PRODUCTIVITY AND HEEBIVOHY IN HIGH AND LOH DIVERSITY TROPICAL SUCCESSIONAL ECOSYSTEMS IN COSTA BICA BY BECKY JEAN BROWN A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE BEQUIREMENTS FOR THE DEGREE Of DOCTOR Of PHILOSOPHY UNIVERSITY OF FLORIDA 1982 ACKNOWLEDGMENTS This research was part ot the University of Florida/CATIE cooperative study. Natural Succession as a Model Cor the Design of New Tropical Aqroecosystems. The research was supported by NSF grants DEB 78-10721 and DEB 80-11136, Dr. John J. Ewel, Principal Investiqatoc. A pilot study to investigate herbivory measurement techniques was funded by a Research Initiation and Support (SIAS) grant from the National Science Foundatior., awarded by the Organization for Tropical Studies. Data were analyzed using the facilities of the Northeast Regional Data Center, University of Florida. I am grateful to Jack Ewel for invaluable guidance and support throughout the project. Ariel Lugo, Howard T. Odun, Edward Deevey, and Dana Griffin provided encouragement and many useful suggestions. I thank Hon Harrell for collaboration in developing herbivory rate equations; Martin Artavii L., Cory Berisii, Chantal Blanton, Don Antonio Coto M., Luis Coto M. , Richard Hawkins, and Norm Price for generous assistance in the field and laboratory; Grace Bussell fcr logistical support; Dawn Green, Laura Jimenez, Chris McVoy, and Doris Randolph for assistance in data processing; and George Fuller for illustrations. TABLE CF CONTENTS ACKNOWLEDGMENTS ii ABSTRACT vi CHAPTER J&age I. INTRODUCTION 1 Related Research 1 The Diversity-Stdbiiity Issue 1 IiBpacts of Hecbivory 6 Direct ifflpacts on uet primary productivity 7 Impacts on species composition and diversity 8 Diversity Effects on Herbivory 11 Research Questions ........ 12 II. METHODS 13 The Study Site 13 Site Preparation 15 Main Treatments 16 Natural Succession 16 Mimic of Succession ..16 Enriched Succession .. 18 Successional Monoculture 18 Plot Layout and Variables Measured . 20 Measurements of Vegetation Structure ...... 23 Leaf Area Index 23 Species Composition 24 Vegetation Height 25 Productivity Measurements 25 Above-Ground Biomass ..28 Litterfall 30 Herbivory Rates ...... 31 Estimation of Hole Expansion 38 Subtreatments 45 Background Herbivory ..U5 Decreased Herbivory 45 Increased Herbivory 47 III. tlESaLTS 50 Veyetatioa Structure ......50 Species Composition .. ..50 Leaf Area Index 65 Herbivory Bates 74 Above-Ground Bioaass Ill Litter 122 Above-Ground Productivity 126 Effects of Decreased Herbivory 133 Sates of Herbivory in Insecticide Plots . . 133 Species Composition ..... 145 Leaf Area Index 150 Above-Ground Biomass 156 Litterfail 162 Above-Ground Productivity 166 Responses to Artificial Defoliation 172 aesuits of Preliminary Study ....... 172 Responses to Repeated Defoliation 174 Changes in leaf productivity 174 Changes in vegetation structure .... 182 Changes in species cciaposition 188 Cassava biomass 200 IV. DISCDSSION 202 Net Primary Productivity 202 Relationship Between Net Primary Productivity and Diversity 202 Continuous Biomass Accumulation in Diverse Systems 205 Continuous Biomass Turnover in Diverse Systems 210 Importance of Standing Dead Biomass .... 210 Herbivory 214 Low Herbivory Sates 214 Absolute Losses and Diversity Not Correlated 222 Percent Losses Correlated witu LAI .... 224 Effects of Plant Species Composition . . . 228 Plant Herbivore Defenses 232 Structural Complexity 233 Herbivory, Diversity and Energy Flow 234 Energy Flow Model 235 Resilience of High and Low Diversity Ecosystems ............. 241 LITERATORf CITED 249 APPENDIX A. CALCULATION OF HERBIVORY BATES 267 APPENDIX E. 3I0MA3S AND LITTEKFALL MEANS BIOGRAPHICAL SKETCH 273 292 Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirenients for the Degree of Doctor of Philosophy PRODDCTIVITY AND HEfiBiyOBY IN HIGH AND LOH DIVERSITY TROPICAL SUCCESSIONAL ECOSYSTEMS IN COSTA RICA By Becky Jean Brown December 1982 Chairman; Dr. John J. Ewel Major Department: Department of Botany Above-qcound net primary productivity (NPP) , herbivory and vegetation structural characteristics were measured in high and low diversity successional and agricultural ecosystems at a wet tropical site near Turrialfca, Costa Bica. Insecticide and defoliation experiments were performed to evaluate the effects of herbivory on NPP in high and low diversity ecosystems. The four experimental ecosystems were enriched succession (natural regeneration augmented by propagule additions) , natural succession (control) , successional mimic (an ecosystem with investigator-controlled species composition designed to imitate natural succession), and successional monoculture (two naize crops followed by cassava) . Plant species richness and leaf area index (LAI) were highest in the enriched, high in the natural succession, intermediate in the miiiiic, and low in the monoculture at 1.5 yr. Net primary productivity, estimated from bioaass increments adjusted for turnover, was not related to ecosystem complexity. The NfP was highest in the most diverse (enriched) and least diverse (monoculture) systems. More than ^'2% of the above-ground production was lost annually through litterfall, plant mortality and herbivory. Standing deal biomass that did not fall into litter traps accounted for a significant fraction of total turnover in ail ecosystems. Herbivores consumed approximately the same amount of leaf tissue per m^ of ecosystem in each of the thrte diverse systems (54-6 1 cm^ m-z ground day~*). Consumption expressed as a percent of total leaf area was higher in the ecosystem with lower LAI (the mimic) . Absolute and percent losses were lower in the monoculture than in the other ecosystems. In the less diverse systems containing cultivars, herbivory had high temporal variability. Species' herbivory rates ranged from <1 to 131 cm^ m-2 leaf day-» and appeared to be related to palatability , ecosystem LAI and species composrticn. Herbivory stimulated NPP over a wide range of herbivory levels in both the diverse system and the monoculture. The stimulatory effect was greater, and Baximum stimulation occurred at a higher herbivory level, in the diverse system. The resilience of the diverse sy^jtem, due to compensatory fluctuations in douinance of co-occurrinq species, has important implicatious for agroecosysteit desiyu. CHAPTER I INTaODUCTION Cofapiex traditiouai agroecosysteais in the huniicl tropics havt: persisted fot many years without the use of pesticides, while introduced monocultures have otten been plagued by pest attacXs that lead to decreased crop productivity. The magnitude of pest probleas in an agrcecosystem may be related to tue degree of similarity between the aqroecosystem and the natural systeo it replaces. The hypothesis is that the natural ecosystem possesses structural and functional characteristics that allow it to survive in its environment, and the more similar the agroecosystem is to the natural system, the greater is its chance for success. The objective of this study was to investigate herbivory and primary productivity in ecosystems structurally similar and dissimilar to a diverse tropical successional systeui. Belated Research The Diversity-Stability Issue In lidition to the goal of maximizing production per unit of energy input, tropical agriculturists are interested in two other properties of agroecosystems; stability and 2 sustainibiiity. A stable ag roecosystem lacks fluctuations in productivity (or variability in yield) over time, and a sustainable agroecosystem has the ability to persist in the face of perturbations (Conway 1982). Many complex traditional agroecosystems have high sustainability and high stability, and it has been suggested that these characteristics are a function of their diversity (Soemarwoto and Soemarwoto 1979, Gliessman et al. 1981) . Interest in the stabilizing effect of diversity in agcoecosystfeiBS is reflected in the expressed need for development of complex agricultural systems for the humid tropics (Holdridge 1959, Dickinson 1972, Tcenbath 1975, Hart 1980), and in the current agronomic emphasis on polyculture cropping systems research (Dalryiaple 1971, Kass 1978). k large body of literature on the theory of diversity-stability relationships in ecological systems bears directly on the guestion of agricultural diversification as a means of reducing pest problems. The traditional belief for many years among ecologists was that diverse systems were more stable than simple ones. Strong support of this view was expressed by most contributors to a symposium volume on the topic (Hoodwell and Smith 1969) . Subsequent work, including empirical studies and development of matheaaticai models (see work cited by Goodman 1975) , did not support the original hypothesis. Goodman (1975) reviewed the development of the diversity-stability theory 3 in detail and concluded that there is no clear relationship between ecosystem diversity and statiility. Empirical studies have yielded inconsistent and contradictory results, partly due to disagreement among ecologists both on the definition of the term "stability" and on appropriate criteria for measuring it. Many empirical studies to test the relationship between diversity and stability have considered fluctuations in numbers of individuals within a single population or trophic level; fewer studies have considered the effects of diversity on ecosystem properties such as energy flow and nutrient cycling. Holling (1973) distinguished between stability (small fluctuations around an eguilifcrium point) and resilience (ability of a system to persist by moving between multiple eguilibria) . Using these definitions the spruce-fir forest of eastern Canada is an unstable system that fluctuates widely in plant and animal species composition. However, because of the instability of populations and the resulting effects on competition, regeneration and forest growth rates, this system has very high resilience (i.e., it persists). In McNaughton's (1977) restatement of the diversity-stability hypothesis, the emphasis was on stability of ecosystem processes rather than stability of population numbers. Process stability and population stability are not necessarily related. As Margalef (1975, 4 page 160) stated, "A system which is highly unstable in species composition may be stable with celaticn to the energy flawing through it." la general, a system will tend toward the configuration of species that best processes the available energy, thus maximizing energy flow (Odum and Pinkerton 1955) . Odua (1975) proposed that the optimal diversity of a system is a function of the sources and guantities of available energxes. He calculated diversity indices from enpiricai data on plant and animal species abundances iu a variety of ecosystems. The freguency distribution of the diversity indices was bimodal. Stressed, selectively managed and subsidized ecosystems had low diversity indices; natural ecosystems where solar radiation was the primary energy source had high diversity indices. Lugo (1978) emphasized the importance of energy drains, as well as energy sources, in determining system complexity. It is generally accepted that ecosystem complexity and efficiency of energy use are positively correlated (see Hargalef 1968), and it has been hypothesized that plant diversity is positively associated with primary productivity (Connell and Orias 1964, Margalef 1968, H. T. Odua 1971). However, the development and maintenance of diversity requires energy expenditures and the complexity of an ecosystem is determined by the balance between energy inputs and energy drains (H. T. OduOi 1971, Lugo 1978). For 5 examplvj, very productive systems with low energy drains have high diversity (e.g., a coral reef), while very productive systems with high energy drains have low diversity (e.g., an estuary with tidal exports of organic matter) . m a natural ecosystem, high diversity of components provides many possible pathways for the flow of energy. When a high diversity system is stressed, either by a fluctuation in the energy inputs to the system or by an increase in energy drains from the system, the dominant energy patnways change, but the system may still be able to process the available energy. High diversity results in more alternative eguilibrium states of the system (Holling 1973) , which provide more options for maximizing energy flow under fluctuating conditions. Diversity, then, is a homeostatic mechanism operating at the ecosystem level that insures continuous energy flow through the system (Seichle et al. 1975). Species abundances change when a perturbation occurs, the decreases in some species are compensated for by increases in other species, and by this mechanism ecosystem functional properties are stabilized (McNaughton 1977). Lugo (1978) proposed that the ability of a system to respond to a perturbation depends on the dynamics of the system's energy pathways, the type and intensity of the perturbation, and the kinds and numbers of pathways altered. 6 Impacts of Hecbivory Herbivocy stresses the ecosystem by draininq energy from plant biomass. In natural ecosysteas, herbivoty is a normal or background stress to which the system is usually well adapted (Lugo 1978). In ecosystems that are not well adapted tc herbivore stress (e.g., many agricultural systems and natural systems with introduced pests) , herbivory may ultimately atfect the ability of the system to persist through its impact on energy tlow. Herbivory may alter energy flow through the primary producers in two ways: (1) directly, by reducing the amount of photosynthetic tissue and by stinulating compensatory growtn in remaining tissue, and (2) indirectly, by affecting structural and functional characteristics of the system, which in turn alter the primary productivity rate. Although insects generally consume only a small fraction of the leaf tissue in a terrestrial ecosystem, the effects of herbivores are greater than simply loss of leaf area (Harper 1977, Whittaker 1979, Lubchenco and Gaines 1981). Herbivory influences ecosystem structure and function by increasing light penetration and reducing competition for nutrients, water, and light. Herbivory may accelerate nutrient cycling through increased nutrient leacning from damaged foliage and increased decomposition rates (Mattson and Addy 1975, Golley 1977, Bormann and Likens 1979, Barbour et al. 1980). Herbivores act as ecosystem regulators 7 through direct and iadirect ttedback loops to the autotrophs (Odum and fiuiz-Beyes 1970, Chew 197a, Hattsoa and Addy 1975, Lee and Jniaan 1975) . The effects oi herbivores on systeo processes aay be positive or negative, depending on the characteristics and state of the systea (Lugo 1978). Direct impacts on net primary productivity. Moderate amounts of herbivory laay stiouiate plant productivity under certain conditions (McNaughton 1979a), and conpensatory growth fcilowing defoliation has been well documented (Alcock 1962, Pearson 1965, Hodgkinsou et al. 1972, Gifford and Marshal 197J, McNaughton 1976, Detliug et al. 1979, Painter and Detling 1981) . Many plants normally photosynthesize at less than their maximum rates. It has been suggested that the relationship between herbivory and net primacy productivity (NPP) is nonmonotonic, and there is an optimum grazing level at which NPP is maximized (HcHaughtcn 1979a) . Although herbivory is usually considered a stress to the plant community, stress may accelerate processes and in some cases benefit the system (Lugo 1978). Stimulation of plant productivity by grazing is an example of a positive feedback loop within the system that amplifies energy flow (Oduia 1977) . Feedback may be negative rather than positive at high herbivory levels, and there is a threshold herbivory level above which plant productivity decreases (Vickery 1972, Dyer 1975, Noy-Meir 1975, Caughley 1976) . 8 Impacts on sjjecies cpinposition and diyersity . Individual plant responses to herbivory may be positive or negative, depending on plant genetics, intensity and frequency of defoliation, the tissues affected, plant developmental stage at tke tine of attack, and environmental factors (McNaughton 1979a) . Herbivory may lead to a variety of physiological responses in the individual plant. These include (1) plant mortality and reduced growth (Kulman 1971); (2) alteration of plant resource partitioning (Gifford and Marshal 1973, Detling et al. 1979) ; (3) stimulation of compensatory growth in residual tissue (Pearson 1965, Hodgkinson et al. 1972, Dyer 1975, McNaughton 1976, 1979a, Detliug et al. 1979, Painter and Detling 1981) ; (4) increases or decreases in plant reproductive output (Jameson 1963, Cavers 1973, Bockwood 1973, Harris 1974, Owen and hiegert 1976, Boscher 1979, irinter and Kalman 1979, Bentley et al. 1980, Stephenson 1981); (5) changes in plant growth patterns, such as increased branching or tillering (Oppenheimer and Lang 1969, Youcgner 1972, Saunders 1978, Simberloff et al. 1978, Owen 1980) ; (6) increased or decreased root growth (Troughton 1960, Alcock 1962, Jameson 1963, Taylor and Bardner 1968, Dunn and Engel 1971, Hhittaker 1979); (7) delay of plant senescence (Chew 1974, McNaughton 1976) ; (8) increased water I'se efficiency, due to reduced transpiration are.. (Daubenmire and Colwell 1942, Baker and Hunt 1961) ; and 9 (9) reduced nutritive quality of remaininq leaf tissue {bchuJtz and Baldwin 1932). Plant responses; to herbivory reflect a ccmplex interaction of tactors. The net result of herbivory at the coffiiiiunity level is a change in competitive advantage among species. As Whittaker (1979) pointed out, the competitive balance among species is altered by herbivory regardless of whether an individual plant is damaged or benefited. Results of numerous studies (e.g., Kalonc 1969, Kafes 1970, Harris 1973, McNaughton 1979b, Linhart and Whelan 1980) support the generalization that herbivory shapes the plant species composition of an ecosystem by altering the competitive balance among species. Instances of successful biological control of plant pests by introduced insects are examples of the impact that herbivory can have on plant species composition (see DeBacL 19714). By affecting competition, herbivory may regulate plant diversity in an ecosystem. It has been suggested that herbivory may maintain local species diversity ty keeping plant populations at low densities and by increasing niche differentiation (Whittaker 1965, Connell 1971, Huffaker 1971, Harris 1973). Grime (1973) predicted that herbivore-susceptible species would he outcompeted at high grazing rates, herbivore-resistant species would be outcompeted at low grazing rates, and therefore nighest species diversity would occur at intermediate grazing 10 intensities. Lubchenco and Gdines (1981) hypothesized that diversity would be a maximuoi at low or interaediate herbivore levels, depending on the nature of the competitive interactions between plants. Harper (1969) and Caughiey and Lawton (1981) suggested that the ettects of predation were determined by herbivore abundance and feeding characteristics and that herbivore activity might increase or decrease plant diversity. Regardless of the direction of the change, the effects of herhivory-induced shifts in diversify on ecosysteo processes may be important determinants of ecosystem stability. HcNaughton (1977, page 516) reiterated the idea developed within the framework of diversity-stability theory that "compensatory iluctuations in the abundances of co-occurring system elements (species populations) in a variable environment can stabilize aggregate system properties." He presentJJ empirical data from a grazing experiment in high and low diversity ecosystems that supported this idea. In the high diversity system, grazing resulted in a change in plant species diversity, but had little effect on the total plant biomass. In the low diversity system, an equal amount of grazing did not affect species diversity, but significantly reduced plant biomass. Thus high diversity provided a homeostatic mechanism that allowed functional stability (maintenance of plant biomass) in the face of a perturbation (grazing). 11 Diversity Effects on Hecbivocy The relationship between herbivory and plant diversity is a two-way interaction. In addition to the effects of herbivory on ecosystem processes, the structural characteristics of the system also influence herbivory patterns. It has been suggested that increased plant diversity results in decreased herbivory, and many investigators have reported fewer herbivores and/or less herbivore consumption in f loristically diverse than in f loristically simple systems (Burleigh et al. 1973, Root 1973, Dempster and Coaker 197a, Smith 1976, Altieri et al. 1977, Altieri et al. 1978, Bacii 1980, Risch 1981). Herbivory reduction in diverse systems has been attributed to the presence of alternative hosts that divert plant pests, greater abundance and diversity of insect predators, and/or structural complexity that interferes with insect movements and makes host plants harder to find (Root 1973, Atsatt and 0*Dowd 1976, Pimentel 1977) . These studies may lead to the conclusion that by increasing pxant species diversity, one increases the resistance of an ecosystem to herbivore attack. However, attempts to relate ecosystem diversity to herbivory patterns have not always yielded consistent results. There is evidence that the buffered environment of a complex ecosysteL may support certain pests not able to survive in a 12 ■ore open monoculture, and that some pest problems may increase with ecosystea copplexity (hart 1974, van Euden 1977, Way 1977). ir"or example, soite investigators havt- reported fewer predaceous insects (Pimentei 1961b, Pollard 1971), lower insect predator efficiency (Price et al. 1980), and greater abundances of some herbivores (Croaartie 1975, Thompson and Price 1977) in diverse systems. Research Questions The primary objective of this study was to investigate net primary productivity and herbivory in higa diversity and low diversity tropical successional ecosystems. The work was done as a part of a larger study designed to test the feasibility of using natural succession as a model for the development of new tropical agroecosystems. Experimental successional ecosystems that lacked, imitated, and exceeded the floristic complexity of the natural successional system provided the framework for investigating four questions^ (1) Does net primary productivity differ in high and low diversity systems? (2) Do herbivore consumption rates differ in high and low diversity systems? (3) How does herbivory affect net primary productivity in high and low diversity systems? (4) Are high diversity systems more homeostatic than low diversity systems when partially defoliated? CHAPTEH 11 METHODS Iii§_^tudjir_Site The research was carried out in the Florencia Norte Forest of the Centro Agronomico Tropical de Investigacion y Ensenanza (CATIE) , at Turriaiba, Costa Rica. The site, located at 9^ 53« N, 83° 40' y, lies at the easteru edge of the central plateau of Costa Rica at an elevation of 650 m. The topography is gently undulating, and the vegetation of the area falls into the tropical preiBontane wet forest life zontJ (sensu Holdridge 1967, Tosi 1969). Long term mean annual rainfall for the area is approximately 2700 mm, with a pronounced dry season from January through ilarch. Mean annual rainfall for 1979-1980 (2169 mm) was somewhat Ijwer than the long term average. Monthly rainfall amounts ranged from 14 mm in March 1980 to 460 mm in December 1980 (Fig. 1). Temperatures ranged from an average maximum of 28.4° C to an average minimum of 17.1° C, with a median temperature of 22.7° C. The 2.4 ha study site is typical of large areas in the mid-elevation warm humid tropics that have been deforested for agricultural use. At the start of the study, the vegetation on the site consisted of 8-9 yr old second growth 13 14 , c^ I D- I D ■^^ I □ £ I ■♦ 1 E O < u — tr pj cvj < HiHS H ei 2^ 2< - 'T 2 cr s; 2 < 2 < m I • [il!?!^ 1 — r o CM o (UiO) TlViNIVa 0) +J en C o (U -5 o tfi 2 O 3 ■P in GJ < -P -3 (T> h- +J -:> O^ (0 S iH iH rH rH < (0 (0 2 •H -H Li. (-1 M -3 -P M C 0) 0 > r (0 Cn -H (X! 15 interplanted with timber trees, and remnants of a 56-60 yr old secondary forest dominated by Goethalsia meiantha. The iaiBediate study area was surrounded by diverse second growth, pasture, and experimental forestry plantings, and overlapped with some of the land where Harcombe (1977a, 1977b) did earlier studies on tropical succession. The soil at the study site, classified as a Typic Dystrandept (Soil Conservation Service 1975) , was an upland soil overlyiny upper Miocene or lower Pliocene rock (Harcofflbe 1973). This deep, freely drained soil is characterized by low hulk density, <50% base saturation, and a moderate to high cation exchange capacity. Site Preparation During the first week of January 1979, the vegetation was felled on six 33 x 33 m plots and several smaller plots, using machetes and a chain saw. Border strips of living vegetation at least 5 m wide were left between plots. Firewood was removed from the site, and the remaining vegetation was left on the ground through the dry season. On 22 March 1975, the plots were burned. The turn was intense and complete, and left the site with a uniform cover of white ash. The impacts ot the slash and burn process on nutrient budgets, soil carbon dioxide evolution, soil seed storage, and piart growth were studied and are reported elsewhere (Ewel et al. 1981) . Immediately after the burn, the four experimental manipulations were initiated. 16 Wain Treatments Three experiaental successional ecosysteias, plus a natural successional system, were studied. The experimental systems were designed to represent two types of f loristically diverse successional ecosysteas and one floristically simple system. Natural succession provided the baseline wita which the other systeas were compared. The four main treatments are described below. Natural Succession In this system natural regeneration began after the burn, and secondary succession was allowed to proceed with no experiirentai manipulations. The natural succession provided an estimate o£ what nature does during early tropical succession. This treatment was used as a control for comparison of structural and functional characteristics of the other three main treatraents- Bimic of Succession In this treatment a diverse successional system was experimentally constructed and maintained. The idea was to try to imitate the structure and function of the natural successional system by substituting species morphologically similar to those found in the natural succession. The species composition of the mimic was completely investigator-controlled. Both careful observation of the 17 natural succession plots and prior knowledqe of tropical successional trends provided guidelines for selection ot species to be included iu the aiiiaic. For exaiaple, herbaceous vines (e.g., Viina uniculata, several varieties of £haseolus vulgaris. Cucurbit a pepg and Sechium edule) ioitated early successional vines in the Cucurbitaceae (e.g., Frantzia pittieri, Momordica charantia) and Leguminosae (e.g., RhYnchosia pyraaidalis, Viana vexillata) . Castor bean (Bicinus communis) and papaya (Carica papaya) were substituted for fast-growing pioneer tree species (£§££0£ij! spp. and Bocconia frutescens) . Large monocots such as plantains (Musa paradisiaca) were imitations of common early succession nonocots (e.g., Calathea insiqais, 5§iicoaia latispatha, and Ischnosiphon pittieri) . Cultivated herbs (e.g.. Capsicum sp.) replaced morphologically similar native herbs (e.g., Solanun niqrescens) . Both cultivacs and non-cultivated species that were not present in the area were included in the mimic. Continuous evaluation of the mimic and regular additions of new species occurred during the 1.5 yr study period. The plots were periodically weeded to remove natural colonizers. The mimic was a key ecosystem for testing whether it was possible to imitate succession in such a way that the productivity and uoaeostasis of the natural system was duplicated. 18 Enciched Succession The enriched succession was a system in which the natural regeneration was suppiementcd by continuous inputs of propaqules of many species not present in the vicinity of the study site. This was a self-design treatsaent in which nature controlled the selection process in an ecosystem in which the limitations of seed accessibility had been reduced. Tais system was used to determine whether or not the removal of some biogeographical constraints would result in an ecosystem -lore diverse than the natural succession, and whether the resulting ecosystem would differ structurally or functionally from the natural successional system. Propagules of both cultivars and non-cultivars were added to the auriched succession plots at approximately bi-weekly intervals. Seeds were scattered on the ground, and stem cuttings and seedlings were planted at randomly located points within the plots. During most months, a minimum of 10,000 propagules of at least 30 species were added to each plot. Successional Honoculture A single species system was included in the study for comparison with the high diversity systems. A series of three monocultures was planted, with the species chosen (1) to resemble the life forms of dominant successional species 19 at that staye in succession, and (2) to represent important croppiaq systems in the area. Maize (Zea ma^s var, Tuxpeno) , an herbaceous inonocot similar to some early successioual grasses, was planted immediately alter the burn (late March 1979) . The first maize crop was harvested in mid-July 1979 and was followed by a second maize planting. After the second maize harvest (November 1979), cassava (Hauihot esculenta var. Japonesa) was planted. Cassava is a tuber crop important throughout the tropics. Cassava was chosen for the monoculture because its woouy growth form was similar to the growth form of the shrubs that were rapidly becoming dominant in the 7 mo old natural succession. The cassava was harvested in mid-September 1980 and was followed ty a planting of Cordia alliodora, an important timber species. Data on the Cordia monoculture are not included in this study. The planting procedures and management of the monoculture plots followed as closely as possible the methods used by local farmers. Maize was planted at 1.0 x 0.5 m spacing, two seeds per hole. The cassava was grown from stem cuttings planted at 1 x 1 m spacing. At plant maturity, the harvestable crop (ears or tubers) was removed from the plots, and the remaining plant material was left on the ground. All monoculture plots were periodically weeded. 20 Plot Layout and Variables Measured The treataent plots were arranged in a raudooized complete block design, with six replications of each of the four main treatments (Fig. 2). Each study plot measured 14 X 1 4 m (196 ra2) within permanent metal markers. An additional border strip approximately 1 m wide was left around each plot, making the actual plot size 16 x 16 m (256 m2) . The study plots within each replLcation were separated by 1 a wide access trails. Buffer strips at least 5 m wide of original, uncut vegetation were left between replications to serve as a source of seeds for the experimental plots. Specific areas within each study plot were designated for particular types of investigations, including the work reported here and the work of other researchers (Fig. 3). Variables monitored during the 1.5 yr study period in the 4 main treatments fall into two categories: (1) vegetation structural characteristics, such as leaf aroa index, species composition, and vegetation height, and (2) productivity aeasutements. The methods employed for each type of measurement are described in detail below. 21 ■Sg^ :«;I ^^^^-^rC rn;rni#&^:;:;: PINE ■:!:;:;:;:;:;:! ^ OLD SECOND SROWTI sJ Bosoue secunoJDio \j '^';;. Figure 2. Map of the study site. 22 + ARTIFICIAL I 1 ^ LITTER COLLECTORS — 16m HERBIVORY SURVEYED CORNER + BIOMASS HARVESTS ACCESS TRAIL TENSION- FREE LYSIMETERS TENSIOMETERS OOOO/ V^ V\i SOIL PIT LIGHT SENSOR o o -f CAPILLARY WATER SAMPLERS D SEED TRAP NON- DESTRUCTIVE SAMPLING AREA + Figure 3. Diagram of study plot. 2J Measurements of Vegetdtiao Structure Leaf .Area Index i.eaf area index (LAI) is defined as leat area per unit qround area. Vaiut.'3 are usually reported as m^ of leaf tiiisue (one side of leaf) per ni^ of ground. In this study LAI was measured usiny a pluiat-Lob mtthod similar to the metuod used by Benedict (197b). A thin line is lowered vertically from tiie top of the voyatation canopy to the ground and the number of leaves touching the line is counted. This method reduces the sampling area to a single point, and the number of leaves above a point (i.e., the number of intersections of line and leaf) is a direct measure of LAI. The intersectioas were recorded by species and height above tae ground. The instrument used to measure LAI was constructed from a rigid extendable metal rod. A fishing reel was connected at its base and a pulley at the tip. A thin nylon twine attached to the rod with a small weight at its end could then be easily lowered vertically through the vegetation. The twine was knotted at 25 cm intervals, and alternate intervals were painted for easy reading in the field. This instrument could be used in vegetation up to 8 m in height. In taller vegetation, it was necessary in a few cases to estimate the numoer of leaves above the rod. Leaf area index measurements were made in all main plots during May 1979, July 1979, November 1979, April 1980, and 2a October 1980. In May 1979, 20 LAI measureiaents were made in each study plot oi; eacu replication. Five locations were chosen randomly in each plot, and four LAI readings were taken at each location by droppinij the line vertically through the vegetation four times. For all other sample dates, 30 LAI measurements were made in each plot. Ten 1 m^ quadrats were systematically located in each plot and permanently marked. Three LAI measurements were madt, in each of these quadrats on each date. The uniform spacing of crop plants in rows created special problems in use ol the plumb-bob method to measure LAI, especially in systems with very low LAI. For this reason, LAI ol the maize monoculture in November 1979 was calculated using leaf biomass/leaf area regressions rather than by using the plumb-bob method. Species data from the leaf area aeasureiaents were used to calculate LAI for individual species, and percent of total LAI was used as an estimate of relative species dominance. Species Composition Species inventories were done in the natural succession, enriched succession, and mimic plots during July 1979, Novembjj 1979, April 1980 and October 1980. For each plot a list was made of all flowering plants and ferns encountered in each of the tea 1 m^ quadrats described above. From the^e data, diversity indices were calculated. In addition. 25 a complete species inventory was made in each 16 x 16 m^ plot in October 1^80. Plant specimens were identified at the National Museum of Costa Rica. Vegetation Height At the same time that the species composition and LAI measurements were made, the height of the tallest plant in each of the ten 1 m^ quadrats in each plot was measured- Average canopy height for each plot was then calculated. Al30, the species and height of the tallest plant in the entire lo x 16 a^ plot was recorded. Productivity Measurements Net primary productivity is one of the principal response variables that was used to compare the four experimental ecosystems. A common metliod for estimating net primary productivity is by using periodic biomass measurements to calculate changes in standing crop over time. However, in fast-growing tropical successional vegetation, the measuLemeat of changes in living oiomass underestimates actual net primary production because of rapid turnover of plant parts and losses to herbivores during the time intervals between harvests. Litterfall and insect consumption are two losses of net productivity that cannot be measured by biomass harvests. In this study, measurements were made of plant mortality, rates of 26 iitterfall, and rates of herbivory, in addition to periodic ffieasurements of above-ground living biomass. The values obtained were used to estimate above-qround net primary productivity. Mean rates of biomass increment (y ra~2 day~^) were estimated for intervals between biomass harvests as B(i)-B(i-1) B = Eq. 1 t(i)-t{i-l) where B(i) = above-ground living biomass at harvest (i) in g/ra^, L(i-1) = above-ground living biomass at harvest(i-l) in g/m2, and t(i)-t(i-1) = number of days between biomass harvests. These rates were plotted at the mid-points of the intervals between harvests, ana the points were connected by straight lines. Linear regressions were then used to estimate daily biomass increments. Increments of standing dead biomass (g b"^ day-*) were estimated as D(i)-D(i-1) D = Eg. 2 t(i)-t{i-1) where D (i) = standing dead biomass at harvest (i) in g/m^ , D(i-l) = standing dead biomass at harvest(i-l) in g/m^, and t(i)-t(i-1) = number of days between harvests. As above, the rates were plotted at the mid-points of the intervals 27 between harvests, the points were conuected by straight liaes, and linear regressions were used to estimate daily increiaents in standing dead uioinass. The turnover rate of standing dead biomass was not known. The conservative assumption was made that turnover was negligible. Positive daily increments in the standing dead biomass category were used as i3tiLit33 of daily production of standing dead biomass. If the turnover rate was high, production of standing dead and net FciiHdt-y productivity would both be underestimated by these methods. Litter fall rates (g m-^ day-^) were estimated for each ecosystem as L(i) L = Eq. 3 t(i)-t (i-1) where L(i) = amount of litter collected during a 4 wk interval (g/m^) , and t{i)-t(i-1) = number of days in interval. These rates were plotted at the mid-points of the intervals, the points were connected by straight lines, and linear regressions were used to estimate daily litterfall rates. Daily herbivory rates for each ecosystem were estimated from three 1 mo sampling periods. Linear regressions were used to estimate daily herbivory rates. Daily net primary productivity rates were calculated as NPP{i) = b(i) *■ l(i) ♦ h(i) + d(i) Eg. H where NPP(i) = net above-ground productivity on ddY(i) in y iu-2 day~*, b (i) = bioaiass increment on day (i) in q a-^ day-i, i(i) - iittertdii on day(i) in q in-2 day-», h (i) = herbivory rate on day (i) in g m-^ day-', and d(i) = production of standing dead bioaass on day(i) in g m~^ day-i. Above-Grcund Bipniass ImiBediately after the burn, randomly located subplots were marked with string and aietdl stakes in the area of each study plot designated for bioaass harvests. Fourteen biomass harvests here saade during the 1.5 yr study period. Early harvests in the natural succession, enriched succession, and mimic of succession were done at frequent intervals (approximately bi-weekly) on siaall (0.24 m^) plots, and later harvests were at less frequent intervals on larger ^lots. Dates and plot sizes for each of the harvests were 14 May 1979, 31 May-5 June 1979, 20 June 1979, 9-10 July 1979 (0.2U a2) ; 1-2 August 1979, 10-12 September 1979, 8-10 October 1979, 19-21 November 1979, 17-19 December 1979, 21-23 January 1980 (1.60 m^) ; 17-19 Karch 1980, 19-21 May 1980, 8-11 July 1980, 28-31 October 1980 (4.00 m^) . At the time of each harvest, one randomly selected subplot was harvested in each study plot (total number of subplots harvested per treatment = 6) . 29 It was decided that the hatvest of individual plants and plant density data, rather than the harvest of vegetation in random subplots, would yield better estimates of biooass in the monoculture treatment where- plants were uniformly spaced. Therefore, from one to four randomly chosen plants of the monoculture species were harvested per plot at each sampling date. Harvests of the morocuiture were made at each date listed above. Additional harvests were made at crop maturity (29 October 1979 and 10-12 September 1930) and during the early growth stage of the second maize monoculture (16 August 1979) . At maturity of each monoculture, samples of the harvestable crop were used to estimate economic yield. Above-ground biomass was harvested by clipping all vegetation within subplot boundaries at ground level. All plants rooted inside the plot were included, even if parts of the plant extended outside the sanpie area. Likewise, all plants rooted outside the plot were excluded. Vines were clipped at the plot boundary. The vegetation from each plot was separated into four classes: leaves, stems, reproductive parts, and standing dead. Vegetation samples were weighed in the field. Subsamples of each vegetation class were taken to the laboratory, weighed to the nearest 0.1 g, dried to a constant weight at 70° C, and reweighed to obtain fresh to dry weight conversions. 30 Data for each veqetation compouent (leaves, steins, reproductive parts aad standin^j dead) and total afcove-qround biomass were analyzed using a randomized complete block, fixed effects statistical model with four treatments and six blocks (replications) . The biomass data did not meet the homoqeneity of variance assumption of analysis of variance. Means and variances were not independent; in most cases, variance was proportional to the square of the mean. The biomass data were transformed usinq the followinq loq transformation: y=ln(x+1) . All analyses of variance and Duncan's multiple ranqe tests were done on the transformed data, usinq the General Linear Models (GLM) proqram of the Statistical Analysis System (SAS) . Reported means and stand?ri deviations are of oriqinal untransf ormed data. Litterf all Three 0.2i m^ litter collectors were located near the soil surface in each replicate of each treatment. Each collector was 1.00 x 0.25 x 0.15 m (length x width x heiqht) and was supported approximately 2 cm above the soil surface by metal brackets. The collectors had wooden sides and fine-mesh screen bottoms for drainaqe. The shape and small size of the collectors allowed the successional veqetation to qrow up and over the collectors rapidly. The collectors were positioned 1 m from the access trail in the portion of each plot designated for litterfall 31 studies (see Plot Map, Fig. 3) . Litter was collected from the baskets at 2 wk intervdis thcouyhout the 1.5 yr study period. The litter firom the three collectors in eacti plot was combined into one composite sample, oven dried at 70° C to d constant weight, and weighed to the nearest 0.1 g. The baskets collected both autochthonous and allocthouous litter inputs to the plots. To calculate net primary productivity of the vegetation in the plots, a measure of autochthonous litter production was needed. Allocthonous inputs were estimated from a single collector (0.25 m^j placed near the other three collectors in each monoculture plot. For eaca of taese 'control' baskets, leaves of the monoculture species in the basket at each sampling date were discarded. Ail other material in the basket was collected, dried and weighed. Hgrbivpry Hates Losses of plant tissue due to herbivory were estimated by monitoring amounts of damage incurred on tagged leaves of dominant species in each treatment. It was not possible to separate losses due to plant diseases (fungal, viral, bacterial) from losses to herbivorous insects, so loss estimates include damage due to plant diseases as well as losses to herbivores. At each of three sampling periods (October 1979, February 1980, and June 1980) the most recent LAI data were used to select the species to be taqqed. The species of each treatment were ranked troia highest to lowest LAI, and those more comaicn species that jointly accounted for at least 80 percent of tne total LkI of that treatnoent were selected for herbivory measureuients. In the portion of the study plots designated for uon-destr active sampling, five plants of each species (three in insecticide plots) were arbitrarily chosen for tagging. Usually no more than one individual of each species was tagged per replication. In a few cases, patchy distribution of a species made it necessary to tag more than one individual of that species within a single replication. & plant stem was considered eligible for tagging if it was unbroken, unbranched, and bore at least four leaves. One eligible stem was chosen on each plant. From four to eight consecutive leaves were selected along the stem, and these individual leaves were numbered from youngest to oldest. Small plastic bands marked with yellow tape were looped around tae stem at two places. Positions ot leaves relative to these bands were used to identify individual leaves at the time of harvest. When the leaves were tagged, the holes present in each leaf were measured by placing a sheet of mm-ruled graph paper under the leaf and counting the uncovered sguares. Bruwn spots on each leaf were estimated visually, and total damage (holes *■ brown spots) was recorded for each leaf. 33 The length ol each leaf was measured to the nearest loio at the time cf tagging - Leaf length/leaf area regressions for each species (developed trom a sample of at least 50 leaves per species) were used to estimate the initial leaf area of each leaf (Table 1). For each species, the best curve fit was obtained by using a quadratic equation for all but very small leaves, and a linear equation through the origin for very saall leaves. These initial leaf area estimates, together with direct measurements of leaf area at the time of harvest, were used to estimate leaf expansion during the interval. In grasses and some herbaceous species with small leaves (mature leaves <'*0 cm in length) , leaf lengths were not measured, and leaf expansion was not estimated. After 3 to 7 wk, the tagged leaves and all new leaves produced on the marked stems during the interval were harvested. Mortality of tagged leaves and number of new leaves were recorded for each plant. In the laboratory, the area of damage on each leaf was traced on a sheet of clear plastic and filled in using a permanent black marking pen. Two categories of damage, holes (H) and brown spots (B) , were drawn separately. All missing tissue, plus damage that left only a transparent layer of leaf tissue, was recorded as holes. All other damage, including damage by leaf-mining insects, damage by rasping insects, fuaqal and viral damage, plus the ne-^rotic tissue around holes, was recorded as brown spots. 34 Table 1. Leaf length:! species. In y = leaf area saf area regression equations for the equations, x = leaf length in in cm^ , common mm and Species Regression Equations R^ Bocconia f rutescens x>125 x^l25 : y=0.00203x^ + 0.303x - 47.779 : y=0.174x 0.96** Borreria laevis x>24: x<24: y=0.000856x^ + 0.0667x - 0.927 y=0.0469x 0.95** Cajanus cajan x>23: x^23: y=0.00431x^ + 0.0475X - 1.909 y=0.0624x 0.97** Carica papaya x>81: x^81: y=0.0172x^ - 1.837X + 56.231 y=0.249x 0.92** Clibadium aff. surinamense x>58: x<58: y=0.00440x^ - 0.228x + 6.995 y=0.147x 0.97** Cordia inermis x>34: x<34: y=0.00318x^ - O.lOlx + 1.895 y=0.0624x 0.98** Crotalaria micans x>22: x<22: y=0.00637x^ + 0.0803x - 3.050 y=0.0772x 0.98** Cucurbita pepo x>32: x£32: y=0.0124x^ - 0.0163X - 6.809 y=0.149x 0.99** Canavalia sp. x>53: x^53: y=0.00473x^ + 0.289x - 20.419 y=0.148x 0.97** Erythrina costaricensis x>171 x_<171 : y=0.00913x^ - 1.061x + 82.992 : y=0.980x 0.95** Frantzia pittieri x>77: x<77: y=0.0115x^ - 0.573X + 27.339 y=0.661x 0.90** Hyptis suaveolens x>19: x^l9: y=0.00361x^ + 0.0633X - 1.042 y=0.0731x 0.99** Hyptis vilis x>10: x^lO: y=0.00402x^ - 0.0148x + 0.339 y=0.0591x 0.97** Ipomoea batata x>29: xf29: y=0.00655x^ + 0.201x - 7.644 y=0.125x 0.95** 35 Table 1 — continued. Species Regression Equations I porno e a sp. x>23: y=0.0117x2 - 0.341x + 4.392 0.97** x<^28: y=0.142x Iresine diffusa x>30: y=0. 00445x2 - O.llOx + 2.373 0.98** x<^30: y=0.101x Man i hot x>51: y=0. 0117x2 _ o.784x + 25.038 0.87** esculenta x<^51: y=0.30 3x Merremia x>53: y=0. 00733x2 - 0.0228x - 10.987 0.92** tuberosa x^53: y=0.152x Phaseolus x>60: y=0. 0135x2 _ o.960x + 29.925 0.98** vulgaris x<^60: y=0.344x Phytolacca x>31: y=0. 00267x2 + 0.0271x - 1.290 0.97** rivinoides x£31: y=0.0666x Solanum x>50: y=0. 00748x2 - 0.227x - 2.224 0.98** jamaicense x<_50: y=0.0989x Solanum torvum x>26: y=0. 00352x2 - 0.00506x - 0.522 0.98** X£26: y=0.0631x x>50: y=0. 001253 umbellatum x<50: y=0.0473x Solanum x>50: y=0. 00125x2 ^ o . 117x - 6.52 0.98** Vernonia patens x>40: y=0. 00154x2 _^ o.221x - 8.702 0.91** x£40: y=0.0610x Vigna sp. x>47: y=0. 00568x2 - 0.0873 - 0.388 0.92** x<47: y=0.169x **p<.01 36 The leaf cemnants and plastic sheets were run through a Lambda Instruments LI-CCH (LI-3000) area meter, which measures the surf ice area ot opaque surfaces to the nearest 0.01 cm2 with an accuracy of * ^%. In a few cases, leaves from a plant were processed as a group rather than individually. For each leaf (or group of leaves) , total damage present, D(t(f)), and gross leaf area, G(t(f)), at the time of harvest were calculated as D(t (f) ) = H ♦ B Eg. 5 and G(t(f)) = k ♦■ H £q. 6 where t{f) = time of leaf harvest, H = holes present at t(f), B = brown spots present at t(f), and H = residual leaf area at t (f ) . Herbivory rates (i.e., loss of leaf tissue per unit area of leaf per unit time) were calculated tor each leaf of each species. Two factors contribute to the total loss due to herbivory: (1) actual consumption by herbivores and (2) loss of potential photosynthetic leaf area due to expansion of damaged areas after consumption has occurred. Since the rate of expansion of holes in a leaf is equal to the rate of expansion of the leaf (Reichle et al. 1973, Coiey 1980) , estimates of percent consumption are not affected by leaf 37 expansioa during the sampiiny interval. Percent consumption (LOSS) waii estimated for individual leaves by the following equation: LOSS = D(t(f)) G(t(t)) D(t(0)) Git(O)) 10000 t (f) - t(0) Eq. 7 where D(t(i)) = damage present at t (i) , G(t{i)) = gross leaf area at t (i) , t (0) = time of leaf tagging, and t(i) - tine of leaf harvest. An absolute consumption rate was then calculated for each species by multiplying mean percent consumption of the species by LAI of the species. The area of 50 leaves of each species was measured using the LI-COR (LI-JOOO) area meter. The leaves of each species were pooled, oven dried to constant weight at 70° C, and weighed. Leaf specific mass (mass per unit area of leaf) was then calculate! so that herbivoty rates could be expressed ou a mass basis as well as on an area basis. Three non- parametric statistics (Wilcoxon 2-sample rank sums test, Kr uskal-Wallis test, and median test) were used to test for differences in herbivcry rates between ecosystems for several plant species. These statistical procedures make no assumptions about the distribution of the data, but do require homogeneity of variance. The level of significance of ordinary 2-saiiple procedures is not preserved if tae variances of the two populations differ (Pratt 1964). The robustness of the tests under departure 38 trom the assumption varies with test used, sample size of the popuidtioas, and magnitude ol departure trom the assumptious. The uoinogeaei ty oH variance assumption was not met by the herbivory data. In general, means and variances were proportional; large variances were associated with large means, and small variances with small means. Therefore the levels or significance associated with test results ace not exact. Estimation of Hole Expansion For those species in which initial leaf area was estimated (using regression equations), it was possible to estimate the loss of potential photosynthetic leaf area due to expansion of tiie holes in leaves. The mathematical equation derived to estimate consumption and expansion is based on three assumptions: (1) the damage expansion rate equalled the leaf expansion rate; (2) the consumption rate was constant during the time interval in which herbivory was monitored; and (3) for a group of leaves on a single stem, leaf growth rate was a constant function. The validity of each of these assumptions is discussed telow. The first assumption (that hole expansion rate = leaf expansion rate) is generally assumed to be valid and has been verified experimentally by Reichle et al. (1973) for a temperate deciduous forest species (Liriodendron tulipifera) and by Coley (1930) for several tropical forest species. In 39 an unpublisaed study of a common successioaal species (Conqstecjia fiittiecL) ia a tropical pre montane wet forest at Monte Verde, Costa Rica, I found that hole expansion rate and leaf expansion rate did not differ significantly (n = 70 leaves) . although herbivory on individual leaves does not occur at a constant rate, the rate of damage accumulation may be assumed to be constant for a population of leaves (assumption 2) . Lilcowise, altuouga tae growth curve of an individual leaf is probably sigmoiJal rather than linear, the average leaf growth rate or a population of leaves of varying ages may remain constant over time (assumption 3) . Altnough these assumptions seem intuitively reasonable, they have not been verified experimentally. If tiie assumptions are not met, bias is introduced into the estimation of the relatxv.; proportion of the total herbivory loss attributable to consumption and expansion. The results of several types or possible deviations from assumptions 2 and 3 are presented in Table 2. If consumption rate (c) and leaf growth rate (G*) are both constant, then assumptions 2 and 3 are Bet, and the method used in this study accurately estimates percent of total damage due to consumption and expansion. If c and/or G' are increasing or decLcasing functions, losses due to expansion (e) may be overestimated oc underestimated by the methods used in this study. 40 Table 2. Comparison of estimated (e*) and actual (e) losses due to expansion, for several consumption rate (c) and leaf growth rate (G') functions; t = time. Case 1 (C constant) Case 2 (C decreasing) Case 3 (C increasing) G'l G' t-> t^ t^ Case 1 (c constant) Case 2 (c decreasing) e* Case 3 (c increasing) t^ 41 Using the assumptions listed above, percent consumption rate (c) and percent expansion rate (e) , both iu ciu^ m~^ day-*, vere estimated for each plant by the foUowinq equations; D(t(f)) - D(t(0)) G(t(f)) G(t{0)) m - + n n-1 i = 1 10000 G(t(f)) n - i(1-r) Eq. 8 D(t(f)) - D(t(0)) - (cXm) 10000 e = X ffl G(t{f)) Eq. 9 where t (0) - time of leaf tagqinq, t(f) - time of leaf harvest, m = t(f) - t{0) = number oi days leaves were tagged, D(t(0)) - damaged area ut t(0) in cm^, D(t(f)) damaqed area at i. (f ) in cm^ , G(t(0)) = gross leaf area at t{0) in cm2, G(t{f)) = qcoss loaf area at t {f ) in cm^, r = G (t (0) J /G (t (f ) ) , and u = the number of sub-intervals (t(1-1) ,t (j) ) into which the time interval (t(0),t(f)) is divided. Ihe derivation of Equation 8 is qiven in Appendix A. In the equation above, D(t(0)), D (t (f) ) , G(t(0)), and G(t(f)) are totals of all tagged leaves on a given plant, excluding tagged leaves that died during the interval and 42 uew leaves produced during the intervdl. Cdlculations of losses due to hole expansion were made using plant totals rather than individual leaf data for two reasons. (1) The precision of the regression estimates of initial leaf areas was not high enough to allow individual leaf eKpansion to be estimated. Although the leaf length/leaf area regressions for most species were quite good (H^ > 0.94 for 19 of 25 species. Table 1), in some cases overestiaiates of initial leaf area led to negative leaf growth rates for individual leaves during the interval. (2) The assumption that leaf growth was a constant function is better fit by groups of leaves of varying ages than for individual leaves. The herbivory rate calculated using plant totals is mathematically equivalent to the mean of the herfcivory rates calculated for individual leaves if all of the leaves are equal in size; if damage area:leaf area is constant for all leaves (i.e., herbivory is evenly distributed among leaves); if tne sums of damage areaileaf area are the same for groups of equal-sized leaves; or if total leaf areas are the same in groups of leaves witn equal percent damage. None of the sufficient conditions listed for equality of the 2 methods are necessarily Jet by the data. Thus pooling individual leaf data for analysis may introduce a source of error. To evaluate the magnitude of the error, hetbivory rates calculated from individual leaf data and from plant totals were compared for six species (Table 3) . Altfiough herbivory 43 Tl C (C a -p (0 TS IH to 0) iH rH m 3 -o -H > -H TJ C ■H e 0 M m T3 Q) 4J nJ rH • 3 CQ 0 0) rH •H (13 O C) OJ a. tfi CO 0) ■P T) nJ a; >-i +J o C (U 0 H •r) (U +J to a g u 3 0 m m C 0 tfl O -P m c -d (0 +J 3 rH m rrt n > cu u C ^^ cu c ^ 05 QJ •— m m Q -H D '>! c fO 0 T) rH m \| T3 rO 0) i+h| 01 :3 +j (rt W fO IT3 0) (fi •H cr; rH CQ > -H 4H(N 1 0) 73 (C B -P c QJ\| m H kIN 1 « B o G >1 O cn (« <-^ 'O -a (0 \ (1) -P MH tfi o n3 (0 H 0) m -P rH 0) GfN 1 +j (0 S m rH X K OiOJ 1 e u UH Cfi 0 -P ,— V c c • (fl 0 rH z &. •H o a ^ -p cn •r^ m cn td C (0 cu nj m 0) cu (U U (0 -P u Til U P4 C u •H (0 cn cn -P 0) fO O •H OJ ;3 (fl 0 rH rH c c 4J C CJ ^ :3 0 •rl o :3 (0 •rl ■H o 4-) > c M ■r—i U c cn >1 -H u m in (Xi (C 0) ^ !-l Q) u u s CM > Oj rates calculated bi the two methods differed considerably for some plants, the two mothods yielded significantly different mean species herbivory rates for only one species (Manihot esculent a) . Consumption rates were estimated by an iterative process in which the time interval (t{0),t(f)) was divided into n smaller sub-intervals (t ( j- 1) , t { j) ) , and consumption and expansion wore calculated tor each of these sub-intervals. In this method, both the expansion of damage present on the leaves at t(0) and the expansion of damage that occurred during the interval (t{0),t(t)) were excluded from the estimate cf consumption. As the number of iterations (n) was increased, the precision of the estimate of c also increased. To select an appropriate value of n, consumption rates were estimated using various n values for nine plants. For each of the plants, an n value of 55 was sufficiently large to insure that the consumption rates {cm^ plant-* day-i) were accurate to the nearest 0.01 cm^. For most of the sample plants, the required n value for this level of accuracy was mucn less than 55. On the basis of these preliminary tests, calculations of damage expansion were done with n = 55. Computer programs to calculate damage expansion were developed using the Statistical Analysis System (SAS) . One program wac developed for use with alternate-leaved species. A modified version of this program was used for opposite-leaved species, in which data were pooled for opposite leaf pairs. 45 Subtreatments In dlditioD to ffldin treatment comparisons, a maioL objective of the study was to evaluate the effects of hecbivocy on net pcimacy productivity, vegetation structure, and species composition in faiqh and low diversity tropical successional ecosystems. To do this, comparisons were made between iii o •H 4-1 tn •H u Q) 4-> D Oi U rti x; u 00 o O 'S" Ln O^ fN (N iH IT) LD iH r- CO tn CM iH O^ ■^ m rn r-i ui 00 rH uD in 00 r- IX) o r-~ VD <" CTi ^ "^ in (Ni m LO H 00 r-- vD ■^ r- n r-~ i en 0) 0) 0) CD 4-1 CD e CD 4-1 Q) >i J-i •H >, H •H 0 0 -H ■H 0 •H > X! 3 U ^ 13 o £ e U ro O E tn tc en CD en CD CD CD i-i Q) Ti ttJ !iT3 Its at? 00 Cb B !2j00 CD iH Q) 0) en CD 0) en CD tH en 0 tn > x) e -t-i e 4-1 J4 E •H U-J O 14-1 O 14-4 o 'd m 14-1 0 14-1 0 0 ■T! 'O 0 0) H 0 CD M 0 (U c 0 E 0 u E tn tn < en (< en fO 'U 14-1 tn tn U U ^A V4 M J )-l V4 V4 CD 00 U 00 CD U CD 01 0) CD 2 2 2 2 2 en w 52 0) J-l 3 +J iH 13 u 0 c 0 T C ^-^ 0 0 ■H in O en •H 0) B o •H c; «^ :^ UJ c t^ 0 U •rH J-) cn •H U 0) -P CJ oi u m x: u o >i e 4-1 •H ro U m (u rH cn(U •H rt! O E e •H C tn 0) 00 CD iH -U -P 0) •H Q) tJ c x:' TCI u c n3 0 to 0) e -tJ u T3 0 rd rd , 00 4-1 rH iw CO CU fO •> CD )H x; Q) ■H ■iH 0 4-) iH u \ X. tn cu •d CO +J CD Cu C -H x: ■H tn rd en u C •H cu 14-: O 0) a. 0 E 'a 5 tn Sh c m rd m M CD IT3 0 £ 4-) ^ Q) ip E rd tn iH tn 3 C CU 4-1 4-> T! C C rd C 0 tn CD 'd (0 •H 4-) •H tn CD 0 (C O CO u > -P 0) CU-P tC tn 'd Ti u ,~, c tn tu CD X rd CD x; tn 4-) Q) •rl 4-1 tn C 'O X o fO ■H C CU CD E e •H tJ a 0 0 tw C tn iH -H rd c •rH MH Xi CU 0 td iH c >1 ■H M-l c 4-1 tn (C UH rd •H ■rl tn QJ o ^ tn CD rH tfi Sh •H -rl u ■ — cu U e • CD > CU CD 0 -^ X! ^ ■^^ Sh Cj M td E CU Ti CD tn M-l > 3 4-1 x: (d C o c S SH T5 tn cu 0 0 Q) tn CU tn C - MH -P fO X iH C ^^ fO u 4J Q) rd C 0) e +J x: ^ d •H CD tn c en (d rH ■p u ■H ■H rd tn 13 tn + > 0) -P •H tn •H rd c Q) ■ H (D E > - • <: 13 •^-' Q) rd E + J iH u CU ttJ 4-1 CD rH CD ^ 4-> > C J2 Sh •H C i U ,-^ iH + CU >H 3 s CD - • CU -P tn \X! W ■ U rd ■H E • Sh • 0) + CD tn 5-) E ~— - C 0 tn CD •H en r-H ^ tn -H 'd o 0 rH \ (N (U 0 00 rH rd — ■ rH CD T3 . o> 2 4J X rd Cu (U tn rH 0 CD tn S-l C 4J II + ^ 3 0 >-l c" 4-> U^ tn •H (D ~— ' 0) tn -- 0 ft5 tn ^ w -C tn <-{ m 0) tn E 1 4-) CD ~^ -H Sh E Q) Q) c rd CU M 4-) II tn c — XJ ■0 CI CU -H cu II -H E 0 0) CD - > — :3 2 u m K 2 [l] U rd C (C X> U tJ CU 53 ecosystems except the monoculture. Species richness at 18 mo (based on a total inventory of all plots) was highest in the enriched succession (159 plant species present ou 1536 a^) , followed by the natural succession (121 speci.es), mimic of succession (82 species), and monoculture (1 species). The Shannon diversity index (H') was calculated as a simple measure to compare overall diversity (richness and evenness) of the experimental ecosystems. An evenness index based on the Shannon index (evenness = HVloy S, where S is the number of species) was also calculated. The diversity index increased over time in the natural succession and enriched succession, but not in the uiioic (Table 4) . Diversity at 18 lao was higher in the natural succession and enriched succession (1.2«4 and 1.26 respectively) than in the mimic (0.92). Of tne possible ranqe of evenness values from 0 to 1, the values in the natural succession and enriched succession were approximately equal (from 0.65 to 0.73), with little change over time. Evenness values in the mimic were more variable (from 0.49 to 0.88). The species composition of the natural succession and enriched succession was very similar early in succession (at 3 mo), ;^ ut less similar at 18 mo. The natural succession and enrichea succession nad 86 species in common at 18 mo. Thirty-five of the species present in the natural succession at 18 mo were not present in the enriched succession. Seventy-three species were present in the enriched 51 succession fcut not lu the natural ouccession, and ot these at least 2^ wece investigator-introduced. Some of the species differences between the natural and enriched succession may be due to random differences in seed availability of native species and to random micro-environiBental differences among plots. However, at least 9% of the 264 species introduced into the enriched succession had become successfully established by the end of the study period. It was possible to increase species richness by propagule additions, and these data suggest that species richness was limited by propagulo accessibility during the earliest stage of succession. This result may be a temporary phenomenon due to the stochastic nature of early succession (Hebb et al. 1972, Horn 1974) and to the continuous rapid changes in vertical and horizontal plant distribution that allowed colonization by new species. Longer-term results of the study will verify whether or not the higher species richness of the enriched succession can be maintained. To compare the degree of similarity in species composition between ecosystems, a coainunity similarity index was calculated for each pair of ecosystems at four dates (Table 5). The index (Gleason 192Q) was C = a{1) *■ a(2) + ... + a{i) *■ . . . *■ a (n) , where i is a species present in at least one of the two ecosysteiss being compared, a(i) is the lesser percent LAI value from the two ecosystems for species 55 cu . o 0) ■H -p Ti ffl C -c •H X >i o 4J (rt ■H Q) U (C c ,-^ o •rH e 6 •H a) U] 4J tn >i >i ■P en -H 0 C O 3 Cl) E fci u 0 0) u c^ c iw o O -H en o in ■H 0) C 'C 0 0) ■H ^ Cfl u w ■H CD i-l u c u M 13 W O O Tt< O O iH O nH O O O O rH O O O o >H n o o o o o O LD o o O iH "* O O -H O 1^ O O d c C c c c c c c 0 0 c 0 o C 0 0 C o o 0 -H •H 0 •H •H C) •H ■H n •H •rH •rH CO Cfi ■H cn cn ■H cn cn •H Cfi cn Cfi tn m en cn cn cn cn cn cn cn cn m (U 0) en CD aj cn -J F= -P !h f= (C C •rH (C C ■rH m c ■H (Ti C •H 2 w s 2 W S IZ w 2 S w s r^ Jh CT\ o^ CD rH Sh nH XJ CD E rH Xi >1 CD •H 0 rH > Sh •P 3 0 &, U 1-0 2 <: o u Q) Xi e CD -P CU CD (XI ct3 56 i, and n is the total number ot species present in the two ecosystems. C may range in value from 0 to 1. Comiaunity similarity was uigh between the natural succession and enriched succession. The values ranged from 0.66 to 0.69, with no significant change during the 18 mo period. Communicy similarity values for other pairs of ecosystems were 0 or very low, indicating little or no species overlap. The natural succession and enriched succession were comprised of a few abundant species and many rare species Figs, ^i and 5) . Most of the abundant species in the natural succession at 3 mo (July 1979) were also abundant iu the enriched succession. Of the species individually accounting for >2% of total LAI iu the natural succession (number of species = 9) md in the enriched succession (number of species =9) at 3 mo, seven were common to the ecosystems (Table 6) . By 18 mo (October 1980) the similarity in dominant species between the enriched succession and the natural succession had decreased. Of the 12 abundant species (those coaprising >2% of total LAI) in the natural succession, only five were also abundant in tiie enriched succession. One of the abundant species in the enricaed succession at 18 mo was an introduced species, plantain (i3ii§§: paradisiaca) . The species composition of eacL of the ecosystems changed during the 1.5 yr study period. The turnover of abundant species from 3 to 18 mo differed in the enriched succession 57 I 5n uj 0 Q. in fe 20 3 MONTHS 1 T 5- 0- 5- 1 I \ \ \ — \ — \ — \ — r 18 MONTHS I \ 1 1 — 1 r 0 .2 .4 ,6 .8 L A I .0 1.2 1.4 ^iqure 4. *Jumber of species in the natural succession by L? I class. Values are based on 180 L.AI neasurements on each date. 58 5^ 0- 5- C/3 LU O LU ?^ 25 20- I 5- 0- 5- 3 MONTHS I I r 1 — r 18 MONTHS 1 r .2 1.4 LA I Figure 5. Niomber of species in the enriched succession by LAI class. ''■'''alues are based on 180 LAI measurements on each date. 59 Table 6. Species accounting for _>2% of LAI in four ecosystems, A dash (-) indicates that a species comprised <2% of ecosystem LAI. Ecosystem Natural Phytolacca rivinoides succession Momordica charantia Solanum nigrescens Borreria laevis Bocconia f rutescens Clibadium aff . surinamense Gramineae^ Panicum maximum Hymenachne amplexicaulis Trema micrantha Frantzia pittieri Acalypha macrostachya Cyperaceae^ Panicum trichoides Vernonia patens Mikania sp. Enriched Phytolacca rivinoides succession Momordica charantia Solanum nigrescens Borreria latifolia Bocconia frutescens Clibadium aff. surinamense Gramineae^ Panicum maximum Vernonia patens Ipomoea neei Musa paradisiaca Ipomoea sp. Mimic of Vigna sinensis succession Cucurbita pepo Phaseolus vulgaris Ipomoea batata Oryza sativa Cajanus cajan Zea mays Cymbopogon citratus Manihot esculenta Crotalaria micans Musa paradisiaca Hyptis suaveolens 37.5 - 5.3 - 2.3 - 2.2 - 10.4 15.7 9.4 7.8 8.2 6.0 5.0 24.7 3.7 2,8 - 5.4 - 3.3 - 3.1 - 2.7 - 2.7 - 2.6 - 2.2 19.7 _ 6.7 - 3.2 - 2.3 - 13.6 10.8 8.2 7.2 8.4 12.8 23.7 27.2 2.2 7.3 - 2.9 - 2.1 - 2.0 25.5 _ 17.4 - 15.9 - 7.5 - 5.0 - 4.0 - 14.0 2.1 4.7 31.4 4.7 13.5 - 15.6 - 3.5 - 2.9 60 Table 6--continued. % of LAI Ecosystem Species 3 mo 18 mo Monoculture Zea mays 10 0.0 Manihot esculenta - 100.0 Includes at least six species of grasses that were indistinguish- able by vegetative parts. Includes at least four species of sedges that were indistinguish- able by vegetative parts. 61 and ^,he natural succession. Two wocdy species (Bocconia fcutescens and Clibadium atf. surinamense) and two grass groups (Panicum fflaxiguffi and a group of 10 grass species) were abuiiu^nt in both ecosystems at 3 mo and 18 mo (Table 6) . However, the enriched succession gained fewer new dominant species (Table 6) , but more species overall (including ail species encountered in the LAI measurements. Table 4) than did the natural succession from 3 to 18 mo. The community similarity index between the 3 mo old vegetation and 18 mo old vegetation was higher in the enriched succession (C = 0.60) than in the natural succession (C - 0 . Ul) . This is due both to the addition of fewer new dominaiit species and to smaller relative changes in specxas abundance over time in the enriched succession. The 82 species present in the mimic plots at the time of the October 1980 species inventory represent Hb% of the 178 species introduced into the mimic plots from March 1979 to October 1980. During the first 3 mo of succession, plant growth and structural development in the mimic of succession egualed or exceeded that of the natural succession. This was due primarily to the early and rapid development of herbaceous species (mainly cultivars) in the mimic. In subsegueat months, development of the mimic was slower. At 18 mo, species richness and plant diversity were lower in the mimic than in the natural succession. In general, the mimic was muca more similar structurally to the natural 62 succession than to tiie monoculture. The structural differences between the luituic and the natural succession indicate that (1) there was a time lag between the dcvclopaieut oi tue natural succession and the development of the mimic, and/or (2) some of the species introduced into the mimic treatment, althouqh morpholoqically similar to the native successioaal species, were not yood functional mimics of the native species. Larqe numbers of relatively uncoBmon species were present in the natural succession, but not in the mimic, at 3 mo (Fitjs. U and 6) . This probably reflects the initial pattern of species introductions in the mimic by the investigators. This difference between the mimic and the natural succession elucidates an important characteristic of the natural succession that was difficult to imitate. The many rare species in the natural succession formed a pool of potentially important ecosystem components that could increase in dominance as microenvironmental factors and the competitive balance of the system changed. In aanaginy the mimic ecosystem, anticipation of the types of species needed and introduction of such species at appropriate times to insure establishment and to maintain a pool of rare species was difficult. Sevoral structural characteristics of the mimic at 18 mo, including species abundance, were similar to characteristics of tae natural succession and enriched succession at a much 63 5n 3 MONTHS I5n 10- 5- 18 MONTHS .8 1.0 LAI Figure 6. Number of species in the mimic of succession by LAI class. Values are based on 180 LAI measurements on each date. 6a earlier age (3 mo). The number of species intersected by LAI measurements in the aiiiaic at 18 ao (32 species) is similar to the aumbers intersected in the natural succession and enriched succession at 3 mo (37 and J5 species respectively). This indicates slower development ot the •investiqator-controlied' treatment (the mimic) than of the ' natare-coutrolled' treatments (natural succession and enriched succession) . For example, there was a time lag between the appearance of woody species in the natural succession and the selection and introduction of similar woody species in the mimic. It is expected that longer-term results will show convergence of structural characteristics of the mimic and natural succession. The mimic of succession had higher turnover of species than the enriched or natural succession (Taoles 4 and 6) . The species composition of the 18 mo old mimic was very dissimilar to that of the 3 mo old mimic (C = 0.15). The Jul/ 1979 monoculture and the October 1980 monoculture had no species in common (C = 0-00). Changes in species composition in the monoculture were not gradual as in the other ecosystems; instead, composition changed completely as one monoculture species replaced another. If community similarity (C) is used as a measure of rate of species turnover in each ecosystem, with lower C values indicating greater changes in species composition during the first 18 mo of succession, then the systems may be ranked by 65 iraqnxtude of chaage as follows: monoculture > mimic > natural succession > enriched succession. Leaf Area Index Leaf area index developed rapidly in both the natural succession and the enriched succession (Fig. 7). The LAI increased rapidly in all ecosysteias during the first 2 mo, but thereafter was lower in the miaic than in the natural succession and enriched succession. Seasonal LAI fluctuations were similar in the natural succession, enriched succassion, and mimic, with maKimum values during the rainy season and minimum values during the dry season. Increase in LAI was rapid during the growth of the first maize moncculture (LAI - 1.22 at 2 mo), but leaf area development of the second maize crop was poor (maximum LAI = 0.5). Cassava LAI after y luo of growth (mean jh 1 s.d. = 2.9 + 2.0) was not significantly different from LAI in the 7 mo old natural succession (3.7 + 2.0). A decrease in LAI occurred during the dry season in the natural succession, enriched succession, and mimic. At 18 mo, mean LAI (+ 1 s.d.) was U.4 + 2.6 in the natural succession, 5.0 ♦ 3.4 in the enriched succession, and 3.6 + 3.C in the mimic. Vertical distribution of j.eaf area was similar in the natural succession and enriched succession (Figs. 8 and 9) , except in tie lowest (0-25 cm) stratum. In this stratum near the ground LAI was consistently higher in tne enriched 66 . c 0 o ■H en - tfi (1) CO U u 3 u 0^ 3 1^ CU m iH !-l 03 3 S-l +J 13 rH 4-) 3 tCj U C 0 c C 0 •H E H -O < c J tC r~- cu u =) irl I V1 67 re • 0) W U CD (C W 0) to 4-> CD C iH 0) C C O -H -H 3-H -H 0) +J (C ■H ^ M 0) u -p -H td S-l H 0) <■; > hJ 68 aj tfi u QJ rrl tfi QJ U-l j::; (C -P CD C iH QJ S-i M-i m O 0-, c c 0 •rn -H -p tn p •H Si •H CD n tr -l-l ftl en -rH £ ^ U fC iH O (0 u 4J •H (t1 +J P M 0) si; Lu) 1H9I3H H •H Cm 69 succession than in the natULal succession. This may be due to the abundaace of introduced propagules in the enriched succession, leading to increased numbers of seedlings. In the 0-25 cm stratum, 0%, 3.9%, and 6.5% of the LAI was comprised of introduced species at 8 mo, 13 mo, and 18 mo, respectively. The LAI in the mimic was concentrated <1 m from the soil surface at 8 mo and 13 mo, and leaf development higher in the canopy was patchy. By 18 mo tiie height of the canopy had increased in the mimic, although more than half the leaf area was still concentrated <1 u from the ground (Fig. 10) . Vertical distribution of leaf area in the monoculture reflected the growth form of a single species rather than the interactions among a large array of species. In the mature cassava monoculture, leaf tissue was concentrated at 1-J m above the ground (Fig. 11). All ecosystems were characterized by rapid growth to an average canopy height oL 3-4 m at 18 mo (Fig. 12). The natural succession and enriched succession contained some emergent plants ^rith heights of up to 10.8 m at 18 mo (Table 7). 70 fC +J 0 H c 0 •H to tn 0) u u P w IW 0 u -H E •H e Q) X! 4-1 C ■H (C . (U Ul ^) d) (0 tn cu M-i 4:: m -p -l m (« 0 a c c 0 ■H •H -M tn 3 -H X! •H 0) u D^ +J fC m •H £ ^ U (C iH 0) (0 C) -P •H (fi +) M H QJ tf t^' iJ •H i^) 1H9I3H 71 (1")1H9I3H r-{ fO -P 0 H -i 3 4J iH 3 O 0 c o g m > (C m m (C u QJ ,C +J C •H tC (U >-l n3 IH to Q) .H M~l 0 c in 0 ^q 13 ■H 72 o 2 (J (/7 (O (f) C/5 (/) 3 • [fl tfi -3 01 O S 00 ■HlX (J) ^i C 0) < (1) S-l 2 0 <]) ■H ;d U- CO (C cu > O u -3 3 • W 0) Q to -P 13 3 -d -P O n3 0 O 0 C E C/) -P c ji: to XT' < 0^ x: 0 -^ 1^ C W m 0 W •r^ 0) -3 2 -P o fO CJ -p 3 0) en en 0) 4-' > 0 t ro oj (LU) 1H9I3H 73 Table 7. Tallest plants in natural succession, enriched suc- cession, and mimic at 18 mo, and in cassava monocul- ture at 10 mo. Height of tallest Ecosystem Species individual (m) Natural succession Ochroma pyramidale 10.8 Vernonia patens 7 . 6 Bocconia frutescens 5 . 8 Enriched succession Trema micrantha 7.5 Vernonia patens 7 . 0 Musa paradisiaca 6 . 9 Mimic of succession Manihot esculenta 5.0 Ricinus communis 4 . 9 Monoculture Manihot esculenta 4.2 74 Herbivory Bates Mean herbivory rates varied widely among species, and among sampling dates for some species (Table 8) . For most species, herbivory rates were not normally distributed. The Kolomogorov-Smirnov statistic to test the null hypothesis that the data were a random sample from a normal distribution was significant in 50 of 59 tests. Sample distributions were skewed to the right in most species studied (Fig. 13). Median losses were lower than mean losses for all species (Table 9) . In three species (Panicuj trichoides, Ery thrina costaricensis, and Maaihot esculent a) , damage distribution was dependent on the type of ecosystem in which the species was found (Fig. 14 - Fig. 16). Of the eight species monitored in both the natural succession and the enriched succession, one species (Panicua trichoides) had different herbivory rates in the two ecosystems. This species had a lower rate in the enriched succession than in the natural succession (Table 9) . For the two species monitored in the enriched succession and in the mimic of succession ( Erythrina costaricensis) and Manihot esculenta) , both had lower herbivory rates in the enriched succession. Hanihot also had lower rates in the monoculture than in the mimic. Some ecosystem characteristics that may affect the herbivory rate on an individual species are species diversity, LAI, and species composition. In addition, the 75 i-l Q) X} B c 3 •H C ^ >H —• 0 Ti ~ m w "" — •H Ix 0) Cu !-l tn nJ 'D en 0) 0) > U) fO m CD 0 iH k:| 1 e C 0) J-i -p (1) Cfi 4-1 >i.H (n to 0 ^-' • u ^ — , OJ (fl m QJ Q) T! > •H C (11 u rO QJ 0) .H a [fi 03 0) m ■H 0 'd u a) 0) !-l > Dj OJ m tfl X! 0) i= iH >i P 1 42 C CU -p cn m ■H Q) ■H cn m o m C D. 0 a, iH •» 0 >i >i (C J-l •n cn 0 \ u > m H •H re m ^ 0) a, iH iH 0) y-j £;(N m E^ (1) C \nH (OCN CD F; M-l S o 0 d 4-4 0 0 •H cn CJ cn H aj e CJ H o jr; 13 cn c ■n 0 CU ■H ^ cn V cn •H a) V4 o c o w 13 CO c 0 rH •rH (« cn S-l cn :=! CD 4J o (« o IZ 3 CX) cn H i-D O^ O "* O O (Ti 00 C^l "XJ cH r~ ^ o r~ v£) n r~ LT) O ID H r-\ ro r- 00 a^ o ix> m iH tN tN 'cr »-' tN tN iH ^ in »* iH iH iH <-{ iH O 'S' iH (N 00 tN 1— I iH tN IX) tTi LT) 00 O O 00 in fs 00 r~ >i ■H ^ M P4 cn c d) (0 o ■H cn C CD o 4J o 13 o i-l 0 m m M-l CD fO cn c (= CD 13 e -H to Ti c to •H Xi M ■H 3 iH cn U 76 O -H ■rH 0) c -r^ 0 Q) ■H j=. en u 02 •H OJ M u c CJ w 13 m C 0 <-^ ■H fO U] S-l cn 13 (1) -P C) (C u 2 3 a) 00 ^ ■>* 00 00 in in o vc IT) >X) oi ro IT) o o r-- i>D in isD 00 o 00 00 o in ijD r- 1^ 00 ro in ,-1 — ^-^ iH oo ro 00 CN 00 . . 0) . 0) CD . • 0) +J -Q c j-i XI c c -P X! c u 0) :3 u (U 3 3 O 0 3 o t^ l-D o tl 1-0 t-3 o IJ-i 1-0 m 1-1 c Q) 0) C u ■H f= en 3 0) rO c: vj •H (0 en -n rH -H M 0 C 0 in u tn aj ^ ■H g 0 3 ^ u u ■iH ■H c i^ m +J cu 0) rC 0) C 77 QJ U =! tn +J Ul <-\ O 3 r-\ U 0 c C o s; c iw o O -H en •H 0) e u CO c 'C 0 0) •H £ en u in -H Q) i-l o c u w :3 m C o rH •H nj [fi Vj tn 3 0) +J u 01 0 S 0 w r^ in 1^ r- cN '^r •^ CTi VD CTi U-) r-l 00 C^ o ^ n m LD O in CN CM ro r- Ln CN rH CM CN rH rH . . CD . CD q; 0) • 0) 4J X! c Xi c c c 4-1 Xl c u 0) 3 Q) 3 3 in o 0) d o Pm *-) fe ^^ h) •d o &H 1-:) X! s d trt 0) d CU 01 frt •H fO 4-1 c u U 4-) (U td 4:: •rl •rl c o R rH 0 X Ti m ro d rH fd 0 Sh M Sh c CU c ■H n (fl Q) (0 XI (U a e jC a rH B fci fc: o o >i O d > 03 s u CO s:: 78 C u-l 0 o •H (fi u Cfi •rH OJ e o -H u 5C 13 W C TJ 0 0) ■H £ m O en ■H 0) M o C 0 W !3 Ui c 0 rH •H to Cfl M m 3 cu 4J C) (H u s 13 W ■ -— • LD •^ OD ro CM ro ■<* o ro IT) 00 fS ^ o o>i o in CO r>j ro VD IX) cTi Ln ro ro t~- o >x> ro un 1X5 ■<-< IX) r- r- LD '^ CN ro — - ^ LD iH a) +j QJ • Q) OJ . • • • • f« J2 C ^ C c; XJ -Q 4-) ■4-1 -p D Q) :3 (U :3 3 OJ 0) u 0 u fa l-D t, 1^ 1^ fa fa o o o m trt Ul H n 6 u 0) 0) u ^ M ;3 0) -p •H (0 U •H Q) N ■H +J -P C -P trt ■H M CL, fa cn ■H tn C tn 0) c ro CJ 0) en CO -P •H iH (U 3 •H tn C 5-1 o != i-i iH !h >■ +) (U rfl en 0) d n 0 (0 m 0 iH +J •H > ^ tri CU cr e ^ 3 (fi -P ITl m rH tn rH ■<-{ O O c^ :3 V4 13 a d to C cn U > tn 0 > xi > OJ ro 01 K w 0,' to 2 79 x>'^ • • • I ID ^ iH r-i— I (NO^ (NO OO iHiH OO iHO OO 001^ ^^(M(NHCN1VDOO •^^ ^-' ■^-- ■— ' •»-' >* n in o> in 00 m 00 [^ o o Q) -P 0) . . • • • CU . • m X! c +J +J Xi -p 42 C 4J XI Q 0) p u u Q) u CD 3 U CU fc •-2 o o fa o Cm h) o fa frt 4-1 C 4-1 C) n iH x: 3 ■H U c m m fl) s m ■p ■H en X2 =s C i-l O c fl 13 O^ rS •r—i U 0) •n m =3 a to u U U c 0 en m 3 n 4-> Oi (T3 n M 4-1 fs ■H >i U u 80 c U-l 0 0 •H CO u W •H (D e U •rH U s 0 W c 'C! 0 (1) •H x: tfi u m •rH d) i-l o C o W d Ui C O iH •H n1 W !-t (fi 3 0) 4-1 u (« o 2 3 w ro o O -^T CM Ln 00 O^ QJ +J (1) • • 0 a; (C C +j x> c c Q d u 0) 3 3 1^ o &H •^ i-D C) tC •H en ■H (rt (Tl Tl (1) -P m o 01 s^ != -P (C 0 to u. ^ (« >i (rt Oj rO 01 -H 04 J-l 01 m 01 iH C C) 01 01 •H -P u S-l 0 •H 03 s-l e U U Q) 0) > > •H •H -P •P 01 n +J -p Q) Q) in tji 0) 0) > > >i >i ^ X 0) Q) iH rH XI X! 0! OJ j; x: en CO •rH •H :3 3 tji cn c c -H •H •P -p CO CO •H •H T! ■ri C c •H •H 0) Q) M S-l Q) (1) s 5 -p -P 01 05 x; x; -p -p CO CO 0) (U CO en CO 'C! 0! 0) u CO CTi tw 14-1 O 0 CO CO OJ CD -rH •H o O (D 0) a, a CO CO S-l X 3 ■rH 0 to M-l •p +J CO CO 01 01 (U 0) <-\ iH -p H-» 01 01 CO CO CD 0) •n • 13 • 3 CO P CO nH H-) H H-) O i-l O U C 01 C 01 M Cu H a 81 Ponicum maximum ■>r. , , n LOSS (cm'/m'/doy) Cyperaceoe n n LOSS (cm'/m'/doy 1 — I 1 1 1 r u. -J i Clibodium aff. surinomense ^ LOSS ( cm'/m'/doy I Vernonia patens l\k4m- n n n n I n i n LOSS (cm' /m' /doy 1 Bocconio frutescens ^HV^ -=\ r^=h LOSS (cm'/m'/doy) Phvlolocco rivinonies n n n — n i n — r-H — r-=l — F=-r 100 150 LOSS (cm'/m'/doy ) Figure 13. Distribution of loss to herbivores among leaves in six common species. 82 E QJ 4-) -P m G >i 0) CO e 0 -u U (tt i.H CD S-< O Ti 0 c fO > -H e ■H £1 ^ (D M C X! 0) — x: Ti en iH c cu 3 (C > 0 •H (C u 'O Q) Q) iH en g c M-) 0 T! 0 en C •H (C ^ S-l CU fO g Q) 3 o S 2 o Dj to \ cne^J en E o-\ 6 U r- o ca r^ iH 00 ro ID ^ ■^ cN IX) C3^ iH ro O rH LD r~ iH in CTi o \X) o IT) LD r~- in ^ n-i (N r- nH IX) CTi X5 in (N! r- CM ro o^ r- ^ (N O 00 r~- ■^ 00 in '?' ro rr (N CN fN) rH -H (^; vx) U-lXl -l u !-l •rH 3 0 M-l rH en en u CD to CU U fO s-l CD >i u T! CD tes CD ■p c 0 ■H x: e •H to c s-l (0 U) s IX) 00 CO (N CTi ^ in IX! cTi in ■^ ^ o r~- in O 'S' »:?• in •<3* rH LO r~- ■* ^ rH O ^ r~- ro oi "sT in H CN) •>* rH ro ro 1X5 H CM rH H CM H H in ro ro H H rH fS H rH VD CM (Ti 00 c C C c C c c c c C C C 0 C 0 c n 0 n c o o 0 c 0 C 0 C 0 0 -H 0 ■rH 0 •H •H •r-t 0 •H •H •H 0 ■H 0 •rH 0 -H •H en •H en -rH en en en -H en en tn •rH en -H cn •H en en en en en en en en tn en en en en en en en tn tn en en CD en CD en Q) a) a) en CD CD CD en CD en CD en CD CD U CD U CD C) CJ o CD O O U CD CJ CD o CD U u u C) O C) C) C) C) C) u O U U O CJ u CJ U U d 0 3 U 3 3 3 O 3 3 3 (1) u 3 o 3 (J 3 3 en 3 en 3 en en en 3 en en tn ^ 3 en 3 en 3 en en tn tn en 3 en en en 13 TS T! Ti 4-1 T! Ti IH 4-) T! T3 Ti rH CD H CD H aj CD 0 .H CD CD 0 rH H CD rH CD rH 0) to x: to £ to £ -C to £ ^ 3 to £ to -C to JS ^4 o s-l U s-l u C) u s-l CJ U CJ CJ Sh U Sh o ^ U 3 -rH 3 •i-i 3 •H •H •rH 3 •H •H •H n 3 -H 3 •H 3 -H 4-1 n 4J s-l 4-1 S-l S-l E 4-> S-l S-l E c 4-J Sh 4-1 Sh 4-1 M to c (0 c to c c •H (0 c c •H n to C to C (0 C 2 W 2 w 2 w u •rr z w w S 2 s W 2 W 2 W en en to CD to CD 4-1 Ti U ri c E •H u •H a; E 3 E n to 0 H 3 E 3 -C rH C 3 O •H O o O •H CJ •H >^ •H •H 4-J > en C fO c Sh > •H CD to E to Cm 4J Xi Sh 83 ro cntN O \ u '3* CT, n CM c C 0 0 -r^ •H m w m en CD f= (D O (1) o u -p u 3 in d en >i en en TJ n >-\ Q) o d j:; w i-l u d -H 4-1 S-l to c 2 W 0) ca •H •H u c aj 0 a. c en u Q) > (U 0) > > ■H •H +J 4-1 (0 (C 4-1 4-1 (U (D en tJ^ (U CD > > >i >i X! X5 (U CD • iH iH 4J £ X! en (C rC 0) £ £ 4J en en •H •H C 13 3 m tji (T" •H C C T3 •H •H 0) 4-J 4-1 6 en en •H ■H V T! TJ JJ • C c en en ■H •H OJ !-l +J ■H Q) CD rO s-l S-l en CI- 0) (D ■H S s rH 4-1 -H (0 -P 4J eo ei) ro (CJ S: iH x: x; 1 +J 4-1 iH 0) (0 -p en en >i •H (U CD en en tJ" en 13 0 ^3 en S-l a CD (0 tiii a en s-l 0 tn •. 4-J -P 4-i' o IW en 0 0 0) en 4-) s-l CD en eu ■H CD en jQ U •H E E Q) o :=! 3 O, CD en c en CI4 en A^ en ^-1 C •H :3 X m 0 •H !-l C 4-1 en (U ^ +J 4-1 tH en en en O^ Q) tes ro e •rH QJ CD rO CJ iH rH en 0) 1 c 4-1 +J (N en (0 to c en en en 0 ■M (D 0) X x: T) • t3 • 0 4J 13 en 3 en u iH 4-1 iH 4J iH s-l o u D s-l ■H 0 C (0 C to g fo H CI, M Oj tC ^ u t3 84 80 -{ 70- 60- 50- 40- 30 CO UJ > 20H UJ I 0 S 0 Panicum trichoides o 40n UJ Q. 30- 20- 0- ENRICHED SUCCESSION 1 — r NATURAL SUCCESSION 50 100 LOSS (cm2/m2/day Figure 14. Loss distribution among leaves of Panicum trichoides . n ii hi CO T y> O tr CJ 2 3 UJ cn o c Ixl I I o C\J S3AV3n (J o ^ j; o JO lN3Dd3d (0 •H (0 c (U D •H )H 1 u &a 14-1 o tn C O E m c o •H X! ■H S-i +J tn •H 'O [fl CO o 3 •H 86 [-§ "-S >s S3AV31 iO XN3Da3d 87 abundance and spatial distribution of a particular species within the system may alfect its herfcivory rate. Few differences iu uerbivory rates between natural and enriched succession were expticted, because these systeas were very siiaila^^ in species diversity, LAI, and species composition. £§lii£liS mchoides had relatively low LAI in both systems (0.04 in natural succession, 0.07 in enrxched succession). Thus differential plant abundance was probably not an important factor affecting herbivory rate for this species. Plant spatial distribution aud/or small sample size may explain the observed difference. Several factors may contribute to the hicjher herbivory rates on Erythrina iu the mimic than in the enriched succession. Abundance of Erythrina was similar in the two systems. Although both systems had relatively high species diversity, tue species similarity between the systems was low. In addition, the LAI of the mimic was lower than the LAI of the enriched succession. This suggests that the kinds of species that surround a given plant, as well as their abundance, may affect the herbivory rate on that plant. Manihot and Erythrina (both cultivars) had lower apparency and greater protection iroia herbivores when surrounded by native successional species in the enriched succession plots, than when planted in plots containing a different array of species including aany cultivars. Manihojt, a relatively unpalatable species, had its highest herbivory rate in the ecosystem with intermediate species diversity and LAI (the miCBic) . The herbivory rate on this species was not linearly related to species diversity. This result suggests that species coaipositioD, rather than diversity £er se, was an important factor influencing herbivory on Maaihot. There was no simple relationship between LAI of a species and that species' herbivory rate. However, the data indicate that in the natural succession, enriched succession and mimic, the very high rates of herbivory occurred on the less common species, and all of the very common species (LAI > 0.5) had relatively low herbivory rates (Fig. 17). The loss rate for each species (cm^ m-2 leaf day-») was multiplied by the LAI of the species to obtain the species' loss rate in cm^ m~2 ground day-*. Some relatively uncommon species contributed significantly to the total ecosystem loss to herbivores (Fig. 18). The coefficient of variation (CV = s.d./mean) of Jheruivccy rates was used to identify trends in the spatial distribution of damage among leaves and plants of several species. A large coefficient ot variation (i.e., s.d. > mean) indicates high variability in herbivory rate among leaves or plants, and implies aggregation of damage, with some leaves or plants receiving very high levels of damage and others receiving very low levels. A low CV value (i.e.. 89 -1 100- - MIMIC •g eo- o N 6°- e E "0- O 20- > . ! 1 1 1 1 1 1 r 1 - 60- ENRICHED SUCCESSION NATURAL SUCCESSION I I I r MONOCULTURE E OH — I — I — I — I — I — I — I — I — I — I — I — I — I — I — I — I — I — I — I — I — I I I I 3-0 .2 .4 .6 .8 1.0 1.2 \A 1.6 1.8 2.0 2.2 2.4 LAI Figure 17. Losses to herbivores by LAI. Each point represents one species. 90 2 Z o o 3 3 UJ cn CO q: n i£ o < 2 O , 1 X! (0 Q) >-) m -d C 3 0 i^ Cn •4-1 ■H C 3 V4 1 H >-( O O 0^ -H en X! ^ Q) 0) C m O I I I I I I I I I I I I I I I I I I I I I I I I 1 I lO O cD O in Q (ADp/punoj6 ^Lu/^uuo) SS0~1 91 s.d. < mean) indicates that spatial variability of damage is low and implies that damage tends to be evenly distributed among leaves oc plants. The CV calculated using mean leaf herbivory rates reflects the damage distribution among leaves of a given species; tue CV calculated using mean plant herbivory rates reflects the daioage distribution among plants. The CV values calculated from leaf herbivory rates were higher, on the average, than the values calculated from plant herbivory rates (Table 10) . This implies that leaf-to-leaf damage variability was higher than plant-to-plant variabili^-y. In other words, most damage from herbivores tended to be aggregated on a subset of the leaves of a species, but ail plants of the species in the same ecosystem were equally likely to have some leaves heavily damaged by herbivores. Both leaf-to-leaf and plant-to- plant variability were high in cassava. This result reflects the foraging pattern of one of cassava's major herbivores, the leaf-cutter ants (Atta cexjhalotes) . These ants selected a few plants of cassava for consumption (leaving many other individuals untouched), and removed some (but not all) leaves of each selected plant almost entirely, leaving only the mid-ribs. Young leaves and old leaves of most species were consumed at equal rates. Percent leaf expansion during the monitoring period was used as an indicator of leaf age (high percent expansion = young leaf; low percent expansion = old 92 Table 10. Coefficients of variation (CV) of herbivory rates by species. Coefficient of variation is calculated: (1) based on individual leaf data and (2) based on plant data. Based on Leaf Data Based on Plant Data Number Number of of Species Leaves CV Plants CV Cordia inermis 62 4.89 10 2.07 Manihot esculenta 140 2.57 33 1.95 Gramineae^ 38 2.21 9 0.81 Phytolacca 178 2.02 27 1.53 rivinoides Cajanus cajan 40 1.94 7 1.74 Momordica charantia 21 1.87 5 1.70 Hymenachne 46 1.71 11 1.21 amplexicaulis Panicum trichoides 35 1.61 10 1.27 Bocconia frutescens 99 1.46 25 1.41 Frantzia pittieri 28 1.39 6 0.68 Phaseolus vulgaris 25 1.26 5 0.81 Panicum maximum 44 1.12 13 0.76 Cymbopogon citratus 39 1.10 11 0.71 Erythrina 46 1.08 10 0.91 costaricensis Vernonia patens 94 1.01 16 0.70 Ipomoea batata 22 0.98 7 0.79 ^Includes at least six species of grasses that were not distin- guishable by vegetative parts. 93 leaf). Linear correlation coefticients between percent expansion and iierbivory rate were non-siqnif icant for the 12 species LosceJ. lu addition to this test, the leaves of each of the 12 species were divided into two groups (leaves that expanded >^0% during the iDonitoring period, and leaves that expanded <^0%) , and mean herbivory rates of the two groups were compared using F-tests. The herbivory rates on young (expanding) and old (not expanding) leaves were not significantly different for 10 of the 12 species. In two species (Phytolacca riyinoicies and Carica papaya) , herbivory rates were higher on old than on young leaves. The LAI and herbivory rates for each species and ecosystem are summarized in Table 11. Herbivory rates on a per-leaf-ar'^a basis were multiplied by species' LAIs to obtain leaf loss rates on a per-ground-area basis. When these values were multiplied by species' specific masses (g/m2 of leaf tissue) , biomass loss rates resulted. Ecosystem herbivory rates were obtained by summing species' per-ground-area loss rates. It was assumed that losses in unsampled species equalled the weighted mean of the sampled species. Ecosystem losses to herbivores (mean per-ground-area rates, averaged over all sampling dates) were equal in the aimic of succession (54 + 44 cm^ m-2 ground day-i, natural succession (61 ♦ 23), and enriched succession (56 i 9) , and lower in the monoculture (11 ±9). Variability among the 94 Table 11 Leaf area index, leaf specific mass, and losses to herbivores in natural succession, enriched succession, mimic of succession, and monoculture. Ecosystem Date Species LAI (m2 Leaf/ m^ Ground) Percent of Total LAI 1.53 37.5 0.42 10.3 0.38 9.4 0.33 8.2 0.22 5.3 0.21 5.0 0.15 3.7 0.09 2.2 0.75 18.4 Natural succession Oct. 79 Feb. 80 Phytolacca rivinoides Bocconia f rutescens Clibadium aff surinamense Gramineae^ Momordica charantia Hymenachne amplexicaulis Borreria laevis Others Ecosystem^ 4.08 Clibadium aff. 0 .71 surinamense 0 Phytolacca .68 rivinoides 0 Panicum maximum .38 Bocconia 0 .27 frutescens Vernonia patens 0 .25 Momordica 0 24 100 19.4 18.7 10.4 7.3 6.9 6.6 charantia 95 Table 11--6 xtended. Leaf Specific Mass (g/m2) Herbivory Rate cm-^/in^ Leaf/ Day ciTi'^/m- Ground/ Day g/m^ Ground/ Day 30.6 16.5 25.2 0.077 37.1 30.5 12.8 0.048 52.4 13.7 5.2 0.027 24.4 0.7 0.2 0.001 9.2 4.5 1.0 0.001 45.8 16.3 3.4 0.016 38.6 2.0 0.3 0.001 18.5 7.4 0.7 0.001 32. 9 = 14. 7^ 11.0 0.036 32.9 14.7 59.8 0.208 52.4 16.0 11.4 0.060 30.6 12.7 8.6 0.026 45.8 6.8 2.6 0.012 37.1 11.9 3.2 0.012 61.2 24.2 6.1 0.037 9.2 131.4 31.5 0.029 96 Table 11 — continued, Ecosystem Date Species LAI (m2 Leaf/ m2 Ground) Percent of Total LAI Natural Feb. 80 succession June 80 Gramineae^ Hymenachne 0.17 0.13 4.7 3.7 amplexicaulis Cordia inermis 0 12 3 4 Frantzia pittieri 0 09 2 6 Others 0 59 16 3 Ecosystem^ 3 .63 100 Panicum maximum 0 .57 24 8 Clibadium aff. surinamense 0 42 18 1 Gramineae^ 0 .29 12 8 Hymenachne amplexicaulis 0 .19 8 4 Bocconia frutescens 0 .08 3 6 Phytolacca rivinoides 0 .07 2 .9 Solanum umbel latum 0 .05 2 .2 Panicum trichoides 0 .04 1 .9 Momordica charantia 0 .04 1 .9 Frantzia pittieri 0 .04 1 .9 Table 11 — extended. 97 Leaf Herbivory Rate Specific cm^/m-=^ Leaf/ cm^/m^ Ground/ g/m^ Ground/ Mass (g/m^) Day Day Day 24.4 29.7 5.0 0.012 38.5 2.3 0.3 0.001 25.8 24.5 38. 6^ 6.3 4.8 23. OC 0.8 0.4 13.6 0.002 0.001 0.052 38.6 23.0 !3.5 0.244 45.8 52.4 13.0 13.5 7.4 5.7 0.034 0.030 24.4 38.6 36.4 11.5 10.6 2.2 0.026 0.008 37.1 30.6 38.3 30.8 9.2 24.5 9.2 9.8 21.4 15.6 13.2 0.7 0.7 0.4 0.9 0.6 0.5 0.003 0.002 0.001 0.003 0.001 0.001 98 Table 11 — continued. Ecosystem Date Species LAI (m2 Leaf/ m^ Ground) Percent of Total LAI Natural June 80 Cordia inermis 0.04 succession Cyperaceae'^ 0.04 Others 0.44 Enriched succession Oct. 79 Ecosystem" 2.31 Panicum maximum 1. 04 Phytolacca 0 86 rivinoides Bocconia 0 59 f rutescens Gramineae^ 0 37 Clibadium aff. 0 36 surinamense 0 Momordica .29 charantia Solanum 0 .14 nigrescens Borreria 0 .10 latifolia Vernonia patens 0 .09 Others 0 .54 Ecosystem^ 4 .38 1.9 1.9 17.7 100 23,7 19.7 13.6 8.4 8.2 6.7 3.2 2.3 2.2 12.0 100 Table ll--extended, 99 Leaf _^ Herbivory Rate Specific cm^/m- Leaf/ cm-^/m- Ground/ g/m^ Ground/ Mass (g/m^) Day Day Day 25.8 12.5 0.5 0.001 46.1 40.1= 21.5 16. 6C 0.9 7.3 0.004 0.029 40.1 45.8 30.6 37.1 16.6 14.7 14.4 25.6 38.4 15.3 12.4 15.1 0.143 0.070 0.038 0.056 24.4 52.4 0.6 17.9 0.2 6.4 0.001 0.034 9.2 25.6 4.5 8.4 1.3 1.2 0.001 0.003 18.7 10.3 1.0 0.002 61.2 35. 8C 46.3 14 .90 4.2 8.0 0.026 0.029 35.8 14.9 i5.1 0 .260 100 Table 11 — continued, Ecosystem Date Species LAI (m^ Leaf/ Percent of m2 Ground) Total LAI Enriched succession Feb, 80 Enriched succession June 80 Panicum maximum Phytolacca rivinoides Clibadium aff surinamense Bocconia f rutescens 0.87 0.69 0.47 0.46 Gramineae^ 0 .23 Musa 0 .14 paradisiaca 0 Vernonia patens .11 Merremia 0 .08 tuberosa Others 0.68 Ecosystem 3.73 Panicum maximum 1 08 Clibadium aff. 0 32 surinamense 0 Gramineae^ 19 Musa 0 11 paradisiaca 0 Erythrina 08 costaricensis 0 Bocconia 08 23.4 18.6 12.7 12.4 6.1 3.7 2.8 2.2 18.1 100 41.8 12.4 7,5 4.1 3.2 3.0 f rutescens Table 11 — extended, 101 Leaf Herbivory Rate Specific cm-/m-^ Leaf/ cm^/m^ Ground/ q/m^ Ground/ Mass (g/m2) Day Day Day 45.8 30.6 52.4 37.1 24.4 81.0 61.2 54.0 42. 8C 42.8 45.8 52.4 24.4 81.0 37.5 37.1 12.8 27.1 9.2 15.0 7 6 0 8 34 .1 1 4 15. 3C 15.3 15.5 16.1 13.5 3.8 14.8 51.0 11.1 18.7 4.3 6.9 1.7 0.1 3.8 0.1 10.4 57.1 16.7 5.2 2.6 0.4 1.2 4.1 0.051 0.057 0.023 0.026 0 .004 0.001 0.023 0.001 0.045 0.231 0.077 0.027 0.006 0.003 0.004 0.015 102 Table 11 — continued. Ecosystem Date Species LAI (m^ Leaf/ Percent of in2 Ground) Total LAI Enriched succession June 80 Mimic of Oct. 79 succession Panicum 0 .07 2.6 trichoides Merremia 0 .07 2.6 tuberosa Phytolacca 0 .06 2.2 rivinoides Cyperaceae^ 0 .05 1.9 Vernonia patens 0 .05 1.9 Others 0 .43 16.8 ]cosys tem^ 2.59 Cucurbita pepo 0 .31 Phaseolus 0 .28 vulgaris Ipomoea batata 0 .13 Oryza sativa 0 .09 Cymbopogon 0 .08 citratus Manihot 0 .08 esculenta Cajanus cajan 0 .07 Musa 0 .02 paradisiaca 0 Others .72 100 17.4 15.9 7.5 5.0 4.7 4.7 4.0 1.2 39.6 Ecosystem^ 1.78 100 103 Table 11— e extended. Leaf Specific Mass (g/m2) Herbivory Rate cm^/m^ Leaf/ Day cm'^/m2 Ground/ Day g/m2 Ground/ Day 30.8 3.5 0.2 0.001 54.0 51.6 3.6 0.020 30.6 3.5 0.2 0.001 46.1 10.6 0.5 0.002 61.2 77.9 3.9 0.024 45. 8^ 17. 9C 7.7 0.035 45.8 17.9 46.3 0.215 46.5 27.7 8.6 0.040 34.2 44.9 12.6 0.043 25.1 28.4 3.7 0.009 45.8 28.7 2.6 0.012 59.6 0.5 0.04 <0.001 45.4 7.3 0.6 0.003 34.7 12.5 0.9 0.003 81.0 1.1 0.02 <0.001 41. 3C 27. 3C 19.7 0.081 41.3 27.3 48.8 0.191 104 Table ll--continued. LAI (m^ Leaf/ Percent of Ecosystem Date Species m^ Ground) Total LAI Mimic of Feb. 80 Ipomoea batata 0.41 19.6 succession June 80 Ca janus cajan Hyptis suaveolens Cymbopogon ci tratus Musa paradisiaca Sorghum vulgare Manihot esculenta Others 0.46 22.2 Ecosystem^ 2.10 100 0.39 18.5 0.31 14.6 0.21 10.1 0.13 6.3 0.13 6.1 0.06 2.6 Cymbopogon citratus 0 .71 53 .1 Hyptis suaveolens 0 .12 9 .1 Erythrina costaricensis 0 .07 5 .0 Manihot esculenta 0 .04 3 .3 Carica papaya 0 03 2 1 Crotalaria micans 0 .03 2 .1 Others 0 .34 25 3 Ecosystem^ 1.34 100 Table ii--extended. 105 Leaf Herbivory Rate Specific cm^/m^ Leaf/ cm^/m^ Ground/ g/m^ Ground/ Mass (g/m^) Day Day Day 25.1 103.7 42.5 0.107 34.7 31.6 62.9 26.2 24.5 8.1 0.085 0.026 59.6 1.4 0.3 0.002 81.0 0.5 0.1 0.001 49.4 45.4 5.6 35.7 0.7 2.1 0 .004 0.010 40.1^ 47. 8C 22.0 0.088 40.1 47.8 100.3 0.323 59.6 31.6 37.5 45.4 0.9 39.0 57.5 1.7 0.6 4.7 4.0 0.1 0.004 0.015 0.015 <0.001 48.5 26.2 52. 8C 3.2 1.9 9.6 0.1 0.1 3.3 <0.001 <0.001 0.017 52. 9.6 12.9 0.051 106 Table ll--continued. Ecosystem Date Species LAI (m2 Leaf/ m^ Ground) Percent of Total LAI Monoculture Oct. 79 Zea mays Feb. 80 Manihot esculenta June 80 Manihot esculenta 0.38 0.85 2.30 100 100 100 107 Table 11 — extended. Leaf Herbivory Rate Specific cm^/m^ Leaf/ cin^Tnv^GroundT g/m-'^ Ground/ Mass (g/m^) Day Day Day 53.1 6.2 2.4 0.013 45.4 11.5 9.8 0.044 45.4 9.1 20.9 0.095 ^Includes at least six species of grasses that were indistin- guishable by vegetative parts. ^Ecosystem values are totals (LAI, percent of total LAI, losses in cm2/in2 ground/day, losses in g/m^ ground/day), and species' means weighted by LAI (leaf specific mass, losses in cm^/m^ leaf/day) . •^Mean of species values weighted by LAI . "^Includes at least four species of sedges that were indistin- guishable by vegetative parts. 108 three sampling dates , based ca comparison of coefficients of variation, was higher in the monoculture and mimic than in the enriched succession and natural succession, possibly due tr insect outbreaks on cultivars in the monoculture and mimic. The lower rates uf herbivory in the monoculture reflect char act ^^Tistics of the individual species planted there. Low rates were expected on cassava, a relatively unpalatable species. Herbivory on the maize was low, perhaps because the plots were located >1 km from ether agricultural experiments, in an area that had not been cultivated for ■any vears. Although leaf loss to above-ground herbivores was low in toe maize, root damage by soil herbivores was extensive, but not measured. Losses per ground area of ecosystem were equal in the three diverse systems (natural succession, enriched succession, and laimic oi succession) ; however, the percent of available leaf area consumed by herbivores differed among systems. For the three diverse systems, percent loss was neqadvely correlated with ecosystem LAI. Consumption by herbivores ranged from 0.5 cb^ m-2 leaf day-i in Cymbopogon ci tratus, a grass in the mimic of succession, to 1J1 ca^ m-z leaf day-i in Momordica charantia, a native vine in the natural succession. High LAI (0. 41) , together with high per-leaf-area herbivory rate (104 cm2 ni-2 leaf day-*, gave Ipomoea batata the highest 109 per-ground-area consumption rate (42 cm^ m-2 gcouad day-i) . The lowest pec-qround-atea rate (0.02 cm^ m-^ ground day-*) was in plantain (Musa paradigiaca) , a species relatively uncommon iu the mimic when sampled in October 1979. Biomass losses ranged from <0.001 g m-2 ground day-i in several species to 0.107 g m-2 ground day-* in Ipomoea batata. HerDivory rate was not correlated with leaf specific mass (r = -0. 10 on a species-by-species basis) . Expansion ot Holes accounted for 6-60% of the total observed damage in the species studied (unweighted species* mean = 3CX) . At the ecosystem level, losses due to expansion comprised 19-4335 of the total damage {Fig. 19). Percent of total loss attributable to hole expansion was lower in February 1980 than at the other two sampling periods in three ecosysteos (Fig. 19). The February sampliu^j toas during the dry season, a time when leaf production and growth (and therefore leaf expansion) were low. Damage due to hole expansion, averaged over the three sampling dates, wat; similar in natural succession (30%), enriched succession (31%), miniic of succession (32%), and monoculture (32*) . 110 I- o o IW 0 c o -p u o Cu o u a. O O o CO o O o CO 3 •H Cl4 (ADp/punoj5 ^uu/^ujo) SSOI Ill Above-G£ound_Bioj!ass Mean bioaass ot each vegetation conpoxieut (leaves, stems, reproauctive parts, and standing dead) and total above-ground biomass (leaves + stems + ceproductive parts *■ standing dead) are presented by treatment and harvest date in Appendix B. Values are means and standard deviations of original uatranstormed data. Biomass differences are based on four-treatment analyses of variance, i.r>., all four ecosystems were included in each analysis of variance (except tne 23 October 1980 analyses). The vegetation iu three of the four treatments (natural succession, dnrijhod succession, and mimic) was egual-aged throughout the study. In the monoculture, three crops were consecutively planted and harvested. Therefore, after the first maize harvest the vegetation in the monoculture was younger than the vegetation in the other treatments. To coapare the three egual-aged treataents (natural succession, enriched successvon, and mimic) , analysis of variance tests were done on these three treatments only (monoculture excluded) . In 6(4 of 70 analysis of variance tests, the differences detected were the same in the analyses excluding the monoculture as in the analyses including the Boaoculture. The analyses excluding the monoculture detected more diiferences among the three equal-aged treatments than the analyses including the monoculture in one case, and fewer differences in five cases. 112 Differeaces among replications occurred on two dates. Total above-qround biomass was higher in replications 1 and 5 than in the other replications on 8 July 1980 (p<.05) . Standing dead biomass was higher in replication 6 than in the other replications (p<.05) on 9 July 1979. Replications 1, 5, and 6 were dominated by grasses throughout the 1.5 yr study period. The differences among replications may reflect growth differences between these grasses and the domii^ant dicots in the other replications. Leaf biomass (Fig. 20) increased at the same rate in all four treatments during the first 12.5 wk of growth. At 15 wk leaf biomass was highest in the maize monoculture (312 g/m2) and lowest in the mimic (51 q/m^) . Leaf biomass of the second maize crop was low (<25 g/m^) . In the natural succession and enriched succession, leaf biomass leveled off at <500 g/m2 after 24 wk. In the natural succession leaf biomass was maintained at this level through 83 wk, but in the enriched succession leaf biomass increased after 51 wk and was significantly high than the other treatments (1162 g/m^) at 83 wk. High leaf biomass in the enriched succession at 83 wk was due in part to the high leaf biomass of some of the introduced species, such as plantain. Leaf biomass in the mimic fluctuated from 24 to 83 wk and was less than or egual to that in the enriched and natural succession, and greater than or equal to that in the monorvilt'ire. The cassava monoculture developed leaf biomass 113 O CD (^uu/D; ssviAioia dV3n 114 very rapidly, and after 16 wk of ^jrowtii had leaf biomass equal to that in the older mimic treatment. SteiB bicmass (Piq. 21) increased at approximately the same rate in the natural succession and enriched succession. Stem biomass in the miiiic equaled that in the natural succession and enriched succession at all but three sampling dates, and «as siqnit icantly lower than that in the enriched successicQ at 83 wk. The monoculture had hiqiier stem biomass than the other treatBients at maturity of the first Baize crop (567 g/a^ at 15 wk) , but the second maize had very low stem biomass (<60 y/m^ at harvest) . Stem biomass of cassava was lower than stem biomass in the other treatments during early growth, but not at cassava maturity. Biomass of reproductive parts (flowers and fruits) was low (<65 q/m^) in the natural succession, enriched succession, and mimic at most sampling dates, with few significant differences among these three treatments (Fiq. 22) . Values were very low (<20 g/m^) in the natural and enriched succession during the first 18.5 wk of growth and slightly higher (up to 61 g/m^) thereafter. One exception to the low values in the mimic occurred at 18.5 wk, when the bioaass of reproductive parts (255 q/m^) was significantly higher than that in the other treatments. This peak in reproauction was due primarily to the reproductive parts of cultivars such as maize, squash, and beans. The biomass of reproductive parts in the monoculture was significantly 115 O CO o ^ O ^ =) LU Z) ^cr> q: _l Qo -J < LU Z) 5 ^ ^ o : O O O o o o CO o o CD O O o o GO O o o S-l (C Oi r- U) CD m -UJ/6) ssvi/MOia l^31S 116 ■z o O o en (/) en en en 00 LU UJ Ul U ej (J (_) o (_> 3 75 UJ -) cn en (/) -> _J < cc — ) Q UJ X u. O ej> _i O 1- tr "? ^ < 7 CJ 2 UJ ^ s o o o o "3" ^ o eo 1^ CM (\J — en +j u m ft > ■H 4J U d 'C o u ft cu o (0 tn to S o •H m (2UJ/6)siHVd 3Aiionaoyd3a do ssvwoie 117 higher than that in other treatments at maturity of the first maize crop (740 q/m^ at 15 wk) and at maturity of the second maize crop (82 g/m*^, 17 wk after planting) . The cassava monoculture had no tlowers or fruits during the first 32 wk of qro«th and had just begun to flower when it was harvested in September 1980. Standing dead biomass in tiie natural succession and eiiriched succession did not differ at most sampling dates (Fig. 23). In the natural and enriched successions, standing dead biomass was <100 g/m^ to 18.5 wk, and fluctuated between 200 and UOO g/m^ thereafter. Standing dead biomass in the mimic was less than or equal to that in the natural and enriched successions throughout the 83 wk period. In thp monoculture standing dead biomass was generally low, but was higher than in the other treatments at maturity 0.1 ^he lirst maize crop, due to dead maize leaves that remained attached to the plants. At 3 3 wk total above-ground biomass was 2078 g/m^ Iq the natural succession, 4854 g/m^ in the enriched succession, and 1233 q/m^ in the mimic (Fig. 24). The highest biomass value in the monoculture occurred at maturity of the first maize crop (1637 g/m^ at 15 wk) ; the second maize crop had low total biomtiss (<200 g /m^), and matute cassava reached a total aDove-ground biomass of nearly 1000 g/m^. At 83 wk, the enriched succession had higher above-ground biomass than the r.atucai succession; at all other dates the differences 118 O O 00 o . 00 o 00 en a\ rH u 0) X! 0 -P O o ^ o ^ I- cr 2 < 2 - Z LlI ^ • O <] cn ■^'Vi o o (^uj/b) ssviAioia av3a oniqnvis m tn cC e o •H XI (C 0) 'X3 Cn C ■H C (B -P C/2 •H 119 O 00 en ^lu/Sm) ssvi^oia aNnod9-3Aogv 13 c; o u tp I > o (0 120 betws-ja these two treatments were not significant. Total above-qround biomaas in the mimic was yeaeirall/ less than or equal to that of the natural succession and the enriched succession. With a few exceptions, total above-ground biomass followed the trend: enriched succession > natural succession > mimic of succession > monoculture. Total above-qround living biomass (leaves + stems + reproductive parts) in the natural succession, enriched succession, and mimic of succession increased continuously durinq the 1.5 yr study period, with a sliqht dry season decrease durinq January and February of 1980 (Fiq. 25). The monoculture was characterized by rapid increment in above-ground living biomass during growth of the first maize crop, poor growth ot the second maize crop, and rapxd growth of the Ci'ssava. Total above-ground living biomass at 83 wk in the pnriched succession (approximately U kq/m^) was double that of the natural succession (approximately 2 kg/m2) , and total living biomass of the mimic (approximately 1 kg/m2) was about half that of the natural succession. 121 1 D □ { w/b^) SSVW0I8 J/6i() SSVAOIB \\\^ \ O \ \ < \ W \ "- ^ 1 ^ ■ < ' <^ ^ ^ . 2 z 1 M ' o o N\\i r CO / CO / V UJ ^ o \ "^V ' ." trt ^v^ < *.^\» -5 < ^<:^v^ — ac ^v^ ~5 z> V — t- ■? 1 1 i 1 1 1 1 < 2 111/ fin) SSVWOIB jm/fiKi SSVWOIB CO to m e o •H c •H > ^3 O Cn I > O tn 122 Litter Tiie litterfall collected iu each ot the plots may be divided into 2 categories: autochthonous litter (litter produced by vegetation in the plot) , and allochthonous litter (litter produced by vegetation outside the plot). The total amount of litterfall (autochthonous ♦ allochthonous) is of interest in the study of the nutrient cycliag processes of the system. For calculating net ptinary productivity and vegetation turnover rates in each eiper itaental ecosystem, autochthonous litterfall is the appropriate measure. In this study, allochthonous litter accounted for 20-31% of the total litterfall (natural succession, 20X; enriched succession, 21%; mimic of succession, 26X; monoculture, 3^%) . This suggests that litter from the older, taller secondary forest surrounding the plots may be important as a source of nutrient inputs. The high allochthonous litter values partly reflect the small size ot the experimental plots. Each replication had an area of approxiffliately 0.12 ha and was surrounded by older secondary forest on all sides. In areas where large-scale clearing of tropical forests for cultivation has occurred, nutrient inputs from allochthonous litter would be much lower. All litterfall comparisons among treatments are based on autochtnonous litter (see Appendix B for table ot means and siguificaut differences by date). The vegetation in the 123 experimental ecosystems began to produce measurable amounts of litter at age 12 w<. From 12 vk to 84 vk, tiiere were several differences in 2-wk litterfall aaounts among the four treatments, with the monoculture in general producing less litter than the other treatments. Although mean monthly litterfall in the natural succession (36 + 10 q/m^) , enriched succession (35 ♦ 14), mimic (26 ♦; 11) and monoculture (2 1 + 11) did net differ statistically, the tendency was for the monoculture to produce less litter (Fig. 26) . Low litterfall values were expected in the maize monoculture, because dead maize leaves commonly remain attached to the plant until they decompose. Mean litterfall in the cassava monoculture (31 * 15 g a-^ mo-») was not significantly different from mean litterfall in the natural succession, enriched succession, or mimic. At five dates (4 July 1979, 11 September 1979, 8 April 1980, 6 tlay 1980, and 3 June 1980), analysis of variance detected significant differences among replications. On each of these dates, litterfall was lo:*er in replications where grasses were very common than in replications dominated by dicots. Litterfall fluctuated over time in all treatments (Fig. 27). In the natural succession, enriched succession, and mimic, litterfall was lower during May and June of 1980 than during other months. These low values coincide with the onset of the rainy season, and may reflect a seasonal pulse 124 40 - 30 - e 20 - < cr UJ 0 - 0 SUCCESSION ENRICHED MIMIC OF MONOCULTURE SUCCESSION SUCCESSION Figure 26. Monthly litterfall means in succession, enriched succession, mimic of succession, and monoculture, Confidence intervals are + 1 s.e. 125 I I I I I I I ' i>) mvjMSiin (»«j . jiu/ 6) niwM3iin C 00 0 a> -H M tfi en J-) cu 0) O Xi o 0 p 4-1 en V o T3 (U 1 x: O (^ -rH r- >-J CTi C nH 0) rH ^ ■H c M 0 Ci •H <; m en >. CD (U o u O 3 d -P to iH 3 iH C) (C 0 M c d o +j e (0 C 'Ti c C to •H ^ rH c rH o (« -H iw en S-l en 0) 0) +J c; +J C) -H d 1-1 en (if«2 jiu/6) nnvjaaiin /6) iiTja3iin 126 of leaf productioa aad rapid vegetative growth, when leaf losses to litter were low. ^bove-Ground Productivity Thirty-day means of net primary productivity, above-ground living biomass, herbivory, litterfall, and production of standing dead bicmass were calculated using the rate equations described in Chapter II, •Productivity Measure me uts' . The curves estimated from monthly mean values and the original data points from field measureiBents are presented it Fi- PRODUCTION OF STANDING DEAD BIOMASS M i M Figure 29. Net primary productivity (NPP) , above-ground living biomass, herbivory, litterfall, and production of standing dead biomass in enriched succession. Triangles are data from field measurements; black dots are estimated monthly means. 129 /^ ABOVE-GROUND LIVING BiOMASS PRODUCTION OF STANDING DEAD BIOMASS 'igure 30. Net primary productivity (NPP) , above-ground living biomass, herbivory, litterfall, and production of standing dead biomass in mimic of succession. Triangles are data from field measurements; black dots are estimated monthly means . 130 MONOCULTURE .MAIZE 2 ^,^ / / • '.^L. ABOVE-GROUND LIVING BIOMASS .V LITTERFALL PRODUCTION OF STANDING DEAD BIOMASS a S 0 N D 1979 1980 Figure 31. Net primary productivity (NPP) , above-ground living biomass, herbivory, litterfall, and production of standing dead biomass in the monoculture. Triangles are data from field m.easurements; black dots are estimated monthly means . 131 calculations. The negative net primary productivity in the natural succession and enriched succession treatments during November 1979-Jaauary 1980 probably reflects the dieback of some of the early successional dominants, including Phytolacca ri vino ides. Second, negative net primary productivity values may indicate periods of time during which plant respiration was higher than photosynthesis. Negative net primary productivity in the natural succession and enriched succession near the end of the dry season (April-aay 1980) may be due to water stress and high plant respiration during the dry season as well as dieback of some important species. A third factor that would lead to negative net primary productivity estimates is translocation of photosynthate from above-ground to below-ground plant parts. Such translocation may have occurred during the dry season, but it was not measured. In the mimic, seasonal fluctuations in net primary productivity were dissimilar to those in the natural succession and enriched succession. The differences probably reflect different life cycle characteristics of the dominant species in each system. Yearly net primary productivity rates (Table 12) were highest in the enriched succession (2396 g m~2 yr-i) and in the monoculture (2267) , and lower in the natural succession (1777) and the mimic (1148). Yearly productivity rates in the natural succession, enriched succession, and mimic were 132 c 0 •H +J (C iH 3 e 3 o 0 (0 (0 en (rt a o -H J3 'd c (0 ^ tn (U w CO 0 H CO m m e 0 ■H • X3 u >1 •> \ X^J +J £ ■H \ > en -H -P 0) o ^ 3 m T! 0 m H 0) 0. ^ iH Tl (H C > D 0 J-i • tr tfi 1 g <\) (]> > +J o tfi X! >i fO [fl o ■P C) QJ Q) C iH r-l 3 (C n 13 m c c a < •H TJ fi c 0 ;3 •H n cn 4-> u w (rt -p CO (« iH 0) 1 e d s QJ o > -H ^ O m u -Q C) <; < -p en 0) > fd > 0 > •H J3 M Q) C tn 0 C tn -H H 03 +J TJ T5 (T3 U C n3 Fi 13 (0 Q) 0 Til 4-> P •H n cn m ^-1 CM iH ^^ (U -P -H 'C >i c +J 3 •H 0 > M -H -P o -P QJ 1 C) ?^ 0) 13 > T3I 0 o Xi M < cu e CI) +j en > en 0 u H fi fi fi OJ o O o M •H •H •H 3 en T! tn ^H tn -P .H eo CD en o en iH (C5 CD ^ 0) CD 3 S-l U CJ u o c; C) 0 CJ •H o ■H C) o 4J 3 u 13 e 3 fi m en c en •H en o iz; H S s 133 Buch higher than y«arly losses to litterfall, standing dead, and herbivory. These systems were cot in steady state, tut were accuaula ting biomass at a rapid rate. In the monoculture, long-term biomass accumulation did not occur because of crop harvests. In all four ecosystems, losses to litterfall and standing dead were higher than losses to herbivory. In the natural succession, for example, 40% of the net primary production was cycled through litterfall and standing dead, and only 3% through herbivory; tae remaining 573! went into biomass accumulation. Herbivores consumed <^% of the total above-ground net primary production in the monoculture, 3% in the natural succession, 3% in the enriched succession, and 5% in the mimic. Herbivor'^s consumed 3% of the total leaf production in the monoculture, 1% in the enriched succession, 9% in the natural succession, and 12!ii in the miaiic. Effects of Decreased Hertivory Bates of Herbivory in Insecticide Plots To study the effects of reduced hertivory on community structure and function, insecticide was applied to a diverse system (the enriched succession) and a simple system (the monoculture). Herbivory rater were monitored on dominant species in the insecticide plots to determine whether or not the insecticide applications reduced damage rates. 134 Leaf area and herbivory rates in the enriched succession and monoculture treated with insecticide are suBKaarized iu Table 13 and Table IU. Insecticide applications to the enriched succession reduced the per- leaf-area herbivory rate by H3% in October 1979, 65f in February 1980, and 65% in June 1980 (mean reduction for all dates = 58%) . Herbivory rates on species common to the insecticide plots and plots not treated with insecticide were compared using non-parametric statistical techniques (Wilcoxon 2-safflple rank sums test, Kruskal-Wallis test, and median test). Because the data did not meet the homogeneity of variance assumption of these tests, the levels of significance reported are not exact (Pratt 1961). Herbivory rates on the four species monitored in both systems (Phytolacca riviagides, Clibadium atr. surinamense, Panicuffi maximum, and Erythrina costa ricensis) were significantly different in the enriched succession with and without insecticide (Table 15). Rates were lower with than without insecticide for each species. The herbivory rates were not signif icautly different ia the oaize monocultures with and without insecticide. The rate in the cassava monoculture treated with insecticide was 56% lower than the rate in the cassava monoculture not treated with insecticide (Table 15). 135 Table 13. Leaf area index, leaf specific mass, and losses to herbivores in the enriched succession and monoculture treated with insecticide. Ecosystem Date Species LAI (m2 Leaf/ m^/Ground) Percent of Total LAI Clibadium aff. 0.86 20.8 surinamense Hymenachne amplexicaulis 0.81 19.5 Phytolacca rivinoides 0.55 13.4 Borreria laevis 0.44 10.7 Gramineae^ 0.28 6.7 Canavalia sp . 0.25 6.0 Vigna sp . 0.08 1.3 Solanum torvum 0.08 0.7 Others 0.86 20.9 Enriched succession Oct. 79 Ecosystem^ 4.22 100 Feb. 80 Clibadium aff 1.14 surinamense Gramineae^ 0 31 Phytolacca 0 28 rivinoides Panicum maximum 0 .25 Solanum 0 .19 jamaicense Vigna sp . 0 .11 Hyptis vilis 0 .08 36.6 9.8 8.9 8.0 6.2 3.6 2.7 Table 13 --extended . 136 Leaf Specific -Mass (g/m^) cm^/m^ Leaf/ Herbivory Rate cm-^/m^ Ground/ Da^ Da^ g/m^ Ground/ Day 52.4 7.6 0.040 38.6 7.6 30.6 16.5 18.5 10.9 24.4 0.1 37.2 0.7 43.3 0.04 34.0 7.4 36. 9^ 8.5C 6.2 9.2 4.8 0.03 0.2 <0.01 0.6 7.3 0.024 0.028 0 .009 <0.001 0.001 <0.001 0.002 0.027 36.9 8.5 35.9 0.131 52.4 4.9 5.6 0 .029 24.4 30.5 1.2 3.9 0.4 1.1 0.001 0 .003 45.8 39.5 6.9 6.2 1.7 1.2 0.008 0 .005 43.3 23.3 0.8 6.5 0.1 0.5 <0 .001 0,001 137 Table 13 — continued, LAI (m2 Leaf/ Percent of Ecosystem Date Species m /Ground) Total LAI Enriched Feb, 80 Hymenachne O.OS 2.7 succession amplexicaulis Iresine diffusa 0.08 2.7 Others 0.59 18.8 Ecosystem^ 3.11 100 June 80 Clibadium aff. surinamense Hymenachne amplexicaulis Gramineae^ Panicum maximum Canavalia sp. Borreria laevis Erythrina cos taricensis Solanum jamaicense 0.11 Others 0.92 26.3 0.72 20.6 0.31 8.7 0.31 8.7 0.25 7.1 0.17 4.8 0.14 4.0 0.11 3.2 0.57 16.3 Ecosystem^' 3.50 100 Monoculture Oct. 79 Zea mays 1.40 100 Feb. 80 Manihot esculenta 1.02 100 June 80 Manihot esculenta 2.76 100 ?able 13 — extended. 138 Leaf Specific Mass (g/m^) cm'^/m^ Leaf/ Day Herbivory Rate cm^/m^ Ground Day g/in'^ Ground/ Day 38.6 18.0 1.4 0.006 35.6 42. 6^ 21.4 5.4c 1.7 3.2 0.006 0.014 42.6 5.4 16.9 0.073 52.4 6.9^ 38. S 6.1 24.4 2.7 45.8 0.9 37.2 13.2 18.5 16.2 37.5 1.5 39.5 4.1 40.9^ 6.3c 6.3 4.4 0.8 0.3 3.3 2.8 0.2 0.5 3.6 0.033 0.017 0.002 0.001 0.012 0.005 0.001 0.002 0.015 40.9 6.3 22.2 0.088 53.1 45.4 45.4 9.8 5.2 2.3 13.7 5.3 6.3 0.073 0.024 0 .029 Includes at least six species of grasses that were indistinguish- able by vegetative parts. 139 Table 13--continued. ^Ecosystem values are totals (LAI, percent of total LAI, losses in cm^/m2 ground/day, losses in g/m^ ground/day) , and species' means weighted by LAI (leaf specific mass, losses in cm^/m2 leaf/day) . ^Mean of species values weighted by LAI . '^Mean of Oct. 79 and Feb. 80 rates. 140 Table 14. Mean herbivory losses by species, in plots with anci without insecticide treatment. Losses are x (s.d.), in cm2/iTi2 leaf/day; n is number of leaves (alternate-leaved species) , or number of leaf pairs (opposite-leaved species) . Species Date Natural Succession n loss Enriched Succession n loss Phytolacca rivinoides Oct. 79 37 16.5 (32.7) 21 14.4 (20.0) Feb. 80 33 12.7 (24.4) 34 27.1 (46.1) Clibadium aff. surmamense Oct. 79 6 13.7 (10.4) 13 17.9 (13.8) Feb, 80 17 16.0 (32.8) 14 9.2 (10.6) Panicum maximum Feb. 80 5 6.8 (8.8) 8 12.8 (13.4) June 80 9 13.0 (25.5) 9 15.5 (11.8) Gramineae^ Feb. 80 9 29.7 (80.4) 6 7.6 (16.9) June 80 11 36,4 (50.5) 9 13.5 (19.6) Hymenachne amplexicaulis Feb. 80 5 2.3 (2.3) June 80 13 11.5 (12.0) Erythrina costaricensis June 80 24 14.8 (20.2) Borreria Oct. 79 laevis June 8 0 141 Table 14 — extended, Enriched Succession Mimic of Treated V7ith Treated with Succession Monoculture Insecticide Insecticide loss n loss n loss n loss n loss 15 16.5 (28.6) 18 3.9 (5.4) 11 8.8 (11.4) 9 4.9 (4.8) 3 6.9 (8.2) 7 0.9 (0.6) 9 1.2 (2.4) 8 2.7 (4.1) 4 18.0 (9.5) 7 6.1 (11.1) 22 57.5 15 1.5 (40.8) ( 1.2) 2 10.9 (5.7) 12 16.2 (34.0) 142 Table 14 — continued. Natural Enriched Succession Succession Species Date n loss n loss Canavalia sp. Oct. 79 June 80 Solanum torvum Oct. 79 Vigna sp . Oct. 79 Feb. 79 Solanum Feb. 80 jamaicense June 80 Hyptis vilis Feb. 80 Iresine diffusa Feb. 80 Zea mays Oct. 79 Manihot esculenta Feb. 80 17 4.0 (4.0) June 80 143 Table 14 — extended. Enriched Mimic of Succession Monoculture n loss n loss Succession Monoculture Treated with Treated with Ins ecticide Insecticide n loss n loss 10 0.7 (1.1) 10 13.2 (26.1) 10 7.4 (8.7) 6 0.04 (0.1) 16 0.8 (2.6) 12 6.2 (5.7) 1 4.1 12 6.5 (4.6) 10 21.4 (51.6) 14 6.2 5 9.8 (7.0) (9.1) 24 35.7 68 11.5 58 5.2 (64.0) (29.4) (16.1) 1 1.7 4 9.1 3 2.3 ( — -) (5.4) (0.3) ^Includes at least six species of grasses that were indis- tinguishable by vegetative parts. 144 4-> 3 to rH rH 0 0) o o Xi +J LO • "sT • 00 +> nj iH V r-^ V o •H -d to • • ^ Q) >. - iH 3 - rH r-i "■ T! H iH CN O IT) O •-\ C (0 to O O o m > • V V • 6 ^ 0 0- - " " >■ " 4J U CM ,-i in <-^ •-i •H 14-1 O o o o o ^ • • • • tn V V tfl OJ • 0) 4-> .H (D Ti 0 aid ■H c ■H g rtJ u to 0^ o a> CO IX) • & rtS e •H •H • • • • • ■P m -P _ Ti n rH o o CN tn C (U U >i Q) Q) •H W £ 0) rtJ g -P x> •• T! C en Tj IX • • • • -p • 0) — ' -H e S-l tn a a; ro 0 E-i c o ID r-i O fO •H en 0) o (NJ r-\ 1^ rH ro H MH iH rH to C (D to 0) 0 m T! 0 iS^ rH 0 -H tj c 1 tn u -H to CTi m cr\ i~- t^ rH CD (D J-l -H u •H • • • • to 4-) +J (D -P •rH T! CO (N •-i r~- in ^ •H (C ^ U -P ,_^ 0) CN tn tn i-i e 0) u >1 ^ ;3 0 13 tn 0) to Sh CL, >i2 C tn 'O « Cu U -H C \ 0 0 H tncM ,— , ^ y-^ — . --^ - > . -p cn e tN rH V£! rH VD 4-> UH •rH -P :3 J2 0 \ ,— ^ • • • • • tn 0 Xi C 0 4-) JtN • (Tl 00 00 in (^ Q) u Q) x: •H e T3 rH ro CN r-^ n 4J Sh 0) 6 -P s o • ^-' — •—^ ~-^ ■— ' Q) £ -P -rH — tn tn Xi (C & -c ■ — >X) ro n •^ o g g C (D \ 0) • • • • :=! 3 fO s^ ^ -p IX •^ in r-{ ro 00 tn C -H -P -P to iH n rH rH ,-{ -a -H tl) A^ tn ^ U ^ W -H 3 >( 0 c cn c >1 ■rH •H 0 to nH cn ^ U to tu to x: Sh s ^ Eh u w s cu cu to Xi 145 Species Composition At 18 mo the enriched succession treated with insecticide had 81 plant species on 248 m^, as compared to 159 species on 1536 ffl2 in the enriched succession not treated with insecticide. Because the insecticide plots were smaller than the other plots, species abundance and diversity comparisons of the systems with and without insecticide were difficult. To make valid comparisons, 25 subplots were randomly selected from all subplots where species composition was monitored in the enriched succession without insecticide. Each of the subplots was 12 m^ (3, IxU m strips of vesjetation) , equal in area and shape to the area monitored in the enriched succession with insecticide. Plant species diversity, evenness, and turnover in the enriched succession treated with insecticide were then coapared to mean values from the 25 subplots in the enriched succession not treated with insecticide. Species richness was hiyher in the enriched succession treated with insecticide (23 species intersected by 36 LAI measurements) than in the enriched succession not treated with insecticide (16 species) at 3 mo, but at 18 mo the values were oot different (Table 16) . This suggests that in the earliest stages of succession the reduction of herbivory allowed a wider variety of species to survive and compete, but increased species richness was not maintained over long periods through application of insecticide. 146 Table 16. Changes in number of species, diversity, and evenness in enriched succession and enriched succession treated with insecticide. Vegetation Without With , Characteristic Age (mo) Insecticide Insecticide Number of leaves intersected by 36 LAI measurements Number of species intersected by 36 LAI measurements Number of species intersected both at 3 mo and 18 mo Number of species 18 (12-29) 11 gained from 3 mo to 18 mo N\imber of species 8 (6-13) 12 lost from 3 mo to 18 mo Species diversity e Evenness Community similarity (C) 0.51 (0.30-0.68) 0.54 between age 3 mo 3 18 161 182 (135-178) (113-258) 149 193 3 18 16 26 (11-21) (19-37) 23^ 22 8 (5-10) 11^ 3 0.87 (0.60-1.05) 1.06 18 1.03 (0.81-1.20) 1.11 3 0.71 (0.54-0.80) 0.78 18 0.73 (0.61-0.85) 0.83 and age 18 mo^ ^Each value is the mean of measurements in 25 subplots randomly selected from the enriched succession without insecticide. Each subplot had area 12 m-^ . Ranges are given in parentheses. h 2 Each value is based on measurements in 12 m of ecosystem. "^Value is outside the 95% confidence interval for the mean in enriched succession without insecticide. '^H' = - E (n./N) log(n./N) , where n. is the number of leaf inter- sections for species i, and N is total number of leaves inter- sected (Shannon index) . ^Evenness = H'/S, where H' is Shannon diversity index and S is the number of species. 147 Table 16 — continued. ■^C = a(l) + a(2) +...+ a(i) +...+ a(n), where i is a species present at 3 mo and/or 18 mo, a(i) is the lesser percent LAI value for species i from the two dates, and n is the total number of species. 148 Seduction o£ lierbivory fcvored sooe species that would not have been able to cotapete successfully under higher herbivore pressure. For example, six species (Solanum toryuiB, Gouania lupuloides, Solanum jamaicense, Vigna sp., Canavalia sp. , and I^ojaoea sp. ) that were abundant (accounted individually for >2% of ecosystem LAI) in the enriched succession with insecticide were not abundant in the enriched succession without insecticide. Although species diversity and evenness were not different in plots with and without insecticide, species composition was different at both 3 and 18 mo. At 3 mo, four of the eight abundant species in the insecticide plots were not abundant (although present) in the plots without insecticide treatment (Table 17). At 18 mo the differences were more striking; eight of the 13 abundant species in the insecticide plots were not abundant in the enriched succession without insecticide. In a complete species inventory at 18 mo, 11 species present in the enriched succession treated with insecticide (248 m^ area) were not found in the enriched succession without insecticide (1536 m2 area) . These species included Iresine celosia. Cola nitida, ln3§ thibaudiana, Mcllinedia cost ar icens is , Heliocarpus sp., X£2S2:§5 sp., and five unidentified species. The community similarity index, C, was used to compare rates of species turnover in the enriched succession with and without insecticide from 3 to 1 8 mo. Species turnover 149 ^ ■p •H S: Ti Ti OJ C en ro -H -P Oj 3 g 0 0 Xi C) +J •H u> 5 -H C CJ o 0) -H a, m w U) 0) (fl u 0 +J 3 m to Si 4-> n OJ en -C (U u +J -H m >-J u c •H 0 T! c C -H •H H 1 <; ■— ' l-l s: m w 0 ro -0 dP CN < A U , 0 +J • m C H cu ri: Cng J c -p •H m e 4J Q) 0) c !-l +j 13 +) 01 0 >1 u Q) tn 0 TS 0 (C •H c; U CD w •H (U +J IW -H CJ 0 u (U 0) tn o\o Oa C cvi CO -H V Q) TS •H U ^ -H +J -P •H O g d) < T3 4J -H d U 0 -H £ -P -P U •H QJ IS tn c H II • • I I I • . I ... I I fvioo cocNro crirHo II ••II • • • I I I I • • • I I I I I iH iH CM • • • • I I I I I • • • . I I I I I I I I cTinoooouDrsioomcM 1 I I I I I I I I I I I (0 ■P •H O c >-( > 5^ M Ho Oi-H cn QJ 0) C 0) o g o H cn •H o x; u • -H cn +J dJ tn c o •H (0 -Q g QJ (0 • rO -r-i cn U g >i O U CO tn tn +j 4-> u >H to (0 a a a; (U > > •iH •H -P JJ (0 (0 -p -P tu CD m ni cu OJ > > >, >i jn Si tn to •H -H T1 -rt C fi -H •H OJ OJ u M Q) 0) 5 ? -P +J (0 ttJ 4:; X! -P +1 cn CO (U Q) cn tn cn Ti (C5 QJ ^1 cn cn MH m 0 0 cn tn Q) OJ •H •H 0 0 QJ < cn < UJ N LlI N c (ti £ -p =1 u o c • 0 vn U •H 153 insecticide were not offset by high leaf productivity as at other times of the year. The enriched succession treated with insecticide had a larger concentration of leaf tissue near the top of tiie canopy than did the enriched succession without insecticide (Fig. 34). This may be because terminal buds and young leaves were protected froio uerbivory in the insecticide plots. k dry season decrease in leaf tissue near the ground occurred in the plots without insecticide (at 13 mo), but not in the plots with insecticide. Mean canopy heights in the enriched succession with and without insecticide treatment were not significantly different at any time during the study (Table 18) . 154 WITHOUT INSECTICIDE 8 MONTHS (5.0) LA I ( m /m ground ) WITH INSECTICIDE 18 MONTHS (5.4) Figure 34 7/////////J 3 1 \ \ 1 .6 0 .2 .4 .6 0 .2 .4 .6 LAI ( m^ / m^ ground ) Vertical distribution of leaf area in the en- riched succession with and without insecticide, Total LAI at each age is in parentheses. 155 Table 18. Mean canopy height in enriched succession with and without insecticide treatment. Values are x + s.d, n=12 with insecticide, n=60 without insecticide. Mean Canopy Height (m) Vegetation Age Enriched Succession Enriched Succession (mo) Without Insecticide With Insecticide 3 1.6 + 0.5 1.9 + 0.4 7 2.6 + 1.0 2.5 + 0.9 12 3.5 + 1.1 3.4 + 0.9 18 3.7 + 1.2 3.7 + 1.2 156 Aboye-Ground Biomass Above-qcound biomass developed at approxiiaately the sane rate in tLe enriched succession and the enriched succession treated with insecticide duriny most of the study (Fig. 35) . At 1.5 yr, however, leaf biomass and total above-ground bioaass were higher in the enriched succession without insecticide than in the enriched succession with insecticide (see Appendix B for tables of biomass means) . Total above-ground living bromass (leaves + stems + reproductive parts) in the enriched succession treated with insecticide was approximately 1.4 kq/m^ at 1.5 yr, less than half the bioaass in the enriched succession without insecticide (Fig. 36) . Thus herbivores did not limit the rate of bioaass accumulation in the diverse successional system. This may reflect the differences in species composition between the enriched succession and the enriched succession treated with insecticide. Some of the species competitively favored by the insecticide treatment were species that accumulated biomass at a slower rate than did the dominant species in the enriched succe:;sion without insecticide. Above-ground biomass was similar in the monoculture plots with and without insecticide treatment for the first maize crop (Fig. 37 - 38) . Maize yields from the first planting with insecticide (mean + 1 s.d. = 2985 ♦ 940 kg/ha, ear fresh weight) and without insecticide (4570 ♦ 1606) did not differ significantly. The second planting of maize grew 157 Figure 35. Total above-ground biomass in enriched succession with and without insecticide, Values are x + 1 s.e. 158 4.0 3.5 3.0 2.5 ^ 2.0 ui ! .5 < g ™ 1.0 0.5 ENRICHED SUCCESSION 2.0 1.0 _ ENRICHED SUCCESSION - PLUS INSECTICIDE < ■5. 2 0.5 A S 1979 1980 LEAVES □ STEMS H FLOWERS a FRUITS Figure 36. Above-ground living biomass by vegetation compartment in enriched succession with and without insecticide. 159 2.0 -I .8 - 1.6 - 1.4 - to 1.2 - .0 - cr 0.8 H (5 0.6-1 0.4- 0.2- -o- INSECTICIDE -•— NO INSECTICIDE liiXJd n — V--T — ^ — ^ — I — I — \ — r MAMJ J ASO N DJ FMAMJ J AS 0 1979 1980 Figure 37. Total above-ground biomass in the monoculture with and without insecticide. Values are X + 1 s. e. 160 2.0 r- MONOCULTURE CASSAVA HARVEST 2.0 1.5 1.0 - MAIZE HARVEST o 0.5 MONOCULTURE PLUS INSECTICIDE 1979 1980 'LEAVES □ STEMS M FLOWERS a FRUITS Figure 38. Above-ground living biomass by vegetation compartment in the monoculture with and without insecticide. 161 very pooLly in the plots without insecticide. Total above-qround biomass was higher in insecticide plots from 7 wk after planting until maize harvest at 17 wk. Yield from the second maize planting was significantly higher with insecticide treatment (4390 + 1918 kg/ha) than without insecticide treatment (1945 + 921 kg/ha). Above-ground herbivore damage to the maize was low. The growth difference in the second maize with and without insecticide was due primarily to differences in below-ground herbivory. Thf^ first crop of maize grew equally well with and without insecticide, possibly because the experimental plots were in a newly cleared area that had not been recently cultivated. Because the plots were >1 km from the nearest cultivated maize, perhaps some agronomic soil pests did not 'find' the experimental maize and build up large populations until the second planting. Bioaass production in the cassava was approxinately the same with and without insecticide application. At two dates (16 and 32 wk after planting) , cassava bioaass was significantly lower in plots with insecticide than in plots without insecticides (Fig. 37). Shading effects are pronounced in cassava, and the lower biomass in the cassava with insecticide may reflect delayed development due to partial shading by vegetation around the plots. At the time of harvest (42 wk) , no significant differences in biomass due to insecticide application were detected. Cassava yield 162 {kg/ha, tuber fresh weigut, mean ♦ 1 s.d.) did not differ significantly between plots with insecticide (8338 j^ 3679) and without insecticide (10,883 + 26«*2) . ititterf ail Althoug^i there were no statistically significant difterences in mean monthly autochthonous litterfall between plots with and without insecticide, several trends were apparent in the data (Pig- 39) . Insecticide application affected litterfall rates in the maize monocultures, but not in the cassava monoculture or the enriched succession. Higner maize bioaass in the insecticide plots during growth of the second maize crop was accompanied by increased litter production. In addition, dead maize leaves remained attached to the plants and resulted in more standing dead biomass in insecticide plots. Mean monthly litterfall in the enriched succession and cassava monoculture, both with and wit.iout insecticide, ranged from 29 to 35 g/m^, and there were no significant differences due to insecticide application. Seasonal litterfall trends were similar in plots treated with insecticide and plots not treated with insecticide (Figs. UO and Ul). Lowest values occurred during May-June 1980, at the beginning of the rainy season. At this time of year, much dead plant material had already been shed during the dry season, and new plant growth was beginning. 163 O ID LlI o C ) III n h- Ci u Ijj 1- C/) • ■ ' o ouj/^ai/b) livjd311in 164 40 30 - ENRICHED SUCCESSION -? 20-1 — I '-I- r 1 \ \ \ 1 \ \ \ \ \ \ \ \ \ \ \ 1 MAMJ JASOND'jFMAMJJASO 1979 1980 ENRICHED SUCCESSION TREATED WITH INSECTICIDE 30 -] J i -I „■ 20 - < 10 .M*^ i V^' 1 \ I \ \ I I \ \ \ \ I \ ] I MAMJ J ASONDIjFMAMJ JASO 1979 1980 Figure 40. Litterfall in enriched succession with and without insecticide, April 1979 - October 1980 165 * 30 - >CORDIA 1980 MONOCULTURE TREATED WITH INSECTICIDE CASSAVA E 20 - ul 10 - Figure 41. Litterfall in the monoculture with and without insecticide, April 1979 - October 1980. 166 Above-Grouod Productivity In both the enriched succession and the monoculture, annual net primary productivity rates were higher in plots not treated with insecticide than in plots treated with insecticide (Table 19) . This result suggests a possible stimulation of productivity due to herbivory, and is consistent with the results oi the artificial defoliation study and the work of others (e.g., Detling et al. 1979). In the enriched succossion without insecticide, higher net priiaary productivity was accompanied by higher turnover rates (littertall, herbivory, and production of standing dead biomass) . In the monoculture, higher net primary productivity in the plots without insecticide was not associated with higher turnover rates on an annual basis (Table 19) or for individual cultivars (Table 20) . Net primary productivity was higher in plots not treated with insecticide than in plots treated with insecticide for the first maize planting and the cassava. For the second maize planting, net primary productivity was tiigher in plots that received insecticide (Table 20). Although above-ground insect damage on maize was not lower in the insecticide plots than in the plots without insecticide, root biomass data (C . M. Berish, unpublished data) and herbivorous nematode data indicate that below-ground herbivory was reduced substantially in the insecticide plots. The productivity differences in the 167 G O -H -P nJ r-l g O o CO ui • m M e > o \ 13 tP (0 0) u - (d D) 0) m tn 3 O rH > (0 (0 lO • g cu O "d •H -H J3 O •H ^-p >i u -P Q) ■H to > c -P O -P o .c: Oi-H •a c -o 3 c 0 td u 1 +J 43 (0 (0 e 0) 0) CO c >. to r-l O Tl C! c 0 3 ■H o 03 +J M tn (rt ■p CO n1 rH 0) 1 e 13 s 0) 0 & > -H 13 0 « U ^ O < < +J tn > >■ u 0 > •H X! U (U c &1 0 c to ■H H tn -P Ti T! (ti C) C CO fi 3 fC 1 C -PI 3 •H 0 > i-l ■H +J CJ -P a) 1 O ?^ (U 3 > TJI 0 0 ^ M «: c^ E (i> -P tn >i tn 0 O w CM CTi o r- CN cx> iH r-i CN (X) i 0) (rt -P T3 fO M o M \ C1)(N > e O \ C !Ji >-i ■ — 13 +J 0) -P d C iH in fC rH > a 13 . C Q) n3 T3 •H >i U ■P ■H •H +J > O •H 0) « -P tn iH O c (U 3 •H T3 T! 0 0 -P a H 3 Ch 0 >i ^ -P T1 4J •H C ■H > 3 15 ■H 0 -P i-i 'O C3 en C 3 1 03 T! CD 0 > £1 i-l 0 ■P Ci ^ ■H (C S: 0 >i cn 1^ rH CD M-l x: ^1 +J 3 -n c 4-1 OJ 0* O -P c m rH ^^ D^ •P E-i o EH iH o > -rH u 1 4J ■rH > ■H -p u o CM ro in ro in O rH CN ID in en O iH O (N in (N in 00 O 00 CN o o in ro m CN o ,_ ijl en c C Q) -H (U OJ -rH T3 -P T3 T! ■P -H C -H -H a o 0) fO O 0) O tfi +J C tn H-> en +J 0 w C u o c u C U o >H -H CD U -H QJ -rH 0) 0 •H en 0) tn en c MH 4-> c; en -P C ■P C o — 3 -H >- 3 -H ra D -H s 0 0 > O OJ X ^ 0) x: x; n3 x:: £ N -P 4J N -P -P en -P -P -rH -H -H •H -H -H en -H -H ro IS S (C 3: & ro S 12 s s U 169 maize with ariu without insecticide ace probably due to the below-ground herbivory differences. Application of insecticide to plots may affect many factors related to ecosystem function in addition to reducing herbivorous insect numbers and damage rates. Soil microorganism populations, decomposition rates, nutrient cycling rates, and insect predator and parasite populations are probably also affected by insecticide additions. Thus the results of the insecticide treatment on net primary productivity cannot be attributed entirely to reduction of herbivore pressure. Changes in other ecosystem processes due to insecticide application were not measured. The net primary productivity curve for the enriched succession treated with insecticide (Fig. 42) was similar to the curve for the enriched succession without insecticide (Fig. 29), with high and low values occurring during approximately the same time periods in both ecosystems. WaxiBum production of standing dead biomass occurred at different times in the two ecosystems. Net primary productivity curves for the monoculture with and without insecticide differed in several ways (Figs. J1 and 43) . Net primary productivity rate increased more rapidly during the first few weeks of growth of the first maize crop in insecticide plots than in plots without insecticide, but the maximum daily rate was higher in the plots without insecticide (45 g m~2 day~*) than in plots 170 ENRICHED SUCCESSION TREATED WITH INSECTICIDE ^ NPP ABOVE-GROUND LIVING BIOMASS PRODUCTION OF STANDING DEAD BIOMASS Figure 42. Net primary productivity (NPP), above-ground living biomass, herbivory, litterfall, and production of standing dead biomass in enriched succession treated with insecticide. Triangles are data from field measurements; black dots are estimated monthly means. 171 MONOCULTURE TREATED WITH INSECTICIDE -M4IZE I ^-(J MAIZE 2 S-*^ CASSAVA 20^ -^ ABOVE-GROUND LIVING BIOMASS /• PRODUCTION OF STANDING /* DEAD BIOMASS Figure 43. Net primary productivity (NPP) , above-ground living biomass, herbivory, litterfall, and production of standing dead biomass in the monoculture treated with insecticide. Triangles are data from field measurements; black dots are estimated monthly means. 172 with iasecticide (25 g m-2 day-^) . Productivity of the second maize crop was high in plots with iasecticide and low in plots without insecticide. The period o£ low productivity in the cassava monoculture treated with insecticide may be the eftect of partial shading of the cassava by plants surrounding the plots. Cassava is very sensitive to shade, and since the insecticide plots were smaller than the plots without insecticide (82.5 n^ instead of 256 a^) , shading by surrounding vegetation was oore pronounced in the insecticide plots. Responses to Artificial Defoliation Besults of Preliminary Study Leaf regrowth was extremely rapid in both the enriched succession and the maize monoculture after removal of approximately 50* of the leaf area of each ecosystem (Fig. 44) . Although between-plot variability was high and no significant differences between defoliated and non-defoliated plots were tound in this pilot study, the observed trends were interesting. In the enriched succession, leaf biomass increased rapidly in the defoliated plots at a time when leaf biomass in non-defoliated plots was decreasing. Similarly, leaf biomass did not decrease in the defoliated maize monoculture as in the non-defoliated maize. Only 6 wk after leaf removal, mean leaf biomass in the defoliated maize was almost equal to that in the 173 720- 600- 480- 360- " 240- 120- .d -I r- 20 40 TIME (WEEKS) T" 60 30- - 20- o^r- is 0) 0 !-i e CD CU 1-1 CU i-i 0) x: 3 -p U) w U-l ■iH o -U Cfi U-J c frt (0 QJ CU rH E ^-^ CU n u to en -U tn C CU 3 =! n r-\ F= trt <: > m 4-) O (T3 C CD nH U (C U -P CD O (Hi Eh + 1 +1 c o ■H Cfi R en 0) CU 4-1 V en u >i 3 en en o u 1—1 w (0 M 13 -P rO S + 1 I -^1 -^1 +\ +1 176 _ o CO CO LxJ O O Z) CO _l Q < LlI 1— cr < 3 _J 1- O < LlI ^L O _ o _ o _ O _ O C • ns in -p ■H II c c o •> •H • tn 0) en • 0) w u O iH 3 , to +1 (0 X > C • •H C O (0 -H H ro to -H iH 4-1 O (0 *4-l <1> 0) ivn 177 < if) cn < ro •H o -H m •H ■p u ^ o to 01 -p 0 o ■p CD -H _ o fd • r^ +J (N •H IS II O (D 0) c u _ O — 3 • 1 in o u m 0 •D C H H+l O LlJ 'T ^ OJIX 1- > tfi !-i cn (0 _ O m ro c :3 _ O X > CVJ c • ■W C 0 _ O fC -H 0) 4-1 rC -H iw 0 - o (1) cu IV1 178 The LAI in the defoliated natural succession plots was 15* less than that in the control plots after the first def oliation-reqroHth period (p<.05), equal to that of the control plots after the second def olia tion-regrowth period, and 11X less than that of control plots after the third defoliation-re-growth period (p<.05). It appears that the LAI in both the defoliated cassava and the defoliated succession would have reached the LAI levels in non-defoliated plots after each defoliation if the reqrowth periods nad been longer. Extensions of the growth curves after tie third defoliation (Fiys. 45 and 46) indicate that the LAI in the defoliated natural succession would equal that of tue non-defoliated natural succession after 24 days of regrowth, while the LAI ol the defoliated cassava would require 35 days of regrowth. Time lioitatious did not allow defoliation experiments with different defoliation intensities and Jlifferent regrowth periods to be performed. Tiie high defoliation level (50") and short regrowth periods (i mc) ».'€rL chosen to ;iiiDulate extreme levels of stress to the ecosystems. The series of defoliations began at the end of the dry season (mid-April 1980) . The onset of the rains approximately 15 days after the first defoliation was accompanied by increases in LAI in the non-defoliated (control) plots. Thus the LAI of the defoliated systems had to increase to levels greater than their pre-def oliation levels to 'catch up' to the non-defoliated systems. 179 Net rates of leaf production (increiaents in leaf area) were used to compare defoliated and non-defoliated ecosystems. The non-defoliated systems «ere growing, but the defoliated systems were growing at a faster rate. For both the cassava monoculture and the natural succession, increment in leaf area was higher in defoliated plots than in plots that ware not defoliated (Fig. 47). Mean LAI increment in the defoliated natural succession was 0.091 m^ m-2 ground day~i, as compared to 0.032 a^ m~' ground day-* in control plots. Leaf growth rates in the cassava were lower. However, defoliated cassava had higher rates (0.054 ffl2 m-2 ground day-*) than non-defoliated cassava (0.016 m^ ffl-2 ground day-*). The data show that defoliation stimulated leaf production in both the high and low diversity systems. However, the amount of stimulation of leaf productivity differed in the two systems. To compare the amount of stimulation of leaf production in cassava and natural succession, the percent differences in leaf productivity rates between defoliated and non-defoliated plots were calculated (percent difference = 100(x-y)/y, where x = change in LAI in defoliated plots and y = change in LAI in non-defoliated plots). In the casf-iva monoculture, leaf productivity was stimulated to more than five times its normal rate after the first defoliation, but the amount of stimulation was <200* after the srronc^ and third defoliations (Fig- 48) . The trend was 180 to o UJ (— u < (O Q CO ir < T (_) 1— Ava a3d ivi (0 (0 S-l -H -P o m m c 0) c ■H iH < CJ UJ CO Q) -H 3 U m -H •H -H 2 IW (0 O 2 m CO o (D 4J CO K rH 3 LU < 0 -1 _J o 0 -P CO li- LlI C 5 o 0 •H T3 ^ (— -P c i Q > m tn 0 (T _j (U u X UJ m G x; o foliation. V to defoliati ations of eac ? CD Q) U o 13 13 -H O z o o UJ < t3 tH X! W (1) J Q) M e >.-P +J (T3 Q) CO UJ ■<-\ U Q) o leaf productiv af production ooled from thr z 0) d. o U-! nH 1- G (C »- 5 c; -p ro C -H fC fr -J O Ti o -H U) u. u. 4-1 0) H UJ 0! W . (2) The estimated time necessary for complete recovery after the third defoliation was less in the diverse system (24 days) than in the monoculture (35 days) . (3) The diverse system, but not the monoculture, continued to respond vigorously after a series of three defoliations. Percent recovery after three def olia'iion-l mo regrowth periods was higher in the diverse system (89%) than iu the monoculture (72*) . Chanaes in loietation structure. The vertical distribution of leaves in the canopy changed after defoliation in both the natural succession and the cassava monoculture. Leaf tissue was removed equally from all layers in the canopy at each defoliation. Leaf regrowth after defoliation was not distributed evenly throughout the canopy in the natural succession or in the cassava monoculture. The natural succession was characterized by increased amounts of leaf tissue near the ground (0-1 a) 183 after successive detoiidtions iFigs. 49-51), while most of the ceqroHth in the cdssdva occurred at the tops of the plants (2-3 m above the ground) . Increases in canopy height were depressed by defoliation. The height of the defoliated cassava increased 0.75 m during the scuay; the height of the defoliated succession increased 0.50 m. These increases were less tnaa those in the non-defoliated cassava and succession during the same period (1 ID increase in each). To deternine whetner the structural changes were due to the defoliation treatment, the changes in the defoliated plo^T vcrr cor-pared to the changes that occurred in the non-defoliated plots during the same period (Fig. 52) . In the succession plots that were not defoliated, LAI increased throngl.c'it the canopy during the 3 mo study period, with the greatest increases from 0.25 m to 1.75 m above the ground. In the defoliated succession plots there was an increase in leaf area near the ground (0-1 a) and a decrase in leaf area higher in the canopy. The most striking difference between the defoliated and nou-detolia ted plots was the greater increase in leaf area at 0-0.25 m above the ground in the defoliated plots. This probably occurred because the defoliations opened up the canopy, more light reached the ground, and seedlings survived that would have died under the shade of the full canopy. 184 i , K [5^ O m in uj J > in CM 0 +J tC Si +J o u en 0; >-l (0 0) (n)lH9I3H (M)lH9iaH N0ISS300nS VAVSSVO < > O I, O tr UUJ u 185 (N) IHSIBH NOISS30DnS O u u 14-1 CD d en •H &4 186 tf> > o u o < o: lO > o s e o z o < o UJ n 6 § o z 8 o < o UJ o: -1 UJ < t- 6 o u. o> < X (- o o q: 8 IT) z o Ife r» ^ UJ m (0 (0 i2 c 0 tn -H -H tn en Qj Q) H U 3 U !T> :3 -H en iM rH x: (0 u S-i 03 s w -p re c • >1 c -a •H d +J Q) tn o c c (d 0 •Xi -H • C -P tn 3 (0 +J XI -H C (0 iH Q) 0 g tn 4-1 0) 0) (D u •H rO 3 o tn 0 i H X! CO +J m C +J 0) c s Q) 0) e V( Q) 3 )-l w :3 fO tn a; m S H < H-1 • • c >i in 0 T3 in -H 3 CN tn ■p tn tn C 0) •H 0 c u 0 to 3 -H c en 4-1 0 m •H iH -H -P (C rH u M 0 0) 3 M-l w +J 0) u (0 T3 Q) C +J n C TS E •H Q) +J n m m n3 •H (1) QJ rH ^ r-{ 0 ■p H-i •H 0) tn •H cn tn OJ M -H a; U X! QJ g a, d •P C o ■H iw ■H •p -P +J 4J c c C C (C (0 (0 fO u o 0 o •H o ■H H •H -H U-l rH m O U-l iW •H • •H • ■H -rH C V C V c c Cn U* cn 04 Cn cn ■H -H •H -H tn tn tn tn -P ■P -P +J 0 0 O O z Z S 2 cu ■p c u in -H O 14-4 • -H V C Dj cn "a- rH O fN iH 'S' O rHVDrHOinrHUJn CN O H in rH criOrHiX)OinOrHCX)I^r-fM V£> OJ 00 in rH n CN OOOr^fNOrHOOOOVDOOtNin'^rHrOrH^rHVD^rH'^fNIVO'il'in rH VC (N rr n OrHtNooroinorHovDOOOrHro rH O rH CN r-i (N OrHCMcT^'a'rHnOrHOOVDfN r^ m CN in CN ro r^oorHoooooroorHrHinroooocNr^'^vDO'^i^rotncNr-- 00 00 rH CN in rH e QJ (8 :3 tn •rl n C rH CO tn o . ^ -H C (d rH td x; CD tn T! •H c fd td o D CU C c >l (d x: to 4J ■H fd tn 0) tn (d tn Q) 3 tn •H ■H 4-> cu ■H >< 0) -p O o •H U ■H cn tn (d •H M M C -a > 0) g tn <-{ tn in c cn c c n tn ■H e •H e ^-^ Ti •H Q) • td JJ O rH . . . 0 g 0 a tn 3 OJ tn J3 to (d I4H 0 g g 04 > +J g 3 04 04 04rH fd •H tn m ,0 u (0 c e c c (d -H >H fd T3 tn -H 04 cn tn tn >i cn cu . (L) e tn rH m f-{ ■H ^H •H (d rH QJ G (D tn 04 fd 3. a , "O cu •H cu Cu •H p ■rl . .l •H •H Q) tn •H ■H O rH ■H ■H r-^ fd cu c c G o u na Id fd D. cu x: tn c M x: o 0. ;3 TJ td fd C fd fd -p N fd ■H C 0 0 0 o fd cu cu cu CO >^ Cu c 0 CU 4J •H 0 ^ (d -H -H o Sh > x; 4-) x; c ■H o CJ u u c 0 0 rH o Q) o U fd Tl u +J X! 'd nS +J cu fd o C 4-> (d g -H ■H H cu cu fd g g m rH ■O u u <-i Sh o cn ■H M M 0 a, c cu fd cu d fd rH rH rH 4-1 cn 0 0 o rH "H 0 0 (d ro Q> i fd u Sh 0 0 M CU o 0) cu >i c 04 04 < < m m m u u U u U u U u u u w Pm o u O ffi ffi ffi ffi X H H H ^OrHCNnc3>inU3'^rH en 4-1 T3 CU en 3 0) C ■rH -H u +J 0) c 04 0 CO u 1 1 m CN m 0) M Q) •H 0) r-\ U X! Xi 0) e fO a, =3 p. en z mr-im'^oo~iroLnoovx)OOor~-r~ ooiHOLn'XJtNfNnofNooJo rHiHiHninfNjooo'^oooocoocNc'^orNinnoo.HooiHOi— I I— I n iH rH oor^oa^':rocx30'<5f rsi CM (N rH . O to 0) EH o x: o o H 'H n3 fd u 0) CU-H 0) u en (U :3 DJfd g -n -P O U o 0. B p B e 3 :3 P C c c fd (0 fd fd fa iH H iH aj o o 0 u en en en H fd +J fd •H • x: en fd s-i en C (U -P fd ClJ-H X eu > iH CN ro ■^ LD IJ3 o, Oj Ol Dj 04 Dj en en en tn en c n3 'O QJ CD •H -rH -rH -H -p -p (1) 0) t! tJ 'O 0) (1) Q) •rH -rH -rH y-H IW 4-1 •H -H -H -P +J -P C d C CD 0) CD 'd 'T3 'O ■H -H -H -H -rH C C G D D D 194 QJ 1 O C tCU 0 rH H Q) iw > ■rH Q) C hJ Cn H cn rH -tJ (rt aj C -p •rH (C Cm H .H 0 IH Q) tH Q to •H -P •U n •H s C H rH m c 15 •rH C) fa 4J (C •rH r-{ 0 rH ^-\ (rt Q) •H D HJ •H C H 0) 6 (C 2 03 0) •rH u 0) a U) m 0) i-i -H CD U XI QJ h CU 3 U2 Zl •^C^r-iOr-iOCDr-tOt-ir-i OOrHOOrHOOOOO OrHOOrHOOOOOO OOOtNOOrHOrHoO OrHCNi^-^inusr- r--00(TirHrHrHrHrHrHrHrH CUC1^C1^C1|0^C1^C1|Q^Q^CU Ch m Cfi cn tfl cn cn Cf) cn cn UJ UJ ^3 "n ^ t; T3 T! ^3 ^a T3 TJ Tl aj 0) Q) (1) (D 01 (1) +J 4-1 4-1 4-1 +J 4-) 4-1 C C c C c C c c c C C fl) 0) (1) 0) OJ 0) d) aj (1) 0) 4-> 0 T! cu QJ C QJ > > 4-) UH •rl •H ^3 cn o 4-1 +J •H 13 to to tJ •r-i (D +J +J 'd cn QJ QJ +J to 3 IT' tT> c 10 Q) QJ QJ g u > > g 0 QJ QJ U Xi >i >i Sh MH X3 Xi :3 Ti cn QJ QJ QJ QJ lO M g rH rH Q) to U JQ -Q e 0 lO to cn H4 X x: t-H QJ M cn cn (0 •H QJ •H •rl c 4J a, S :3 •H •rl cn cn 1+4 rH +J C C -H 0 •H •H 0 X! C 4J +J 4-> to cn cn X ••-' •rl •H rH 0 cn TS 'd to Sh QJ c C •H eu •P •rl •rl +J •rl QJ QJ QJ C • +J U jh •H cn (0 QJ QJ +J CJ 5 ■? g 0 -H 0 i-l rd 4J 4J SH & c to fO MH ■H X -C na 4-) +J cn QJ cn C ■P M cn cn 0 fO C QJ QJ •rl •H (0 cn cn 4-) rH rH TS cn 0 0 m QJ fO QJ 144 cn u cn QJ cn u -d • MH 0) 1 g 0 MH 4J C 0 0 d 0 T! cn •rl C QJ QJ cn QJ •rl 0) IW d !H u •rl 0 C m 0) u to & QJ cn 14H cn Cu U ■d 0 cn QJ QJ !h S 4-) QJ 3 X to QJ 0 •rl 3 •H U m cn c rH cn 0 QJ 4-) 4-) c: U TJ • n3 • £: u cn :d cn 3 cn E-t 0) QJ rH rH +J rH 4J 4H +J rH U Sh O U .. <4H to c to C to O^H(~M g to cT TJ X LO 195 non-defoliated plots. The two species that increased to hiqh levels of dominance during the 3 mo period were the saoe in the defoliated and non-defoliated plots {Bocconia frutescens and Clifaadium aff . sur inamense) . Most species whose LAI increased in the control plots also increased in LAI in the defoliated plots (Table 24) . Similarly, most species whose LAI decreased in the control plots also decreased in LAI in the defoliated plots. However, five species increased in LAI in the non-defoliated plots and decreased in LAI in defoliated plots: Panicum trichoides, Cyperaceue group, Lasiacis procerrima (misideutified, but referred to in this study, as Hymenachne amplexicaulis) , Incia edulis, and Frantzia pittieri. The LAI of three species (Borreria laevis, Canavalia sp., and Heterocondylos yltalbis) decreased in control plots and increased in defoliated! plots. This group of eight species that showed opposite trends included comiaon species (Pduicum, Hymenachne, and Frantzia) and relatively uncommon species (the other five species) , and jointly accounted for approximately 2U% of total LAI in defoliated and non-defoliated plots. Three of the above differences in growth patterns were significantly different by chi square tests (Hyaienachne , p<.05; Frantzia, p<.01; Panicum, p<.10). For the other five species, either the differences were not significant or sample size was too small to perform the test. 196 Table 24. Numbers of species that increased and decreased in LAI during the 3 mo defoliation study in de- foliated and non-defoliated natural succession. A "+" indicates an increase in LAI; a "-" indicates a decrease in LAI . Non-Defoliated Succession Defoliated succession Total 31 36 11 Total 34 13 47 197 In addition to species that differed in magnitude and direction of change in LAI in defoliated and non-defoliated plots, three species differed significantly in magnitude but not in direction of change. The LAI of Phytolacca rlyinoideg decreased more in defoliated than in non-defoliated plots, and the LAI of Ipomoea neei and Cordia ipermis increased less in defoliated than in non-defoliated plots (Table 2J). Because the changes in LAI of individual species differed in direction and magnitude, the relative percent dominance of the species in the ecosysteia also changed. A species can increase in percent dominance in the system because the species itself increased in leaf area, or because the other species decreased in leaf area relative to it. The changes in percent LAI of all species in the defoliated succession are ranked in Fig. 54. A few species showed large increases in percent LAI, a few showed large decreases, and many species changed little in percent LAI during the defoliation experiment. The percent changes in LAI that occurred in the defoliated plots were dissimilar to the cnanges that occurred in the nox»-def oliated plots during the same tine period (Fig. 55) . The comaunity similarity index C was used as a measure of the overall change in plant species dominance during the 3 mo defoliation study. Similarity values were egual in defoliated and non-defoliated plots (C=0.71), indicating 198 UJ Sz po uJb m<- is to to UJ UJ*-* ^ o tea: o u Q.UJ UJ £L tn"- Q. to nilir cn rT< 1— ^Q ^ to n z UJ o w< z _i tn 3 cn x: u i-H m (0 w >H 13 -P • (0 CO c -d 0 c -H ■H ^^ (D H Cu < ^ x: +J •p S c O OJ M u cn s-l (U 0) s-l & 1 c c 0 ■H •H +J U) m o ■H CTlrH c 0 (ti m x: OJ U t3 IV"I 7o "IVIIINI - ivn % IVNId 199 (0 dJ •H o -i rH 0. 0 QJ y-i s-l 0) • '0 u ro (Tl CM c ja ■H CU rH rH tn r3 ja QJ o rc ■H •H H U +J cu u C a< 0) -H en > tn c x: M o u cu iTi X! g W fc 0 3 u c 50S of total annual NPP went into development of permanent structure (Fig. S6) . Ihe rate of biomass accumulation during early succession was accelerated by propagule enrichment. Above-ground biomass of the enriched succession was higher than that of che natural succession at 1.5 yr, and this was due in part to the biomass of an introduced species, Musa paradisiaca. The lower biomass in the mimic of succession may reflect a time lag in development of the mimic. 20fi NATURAL SUCCESSION NPP » 1777 q/m^/yr ENRICHED SUCCESSION NPP = 2396 g /m^/yr -HER8IV0RY (60) •^HERBIVORY (72) MIMIC NPP =1148 q /m^/yr MONOCULTURE NPP » 2267 g /m^/yr HERBIVORY (56) HERBIVORY (15) T^iqure '^^. '^ionass accretion anH turnover in natural succession, enriched succession, min.ic of succession, and monoculture. 207 Estimates of piaot bionass in early tropical succession are more numerous than estimates of productivity. Above-ground biciaass estimates in the natural succession and enriched succession were high compared to literature values from other sites (Table 26) . The Turrialba site was on relatively nutrient-rich volcanic soil. Also, nutrient availability was high after the initial slash-and-burn site preparation (Ewel et al. 1981) . Above-ground biomass in the natural succession at 12 mo (1522 g/mZ) was higher than the above-ground biomass at 12 mo (1113 g/m^) reported by Harcombe (1973) on the same site. These differences may reflect effects of the burn (Harcoabe's plots were not burned) and year-to-year rainfall differences at the site. Continuous Diomass accumulation is a key characteristic that distinguished the three diverse ecosystems from the monoculture. la the monoculture, living oiomass was cut at each harv.ist. The soil, left without a protective vegetative cover, was vulnerable to erosion and nutrient leacning. in the diverse successional systems, the permanent structure of the systems buffered microenviLonmental fluctuations and allowed the development of complex biological interactions that may enhance sustainability. 208 u a, o u •4-) c o M Cn I > O tn T? tn C fO 3 g 0 0'- iJ -HfM C cc g 1 \ OJ Cn en > C O -rH XI > . Q) c» O a> C i-i (1) Ch >i 0 ^ o ro a, Ch r- r- u a^ LD CT\ Ti 3 rH VD rH c ?; IJ\ <-^ (0 U rH 0) r~- ^3 Q) X3 cn m C M w F= rH fO (0 m (0 O tn T! en u rH 1-1 rH d) Sh u (1) 0) £ C 0) (d :^ H D cn H K w >1 u s c (d c 0 0 •H -rH u tn CO m OJ (U o U u U 3 ..^ 3 en c 0 0 m 0 (0 o (0 CO o rd fd to ■^3 (d MH •H fS fO fO fO o •H o o •H U 0 rH o 0) u 0 cn Id rH rH r-^ rH -H O::; ■H ■H Pi •H ■H (d -H x; ■H (n rH r3 0) to (C Pi Pi Pi Pi Pi Jh Pi u Pi U 0) Q> e :3 E e 0 o 13 •H ■H U 3 CD N CD 0) m 4-> rd fd -p td (d 4J fd M fd e 0 N ■p Q) +J 4-) +J >H 4-) ■4-) u -P •4-1 fd 4-) C 4-) -H d 0) (C c Id (0 in (D en en Q) m to c CO OJ tn e en C =) (U ^ 3 o 13 o 0 0 0 0 •— 0 0 0) c; ^ o o u cu u u cu u u u u > 209 en T3 en C fC 3 g 0 0-- u m e 1 \ > c^ 0 -H ^ > ft! (t! -- 3 4-1 C C q -H-- iH r^ iH fC 5 (1) (U 4-1 rH f= -p Q) r- M o a) (T> 0) >H >i iH Ai o >i QJ nJ ^ 0) rH iH Ti 4J iH <-\ 0) (1) M iH O 5 C nj O O w cn m u O &i d 0 (0 u r« i-f •rH fC c X! (T3 Fi rC (rt e != 30T. of annual plant turnover in the three diverse systems. Excluding the standing dead biomass component from the productivity calculations would reduce the estimates of NPP by >15% in the diverse systems. Litterfall rates have also been used to estimate gross nutrient cycling rates in mature tropical forests (Golley 1975). Ihis method fails to account for standing dead biomass. Standing dead matter decomposes above the ground, without contact with soil, roots or mycorrhizae. If the decomposition rate of standing dead biomass differs from the decomposition rate of organic material on the ground, significant amounts of standing dead biomass may have 214 important implications in ecosystem nutrient cycling processes. Herbivory Low Herlivory Bates In all four experimental ecosystems, herbivory was a less important pathway for cycling of organic material than were litterfall and plant mortality. Only 0.7 to 5% of the annual NPP was consumed by insects. Estimates of consumption by herbivores in other tropical and temperate ecosystems range fjLum 0.i% of NPP in slash-and-burn agriculture in Venezuela to 38. JX of NEP in a short grass area of the Serengeti (Table 28) . The highest rates reported were in grasslands with large herbivores (Andrews et al. i97U, Sinclair 1975) and temperate old fields (Odum et al. 1962, Van Hook 1971, Boring et al. 1981). The lowest rates were in a Liriodendron forest ic Tennessee (Reichle et al. 1973), a tropical palm savannah (Lamotte 1975), and tropical slash-and-burn agriculture (Uhl and Murphy 1981). Methods for estimating herbivory rates and productivity rates varied widely among studies. This accounts for some of the differences in percent consumption reported in the literature. Herbivory rates are often difficult to estimate precisely because herbivory is extremely variable both temporally and spatially (Janzen 1981). 215 T3 fi (C ^ 0} 0) , ^ IS) 0 e > 0) •rH +J XI m S-l >, dJ tfl £ o C) >i g; 43 OJ c -p 0 (C -H s-l -p a; a, a e E 3 (D m 4-) c 0 "U u c to ^ ,—, rH Oi (C PL, u s -H ^-^ a o >1 ^ -p -p -H > iH •H m -p u CJ 0) p > t:! 0) C) Cfi >-i a c •H >^ U T! (Ti Q) e e ■H 3 U tn a c 0 +j o QJ c a< :i^ -o S c; 3 0 0 s-l XX 0) < a o -p C D^ 0) Cu U 2 5-1 0) Ph c O 03 '- -H OJ J-l -P !-i H Cu O \ O CU u d: 13 c 0 s-l ^1 > 1 aCNJ > \. O Cn| 43 1^ < O H cn CO CO S w w U Eh U CO ID >H CO CO o >1 45 S-l 13 C n3 Cfi o; s-l ^ O (T! M-l M -p P (U N c s-l 0) >l> Cfi 0) o u (fi -p OJ s-l ^ O (0 U-l -H 0 •p 3 (1) N c s-l OJ >1> •p CO c o ■H CO CO Q) u u 13 to -p CO 0) s-l o — M-l tC o -P -H QJ Ci M -P >- Cfi o LD CJ c o [n OJ u o (0 Q) 4:: — O (0 •H U ^-1 -H C K 0) (C S-l -P >, Cfi O LD U O (0 ■H U •H a e (D S-l -P >i Cfi 0 LD U 0) o u to 4-1 to 0) s-l o — 4-1 n3 CJ -P -H 0) CC IT3 s-l +J >-i CO O in u M-l o C O •H (fl to cu o u CO Ti QJ 4: — U to ■H U U -H C « Q) to S-l +J >i to 0 un (J 216 4-1 U 2 C S-l Oh O t/) ^^ ■H OJ )-l -P U >i C S-i Cn O Q) U E 13 C O c a. \ O 2 g > \ O !Ji < O H cn in W pq U Eh U Cfi D >^ C < w u H o U (0 •H U e-H -H Pi E (0 i-l -P >! tn O in o Eh D S u w H Eh O >H 1 CT\ TJ iH 3 +J C to 0 +J (0 c •H fO x; r-\ tH m c ■H -P c N > (M U I ^- I >1 (« cu ^ u i-l -H 4-1 CD tH 5 to 3 O 4-1 O -H O c o u O U-i -^ 0) i-i o y-l ^ 0) Sh 4-) iH U o c o e 4-1 (0 tn n3 > (U 4J to > tfi m u o to (0 u n3 £ -- U a O OJ 14-1 > 0) Sh -Q to u u P 0) 4-1 4-1 tH 4-1 3 to U '— O HD to ecu O (0 -H to tn to > C -P to -H en m u o en 13 u to 'T! — u >1 M 3 s C to X! X! >i 4-1 iH C O U QJ CO 217 CO CM d VD m CTi > u-) rH w H r^ U-) r^ a^ r- • ■ri o^ nH CTi iH c rH iH fO (0 S-l ^ •H cu 4-1 +j n (0 = r -p (D S-l n iH 4-) Q) r- m O 0 e Cnvo C e =1 cu en c ■r-\ fO T! •H tH (fl CO ij o s > 0 0) in ■i-> E • c a. d IX> tX) ro rr r- (N •^ [^ 0) Oi w 1 • 1 . • • . o 2 c o r- 00 ^ o n >-< CTi u 0 • (N rO rH r-i 0) u n ex c U) 0 0 'u •H S-l >1 4-) 0\ a >CM 6 >i •H E ID ro VX) in o >.D -a* r-\ in ,-{ =1 Xi x: \ in (N iH 1J3 00 C30 <-{ •^ ro in S-J Cn rH rH c 0) 0 E u Ti c 3 ,-, 0 S-l S-l >i o 1 0^ \ 00 O 00 \ CM 0 Cn X ^-^ H CO 4J en 0) s-l 4-1 "o o •H Pi 0 4J S-l 4J in (T3 0 u t-:i O M CO to •H o Q) CQ tn OJ C c QJ O 0 in QJ ^ (0 Sh CO CO -H Eh U M-l 0) d c (0 cu 0 K S -d s w ^4 Oi re QJ U > U PJ in ^^ 0) 0 •~^ rH U (TJ M y Eh 'C! re TJ a c M-l Ifi rC ■ — CO rH c T3 C cc •H ^^ ,— , t3 in en (13 w >H Q) •H rH re D 4-) 0 0) 0 C rC in in QJ Xi CO •H rH 0) rH ^ C o c u (C Sh O in (0 Sh ca w o UH 0 •H in 0) ■H 0 -H rH iji ,~. to Sh re c Eh u Sh H-i in ^ aK 4-1 cr; T! rO Sh CTi c < W T! re re s-l in O •H ■H tn en rO K rH U ^3 Sh ij 0) 0 OJ 0 0 ■p C 4J 0 > H O •-t CP < en 4-) e 4-) s QJ in tn Sh -r-i rO Oi £ 0 C) 5-1 ■H s-l tJ^ N C 0 a en S Sh 4-> Sh M 4-) 0) rH 0) c C c 0 £ O w >( d Sh >i Ol in d p •H CU (C i-q CO fc^ e E-: 0 >i o •rH Oi >ieu MH Sh Eh ,-{ r^ C/^ in Pi 0 ^^ u ^— rH CU ~— re 1 •~-' O r ^ ^ Q W CO CU T-\ n »* 218 14-1 O TJ 0) 0) D^ w use i-* o OJ u c C W '— ■H (D i-i -p i^ >1 a o \ e > >fN p X! -H e tfi X! \ O 0) 1 P-c\i > \ O CTi < vo • CTl iH iH fO ^ 4J CJ (U ■H c CP -H c x: -H c M (U 0 s m iH 00 rr CN in Ln O O CN ■rj ■<:r 00 r- >1 -< 'D fC ,-^ e o ij m T! Q) QJ < E -p (0 <-i iH 2 K iH S-l -P +J c ^ tC 0 -P -P M C/-J U) r-i (D (0 0'. >H 0 O V crj CO cn fC u Q) K s; +j o J-< r-i >. >i u w (0 u o -P £ ^ +J u t- QJ w iH (U •P tfl O Uj ,— , U ^^ M-i -H (B ^3 TD (D U1 >'■ to 0 03 w •H s^ O OJ (P ^ U. TD c U-l C p:^ C •H N N 0 w c C •H •H P S-l ro >• (0 (0 4-i t^ u n3 rH ^ I— 1 ^ (D s-l X3 s-l s-l , >, s-l (U pu oi u 'C •H u W fd CO N rH iH TJ tn 2 N ^^ tn t- (C •H fC (C -P -H C CO w 0) /: 03 CO x; ''^ c CT. s-l S-l x: > 0) OJ E-. Tj -p £ 0) -p Pd •rH S-l Cr> !T' tJi ffi TJ c CD ;3 u i-i t; -p 0 +J C -H OJ 0 c a o S-l o o p. s-l 0) s-l D (J K ■H (D Cfi an >- 3 s s (C c O s-l ir< (1) ^^ U5 — - K D. — - x:. ■H -^ M rH ^ Uj U) K^l 219 g 0) P- tn U 2- -p C Oi O tfl — • •H 0) i-l -p >-l > Cu OX E > >oi 13 X! -H E o q; C o S-l > X O D^ CT^ in rH "O r» c (yi r- • to H 1X1 iH "O CTi rd N -H C -H U Ai fC ^ -P -H Cfi V^ VD (U & c tH 0) 0^ 0) -H nH a; iH en Cfi N OJ (C C P TS 5 -p >i •H 1^ 0 ^3 -P (0 i-i -P M 0 -H }-( 0 OJ U 0 £ PQ m Cl< :s s CO s: K E-: CTj >-i -0 CO C CD O rc -P O 1— i ca w tn T3 I— 1 -p ^^ 0 w en (C 0 V ec CD •P c S -p Cj i-i Cfi -H cn E-! 0 QJ rH ,i< CD rH < iw i-l 0 to i-l to S O U 0 0 c -- cn — ^ ^-t fd M-. 0 Ai t. 3 0 U cn ^~. -H i-l ^ tn •H -a ^ T! CD cn 0 g (0 V^ 0 X 0 C C tn >H fc (C 0 -P 13 tC3 •H CD K 1 -P 5 >H 13 M a u 5 a !-i c T3 0 •r-i 0 1 O CD s d) o U 2 CJ 0. x 3 2: K U ^-' fO — CD • — • to tn ^ t^ < K Q o 1 0 o u •H i-l • 0 — iH in to I-- O CTi iH CD — rC -P U •H C7> to C M •H U cn c 3 -H CO -o (D >^ •P X! i-l CD 'C > CD C > 0 0 U rH a CD g iH CD CD 5 — ■P -~x; iH cn 00 -H a\ CD • d ^ tn CD >. > -P iH to iH -C CD CD iH cn cn CD\ to -H rH > E- to to O cn T3 « cn C to to t^ u m U rH y-i CD . o IP T UH --' • c -H c 0 CD tn 0 ■H MH tn •H 4-> pL, to 4J a >H u g >i Cr> 3 3 X! T3 tn iH 0 c t3 0 i-l o CD MH a u -H )H +j >. iH 0 0 1— 1 to +J 0 c g o i-l 0 g to 3 Uh tn cn cn CD CD c 13 73 cn 0 :3 3 CD -H 1—1 (H 3 tn o u rH ^^ c c to CD M M > > fC3 X! u 220 00 tn CT^ 0) I— 1 ^ •~^ 0 • > >. -P ■H 1— f U X! c (D J-l o ij' CD Q) jr. tn •H +J 3; 0 c -P n3 'V C Tj Cn tC o C 0 •H u 14-; +J 0) -M U-. w 3 IM to O •H 1 QJ Q) U-^ U^ 1— i ro p., X2 CD tC iH >- rH £ •H (0 (C -p Ti > 4-) (D (C <: N •H O^ >, S-l Ph X2 fO 2 e c e U-l 0 3 0 -H tfi -P • c Cb CD +J 0 e (U £ -H 13 p tr -p tn 1-H ■H J-l C (0 CD 0 0 > 15 D^ u 'C' CD IW 221 The hecbivory rate measured in the monoculture oay underestimate the true rate because of the sampling techniques used. Monocultures are characterized by pest outbreaks in which much damage is very localized and concentrated over a short time interval (Pimentel 1961b). Such outbreaks cay be missed completely by non-contiuuous monitoring of randomly selected plants. Field observation confirmed that above-ground damage rates in the maize monoculture were actually quite low. In the cassava monoculture, true herbivory rates were probably higher than the rates reported here. Leaf-cutter ants (Atta cephalotes) were the principal herbivore in the cassava monoculture. Because jVtta's foraging activity was intense and concentrated on a few plants, monitoring herbivory rates ou a small number of cassava plants missed most of the Atta damage. In a study of leaf -cutter ants (Atta cephalotes) at the same site, Blaaton (1982) found that the ants removed an average of 88 cm2 leaf m-2 ground day-* iu the cassava monoculture. Blanton's values, obtained by monitoring activity on leaf-cutter trails, are more than four times the values I obtained by measuring damage rates on randomly selected plants. In the diverse systems, individuals of all dominant plant species were tagged. The number of plants (and total leaf area) monitored for damage was much greater in the diverse 222 systems than iu the monocultute. Larger sample size reduced the underestimation proi)lem due to uon-random foraging by some herbivores, and the herbivory rates measured in the diverse systems are better estimates of true loss rates. Absolute Losses and Diversity Not Correlated Herbivory rate was calculated as an absolute amount of leaf tissue consumed per unit area of ecosystem per unit time and as a percent of total leaf area consumed per unit time. Absolute consumption rate was not related to ecosystem diversity (Fig. 57) . The three high diversity systems incurred approximately equal amounts of damage (54-61 cm^ m~2 ground day~i) ; the monoculture incurred less damage. If Blanton's (1982) herbivory rates for the cassava monoculture are used, absolute consumption rates in the monoculture were slightly greater than in the diverse systems. These data indicate that herbivores consumed approximately the same amount of leaf tissue per unit of ground area, regardless of system diversity. These data do not support the dogma that diverse systems receive less damage from herbivores than do simpLe systems. The timing of herbivory may make the damage more apparent in low diversity systems. In this study, herbivory was much more variable temporally in the less diverse systems. High concentrations of insect attack occurred during short time 223 I I r o 1 — r o O ^ Q UJ z o o > > u e cu ■p w >1 tn 0 u (D U-J 0 (0 0) M (0 -o C 3 0 i-( Cn U c -< 0 (0 > > •H X! M-l ^ 0 0) x; -p c C) OJ -p •H o 0) •H ti) iw tn IW in 0) 0 0 u u (Xop/punojS ji^/eiJuo) SSOI 22a intervals. In the diverse systems herbivory occurred at a fflore constant rate, so damage was loss noticeable. Percent Losses Correlated with LAI The mimic ecosystem had lower LAI than the other two diverse systems. Althouga absolute losses did not differ, percent losses to herbivores were higher in the mimic than in the enriched succession and natural succession (Fig. 58) . Using my data, peccent losses were low in the monoculture; if Blantoa's (1982) values are used, percent consumption in the monoculture was at least as great as in the mimic. Herbivory had most impact on the systems with least leaf area. It has been suggested that diversity per se is not the factor that controls herbivory in an ecosystem, and that herbivory patterns are better explained by examining ecosystem structural properties that influence insect behavior (Feeny 1974, van Emden and Williams 1974, Murdoch 1975, Boot 1975). Leaf area index is a structural property affecting herbivory that is often, but not always, correlated with system diversity. Diverse systems may maintain higher LAI than simple systems because many species witu different growth forms are able to utilize the available space in the ecosystem mote fully than can a single species (Trenbath 1974) . However, LAI is not always correlated witu diversity; jouocultures, as well as diverse systems, may have high LAI (Swel et al. 1982) . 225 0.5- 0.4- -1 0.3- n r- < CO g 0.2- -1 r- - - 0.1 - - -| r o _ - - MIMIC SUCCESSION ENRICHED MONOCULTURE Figure 58. Percent of total LAI lost to herbivores in four ecosystems. Values are x + 1 s.d.; n = 3 sampling dates. 226 In this study plant diversity and LAI were positively correlated. The systems without investigator -controlled diversity developed both hiyh diversity and hivjh LAI. In the systeas where diversity was investigator-controlled, LAI was lower (Table 29) . The aiqh diversity, high LAI systems had lower percent losses to herbivores than the low diversity, low LAI systems. Because the effects of diver^J-ty and LAI ware confounded, it was not possible to separate the single effects of these two factors. Although both host plant density and overall ecosystem LAI are recognized to be important factors that affect herbivory patterns (Piwentel 1951a, Boot 1975, Bach 1980, Rauscher 1981, Solomon 1981), few researchers have attempted to separate density effects from diversity effects (e.g., aauscher 1981, Bisch 1981, Bach 1980, Benedict 1982). Bach (1980) found that plant density had no effect on herbivory and that the difference la beetle abundance on cucumbers in monoculture and polyculture was a function of plant diversity. Pimentel (1961a) found more herbivores per unit leaf area in dispersed and sparse plantings of Brassica species, but more herbivores on a per unit ground area in dense plantings. Solomon (193 1) reported that horsenettle plants at low density had more moth larvae per plant than plants at hign density, but the numbers of larvae per unit ground area did not differ. Hisch (1981) reported equal numbers of herbivorous insects in monocultures and 227 1 0 u a >1 u trt g -H U a, ■p (U c -0 c (d >i M 0 > • •H (0 ^ g V^ OJ Q) +J x: to >1 -p tfi c 0 cu u u Q) M 0) M cu m -P b C _ 0) H E rt; •H h^ U -* — ■ CI) a X X Q) ^ c (0 ■H M-l ,_^ IC 0^ 0) PM rH 2 *— ' ^ >s >- +J 4J -H ■H tn > M H 0) -P > O •H P P -a -p ^-1 c o Q) > U -H S-J ^ QJ S-i a< cu x; •iH £ en "U 15 tr -H CU 0 •H w s i-:i ffi +> CO c (U 5 x: ^ ■H en CT Ti 5 •H •H CU 0 ffi K s •4 to E cu d x; 4:: ■rH en Cn ■o is -H •H 01 0 K ffi 2 i-:i C C 0 q 0 -H 0 •H (fi •H CO to CO to cu to cu u cu u 0 u u 13 0 3 CU to p to iH tn :3 -O >4-4 -p CU r-i 0 tH X n3 3 0 M 0 CJ •H 13 •H 0 k +J E c c (C -H 0 H 2 S s 228 diculturt's cf corn and sweet potato, but because plant density was higher in the dicuitures, the numbers of insects per unit leaf area were lower in the dicultures. Ewel et al. (1982) found that herbivore consuaptiou was a nearly constant proportion (2-10*) ot the leaf area present rather than a constant amount per unit ground area. Those data were based on amounts rather than rates of damage and are not directly comparable to the results of this study. The results of this and most other studies suggest that LAI affects herhivory rates. This may be due to physical interference with insect movement patterns and reduced apparency ot host plants in structurally complex systems (Boot 1973, Atsatt and O'Dowd 1976, Pioeutel 1977). To minimize the impact of herbivory on the ecosystem, maintenance of high LhI is an important design consideration for high diversity agroecosystems. Effects ot Plant Species Composition Diverse systems have certain characteristics (such as aicrohabitat coaiplexity, diversity of plant herbivore defenses and high LAI) that affect herbivocy patterns. However, two equally diverse systems containing different plant species may have very different herbivory tates. The particular species that ace found together in an ecosystem and their relative abundances have an iisportaut influence on herbivory patterns. 229 Ecosystem herbivory rates reflect characteristics of the coaponent species that comprise the system. Monocultures may have high or low herbivory rates, depeudiiig on characteristics of the single plant species in the system. For example, palatable species in monospecific stands have higher herbivory rates tnan unpalatable species in monospecific stands (Ewel, Brown and Q-jima, unpublished data) . Similarly, diverse systems may have different herbivory rates because of differences in plant species composition. For instance, the median species herbivory rate was much higher in the mimic of succession (23.5 cm^ u-z leaf day-*) than in the natural succession (12.9). On a species-by-spucies basis, the species in the mimic incurred higner herbivory rates than the species in the natural succession. This is partially due to the high palatabilitien ot many of the cultivars introduced into the mimic plots. In Lue natural succession, the enriched succession, and the mimic, very hijh herbivory rates occurred only on species with LAI <0.5, while very abundant species (LAI > 0.5) incurred lower than average rates. The same trend has been reported ir* other studies of herbivory in successional and agricultural systems, in which the more apparent, relatively abundant species in the ecosystem were the least consumed (Reader and Southwood 1981, Ewel et al. 1982). Low 230 hecbivory rates may partially expain the dominance of some species in the diverse ecosystems. The minic had only one species with LAI > 0.5 (Cyabopogon £it£§i]is) . 8ith respect to herbivory patterns, the paucity in the miinic ecosystem of abundant species with low herbivory rates was a aajor difference between the mimic and the natural succession. Such species help tc maintain high LAI in the ecosystem and reduce the apparency of more palatable species in the system. The data suggest that unpalatable species may be essential components of stable, complex agroecosystems. Tahvanainen and Soot's (1972) hypothesis that a plant species may have increased resistance to herbivore attack through association with other species is probably valid in naturally diverse ecosystems where the abundant species are the less consumed species. However, my data suggest that the degree of associational resistance gained by a species in the system is determined by the relative consumption rates of the species in the ecosystem. In some cases, the association may be negative instead of positive. A relatively unpalatable species may experience •associational susceptibility' to insects rather tnan 'associational resistance. • For example, cassava (a species with a low herbivory rate) incurred more damage from herbivores in the mimic system, surrounded by an array of heavily consumed species. 231 than in the enciched succesiou or in the monoculture. In the enriched succession, the cassava plants were surrounded by many little-consuned successional species; in the monoculture, each cassava plant was surrounded by little-consumed cassava plants. Similar results have been reported by others. Bach (1980) round more beetles on corn when grown in a polyculture with cucumbers (a heavily consumed species), than tfuen ■jrown in monoculture. Risch (1981) observed that numbers of beetles were lower in polycultures containing at least one non-host species, and higher in ^iolycultures containing all host species. In an unpublished study by Ewel, Brown and Ojima, palatable species were consumed less in a diverse successional ecosystem than when grown in a monoculture, but unpalatable species were consumed more in the diverse ecosystem than in monoculture. Another species (Erythrina costaricensis) was damaged less in the enriched succession than in the mimic. In addition to the different range of consumption rates for species in the mimic and in the enriched succession, tiie enriched succession was more f ioristically diverse and had higher LAI than the mimic. All of these factors may have influenced the herbivory rate on Erythrina. The data suggest that to build associational resistance rather than associational susceptibility into an agroecosystem, the plant species must be very carefully 2 32 selected. Species that are relatively unpalatablae to herlivores are important in providing associational protection to the herbivore-susceptible species in the system. Plant Hertiyore Defenses Diverse ecosystems ccntaiu both palatable and unpalatable species, with a wide range of cneaiical and physical herbivore defenses. In this study, mean species herbivory rates varied by more than two orders of magnitude in the natural succession (0.7 to 1J1.U cmz m-^ leaf day-*), enriched succession (0.6 to 77.9), and mimic of succession (0.5 to 103.7). Other investigators have reported herbivory rates that raaged widely among tropical pioneer and persistent species (Coley 1980) and among species in three subtropical and one warm temperate forest (Benedict 1976). Although herbivore defenses were not measured in this study, the wide range ot herbivory rates suggests that herbivore defenses varied aiaong successioaal species and among successional mimic species. The diversity of secondary compounds in successional herbaceous species is high (Feeny 1976) . Small amounts of toxic compounds, i.e. •qualitative' chemical defenses, are common in successional specie? (Feeny 197U) . Low herbivory rates on some species in the diverse ecosystems may have been due to the presence of chemical defenses, the presence of physical defenses, or 233 ' associational resistance'. Chemical and physical defenses are intrinsic properties of a plant species; 'associational resistance', the resistance of a species to herbivore attack due to characteristics of the species around it, is an ecosystem attribute (Tahvanaiuen and Foot 1972, Atsatt and O'Dowd 1376). Associated plants may function as insectary plants that aaintain predator and parasite populations; as insect repellaats with spines, toxins, or olfactory deterrents; or as attractant plants that serve as alternative prey for herbivores (Atsatt and O'Dowd 1976). Structui.al Complexity ^loristically diverse systems are generally more complex in structure than are f loristicaliy simple systems. The diversti successional systems in this study contained species with many different growth forms, including herbaceous dicots, grasses, erect woody plants, and climbing vines. Because of the variety of growth forms, the diverse systems had a more even vertical and horizontal distribution of leaf tissue than did the monoculture. Greater variety in plant physiognomy leads to a diversity of aicrohabitats in the ecosystem (Pimentel 1961a, Dempster 1969, Dempster and Coaker 1974, Smith 1976, Bach 1980), and these 'enemies' may keep herbivore populations at low levels (Root 1973). In addition, the structure of diverse vegetation may create physical barriers that affect insect Kovements and make host 234 plants natdec to tind (Boot 197J, Pauschcr 1981, Kisch 1981, SoiOBion 1 981) . Her bi V or X , Diversity and Energy Flow Many studies of herbivory have considered responses of single plants or species to increased or decreased herbivory (e.g., see review by Jameson 1963). This study ditfers in that entire ecosysteas were manipulated under field conditions. Herbivory was experimentally controlled by use of iniirfcticides and artificial defoliation, and responses to increased and decreased herbivory were monitored. Vegetation structure, species composition and net primary productivity were affected by changes in herbivory, and responses difftred in high and low diversity systems. Interpretation of the results must consider the design of the experiments. The insecticide experiment was a long-term study (1.5 yr); the defoliation experiment was a short-term study (3 mo). The application of insecticide affected all types of herbivory, including damage from stem borers, piercing insects and root herbivores; in the defoliation experimeat, only one type of herbivory, removal of leaf tissue, was simulated. Herbivory rates ou all species were increased or decreased nondif ferent ially, imitating the effects of generalist herbivore activity. The results have practical implications tor agricultural systems where insect pests are 235 qeneraLists. The results are not comparable to theoretical predictions based on the assumption that changes in herbivore intensity have difierent effects on palatable and unpaxataijxt specie^,. Energy Flew Model The model in Fig. 59 shows soae of the energy flows that affect the relationships among plant productivity, plant species richness, and herbivory in an ecosystem. This study involved manipulation of plant species richness in several successional ecosystems. Investigator control of propaguies and the size of the species pool that determined the richness of the ecosystem varied among the four ecosystems. In the natural succession, no manipulations were imposed; the species richness of this system was determined by naturally occurring seed inputs from outside the system, plus reproduction of plants in the system. In tae enriched succession, these two sources of propaguies were supplemented by artificial seeding of additional species. In the mimic system, artificial seeding of many species, outside seed sources, and reproduction in the system provided propaguies. However, in this system only the artificially seeded species were allowed to grow and reproduce; all other species were weeded out. The monoculture was seeded with a single species, and other species were removed by weeding. 236 >i Q) C W 237 How miqht these differences in availability of propaqules affect the primacy productivity of the systems? If the enerqy available vj the system is not used completely by the specie^a present, additional propagules provide a pool of species that may be able to utilize the enerqy more fully, because they have ditferent growth forms and growth cequi^'e meats. lierbivory stimulates primary productivity; however, the amount of stimulation is influenced by level of herbivory and plant diversity of the system (see Chapter IV, •Resilience of High and Low Diversity Ecosystems'). When plant biomass is lost to herbivores, the system may respond in at least two ways. First, plant growth may be altered by a viriety of physiological mechanisms (see Chapter I, •Impacts on species composition and diversity'), including stimulation of photosynthesis in residual leaf tissue. Second, changes in species composition may result from compensatory interactions among co-occurrinq species and from addition of new species to the system (when seed sources ate present). The process is one of adjustment in species and numbers, and the result is a new complement of species utilizing the energy available to the system. In the monoculture, plant species richness was tightly controlled; the fluctuations in species composition that contributed to the high resilience of the diverse system were not allowed. Although productivity of residual plants 23 8 was stimulated in the monccuiture as well as in the diverse systeai^, overall stimulation in the monoculture was less (see Chapter IV, •Resilience of High and Low Diversity Ecosystems'). Thus, productivity in the monoculture was primarily a function of the growth rate of the species and outside energy subsidies. Outside subsidies were low, but information that went into planting (e.g., propagule selection, spacing, timing) may be considered an energy subsidy. Fossil-fuel based subsidies are often very important in modern agriculture; tiiis flow is often so large that high net productivity of the agroecosystem is maiDtained regardless of other processes occurring in the system. Tae monoculture had overall NPP almost as great as the NPP of the most diverse system studied (the enriched succession) . From the model, one might predict that productivity of a monoculture without energy subsidies would be less than productivity of a diverse system because it lacks the compensatory mechanism provided by diversity. Three factors help to explain the high monoculture productivity. First, the species planted in monoculture (maize and cassava) had high growth rates and low losses to herbivores. Second, the monoculture was a subsidized system in the sense that the species were carefully selected, planted in rows, and maintained under conditions favorable for rapid growth 239 (i.e., kept tree of competition from othec species by weedinq) . Third, the monoculture, but not the other ecos/stems, was periodically harvested. Because the plants were harvested at maturity, the senescence staje of the life cycle (a period of low NPP) was bypassed. Although overall productivity of the monoculture was hiv^h, Ni"? varied widely amonq plautinq^. In the first maize planting, losses to herbivores were low; NPP was high. In the second maize planting, losses to soil herbivores (not measured in this study) were apparently quite high; NPP was low. The NPP was high, however, in the second maize monoculture treated with insecticide (i.e., when the system was subsidized) . .lany factors determine the size of the herbivore population, and the details of this are not shown in the model. fls a result of other processes (not shown) , herbivon populations fluctuate and the rate of herbivore consumption varys temporally. Although the data from this study did not indicate that diverse systems lost less biomass to herbivores than simple systems, the simple system showed mere temporal variability in losses. Higher variability botn in herbivory rates and in NPP in the monoculture are key characteristics that distinguish it from the diverse systems. One flow that is not shown in the model is a possible effect of herbivory on plant diversity. Some researchers 240 have leported reductions in species cichaess of marine orqanisms after predator reiBovai (Paine 1971) and reductions in plant diversity after herbivore exclusion (Harper 1969). Others have reported increases in plant diversity after applications of insecticide (Malone 1969, Shure 1971). The results from this study coucerniny changes in plant diversity due to herbivory are inconclusive. The experiments were not designed specifically to test the hypothesis that herbivory causes changes in diversity. Altliough the data from both the insecticide experiment (decreased herbivory) and the defoliation experiment (increased herbivory) suggest that plant diversity may decrease as herbivory increases, the indication is not strong. For example, in the defoliation study species richness increased in both defoliated and non-defoliated plots. However, species richness increased less in defoliated plots. It is unclear whether this result should be interpreted as a positive or a negative effect of high herbivory on diversity. Thus, the model diagrammatically summarizes important relationships (among plant production, herbivory, seed sources, and energy subsidies) that are suggested by the data. Although it does not demonstrate the complex interactions between plant species richness and herbivory, it is consistent with findings on ecosystem resilience, the topic of tiie next section. 241 Besilience of High and Low Diversity Ecosystems It has been proposed that stability has two couponent parts: resistance, or the lack of fluctuations after a perturbation, and resilience, the ability to return to an equilibrium point after perturbation. This definition of stability implicitly assumes that a system is fluctuating around a single equilibrium point. Hollinq (1973) approached the problem of stability by recoqnizinq that more than one equilibrium point may exist in many systems. He defined a stable system as one with small fluctuations and rapid response to a state of equilibrium after perturbation, and a resilient system as one able to adapt to perturbations by moving amonq multiple equilibria. Boiling pointed out that increasing the stability of a system (e.g. , usiaq insecticides to reduce insect population fluctuations) might in fact decrease the resilience of the system. Resilience is a measure of the functional stability of relationships between populations or state variables ^n the system (Hollinq 1973). Are diverse syst3ms more resilient than simple systems? Few researchers have addressed this question directly. In this study the effects of a perturbation (herbivcry) on an ecosystem procer-.s (NPP) in a diverse and a simple system were investiqated. Fiq. 60 summarizes the results and is based on data from che insecticide experiment (reduced faerbivory) , routine measurements of herfcivory (background 242 O • < D X /^ > q: o > m a: UJ X -P >i cn O o (U u > •H c . -H >i !h ,— . O Ci< > Cu •H 2 X5 ' ■ Vh QJ >ix: -p •H U-l > 0 -H -P in U M S Q> T! > 0 i c M Q) m >-l g 0) •H H-i >-l iw D4-H -O +J 0) -P 3 m ddN 2U3 levels), and the three defoliations (incceased herbivory). Quantitative coiuparisons were not possiLie because the units and time scale for measurenent of NPP and herbivory differed among experiments. However, the figure was derived from data and does show qualitative relationships between the variables and between Gcosystems. Herbxvory (abscissa) is based on absolute amounts, rather than percent, of leaf tissue consumed. Because LAI and leaf specific mass were higher in the diverse system than in the monoculture, each 50% defoliation removed more grams of leaf tissue from the diverse system. This is reflected in Fig. 60 as higher herbivory at one, two, and three defoliations in the iiverse system than in the monoculture. Background (naturally occurring) herbivory was higher in the di^cirse system than in the monoculture. When herbivory was reduced by insecticide in the diverse system, herbivory was still as high as background herbivory in the monoculture. This resulL does not support the idea that high diversity systems incur less damage from herbivores than do low diversity systems. Monoculture herbivory in Fig. 60 includes data only from the cassava monoculture (maize monoculture excluded). Cassava leaves contain cyanogenic glycosides and are relatively unpalatable to most leaf-feeding insects (although not to leaf-cutter ants; see Blanton 1S82). This intrinsic resistance to herbivores may be one reason for the widespread cultivation of cassava in 244 fflany tropical areas. It is likely that herbivory would have been different had other species been planted in moQOCuiture. Also, the true herbivory rate on cassava aay have been higher than the value reported here (see Chapter IV, •Low Herbivory Rates'), but was probably not significantly greater than the rate in the diverse system. If diverse systems incur as much daaiage from herbivores as do simple systems, why do polycultures appear to have an advantage over monocultures with respect to herbivory? The answer may be in the different responses of diverse and simple systems to herbivore attach. In this study the responses of diverse and simple systems to herbivory (summarized in Fiq. 60) wt?re similar in one respect and differed in two respects. In boch systems herbivory stimulated NPP over a wide range of herbivory levels. At the highest herbivory level (three SOX defoliations) , more than five times the annual background loss to herbivores was artificially removed. Even this high herbivory level stiaulated NPP iu both the simple and diverse systems. Compensatory growth following grazing has been reported for a wide variety of plant species (see summary of previous work in Chat:ter I), and several researchers have suggested a noiimonoconic pruducLivity response to grazing (Vickery 1972, Dyer 1975, Noy-Meir 1975, Caughley 1'^76, McNaughton 1979a). The results of this study suggest that the nonmonotonic form 2a5 of the NPP response curve may be appropriate for plaat coBHJUuities comprised of maay species as well as for sinyle species. In both systeas there was au herbivory level at which maximuo stimulation occurred, and at higher herbivory the stiauiatory effect decreased (Fig. 60) . The NPP response to herbivory in the high and low diversity systems differed in two ways. First, over Bost of the horbivocy range studied, and particularly at high herbivory levels, the stimulatory effect on NPP was greater in the diverse system. Given equal herbivory, NPP was higher in the diverse system than in the monoculture, except over a narrow range of low herbivory rates. At these low rates the monoculture had higher NPP than the diverse system. The implication for agriculture is that a polyculture may be better able to maintain high NPP than a mcnoci'lture under heavy herbivore pressure, but at low herbivory levels a monoculture may perfcrm equally well. Second, maximum stimulation of NPP occurred at a higher herbivory level in the diverse system than in the moQoculture. If the two curves in Fig. 60 are extrapolated to the right by drawing straight lines through the last two points on each curve, the monoculture curve reaches the abscissa at a lower herbivory level than does the curve for the diverse system. Although the exact shape of these curves is not known, the data suggest that the stimulatory effect of herbivory on NPP spans a much wider range of 246 herbjvory levels in the diverse than in the simple system. A hiqh herbivory level that has a negative effect on NPP in a monoculture inay produce a positive response in a diverse system. Positive response to a wide range of herbivory levels in the diverse system indicates nigh resilience. Hollinq (1973) proposed that diverse systeas should be more resilient than simple systems for the following reason. A system with many species has many equilibrium points, each with its own domain of attraction. Althouqa fluctuations in population numbers will jove the diverse system from one domain of attraction to another, system function will be maiutaiued and the system will persist. McNaughton (1977) gave examples of ompiricdl studies in which fluctuations in the species composition of diverse systems had a stabilizing effect on ecosystem processes. Shifts in diversity are common responses to perturbations such as insecticide application and nutrient enrichment (see, for example, Shure 1971, Harcombe 1977a) . In this study both increased and reduced herbivory levels in the diverse system resulted in changes in the dominant plant species. The wide positive response range to herbivory in the diverse system was probably due to compensatory interactions among the co-occurrinq species. Changes in species dominance favored those species best able to respond to the perturbation. Species varied in their 247 responses to hezoLvocy , depeuding on tiniinq and intensity of tiie hecbivocy relative to the life cycle of the plant. However, because many complements of species could utilize equally well the available space and resources, positive response occurred over a wide range of herbivory levels. The diverse systeu was able to maintain energy flow throuqn the system (NPP) by species substitutions, but the monoculture was limited by the regrowth capacity of a single species. For example, the effect of hertivory on vertical distribution of leaf tissue differed in the diverse ecosystem and the monoculture. Defoliation allowed greater light penetration through the canopy. In the diverse system the result was increased growth of understory plants and an increase in leaf area near the ground. After defoliation of the cassava monoculture, leaf tissue developed at the top of the canopy ratiier than near the ground. This reflected the growth form of tlie cassava and the lack of understory plants (due to weeding) to take advantage of increased light transmission. High resilience of diverse ecosystems nay have importaat implications for the design of agroecosystems. Although diverse agroecosystems and monocultures may incur equal amounts of damage from herbivores, the wider range of response to herbivory in diverse systems makes them more sustainable. High resilience of complex agroecosystems that 248 imitate succession translates into a reduction of the risk of total crcp loss by the farmer. As in tae natural system, compensatory species substitutions may occur in complex agroecosysteias. In ag roecosystems these substitutions are controlled by management, but the principle is the same: compensatory effects result in maintenance of energy flo« through the system. Because minimizing risk is often more important to a subsistence farmer than maximizing yield (Barlett 1980), incorporating resilience into agroecosystems by crop diversification is a critical design consideration. LITERATURE CITEC Alcock., n. B. 1962. The physiological significance of defoliation on the subsequent reqrowta of qrass-clover mixtures and cereals. Pages 25-41 in D. J. Crisp, editor. Grazing in terrestrial and marine euviron- mants. Blackwell Scientific Publications, Oxford, Enqland. Altieri, M. A., C. A. Francis, A. Van Schoonhoven, and J. D. Doll. 1978. A review of insect prevalence in maize (2ea ma_xs L. ) and bean (Phaseolus vulc|aris L.) polycultural systems. Field Crops Research 1: 33-U9. Altieri, M. A., A. Van Schoonhoven, and J. D. Doll. 1977. Tuu ecological role of weeds in insect pest management systems: a review illustrated by bean (Phaseolus vulraris) cropping systems. Pest Articles and News Summaries 23: 195-20^. Andrews, E., D. C. Coleman, J. E. Ellis, and J. S. Singh. 1^/4. Energy flow relationships in a shortqrass P'-airie ecosystem. Pages 22-28 in Proceedings of the First International Congress of Ecology. Centre for Agricultural Publishing and Documentation, Hageninqen, The Netherlands. Atsatt, P. R. and D. J. O'Dowd. 1976. Plant defense guilds. Science 193: 24-29. Bacd, J. E. 1980. Effects of plant diversity and time of colonization on an herbivoro-plant interaction. Oecoloqia 44: 319-326. Baker, J. N. and 0. J. Hunt. 1961. Effects of clipping treatments and clonal differences on water require- ments of grass. Journal of Range Ranagement 14: 216-219. Barbour, «. 3., J. H. Burk, and H. D. Pitts. 1980. Terrestrial plant ecology. The Een jamin/Cummings Publishing Coapany, Menio Park, California, OSA. 249 250 Barlett, P. F. 1980. Adaptive strategits in peasant aqricultucai production. Annual Beview of Anthrcpoloqy 9: 5U5-573. Bartholomew, H. V., J. Meyer, and H. Laudelout. 1953. Mineral nutrient immobilisation under forest and grass follow in the Yangambi (Belgian Congo) region. Ser. Sci. No. 57. I.N.E.A.C., Brussels, Belgium. Benedict, F. F. 1976. Herbivory rates and leaf properties in four forests in Puerto Rico and Florida. Thesis. U'liversity of Florida, Gainesville, Florida, OSA. Benedict, F. F. 1982. Structure, function, and stability of intercropping systems in Tanzania. Dissertation. University of Florida, Gainesville, Florida, USA. Bentley, S., J. B. Whittaker, and A. J. C. Malloch. 1980. Field experiments on the effects of grazing by a ChrysoBielid beetle (Gastrophysa viridula) on seed production and quality in fiuaex cbtusif olius and K'll^E crispus. Journal of Ecology 68: 671-67U. Blanton, C. M. 1982. Patterns of leaf-cutting ant herbivory in simple and coaplex tropical successioaal ecosystems in Costa Rica. Thesis. University of Fl:>rida, Gainesville, Florida, USA. Boring, L. R., C. D. Monk, and W. T. Swank. 1981. Early regeneration of a clearcut southern Appalachian watershed. Ecology 62: 12U4-1253. Bormann, F. H. and G. E. Likens. 1979. Pattern and process in a forested ecosystem. Sprinqer-Verlag, New York, New York, USA. Boscher, J. 1979. Modified reproduction strategy of leek Allium fiorrum in response tc a phytophagous insect, Acrolepiopsis assectella. Oikos 33: ii51-U56. Bray, J. R. 1961. Primary consumption in three forest canopies. Ecology 45: 165-167. Bray, J. H. and E. Gorham. 1964. Litter production in forests of the world. Advances in Ecological Research 2: 101-157. 251 Burleigh, J. G., J. H. Young, and R. D. ttOLcison. 1973. St rip-cropping 's effect on beneficial insects and spiders associated with cotton in Cklahoma. Environmental Entomology 2: 281-285. Caughley, G. 1976. Plant- herbivore systems. Pages 94-113 in fi. (1. May, editor. Theoretical ecology: principles and applications. W. B. Saunders Company, Philadelphia, Pennsylvania, USA. Caughley, G. and J. H. Lawton. 1981. Plant-herbivore systems. Pages 132-166 in R. H. May, editor. Theoretical ecology: principles and applications. Second edition. Blackwell Scientilic Publications, Oxford, England. Cavers, P. B. 1973. The effects on reproduction of roiBoval of plant parts by natural or artificial means. Pages 140-144 in P. H. Cuun, editor. Proceedings of the Second International Symposium on Biological Control of Weeds. Commonwealth Agricultural Bureaux, Farnhaa Royal, England. Chew, R. M. 1974. Consumers as regulators of ecosystems: an alternative to energetics. Ohio Journal of Science 74: 359-370. Coley, P. D. 1980. Effects of leal age and plant life history patterus on herbivory. Nature 284: 545-546. Connell, J. H. 1971. On the role of natural enemies in preventing compstitive exclusion in some marine animals and in rain Eorest trees. Pages 298-312 in Dynamics of populations. Centre for Agricultural Publishing and Documentation, Wageniagen, The Netherlands. Connell, J. H. and S. Orias. 1964. The ecological rejulation of species diversity. American Naturalist 93: 399-414. Conway, G. R. 1982. Identifying key guestions for the development of tropical agroecosystems. Paper presented at the Cuinese Environmental Protection Of f ice/East-Hest Environment and Policy Institute Workshop on Ecosystem Models for Development, Kunming, Yuuau Province, People's Republic of China. Cornforth, I. S. 1970. leaf -fall in a tropical rain forest. Journal of Applied Ecology 7: 603-608. 252 Cromartie, H. J. 1975. The effect of stand size and veqetationai background on the colonization of ccuciferous plants by hecbivorous insects. Journal of Applied Ecology 12: 517-5J3. Dalrymple, D. G. 1971. Survey of multiple cropping in less developed nations. Foreign Economic Development Service, U. S. Department of Agriculture and D. S. Agency for International Developmeat, Washington, D. C. Daabenmlre, 2. F. and U. E. Colwell. 1942. Some edaphic cnanges due to overgrazing in the Agropyron-Poa prairie of southeastern Washington. Ecology 23: 32-aO. DeBach, P. 1974. Biological control by natural enemies. Cambridge University Press, London, England. Dempster, J. P. 1969. Some effects of weed control on the numbers of the small cabbage white (Pievis rapae L.) on Brussel Sprouts. Journal of Applied Ecology 5: 339-345. Dempster, J. P. and T. H. Coaker. 1974. Diversification of crop ecosystems as a means of controlling pests. Pages 106-114 in D. P. Jones and M. E. Solomon, editors. Biology in pGst and disease control. Blackwell Sci<^ritific Publications, London, England. Detling, J. K., K. I. Dyer, and D. T. Winn. 1979. Net photosynthesis, root respiration, and reqrowth of Bouteloua gracilis following simulated grazing. Oecologia 41:' 127-1J4. Dickinson, J. C, III. 1972. Alternatives to monoculture in t'je humid tropics of Latin America. Professional Geographer 24: 217-222. Dunn, J. H. and R. E. Engel. 1971. Effect of defoliation and Loct-pruning on early root growth from Werion Kentucky Dluegrass sods and seedlings. Agronomy Journal 63: 659-663. Dyer, M. I. 1975. The effects of red-winged blackbirds (A^elaius phoeniceus L.) on biomass production of corn grains (Zea oaj^s L. ) . Journal of Applied Ecology 12: 719-7261 253 Edwards, P. J. 1977. Studies of mineral cyclim^ in a montane rain forest in New Guinea. IT. The production and disappearance of litter. Journal of Ecology 65: 971-992. Einden, H. F. van. 1977. Insect-pest nanatiement in multiple cropping syr.tems--A strategy. In Proceedings of SyEposiua: on Croppiiig Systems Research and Development for the Asian Farmer. International Rice Research Institute, Los Banos, Philippines. Enden, H. F. van and G. F. Williaois. 197U. Insect stability and diversity in agro-ecosystems. Annual Review of EntOBOlogy 19: 455-U75. Ewel, J. 1971. Experiments in arresting succession by cutting and herbicides in five tropical environments. Dissertation. Ouiversity of North Carolina, Chapel Hill, North Carolina, USA. Ewel, J. 1976. Litter fall and leaf decomposition in a tropical forest succession in eastern Guatemala. Journal of Ecology 64: 293-308. Ewel, J., C. berish, B. Brown, N. Price, and J. Raich. 1981. Slash and burn impacts on a Costa Rican wet forest site. Ecology o2: 816-829. Ewtl, J., S. Glicssman, M. Amador, F. Benedict, C. Berish, R. Bermudez, B. Brown, A. rtartinez, R. Miranda, and N. Price. 1982. Leaf area, light transmission, roots, and leaf damage in nine tropical plant communities. Agro-Ecosystems 7: 305-326. Peeny, P. 1974. Biochemical coevoluticn between plants and their insect herbivores. Pages 3-19 in L. E. Gilbert and P. H. Raven, editors. Coevolution of animals and plants. University of Texas Press, Austin, Texas, DSA. Feeny, P. 1976. Plant apparency and chemical defense. Recent Advances in Phy tochemistry 10: 1-UO. Folster, K., G. de Las Salas, and P. Khanna. 1976. A tropical evergreen forest site with perched water table, Magdalena valley, Colombia. Biomass and bio- olement inventory of primacy and secondary vegetation. Oecologid Plautaruia 11: 297-330. 25a Gifford, R. n. and C. Marshal. 197J. Photosynthesis and assimilate distribution in Lolium multif locupa Lans. followiny ditterential tiller defoliation. Australian Journal ot Biological Sciences 26: 517-526. Gleason, H. A. 1920. Some applicaticns of the quadrat method. Bulletin of the Torrey Botanical Club 47: 21-33. Gliessman, S. R., R. Garcia E. and M. Amador A. 1981. The ecological basis for the application of traditional aoricultural technology in the management of tropical agro-ecosystems. Agro-Ecosystems 7: 173-185. Golley, F. B. 1975. Productivity and mineral cycling in tropical forests. Pages 106-115 in Productivity of world ecosystems. National Academy of Sciences, Washington, D. C, DSA. Golley, F. B. 1977. Insects as regulators of forest nutrient cycling. Tropical Ecology 18: 116-123. Golley, F. B. , J. Ewel, and G. i. Child. 1975a. Vegetation bionass of five ecosystems in northwestern Colombia. Tropical Ecology 17: 16-22. Golley, F. B., J. T. McGinnis, R. G. Clements, G. I. Child, and M. J. Duever. 1975b. Mineral cycling in a tropical moist forest ecosystem. University of Georgia Press, Athens, Georgia, USA. Goodman, D. 1975. The theory of diversity-stability relationships in ecology. Quarterly Review of Biology 50: 237-266. Grime, J. P. 1973. Competitive exclusion in herbaceous vegetation. Nature 242: 344-347. Harcombe, P. A. 1973. Nutrient cycling in secondary plant succession in a humid tiopical forest region (Turrialba, Costa Rica). Dissertation. Yale University, New liaven, Connecticut, USA. Harcombe, P. A. 1977a. The influence of fertilization on some aspects of succession in a humid tropical forest. Ecology 58: 1375-1383. 255 Harcombe, P. A. 1977b. Nutrient accumuldtion by vegetation during the first year of recovery of a tropical forest ecosystem. Pages JU7-378 in J. Cairns, Jr., K. L. Dickson, and E. E. Herri.cks, editors. Secovery and restoration of damaged ecosystems. Jniversity of Virginia Press, Charlottesville, Virginia, OSA. Harper, J. L. 1969. The role of predation in vegetational diversity. Pages U8-62 in Diversity and stability in ecological systems. Brookhaven Symposium in Biology 22. U. S. Atomic Energy Commission, Clearinghouse for Federal Scientific and Technical Information, Springfield, Virginia, USA. Harper, J. L. 1977. Population biolcgy of plants. Academic Press, London, England. Harris, P. 1973. Insects in the population dynamics of plants. Pages 201-209 in H. F. van Emden, editor. Tusect/plant relationships. Biackwell Scientific Publications, Oxford, England. Harris, P. 197U. A possible explanation of plant yield increases following insect damage. Agro-Ecosystems 1: 219-225. Hart, R. D. 1974. The design and evaluation of a bean, corn, and manioc polyculture cropping system for the humid tropics. Dissertation. University of Florida, Gainesville, Florida, USA. Hart, a. D. 1980. A natural ecosystem analog approach to the design of a successicnal crop system for tropical forest environments. Biotropica 12 (supplement) : 73-82. Hodgkinson, K. C. , N. G. Smith, and G. E. Miles. 1972. The photosynthetic capacity of stutble leaves and their contribution to growth of the lucerne plant after high level cutting. Australian Journal of Agricultural Research 23z 225-238. Holdridge, L. R. 1959. Ecological indications of the need foL a new a^-prcach to tropical land use. Economic Botany 13: 271-280. Holdridge, L. R. 1967. Life zone ecology. Tropical Science Center, San Jose, Costa Bica. 256 Holling, C. S. 1973. Resilience and stability of ecological systems. Annual Review of Ecology and Systeoatics 4: 1-23. Horn, H. S. 1974. The ecology of secondary succession. Annual Review of Ecology and Systematics 5: 25-37. Huffaker, C. B. 1971. The phenomenon of predation and its role in nature. Pages 327-343 in P. J. den Boer and G. a. GraJwell, editors. Oynaaics of populations. Centre for Agricultural Publishing and Documentation, Wageningen, The Netherlands. Jameson, D. A. 1963. Responses of individual plants to harvesting. Botanical Review 29: 532-594. Janzen, D. U. 1981. Patterns of herbivory in a tropical deciduous forest. Biotropica 13: 271-282. Jordan, C. F. 1971. Productivity of a tropical rain forest and its relation to a world pattern of energy storage. Journal of Ecology 59: 127-142. Kass, D. C. L. 1978. Polyculture cropping systems: review and analysis. Cornell International Agricultural Bulletin 32. New Ifork State College of Agricultural and Life Sciences, Cornell University, Ithaca, New York, USA. Kellman, M. C. 1970. Secondary plant succession in tropical montane Mindanao. Publication BG/2. Australian National University, Canberra, Australia. Klinge, H. 1977. Fine litter production and nutrient return to the soil in three natural forest stands of eastern Amazonia. Geo-Eco-Trop 1: 159-167. Klinge, H. and W. A. Rodrigues. 1968. Litter production iii an area of Amazonian terra firma forest. Aaazoniana 1: 287-310. Kulman, H. M. 1971. Effects of insect defoliation on growth and mortality of trees. Annual Review of Entofflology 16: 289-324. Lamotte, M. 1975. The structure and function of a tropical savannah ecosystem. Pages 179-220 in F. B. Golley and E. Medina, editors. Tropical ecological systems. Springer-Verlag, New York, New York, USA. 257 Lee, J. J. and D. L. Inmau. 1975. Ihe ecological role of CO nsuiuers--aa aggregated systems ?ie«. Ecology 55: 1455-1458. Linhart, Y. B. and R. J. Mhelau. 1980. Woodland regen- eration in relation to grazing and fencing in Coed Gorswen, North Wales. Journal of Applied Ecology 17: 827-840. Lubcheaco, J. and S. D. Gaines. 1981. A unified approach to marine plant-herbivore interactions. I. Popula- tions and coamuiiities. Annual Review of Ecology and Systematics 12: 405-4J7. Lugo, A. E. 1978- Stress and ecosysteas. Pages 62-101 in J. H. Thorp and J. M. Gibbons, editors. Energy and environmental stress in aguatic systems. Department of Energy Symposium Series. National Technical Information Service, Springfield, Virginia, USA. Halone, C. E. 1969. Effects of diazinon contamination on an old-field ecosystem. American Midland Naturalist 82: 1-27. Margalef, a. 1968. Perspectives in ecological theory. University of Chicago Press, Chicago, Illinois, USA. Margalef, E. 1975. Diversity, stability and maturity in natural ecosystems. Pages 151-160 in W. H. vanDobben and S. H. Lowe-McCouneli, editors. Unifying concepts in ecology. Df. «. Junk B. V. Publishers, The Hague, The I.'etherlands. Mattson, W. J. and N. D. Addy. 1975. Phytophagous insects as regulators of forest primary productivity. Science 190: 515-522. McNaughton, S. J. 1976. Setengeti migratory wildebeest: facilitation of energy flow by grazing. Science 191: 92-94. McNaughton, S. J. 1977. Diversity and stability of ecological communities: a comment on the role of empiricism in ecology. American Naturalist 111: 515-525. 258 McNauyhtou, S. J. 1979a. Grazing as an optiaization process: qrass-unguiate relationships in the Serengeti. American Naturalist 113: 691-70J. McNauqhton, S. J. 1979b. Grassland-herbivore dynamics. Pages 46-81 in A. B. E. Sinclair and M. Norton- Griffiths, editors. Serengeti: dynamics of an ecosystem. University of Chicago Press, Chicago, Illinois, USA. Menuinick, E. F. 1967. Structure, statility, and energy flow in plants and arthropods in a Sericea lespedeza stand. Ecological Konographs 37: 255-272. Murdoch, fi. W. 1975. Diversity, complexity, stability, and pest control. Journal of Applied Ecology 12: 795-807. Noy-Meir, I. 1975. Stability of grazing systems: an application of predator-prey graphs. Journal of Ecclcg/ 63: U59-481. Oduffi, E. P. 1971. fundamentals ot ecology. Third edition. H. B. Saunders Company, Philadelphia, Pennsylvania, USA. Odum, E. P. 1975. Diversity as a function of energy flow. Pages 11-14 in W. H. vanDobben and R. H. Lowe-McConnell, editors. Unifying concepts in ecology. Dr. H. Junk. B. V. Publishers, The Hague, The Netherland s. Odum, E. P., C. E. Connell, and L. B. Davenport. 1962. Population energy flow of three primary consumer components of old-field ecosystems. Ecology 43: 88-96. Odum, H. 1. 1971. Environment, power, and society. John alley and Sons, New York, New York, USA. Oduffl, K. T. 1977. Energy, value, and money. Pages 174-196 in C. A. S. Hail and J. W. Day, Jr., editors. Ecosystem modeling in theory and practice: an introduction with case histories. John Wiley and Sons, New York, New York, USA. 259 Odum, H. I. and R. Tinkerton. 1^55. Time's speed requldtoc: the optiaiuni efficiency for raaxiiBum powei output in physicaj. and bioloqical systems. American Scientist U3: 331-3U3. Odum, H. T. and J. Ruiz-Heyes. 1970. Holes in leaves and the grazing control mechanism. Pages 1-69 - 1-80 in H. 'I. odum and E. F. Pigeon, editors. A tropical rain forest. U. S. Atomic Energy Commission. National Technical Information Service, Springfield, Virginia, USA. Oppenheimer, J. a. and G. E. Lang. 1969. Cefaus monkeys; effect on branching of gustavia trees. Science 165: 187-188. Owen, D. P. 1980. How plants may benefit from the animals that eat them. Oikos 35: 230-235. Owen, D. F. and a. G. Wiegert. 1976. Do consumers maximize plant fitness? Oikos 27: 488-492. Paine, R. T. 1971. A short-term experimental investigation of resource partitioning in a New Zealand rocky intertidal habitat. Ecology 52: 1096-1106. Painter, E. L. and J. K. Detling. 1981. Effects of defoliation on net photosynthesis and regrowth of western wheatgrass. Journal of Range Management 34: 08-71. Pearson, L. C. 1965. Primary production in grazed and ungrazed desert communities of eastern Idaho. Ecology 46: 278-285. Petrusewicz, K. and W. L. Grodzinski. 1975. The role of herbivore consumers in various ecosystems. Pages 64-70 in Productivity of world ecosystems. National Academy of Sciences, Washington, D. C, USA. Pfeiffer, W. J. and R. G. Wiegert. 1981. Grazers on SBartina and their predators. Pages 87-112 in L. R. Pomeroy and R. G. Wiegert, editors. The ecology of a salt marsh. Springer-Verlag, New York, New York, USA. 260 Pimentel, D. ISbla. The influence of plant spatial patterns oe insect populations. Annals of the Fn tonioiogical Society of America 5H ;3 ai ^ 0 C o M ■""^ 0 Pi 0) 13 en iH u c C Q rO 0 >i , , 13 en JQ CL, S-i 4-> 03 03 ^3 0 s: 13 Q) - +J ■p CU > OJ •H en ■H M C £ en ^1 rH 0 u OJ 0) •H g p S-l T! > IS en m Q4 X Cfl (C n3 en cu tC cu S iH 5 s >i 03 > cu Xi .— . •« u S-l = q; 0 (U >1 c en > +J 03 0 QJ •H C g ■H nH Xi •H ,-^ tn r-l n 1 iH 0 C 0 0) X! 4-) 44 (T3 u £ 3 a en 4J 0 X >1 4:; 03 w (U u •H iH 03 en en 1 (U Xi X5 13 cu cu ^^ iH tr. s tX > !-l 44 0 (D (U 03 0 +J K rH S-I CU > H Q) G iH •H 0 4-1 < > ^,. — -^ -Q 0 0 0 4-1 4-J S-l T3 II 4J 0 13 4-1 0 >i ^ 03 +J ■P H T3 0 <3 4-1 X5 •H 5-1 (C fd (U (U 5-1 - >-^ 44 e e +J Ti tn ^-> ^-^ 0 c:) c ^ ■H 0) c -rH 0 0 0 4-1 4-1 -P jn •H > CU 4-1 44 + ■H 03 5 m -p ■H s: — - ^-• 4-J 0 w rO -' cu -p (D Ui .. en + Q 44 rH cu OJ e s (U 0 II cu 0 en s 4-1 0 X3 -P i-i ,-~. + en 03 ^-^ S-l iH -■' ■f—i a 0 cu en !-i fO iH , — . 4J +J 4-1 0 CU 00 a 0 4-1 -P - cu — - < 03 5-1 > 0) 4-1 -P 0) rH en D U U Cn 03 c Q •> J 1 03 0 ■— ' W 0 •r~) e II II II II II -H fC -p •P 03 -P tH ^— • Ti . — . U ._ . (0 iH V en 1 iH •H 0 3 (U c -p •r— 1 £ -^ U w 0 M PQ i-:i 03 -P II E-i Q CU 5 0 . . 4-1 4-1 267 268 ^^ -p iH o -p -p 1 - ^ <] c o CT - +J o ^-- + o 269 ^ u -p u < 0 0 m II en c ^.ms. ■H M-l > ■M iH — ' 0 Q Ui C v| -P U-l -P +J ■ — — - n O ■n 1 4-1 -P 270 iH 1 03 1 t=: C C C + (T3 •n o o -p 0} 271 H I h m 11 0 n c -p o •H 1 -p 3 4-1 +J 4-1 -H 4-1 II cn XI i= 3 w (D >< i-l m 1 w c e + e 1 APPENDIX B BIOMASS AND LITTERFALL MEANS c rH 0 Q) •H > m a; U) I— 1 QJ en O C in u rO O 0 0) • en b 0) IW -x; n 0) 4-1 +j t) (T3 -p •rH Tl fC 6 ■H £ -u 6 u c ITJ (D •■ 0) !h c 0) 0 -P M-l ■H < MH en •H in t3 QJ O U3 >1 C) H 3 II -P en C C en T! O aj •- -H ^ — U-l u • -p-l •H T3 C !-l • Cn C en -H QJ ^ en ^ IX -P C O 0 0) C ■H S-l en (C OJ en u (1) en rc U Q) O d U d :-\ eD en fC -P > -p iH QJ m en i-i S-l en d (13 eu -p S E (C n3 c en c (U QJ •H !h X^ d -P en 4-) en rH >1 frt 3 £! f= U 0 0 13 •H C 0) -Q 0 :? g 0 4-1 <-^ e« 'O rH 0) C 0 iJ (13 4H c 4H O O -H en u tn •H OJ e o ■H U c 'O o x: en u en rd en Sh en ^ ^ r-^ Ti rC 0) o U X^ •H d O S •P -rl •H 03 U T. S C W o o ^ rH < ^ u — — u -- OOO fNrH 02 m CO CO rH r- CTi CTi rH O^ CTl LD rH -JT 00 LD < :-^ < (» o O cn CO in 03 in in O (N in n n 00 iX) 00 o -* ■>:}< CO rH r- '3' CTl in CT\ CO en ^-- rH ^^ rH rH in r-\ fO CM 1X> rH IX) «^ 00 (Ti (^00 00 ■^ o r-~ 00 1^ CD r- 00 o r- rH 1^ o> rH i-H in in CTi 00 CO r~ in 00 CTl 00 CO (N in r- in it3 0) (1) 4J >i >i c fO 03 (« d P 2 S ^0 C\ !h o> r-- 42 r~- +J e U -P en OJ (U en d 4-) X! d Cn Cu o IT. d 0) 4-) d < CO 0 < o 273 274 13 o > o > o > o > o > h • fO • fO • fd • (C • 03 • 0! 00 m 00 m ix> Ul in m CN m CM cn T! m m iH tn fvj en 00 m •^ in C TCl n3 03 m 01 01 CN o O O 0 u U •^ CO CM O iH in en -. ^ CM in ■^ 00 00 OM r- CO o 00 cr> <0 1.D ':r (T\ ^o in o t^ UD iH "^ 00 in ■^ r- r- 1 00 X! u X! ^ ^ o o 1= U C u >i >1 D. -P 0 Q) 03 03 03 H a) U Z a l-D s S 3 1-1 en o 275 T3 aj :5 - C.) c iH 0 -H •r^ o (fl IW m Q) en u c u m :3 QJ tfl e QJ 4-1 ■> > 0 0) 0) 4J iH 0 (d •H T3 LD i= O ■H £ • g u in 0) - a) x^ c •4-1 n -p ■H < -P en (fl en aj • -P u «D c u a) 3 II u en (U c q-i 13 M-l (U • *. •H ,G ,-, ^ u 1 ■iH 'O >i S-( • rH c en -p 0) "-^ eel - X U c •H o (1) 4-1 ■rH S-l ■H U} rrt C en m a; en •r^ u QJ en u 3 3 M -P en en n > c rH a en a) u en M 0 trt (fl -p ?: (TJ !-l c (D c i W (0 Xi O -H en U en ■H (D e o ■H O CO C TS O (1) -H ^ en u en H (U u u C u w ;3 en o H -H 03 en !-) en -P o 03 u en V » rH T! 03 0) o U £ •rH 3 u F; -P -H •H 03 Sh T. S c W < ^ ro rH o o OO «DCTi LO'* rHCO O rH r^ in U3 r- o rH in in 00 CN U3 •^ CM r- in cr\ o in ro CM CO in en •sa" <: -- en in in 'Xi CO C3^ < ^ (Ni in CO IX) •^ rH 00 cTi in 00 (N 00 rH r^ U3 '* CTl m CTl CN CN CM '-' ■-'CM CO ro in CN CO in in 00 'J3 en r^ X) o 03 in 03 01 in 03 e • e ro ■rs 'C c C 0) QJ -P >i >i C 03 0) 03 :3 D S S •-3 >1 CJl Sh C> CTl r~- QJ r~- -P c J-l •P en QJ QJ en 13 -P XJ Hi Cn a n tji 3 QJ -P :3 < CO o < o CN CO CM CM 276 H T3 (13 0) U :3 o g c w o o PQ ^ 00 CM n ^X) o c m 0 <; 0 -H <: ^-^ < .■^ --^ <: ^ < ,— > < ,— ^ m — cn 1X> on ^ VD (X) (N CTi * o 00 rr 00 u en • • • • • • • • • • • • H (U in 00 (N ro CO CT\ rg en ■^ 00 in (M Osl c^ u o o m a> ro rH •^ 00 "d* (J^ o> o O OM VD C o •^ (N •^ 00 ■<* CN VD VD CO in VD in in CO a! 3 — - ^-^ — - — - ^^ •—' fM H w ' c 0 CQ -H -H < ^^ < -^ < ^-~ < ^—. < -—. < —. < ^ (13 tn CO o VD r- n iH VD VD O 00 r-~ rH VD 1^ !-i tn • • • • • • • • • • • • 3 0) r-i UD iH ■5}" CTi (^ iH a\ 'S' o o M* H rH +J o r-l OJ (N VD r- ro VD 00 r~ VD 00 CT\ rH CN m u VD (^ VD Osl ■* .H iH CTi r~ ■^ CO 00 CM 0> 3 :3 ^-' ~-' '-' H "-^ ■^ ^-' rH -^ a 0) 0 N Jh ■H o O (0 • s I^ tn rH 'a M C 5 OM (0 03 (t3 03 03 > o > o > o > o > (13 • (C • 03 • 03 03 tn 00 tn VD tn in tn CM en en tn rH tn Osl tn 00 tn 03 03 03 03 03 u u U u U U X) e CU a) +J > 03 n p 2 u 03 03 03 u 0) x; e Q) 4-1 0) CU +J tn 0) > u 03 >i rH C o 0) Sh -P iH 13 O O c o 03 277 4-1 ■p C rf QJ <. M c 0) o 4-4 •H 1^ 4-4 to ■rH m II ^3 0) u c >i 0 iH 0 4J m ^^ T! Ti u 0) • •H x: CO 4-1 o ■— ' •rH •H c !-ll X m c •rH Q) 0) CO •. m 4-1 c n 0 U) c ■rH OJ m 13 CIJ m iH j-i 0) rc ffl u > u S-l 3 tn (1) U) en 4-) tti 4-) .H SI CI) 4J OJ F= (C S-( m c 3 -P CO C r-t cu ■H 3 4: U 4-1 0) n J-) c >1 V^ 0 ^ (fl e 04 ■n T3 Cl) 0) c 15 > n3 0 -H <-{ -P K rH u c 0 3 n 4-1 TJ rH 0 m CO t-i ^1 m c aj a, QJ 05 > 0) U a) (1) ^1 u 3 e iH -H ■<* CO r- n i-H ^— r- in o o PQ ^ rH O ■^ m in ro o o O H CTi O 00 ■^ O 00 CN in ro LD r~- tH on iH ro iH •5J< 00 CN ro ro H (N <: ^ CN in O rH CO O rH ro I^ r-~ (N CT^ in vD (N rH u o in 1 C7\ U tn r- 0) 4-1 E 4J cn QJ CO 3 4-1 3 Cn C-, tn 3 (U 3 < en <: 278 4-1 0 0 -rH 03 u — < — < ^ < -- X) •>* (N LD ^ U") t^ r-\ (N CS iH CN in r-\ m VD IT) o o u — o o o o CQ — o o u — o o (N n 00 "=}- o o o o o o o o o o o o U o o U pq ^ (N 00 CQ a^ in i-H fM »* in •^ in CO CO r-{ Oi PQ ri: ^ < ^ ro ro O ro Sh 0) -§ aj 0) 4J > rO 0 O s ra CQ <;^^ <— <--. o > o > o > o > o > fa • tC • (fl • ITS • ro • 10 to ro en CO en 1X1 tn in tn CN tn CN m Ti tn en ,-] en (N tn 00 en ^ en c 03 ftJ t« (0 10 (0 CN U O u u U u >i 00 Sh fO ^ 3 U C u rO to t-D s Sh CO 0) XJ !h F. 0) Q) X! -P 0 Oj -P ei) 0 CO o a) +j en a) > !h m x: >i rH C o tu !h -P rH U O c o to 279 H-) m rO 0 n i _. ,-t Tl • -P OJ Ti c ^ • ta u cn o ■H ^- •rl M MH C X -rH O c 0) tp ^ in •H c (fl tn o •H M ■P en 0) o tn 3 C 0) H o 01 d) u > Sh 0 (fl tn U3 cn Sh iH rrt 0) frt ?' H-J M +J 13 0) -P • rH (13 (U C S-l 0) p e C H-1 n3 •rH rH =3 U) tn O (U u: 0 x: (rt c H-) S 0 0 h >1 •rH Si ^ 'O C T! T! (rt (U re 15 CI) - n T! C rH . 0 rH rH fTi •H 0 0) C m MH > •H m 0) T! d) 03 rH C o c m u fC in +J 0 CD O CO m e • u tn ■H (1) e o c t3 O CU -H U en c o •-I -iH n3 en Sh in 13 (U ■P O (C o CO ^ ^ 1 r-l T3 (0 (1) u Sh ^ •H 3 U g •P -H •H m ^ ?: 2; c w o o r- r- CM O^ CO •^ o o en — o o CQ rH a> CO IX) r- CM rH 00 LD CN o o ffl ^ o o o o CQ ^ LD O O IT) o o CQ — o o o o CQ -- CN r-i in rj- <-i o 1^ [~-- O rH 00 00 o e tc3 in g o g Tj m T3 u — n CO ro r~- '^ in CN VD in LD O-l IT) OO fM CQ "^ rH m rH in 00 rH rH O g O g CTl Sh i u en i-i ■H OJ c S-l u c o w 3 CO c 0 rH •iH (« cn !-i tn 3 0) -U 0 (Tl u S 3 d 4-1 u o .^ ^ H Ti n3 0) u u s: ■H 3 o 6 +J -H •H (0 J-l T. s c w iH in o o U ^ o o o o 00 o r- CM o n CO in ro in n CO in n in i-D 00 00 1^ ^o IX) in 'S' fH in ^ ■^ iH CM ivD tx> 00 00 •=j< in cTi IX) r~- '3* a\ n o r- -a- O 00 in CN rH o 00 o in CTi O r-{ ^ IX) (N <-^ CN) iH CN CN "* M' ro iH ro tN 00 o > o > o > o > o > fc • e • tC • ro • m • n3 • rd . fd r- ro cn 00 cn IX) cn LO cn CN cn CN cn t3 rH T3 tn cn iH cn CN cn ro cn 'S' cn C c (t3 fd m m m m CN CN u O u o o o m ^~. 1X3 CN 00 ^ r- iH (N ■^ ^ IX) ro CO in ro r- u s-l cc o 5h 0) 00 5-1 0) CD >1 CO X! 5h 0) ^ ^ 5-1 o o e d) ^ B g cd x: 00 00 a) ^ 0 (U (D 3 U 4-J 0 0) 4-) > O C u >i >1 a. 4-1 4J o 0 Q) (d rd rd rH (U 0 (0 o 2 a 1-0 S 2 P cn o D l-D 281 >1 Xi t: Q) iTTJ s ceo •H fO iH Ti ^ C - 0 fC C U-l •J-" 0 CO -iH CO u) c + CO n3 1 0) 0) .H en 0 e a) ■P o > s-i :3 - 0) rO CO (U nM a -p y-j (d LT) 0) 0 t3 O > • ■H U £ -P -H O Q) O g nJ 4:; 3 -H 0) +J TJ g 0 4J -U 1-1 -<: (0 Qa c (U 0 -P U -H • c CO VD 0) + en M 0) II QJ m u M-l 6 o c 14-1 0) 3 ■H -P CO •» T3 CO --- T3 • >1 + (U 'O .H ^ • 4-t CO U CO C QJ -H — m > !-l 0 (0 C IX •H 0) 0) 14-1 T-\ Q) •H ^ ^ u c C re CTi CO 0 •H CO -H CO CO fC cn (U g CO 3 4-) 0 0) -H n ■H O to c XI u > 3 cu Ti CO CO M C CO to :i •-{ ai 0 (« 2 1-1 i-i 1-1 fl) ci d -p 1 -P ■ -U OJ tC (U OJ > c s^ H 0 :3 X3 C -P 0) to -H rH g 3 to rH ^ O CO to TJ 0 ■<-> oi c OJ 0 0) 0 s: t^ T! g ■p c 14-4 O O -H CO U CO g O H U cn T! O £ Cfi U CO rH d) c o to CO U CO 3 CD -P U to o to 0) o u s: -M 3 o g -P -H -H to H 2 2; c w <; -^ fi, '—■ < ^ ■^ (^ CO n cxD r~- O CN CM IT) 00 LncN oco ^n 00 -cr LT) (N in CO rsi (Ti o i-D r-- CN ^ o 00 00 en ^ m C3C < ^ m ,_^ CQ ^^ cr^ r- nH 0 ■^ .H H o> iH (T. (N in in in ■^ \a IT) 00 CM X) [^ 1-4 en oi in ^^l o •^ in 00 CO a^ 'S* •* 00 LO vD in CO CTi in (Ti O CO O (N o~i a^ CO in U3 r— I O rH (0 in to g • g CT\ -p -P CO CO to 0 to g • g ^ T! r-i -a C c en u (Tv [^ cn r- Q) 1^ en r- Xi 1-1 a^ en r- cn 4-) g M i >i C >i H! Cn Cb 0 -p -p to to d -H Cn ;3 (D -P u to 2 1-3 D P <: CO CJ 0 D 1-1 < 0 282 CD < C O •rH +J to -p 0) 01 O -H en u en rH (U e o H O CO c na O CU -H x; M O en H (U M U c u y 1=1 en CO ^ TJ CU u ^ •H o 6 •H •H S-l ;^ c; w u — pq — U — o o in 00 00 CJ m ,-> ^~. m •"^r 00 ■^ (T> J m en Pi IX) O < ,— , o ^ fN in CN rH rH CM rH r~ ro ro ■^ rH ** o 'S' rr ■^ in nj n3 fO It! rrt n3 > o > o > o > o > o > (C • fO • m • teJ • ro • (13 en 00 en X) en in en tN en CN en en en rH en CN en ro en ^ en td (C ftJ eel in (0 U O O U u U u -- 00 IX) CN CM ro , CO o Sh £! ^ Sh o o g CU e s rej x: CO 00 Q) -Q (U 0) (U 3 u -P n ■p > u C Sh >i >i CU +j n1 0 tu fO rej fe3 rH (1) o P 2 Q i-:i ^ ^ CO o 283 O •H S-l c • 0) ro 0 -o •H • X! en T! IX c; 3 3 0 H -Q (fl <: > en c •H T3 to Q -P ■4-J O 0 (t3 O Q O en -P 0) 0) > o o o o o o o o o o o o o o n <■ O O ^X) L o o o o o o o o 00 ^ o o i^ c uD a-. in m tN r~ rsi CO r- r 00 -^ (y\ m 00 CM 00 00 CM ' .H (N iH r- in r-- ■^ ro - 0) r-~ i>D CM in ixi ro 00 00 > • • • . tn un ^ CO CT^ in CO o CO r- ^ QJ Q) >i i >i c >i :=! C. 0 > o c (l! to 3 r-l CJi CD -P 0 Q) (D 2 S 1-1 3 t-3 3 < cn o o s c hD 284 Cn r-H ,^ ,_^ ,_, (t3 -^ m ■^ o O 00 r- LT) ■^ in -tJ • • • • . n iH IT) ro m 'S' ^ iH (Ti H O IX) CO ■rr "g- in ro CM o CO ro rH UD in 00 r-^ o r~- O i-l 00 iH r- ^ iH O r-- o ,H ro iH M CO r^ r- ^ ■X) tN ro .-H CN 1— 1 (N (N (N rH 00 S-l o o 0) Xi CO CO £ O o u >1 >i 4-) (H rfl M U s S 1^ o 285 0) 'O ■H u ■H +J 0 0) en c •H X: -P •H 5 T! OJ -P fO Q) U 4-1 (fl 0) U 3 +J H d U n c: n e ^ m m ro e • 0 ro •H II ^ C 13 - C -^ P • 0 t3 U • Cn cfi 1 ^- 0) IX > 0 (U ^ s^ m ■H 4J 0 Cfi en :3 +J en ■a u ra 0 to ^. M d. Cu > Q) u A a c ^^ 0 Cl. -^^ 0 -P ^ trt 0) o +J m — - 0) < tn en aj 44 > S CO 00 CM o '3* -— ' CM to ,-. (0 _ to ,-^ lO _ in o^ o CN iH »^ O 00 iH o CT^ o r- 00 CN CN ^-' iH (N (N rH O (N iH — - D O ix> to CN 0> cr> (N (0 ^ CTi O 0) o <-\ oc -^ \a ^ ro 'S' +J LD CN in \D in "^ D CN CTn to ^ CN CN CN r~- O CN to in to o to O to o lO g • e • e • E • e fO r~~ iH 'fl' T1 T3 'C ■H T5 ^ T3 c C c C C r~- cn r- 0) r- en r-~ -Q 1-1 tTi CT^ r- CJ> -P 6 i-i 0) f-~ r- r- -P en 0) 0) X) 0) en 3 -p ^ 0 >i >i c >i d Cn a 0 -p to to :3 iH tr> m 0) +J o s s l-D 1^ 3 < < en u o o 286 en c •H Ti C 0) 03 Q 4-1 O -P 13 in O Ot U a, 0) Pi C 0 aj ■H o ■p u (C 0) o en to > 5 n3 ^^ ^^ ,_ X! -- _ Xi -. ^^ 00 1^ o o r- r- O ro ■^ Ln ro LT) r- cTi ^ 00 o o -H ro 00 rH ro ro ro y3 ro 00 r- CN — - ^-^ ^-^ r- 00 00 a> iH in in ro "* iH — - ■^ (N rsj iH 00 in (13 ^ CO 1X> o o o o o o o in ro 00 O U3 1X1 CO o o o o o o (N O O LT] i-H iH in 00 r-l --' ocro oo oo oo oo oo rsiro oc'^ oo oo oo oo oo oo inix) -^ '^ --^ ^— ~-' ^-^ _^ ,^ ,_, X! ,^ ^^ X! — ,-^ un 'S" o o in ^ ro ix> •«*• in (Ti iH 00 o a^ "^ o o ■^ rH xi ro ro r^ CTi OsJ ^ r~- "vT r- — - ^ 'J- in -^ .H r- in o ro iH ^' ^-' ro tN iH iH r- ■<* 00 ,-{ o o (N (N Xi r- (T, 00 O O r- CM 00 o n3 • e U ca o G s 00 LO 00 CN ro tT 00 r^ tN oj ro o^ m 03 n3 03 03 03 > o > o > o > o > o > (0 • fC • tC • to • 03 • 03 05 00 en UD tn in tn r\i en OM tn (n en rH en CN en ro en ■^ en m tc n3 03 03 03 u 0 U U U U >1 u 03 d c 03 1^ 1-1 0) X2 e 0) -p 0 > • a) rH rH (1) > in 0) o rH in Q) o £ • +J 0) -P x; 03 4-1 0) +J T3 03 •H U 0) ■H T! 4-1 •H U u (1) ■H tn 4-1 c U ■H 0 m 4-1 C 3 ■H n ^ +4 4-t 3 •rH o /5 x: 4-1 0) •H S-J :? 3 4-1 0) rH M 3 13 C) 4-1 n iH c 3 o U F= 0 c c 0 03 6 x: 4-1 c 03 S-l x: (1) 4-1 4-1 01 tn 0) tn M tu en rH >i >l rH rH 4J +J c C Oi 03 u U •H ■H 4-1 4-1 ■H -H C C tJi tJ> ■H •H tn cn 03 42 287 c 0 • 0 IW iH •H a) m to > m C Q) (D (TJ .H u cu 0 e ^ =) 0 en » • QJ m -P dJ 0 (C x: 'a j-> C) ■H ^ -p s U It •H tC f: q; 4-j c - -P (D c <; ^4 0 CD •H 4-1 CO • 4-1 U) IX) -H 0) II T3 u C u >i 3 -iH CO -P -P £ C Tj Cn (C QJ ■r^ 0 4:: (D -H u :5M-i •H •H Vl >i C c S-l tji CD 13 -rH CO « c C 03 -P n cu 0 •H e c CO CO 0) OJ (I) u u u to (0 0 3 CO !-i CO 0) rH nH +J (CJ fd OJ M > H 3 4-1 OJ m • R C 0) ra U CO c 3 ■H -P 0) -H ^ i-l 3 -p i-H 0 (0 0 >i IW C X! i-l 0 (U 1= T7 -U (U -P TJ 5 ■H C 0 iJ (C rH O -H u m H 0) ^ o -C CO O c/l rH OJ U V C O w d en tC CO s-l ifi 2: 3 en ^ ^ H T! (0 (U U ^ j:: •H 3 D e +J -H •H (C M ?- 3 C W N N N N N N H •rH •H •rH •rH ■H m fO to to to tC 6 e e g e e (N «* ^X) rH n -p M -P rH +J rH -P Ti Ti CO to CO t/) C c <7i 0^ r-- CT\ r~- r~- -P rH rH a^. r- r-~ r- U) (U •H •H r- (U >i . >i 3 -p !h i-l >i 0) c >i rH rH &i to a Gj >1 to c 13 r-^ 3 3 3 a < < to ^ 3 1^ Id 3 Id ^D Id <: 288 Cn < C O ■H 4-) (0 J-) 0) Cn| 0) > d) u :3 -p iH 3 u o c n s c ,-v y-i 0 X 0 ■rH :$ w o cn fNj •H tl) x 6 u N •H t) S Hi CO cn c iH •n 0 iH 0) •H rO £; cn m u Cfi i-i ■H 0) 0) S-l CJ +J c C) -p w 13 ■H en iJ c 0 rH ■w (d cn S-( cn 13 Cl) -p C) (C U s 13 •^ 'O 0) u x: •H u E ■H ■H S-l c w 4-1 cn 0 Id 0) 0) ■4-) -p a. cu m tc3 rc3 > > > tc3 IT! fC cn ro cn Ln cn cn cn cn (T3 m (C3 U o U 5-1 00 ^ 0) Q) X5 >i Xi e s-l B cu CC3 Q) u 13 U (U C OJ p m Q 1^ 289 o tn ■H 0) e o S 3 en 'Ci o 4-1 U fa o 01 ^ » nH Tl n3 0) u ^1 x: ■H 3 u e +J -H •H fO J-i T 2 C w fd fd fd fd (t3 td (d fd td IT3 > > > > > > > > > > fd fd 03 fd fd fd (d fd fd fd U} cr, m iH 01 ro U) LT) tn r- m (Tv m M en (^ en LD tn Cfi m H U) iH tn ■H en iH CO iH « CM en (M m CM m fd fd fd fd fd rd rd td fd fd o O U u o U O O u U >( >i o o o >i >i S-l M OD CO o CO !-l 5-1 fd fd 00 o fd (d d :=! s: x: iH O CO P 3 S-l M u o H •H CO c c ^ x: S-J S-i ■H !-l >l iTi fd (U CU fd fd !-l &. >i fd 1^ 1^ fa h S g- < < rd 5; 29 0 O tn u en ■r-i (D e o ■H U tn c T! O > > > > > > > > 03 rt! (fl (IS tC3 fd fO m fO rd ■H r- cn cTi cn iH cn ro cn LD cn r~- tn (T\ cn rH tn cn tn .- 'Xj CM cn l >l 3 3 a; 4J X! 1 rH tH cn cn 4-1 a o C 3 r-l 3 3 13 =3 Ch CD -p 13 l-D 3 f-D 1^ < -< tu cn u l-D l-D cn o 291 O -H o en ■H dJ e u CO c T3 O o cn U O C U c 0 rti en s-i tn :=! 0) -u u fO u IS 3 en V .. -H T1 fC 0) U !-i j:: •rH ;3 u e J-l -H ■H (C i-i ?; IS c K •H O u (T3 •H in 13 O U O 0 12 BIOGfiAPHICAL SKETCH Becky Jean Browu was born in Texas in 1948 and raised in Texas and Georgia. She received a B.S. iu fflathematics education from the University of Georgia in 1970 and an M.S. in statistics and biometry from Emory University in 1974. In January 198J, she will assume the position of assistant professor at the University of Wisconsin-Madison, Institute for Environmental Studies/Department of Botany. 292 I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. e^ lytc^ £^t^. f^ John J. Ewel, Chaxr Professor of Botany I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Edward S . Deevey Graduate Research Professor of Botany I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of jiocto:^ of Philosophy. Dana G. Griffin III /Professor of Botany I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. l^^^^^-(.., f A^ Ariel E. Lugo Associate Professor of Botany I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Howard T. Odum Graduate Research Professor of Environmental Engineering Sciences This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1982 Mic-t cA ^/vt; Dean //College of Apiculture i//College of A( Dean for Graduate Studies and Research UNIVERSITY OF FLORIDA 3 1262 08666 296 1