â€open areaÂ and leaf height and total leaf area in both the edge and forest sites were normal as defined by the perimeters of the Kolmogrov Smirnoff Test. __________________________________________________________ ___________________________ Edge Continuous Forest Height Â€Open AreaÂ Height Â€Open AreaÂ Sample Size 144 144 144 144 Chi square 0.889 6.125 2.0 6.125 P value >0.999 0.0935 0.0735 0.0935 Height Total Area Height Total Area Sample Size 108 108 108 108 Chi square 0.463 3.630 0.463 4.167 P value >0.999 0.3257 >0.999 0.249 _____________________________________________________________________________________ TABLE 2. Several pairwise comparisons of leaves on M.tenuis showed significant differences in leaf surface area. Leaves farthest from each other on the plant often showed the most deviation between surface area values . Significant values are included on the table. _________________________________________________ ____________________________________ Leaf Number Leaf number Mean Difference Critical Difference 1 12 146.556 115.225 2 11 121.98 101.688 3 9 130.352 99.072 3 10 140.688 98.103 3 11 169.559 97.261 3 12 212.733 97.261 4 11 97.826 97.261 5 12 133.222 92.225 6 12 113.833 92.225 7 12 118.65 89.601 8 11 91.076 89.601 8 12 134.25 89.601
_____________________________________________________________________________________ TABLE 3. Several pairwise comparisons of leves on M. tenuis showed significant differences. In Â€open areaÂ . Leaves furthest from each other on the plant often showed the most variation between respective Â€open areaÂ values. Only significant combinations are present in the table. ________________________________ ______________________________________________ Leaf Number Leaf Number Mean Difference Critical Difference 1 5 0.041 0.03 1 6 0.047 0.03 1 7 0.052 0.03 1 8 0.067 0.03 1 9 0.05 0.03 1 10 0.063 0.03 1 11 0.063 0.03 1 12 0.076 0.03 2 8 0.045 0.03 2 10 0.041 0.03 2 11 0.041 0.03 2 12 0.54 0.03 3 7 0.034 0.03 3 8 0.049 0.03 3 9 0.032 0.03 3 10 0.045 0.03 3 11 0.045 0.03 3 12 0.058 0.03 4 7 0.032 0.03 4 8 0.047 0.03 4 9 0.03 0.03 4 10 0.043 0.03 4 11 0.043 0.03 4 12 0.056 0.03 5 12 0.035 0.03
FIGURE 7. The interaction between leaf number and plant location though interesting is not statistically significant Two way Anova, F 11,264 = 0.728, P Value = 0.7112. Monstera tenuis display a general trend of increasing mean of Â€open areaÂ with increasing leaf number on plant and a higher average mean in the edge site. FIGURE 8. Leaf number increases in M. tenuis with mean leaf area, in the edge and forest sites. Two Way Anova, F 11,193 = 2.787, P Value = 0.0022. Site does not have a significant effect on leaf area. Two Way Anova, F 1, 193 = 0.608, P Value = 0.4365.