Shade tolerance within the context of the successional process in tropical rain forests Gerardo Avalos1,2 1. Escuela de Biología, Universidad de Costa Rica, 11501-2060 San Pedro, San José, Costa Rica; gerardo.avalos@ucr.ac.cr 2. The School for Field Studies, Center for Sustainable Development Studies, 100 Cummings Center, Suite 534-G Beverly, MA 01915-6239 USA; avalos@fieldstudies.org Received 30-IX-2018. Corrected 26-II-2019. Accepted 26-III-2019. Abstract: Shade tolerance (the capacity to survive and grow over long periods under shade) is a key component of plant fitness and the foundation of current theories of forest succession in tropical rain forests. It serves as a paradigm to understand the optimal allocation of limited resources under dynamic light regimes. I analyze how tropical rain forest succession influences the expression of ecophysiological mechanisms leading to shade tolerance, and identify future areas that will increase our understanding of the ecological and evolutionary conse- quences of this phenomenon. Shade tolerance is a multivariate, continuous functional trait reflecting the growth- mortality trade-off of investing resources under limited light vs. exploiting high light conditions. I propose the life cycle successional trajectory model of Gómez-Pompa & Vázquez-Yanes as an integrative tool to understand tropical rain forest succession. This model shows how species distribute along the successional environmental gradient based on their degree of shade tolerance and represents a more integrative paradigm to understand the interface between different aspects of species diversity (ontogenetic variation and functional diversity) throughout succession. It proposes that different trait combinations determining shade tolerance are expressed at different stages of the life cycle, which affects how and when plants enter the successional trajectory. Models explaining the expression of shade tolerance (resource availability, carbon gain, CSR, resource competition) are based on whole-plant economics and are not mutually exclusive. The analysis of shade tolerance is biased towards tree seedlings in the understory of mature forests. Other life stages (juvenile and adult trees), life forms, and microhabitats throughout the forest profile are almost always excluded from these analyses. More integra- tive explanations based on the distribution of functional traits among species, ontogenetic stages, and the nature of the environmental gradient are being developed based on long-term data and chronosequence comparisons. In summary, shade-tolerance is a complex phenomenon, is determined by multiple characters that change ontogenetically over space and time and entails considerable plasticity. Current methods do not account for this plasticity. Understanding the nature of shade tolerance and its functional basis is critical to comprehending plant performance and improving the management, restoration and conservation of tropical rain forests given the combined threats of global warming and habitat loss. Key words: environmental filtering; functional traits; gap phase; leaf-economics spectrum; niche differentiation; ontogenetic niche shifts; plant-economics spectrum; secondary succession; shade tolerance; regeneration niche. Avalos, G. (2019). Shade tolerance within the context of the successional process in tropical rain forests. Revista de Biología Tropical, 67(2) Suplemento, S53-S77. One of the fundamental goals of tropical what determines such high levels of species rain forest community ecology is to explain diversity across different groups (Denslow, the high level of species diversity and the lack 1987; Hubbell, 1997; Givnish, 1999; Wright, of dominance of a single species. Tropical 2002) by addressing the historical (Hille- ecologists have long pursued the question of brand, 2004; Mittelbach et al., 2007; Jansson, Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 53 Rodríguez-Castañeda, & Harding, 2013), phys- from seeds to adults (sensu Poorter, 2007). ical (i.e., Huston, 1999), and biological com- Understanding the evolution of the regenera- ponents of the latitudinal gradient in alpha and tion niche, expressed as ontogenetic changes beta diversity (Novotny et al., 2006). There are in functional traits (mediated by biotic interac- multiple explanations for the diversity patterns tions such as competition and plant-herbivore of tropical rain forests (i.e., Wright, 2002), relationships), is critical to understanding not but most fall into two general categories: only the causes of tropical forest diversity, but deterministic and stochastic models. Deter- the very nature of ecological communities. ministic models make emphasis on biological This knowledge goes beyond an academic or environmental factors predicting species discussion and is urgently needed to ensure abundance and diversity across space and time, that conservation, management, and restora- whereas stochastic models assume the influ- tion strategies coincide with the regeneration ence of random conditions affecting species requirements of the plant species involved. diversity. Niche divergence and environmental This goal is increasingly important considering filtering constitute two of the most common the multiple threats to tropical diversity (e.g., deterministic mechanisms explaining the low habitat loss, increased fragmentation, and the abundance of most plant taxa (Kraft et al., fast pace of climate change; see Bradshaw, 2015). In plants, resource partitioning in a nar- Sodhi, & Brook, 2009). These threats are row portion of the gradient, especially during changing the evolutionary rules of slow-growth the early phases of succession, results in niche taxa thriving in resource-limited conditions. divergence at different stages of the life cycle Shade tolerance, traditionally defined as (Ackerly, 2003; Poorter & Arets, 2003; Kraft, the capacity to survive long periods under deep Valencia, & Ackerly, 2008). Niche divergence leads to the differential distribution of plant shade, is a critical component of plant fitness species throughout their ontogeny into spe- under dense canopies (Valladares & Niinemets, cific environments (environmental filtering), 2008; Valladares, Laanisto, Niinemets, & reflecting the adaptive role of functional traits Zavala, 2016). This concept must be expanded matching organisms with the physical condi- to include the capacity to compete and grow tions of the successional gradient (Reich et in the shade. Shade tolerance is a complex, al., 2003; Wright et al., 2010; Sterck, Markes- multifactorial and continuous plant response teijn, Schieving, & Poorter, 2011; Spasojevic, directly related to life history adaptation. How- Yablon, Oberle, & Myers, 2014). Other deter- ever, our understanding of the extent and con- ministic explanations stress biological inter- sequences of shade tolerance is biased towards actions (i.e., predation and plant-herbivore instantaneous traits (i.e., photosynthetic rate), interactions), such the differential mortality which show faster responses to changes in light hypothesis or Janzen-Connell model (Janzen, conditions and are relatively easier to analyze 1970). Stochastic models include neutrality compared to traits that require more time to theory (Hubbell, 2001), dispersal limitation, stabilize (i.e., relative growth rate and alloca- and the role of the initial species composi- tion patterns). This bias also includes a few tion (Weiher et al., 2013) determining species –and usually the most abundant– ontogenetic abundance and diversity throughout succes- stages and life forms, such as tree seedlings, sion (Webb, Cannon, & Davies, 2008). Both and a limited range of microhabitats (i.e., the deterministic and stochastic factors come into understory of mature rain forests). However, a play to define a species´ regeneration strategy thorough understanding of shade tolerance and (regeneration niche), integrating physiological how it varies during the successional process trade-offs as well as biotic and abiotic filters is critical to comprehending plant performance (Weiher et al., 2013) into a life history tactic and the evolution of functional traits, and of habitat selection across forest succession, key concepts driving plant and community 54 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 structure and evolution, such as competition the disturbance clearly affects the direction of and succession (Valladares et al., 2016). succession, its duration, and the species com- Shade-adapted or shade-tolerant species position (Arroyo-Rodríguez et al., 2017). It is have a slower growth rate and a more restricted not my intention to cover all possible scenarios, capacity to respond to sudden environmental especially for human-disturbed forests, but to changes (Strauss-Debenedetti & Bazzaz, 1996). concentrate on mature tropical forests after a Since many tropical rain forest plants spend natural disturbance. This does not disregard some time in the shade, either as seeds, seed- the fact that succession takes place under other lings or saplings, juveniles, or adults, under- conditions; discussing these alternative sce- standing their adaptation to variable periods of narios is beyond the scope of this review. shade during succession is of vital importance to the management of the remaining tracts of THE MECHANISTIC BASIS tropical rain forests (Melo, Arroyo-Rodríguez, OF SHADE TOLERANCE Fahrig, Martínez-Ramos, & Tabarelli, 2013). My objective in this review is to analyze Shade tolerance is a multivariate, con- the ecological and evolutionary consequences tinuous trait: Shade tolerance facilitates sur- of shade tolerance within the context of tropi- vival under deep shade by maintaining the cal rain forest succession. To understand the carbon balance and the growth rate close to relevance of shade tolerance, it is necessary to zero (Kobe & Coates, 1997). Shade tolerant life review successional dynamics and the adapta- stages remain suppressed under low light by tion to successional gradients. This is ultimate- keeping a balance between high survivorship ly linked to theories explaining community and slow growth (the competition-colonization structure. This analysis will be restricted to the and growth-mortality trade-off). Under shade, rain forests sensu lato following the classic cri- the strategy of biomass allocation must target teria of Richards (1952). Following these cri- light interception and investment in defenses at teria, tropical rain forests consist of evergreen, different organismal scales, from physiological hygrophilous, and structurally complex vegeta- processes at the cellular level (i.e., activation tion up to 30 m in canopy height, and complex of phytochromes, instantaneous photosynthetic in life forms in addition to trees (epiphytes, responses, and regulation of morphological lianas, palms, understory herbs). Rain forests processes) to changes in plant architecture and are concentrated in tropical areas with abun- resource allocation (Valladares & Niinemets, dant rainfall, warm temperatures, and weak 2008). However, the concept of shade toler- climatic seasonality. In other types of tropical ance must incorporate not only the ability to forests, such as tropical dry forests, light limita- survive and endure shade but also the capacity tion does not determine the idiosyncrasy of the to grow and compete under other sources of successional process as strongly, compared to stress, such as drought or herbivory (Valladares factors such as water distribution, temperature, & Niinemets, 2008). Different degrees of shade and air humidity (Lebrija-Trejos, Pérez-García, tolerance, expressed at various stages in the Meave, Poorter, & Bongers, 2011). life cycle, could determine the phase in which My goal is to focus on the successional a species enters the successional trajectory. process of mature tropical rain forests after Valladares and Niinemets (2008) compiled a a natural disturbance (i.e., tree-fall gaps). list of traits conferring shade tolerance, from Other scenarios where succession is important cellular and leaf processes to whole plant include disturbed secondary forests, extractive scales. This list is not exhaustive and being reserves, abandoned agricultural fields, forest limited to the physiological measurement of fragments, and forest regeneration after con- shade tolerance at the plant level, it does siderable disturbances or catastrophes, such as not include the demographic or ecosystem- extensive fires and hurricanes. The origin of level implications of shade tolerance (but see Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 55 Valladares et al., 2016). Shade tolerance varies mass (Kitajima, 1994; Poorter & Rose, 2005) from instantaneous, leaf-level responses (i.e., is the best predictor of shade tolerance since it photosynthetic rate and stomatal conductance) integrates different selective pressures, such as to long-term, whole-plant responses such as dispersal mode, impacts of pathogens and her- changes in growth and resource allocation (i.e., bivores, and the overall capacity to cope with relative growth rate, fecundity, longevity, car- reduced resources. The combination of traits bon distribution). Since the distinction between directly related to biomass production (i.e., shade-tolerant and light-demanding species is seed mass, leaf dry mass per unit area, pho- rooted in the demographic trade-off of growth tosynthetic capacity, and whole plant relative vs. mortality, traits that more efficiently inte- growth rate), determines the capacity not only grate plant fitness over longer periods (i.e., to survive the shade but to grow and reproduce biomass production) should have a higher in low light (shade performance). predictive value for shade tolerance than traits There are different strategies to withstand expressed over shorter time scales (i.e., photo- the shade. Some plants complete their life synthesis rate, stomatal conductance). Hence, it cycle under shade (i.e., some understory palms, is necessary to extend the analysis of shade tol- ferns, and clonal shrubs) whereas other groups erance to demographic and ecosystem scales, can survive extended shade periods but need to facilitate the understanding of the influence to reach the canopy to reproduce. Wright et of this process on succession dynamics, and al. (2010) found that wood density explained ecosystem processes such as nutrient cycling more than 80 % of the variation in the distri- and carbon sequestration. bution of species along the growth-mortality Over short time scales the degree of shade trade-off. Wood density is an efficient integra- tolerance is traditionally measured using the tor of ontogenetic changes in plant allocation parameters of the photosynthetic light response strategies for most woody plants and is directly curve, such as light compensation point and res- related to demographic parameters such as piration rate (Valladares & Niinemets, 2008). mortality and relative growth rate (Poorter et Over longer time scales seedling or sapling al. 2008), although it might not as important survival, leaf lifespan and defense, and biomass for palms (Tomlinson, 2006) and lianas (Putz, partitioning become more important (Poorter, 1984), which have different growth strategies 2009; Kitajima & Poorter, 2010). More integra- and tissue properties relative to trees. There tive approaches incorporate fitness components is clearly a bias in measuring the responses at the whole-plant level, such as relative growth of short-term characters relative to plant traits rate and the amount of total biomass accumu- that integrate responses over extended periods lated as a function of light magnitude (i.e., and are more significant within a specific life the whole light compensation point of Baltzer stage in the life cycle of tropical species. For & Thomas, 2007). All these approaches are instance, Poorter et al. (2008) found leaf traits consistent with processes favoring survival in to be more important influencing the growth/ the shade, such as the maximization of carbon survival trade-off in seedlings, whereas wood gain under low light (i.e., achieved by adjust- density was more effective in predicting mor- ing plant architecture to decrease self-shading, tality rates in adult trees. Chazdon, 1986), increase in carbon storage Since multiple environmental factors which is eventually spent in pulses to favor affect suites of traits that determine growth and growth and reproduction in understory plants survival in the shade, shade tolerance should be during times of increased light, and investment analyzed as a multivariate trait or performance in traits favoring seedling survival in the shade trade-off (i.e., the growth-mortality trade-off, (increased leaf toughness, higher concentra- Gravel, Canham, Beaudet, & Messier, 2010) tion of secondary compounds, and longer leaf in which increased fitness under shade favors lifespan, Poorter, 2009). Of all these traits, seed growth and survival under limited light. As a 56 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 result, the plant´s capacity to respond to sudden sunflecks is responsible for the high variability light increases, specifically growth and sur- in the light regime of the forest understory. vival under high light, is restricted (MacArthur Under deep shade, plants maximize carbon & Levins, 1964; Kitajima, 1994; Wright et al., gain but minimize respiration costs (Kitajima, 2010). The evolution of shade tolerance centers 1994). To achieve this, respiration rates are low, on the growth-survival trade-off; this process congruent with low values of net photosynthe- maintains high species diversity because it sis rate. In addition, slow growth is punctu- favors niche divergence among species and ated by peaks in biomass accumulation when ontogenetic stages through the partitioning resources are favorable (growth and reproduc- the light resource at very fine scales (Wright, tion take place in pulses and sometimes out 2002). This concept is critical to predicting of population synchrony, Sylvester & Avalos, changes in community organization and suc- 2013). In shade tolerant plants a low reproduc- cession, and ultimately, ecosystem function. tive output is sustained over long periods, and The mechanistic basis of shade tolerance can combine with increased clonal propagation constitutes the foundation of classic works in (i.e., the genus Piper; Greig, 1993), increased plant ecophysiology (i.e., Chazdon & Fetcher, biomass allocation targeting higher light inter- 1984); for instance, photosynthetic induction ception (in the palms Geonoma cuneata and in response to sunflecks (Chazdon & Pearcy, Asterogyne martiana; Chazdon, 1985), high 1991), and cytochrome-mediated responses efficiency in leaf area distribution while mini- affecting plant morphology and gas exchange mizing leaf overlap and increasing light har- (i.e., Lee, Baskaran, Mansor, Mohamad, & vesting efficiency (in the palm Calyptrogyne Yap, 1996). These responses affect instanta- ghiesbreghtiana; Alvarez-Clare & Avalos, neous carbon fixation and long-term patterns of 2007), and increased allocation to storage roots biomass allocation and architecture. Sunflecks (i.e. Asterogyne martiana, Alvarez-Vergnani & represent transient, intermittent, and intense Avalos, in prep.) Although there is great varia- pulses of high levels of photosynthetically tion in seed mass, shade-tolerant species tend active radiation (PAR) that reach shaded envi- to have large seeds, storage cotyledons, low ronments without being filtered by the canopy. relative growth rate, low specific leaf mass, and The light energy received during a sunfleck low leaf area ratio (Kitajima, 2002). could be two orders of magnitude higher than The list of functional traits associated the average diffuse light levels of the forest with shade tolerance discussed by Valladares understory. Most sunflecks are extremely brief & Niinemets (2008) leads to a relative general (60 % of sunflecks lasted less than 30 s in shade-adapted syndrome, reflecting the covari- Chazdon & Pearcy’s 1991 study) and are clus- ant nature of many traits that determine sur- tered temporally. However, they account for up vival under low light. Valladares and Niinemets to 75-80 % of total daily photon flux density (2008) also showed that, although the shade tol- in the shade (Chazdon & Fetcher, 1984; Chaz- erance syndrome is consistent across species, it don & Pearcy, 1991). The energy contained in can vary following ontogenetic changes in sunflecks is responsible for most of the daily growth and mortality across size classes within carbon fixed by understory plants. Shade- changing environmental conditions and differ- tolerant plants require minutes to hours to be ences in growth forms (Dalling et al., 2001, fully induced. Once induction is reached, sub- Niinemets, 2006; Santiago & Wright, 2007; sequent sunflecks are utilized more efficiently. Wright et al., 2010). The final shade-adapted At the level of seconds to minutes, plants adjust phenotype, and thus the nature of the shade chloroplast orientation and the concentration response, depends on the covariation of mul- of photosynthetic enzymes and electron carri- tiple morphological and physiological traits as ers to benefit from sunflecks (Way & Pearcy, plants move across different ontogenetic stages 2012). The spatial and temporal distribution of and light environments. The list of Valladares Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 57 and Niinemets (2008) overlooked some key shade adaptation and are inherently economic traits, such as delayed greening (Prado, Sierra, since they implement a cost-benefit analy- Windsor, & Bede, 2014). Delayed greening is sis (i.e., the survival-adaptation trade-off) to the production of new leaves that are red or explain how plants allocate limited resources light green, tender, and have variable levels of to adapt to changing environments (Diaz et al., secondary compound defenses. Leaf greening 2004; Reich, 2014; Lohbeck et al., 2015). The takes place once the leaf is fully expanded, and sun-shade dichotomy has been the classic sub- mechanical defenses replace chemical defenses ject for explaining species differences in forest (Kursar & Coley, 1992). This strategy is fre- succession, light acclimation, photosynthet- quent in shade-adapted, understory species, ic performance, and evolution of life-history which can sustain heavy herbivore damage trade-offs, but as discussed above, it is an over- during leaf expansion, but since the lost leaf simplification of the shade tolerance continu- area lacks nitrogen and photosynthetic com- um. These models are not mutually exclusive pounds, the strategy is relatively cost-efficient but revolve around the concept of changing (Clark & Clark, 1991). resource availability and the time necessary to In summary, shade tolerance has a genetic express adaptation. In other words, a plant will basis (Gommers, Visser, St Onge, Voesenek, adapt to shade depending on the frequency in & Pierik, 2013), is subject to natural selection, which it experiences light limitation throughout is context-dependent, can be understood as a its life cycle (i.e., the opportunity for adapting shade-adapted syndrome, and varies across to low resources depends on the frequency in space and time. The final response to shade which a species experiences a limited environ- will depend on the combination of other stress ment during its life cycle). The frequency of factors such as drought (Markesteijn, Poorter, exposure to shade determines the magnitude Bongers, Paz, & Sack, 2011), nutrient deficien- of plasticity, measured as different phenotypes cies (Niinemets, 2010a), herbivore damage expressed over a physical gradient over time, as (Boege & Marquis, 2005), and competition reflected in the trajectory of the reaction norm. (Wright, 2002). The time in the ontogenetic trajectory where the observation is made is critical in determin- UNDERSTANDING ing the categorization of high vs. low plastic- SHADE TOLERANCE ity, assuming that responses have stabilized in the new environment. Some species require Since shade tolerance is a multivariate more time to adapt to a sudden change in light, trait, an integrative approach considering life which does not necessarily indicate less plastic- history trade-offs and multiple morphological ity relative to a species expressing adaptation and physiological traits (or functional trait more quickly (Valladares, Sánchez-Gómez, & spectra, following Lohbeck et al., 2015) is Zavala, 2006). mandatory. Different theories serve to explain In the following paragraphs, I review four the evolutionary development of these trade- of the most common conceptual models relevant offs (Diaz et al., 2004). These theories form the to understand plant adaptation to limited condi- foundation of classic models of plant evolu- tions, one of them being shaded environments. tion, resource allocation, and forest succession, serve to organize the complexity of shade toler- Resource availability hypothesis or opti- ance, and help to comprehend overall plant per- mal defense: This model represents the most formance and growth strategies within a given general explanation for resource allocation context of resource availability (i.e., Tilman, under variable environmental conditions. Orig- 1994; Pacala & Rees, 1998). Any model aiming inally proposed to explain differences in the to explain plant growth in response to changing quality and quantity of defenses against her- resources can be used to explain differences in bivores (i.e., Coley, Bryant, & Chapin, 1985; 58 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 Bazzaz, Chiariello, Coley, & Pitelka, 1987), is analogous to the mechanism controlling the the model serves as a suitable framework to generation of defensive compounds. describe differences in growth related to shade The work of Kitajima (1994) on Barro tolerance. It is compatible with more general Colorado Island, Panama, was one of the first hypotheses based on the economics of trade- to analyze in detail the optimality trade-off offs in resource allocation (i.e., optimal alloca- under limited resource conditions. Kitajima tion theory and allometric biomass partition (1994) found that seedling survival in the shade theory, see McCarthy & Enquist, 2007; or the was related to morphological traits increasing “worldwide leaf economics spectrum”, Wright defense against herbivores and pathogens (high et al., 2004). The model proposes that the level leaf construction costs, low specific leaf mass, of resources available –and the type of resource low leaf area ratio, high root-to-shoot ratios, limiting plant growth– determines the kind and and low whole-plant carbon gain). The expres- amount of defenses. The quality of defenses sion of these traits supports the dichotomy and their metabolic cost is a function of the between shade-tolerant vs. light-demanding amount of resources available to maintain species because it represents a trade-off in car- existing defenses and generate new ones. For bon investment. In an environment where over- instance, if resources are in short supply, plants all resources (light and nutrients) are scarce, will invest in permanent, nitrogen-based, quan- plants allocate more to defenses than plants titative defenses that have high initial costs but specialized in exploiting high-light conditions that are more efficient since they require less and nutrient pulses in gaps. Maternal effects maintenance and synthesis (in other words, are on seedlings also determine seedling structure more cost-effective). In contrast, if resources and function. Many shade-tolerant species have are abundant, plants will invest more in mobile, large seeds, and thus, are able to produce large qualitative defenses that have low initial costs seedlings that require more structural support, but require continuous synthesis and mainte- and have low leaf area ratio and specific leaf nance (i.e., alkaloids, cardiac glycosides). The mass (Poorter & Rose, 2005). level of investment in defenses varies with The survival capacity of shade-tolerant resource quality, the intensity of herbivory seedlings is associated with structural charac- (defenses can be induced; Schaller, 2008), and ters providing resistance to physical damage, plant ontogeny (Boege & Marquis, 2005). such as stem and leaf tissue density (Alva- Similar principles apply to shade toler- rez-Clare & Kitajima, 2009). Shade-tolerant ance. If resources are scarce (i.e., limited light), species are structurally better defended than plants will invest in long leaf lifespans, more light-demanding species, even in very young efficient crown architectures with less leaf over- seedlings, although this comes at a price of lap, and a decreased relative growth rate. Plants slow growth and slow biomass accumulation also grow in pulses (will accumulate resources (Kitajima, 1994). Structural traits are corre- under shade and will spend those resources lated with large seeds and large seedling sizes, in growth and reproduction when light condi- the presence of storage cotyledons, carbohy- tions improve). The relative importance of this drate reserves associated with shade tolerance allocation strategy, and thus, of the intensity (Wright et al., 2010), and the long-term surviv- of shade tolerance, varies with ontogeny. By al of seedlings and saplings under shade (Kita- increasing in size, plants move into a different jima & Poorter, 2010). Stress tolerance is thus light environment while changing simultane- associated with the level of available resources ously their allocation relationships (i.e., invest- and explains patterns of seedling survival in the ment in reproductive structures could be more shade, and strategies facilitating initial habitat important than increased light interception). colonization and determining patterns of spe- In summary, if light is limited, plants will cies distribution and abundance in gradients of express the shade-adaptive syndrome, which light availability. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 59 The carbon gain model: This model was positioning plants inside a triangle determined proposed by Givnish (1988) who provides by the relative importance of these three fac- a holistic overview of how plants adapt to tors. For instance, a plant exposed to a habitat changing light conditions. Givnish (1988) sug- with a high level of disturbance and low stress gested the examination of integrated responses will likely develop the “R” or “ruderal” strat- at the organismic level to changes in the light egy, in other words, will evolve to benefit from environment, rather than examining minute, frequent disturbances characterized by initial isolated physiological processes at the leaf low competition and lack of environmental level. The fundamental concept of Givnish stresses. The classification could include many (1988) is that plant responses are context- combinations of the CSR continuum, giving dependent (including the physical and the room for niche divergence along these three biological environment) and are adaptive in axes. One appealing attribute of this model is the sense that natural selection will favor the its simplicity and generality. Grime accom- maximization of carbon gain in congruence modates for a continuum of responses since with resource quality following a cost/benefit the three main axes correspond to continuous model. In other words, maximization of car- variables, and thus, the position of a plant bon gain under low light requires a strategy species within the triangle can be temporary, of resource investment with adaptive benefits giving room for ontogenetic adaptation and in the long run. Under restricted light, plants size-mediated (or allometric) responses (Wein- will show a shade-adapted phenotype charac- er, 2004). This possibility accommodates some terized by slow growth, low photosynthetic of the major criticisms of this model (Craine, rates, low leaf area ratio, and low specific leaf 2007), including the static categorization of mass. This phenotype is more advantageous plant strategies and the lack of consideration of compared to that of light-demanding species scenarios in which plants move out of the shade under shade. Givnish (1988) emphasizes the and acquire a different strategy. The intensity examination of the impact of individual func- of competition, environmental stresses, and tional traits on whole-plant performance, open- disturbance also covary across space and time. ing the possibility of analyzing integrated Finally, transient combinations of this model, responses to multiple stressors, not only to low as dependent on ontogeny, could lead to accli- light. Since the carbon gain model integrates mation to different stress factors, in addition responses over time and considers whole plant to light acclimation (Kozlowski & Pallardy, allocation and allometric strategies, the con- 2002). The debate about the functionality of sistency of this model with instantaneous and the CSR approach still continues (Craine, 2005, short-term measurements of photosynthetic 2007; Grime, 2007). performance is limited. Resource competition model: This model CSR (competitor, stress tolerator, ruder- is based on Tilman´s (1977, 1990) notion of al) model: This model links a plant´s strategy, asymmetric or exploitative resource competi- or syndrome, with functional performance and tion. As in the case of Givnish (1988), Tilman´s population biology, and recognizes strategies work was generated in temperate areas in sim- of evolutionary specialization associated with pler systems under controlled conditions very stress responses (Grime, 1989). Following clas- different from tropical communities. However, sic CSR theory (competitor, stress tolerator, the conclusions of such studies can be applied ruderal) shade tolerants are classified as “stress to the analysis of succession in tropical rain tolerators”. However, Grime (1977) distin- forests, especially in the early successional guishes several alternative strategies depend- stages when seed dispersal and seedling spatial ing on the relative importance of competitive distribution are critical. Given two competing interactions, stress conditions, and disturbance, species, the winner is the species able to survive 60 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 with the least resource requirements. Species mechanisms responsible for the maintenance diverge in resource use through the expression of a high level of species diversity (Rosindell, of different tolerances for a lower amount of Hubbell, & Etienne, 2011). By creating hetero- the resource compared to competing species. geneous conditions and influencing resource Furthermore, a species can be a good competi- distribution (especially of light and nutrients; tor for one resource, but it might not be as good Prescott, 2002), gaps influence stochastic and a competitor in other aspects of the competitive biotic factors, facilitating the coexistence of performance, such as dispersing seeds or fight- species with different regeneration niches that ing predators. Different species express a vari- specialize in a narrow portion of the gradi- ety of tolerances for lower amounts of available ent (Poorter, 2007; Wright et al., 2010). This resources (see Rees, Condit, Crawley, Pacala, mechanism constitutes the “gap hypothesis” & Tilman, 2001). There is an evident simili- of species diversity (Schnitzer, Mascaro, & tude of this model with the recruitment limita- Carson, 2008). Therefore, multiple regenera- tion and neutral models of species diversity tion niches are possible given the high number in tropical rain forests (Hubbell et al., 1999). of plant species, ontogenetic stages, and life Competing species balance out their shortcom- forms distributed over a complex environmen- ings by exploiting alternative resources along tal gradient, which filters species based on their the environmental gradient. In tropical rain capacity to withstand gaps of different sizes forests, the limiting common resource is light. and shaded understories. Although variation Divergence takes place because plants differ in the regeneration niche is clearly continuous in their light limitation and because plants (i.e., Augspurger, 1984), the literature (i.e., develop alternative adaptations to decreased Swaine & Whitmore, 1988) has traditionally light, such as increased allocation to light inter- divided plant species into two opposing groups: ception, increased constructions costs, or better shade-tolerant (late successional or climax) and herbivore defenses. As discussed by Kitajima light-demanding species (early successional, & Poorter (2008) the competitive advantage gap-dependent, shade-avoiders, nomads, or of one species over another along the light pioneers). This classification is biased towards gradient is affected by stochastic factors (gaps, the segregation of tree seedlings along the light mechanical damage) and trade-offs allowing gradient based on physiological and demo- plants to exploit different levels of the light graphic trade-offs between slow growth and resource at the seedling stage (and throughout low mortality in the shade vs. fast growth and ontogeny in general). Light competition forms high mortality under sun (Kitajima & Poorter, the basis of divergence along the regeneration 2008). This classification has rarely considered niche (Pacala & Rees, 1998), but other limit- other life stages and life forms and regards light ing resources, such as nitrogen availability, as the primary limiting factor, although chang- affect plant performance and influence how es in forest structure affect the distribution of plants partition the light resource (Campo & other critical plant resources such as nutrients Vázquez-Yanes, 2004). and water, vapor pressure deficit, humidity, and temperature (Guariguata & Ostertag, 2001). THE REACTION NORM APPROACH Although over simplistic, the classification has (GRADIENT ANALYSIS) AS practical significance since it guides forest res- A FRAMEWORK TO UNDERSTAND toration and management practices. However, SHADE TOLERANCE it should not override the dynamic nature of a species´ regeneration niche, which follows The rich environmental complexity of an ontogenetic trajectory over time (Clark & tropical rain forests rests on the temporal and Clark, 1992). Many species could adjust or spatial dynamics of the aperture of canopy gaps, even reverse their regeneration niche during which constitute one of the main ecological their ontogeny (Dalling et al., 2001; Poorter, Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 61 Bongers, Sterck, & Wöll, 2005); whereas other Enquist, Weiher, & Westoby, 2006), and facili- species have intermediate light requirements tating the exploration of plastic responses and benefit from moderate light increases early without having to replicate genotypes across in life (Augspurger, 1984; Wright, Muller- environments (Valladares et al., 2006). In this Landau, Condit, & Hubbell, 2003). There is manner, individual phenotypic variation is evidence for ontogenetic concordance (i.e., the directly connected to functional performance fundamental regeneration niche is conserved in a given environment, and thus, is tied to throughout ontogeny; Gilbert, Wright, Muller- natural selection, and eventually, adaptation. Landau, Kitajima, & Hernandéz, 2006; Poorter, Since functional performance changes across 2007; Kitajima & Poorter, 2008; Avalos & ontogeny and varies following spatial and Mulkey, 2014), although the expression of temporal gradients in resource distribution, it ontogenetic niche shifts could be the standard is evident that ecophysiological studies must (Poorter et al., 2005; Gilbert et al., 2006), consider integrated responses at the whole despite the paucity of data on the variation in organismic level to determine how sets of func- microsite conditions and how it determines tional traits co-vary following a reaction norm the growth responses of ontogenetic stages across the successional gradient, and how the other than seedlings and adults (Martínez- adjustment capacity is affected by ontogeny Ramos, Alvarez-Buylla, & Sarukhan, 1989; (Givnish, 1988). This research implies consid- Clark & Clark, 1992; Niinemets, 2006; erable logistic challenges for species with long Wright et al., 2010). lifespans and complex spatial distributions, and The regeneration niche is multifactorial for growth forms not easily accessible for long- and is contingent on the combination of mul- term observation (i.e., canopy trees, lianas, and tiple functional traits whose expression var- epiphytes; Clark & Clark, 1992; Condit, Hub- ies ontogenetically (Schlichting & Pigliucci, bell, & Foster, 1996; Hubbell et al., 1999). 1998). The application of the reaction norm approach or gradient analysis (i.e., the exami- LIFE CYCLE SUCCESSIONAL nation of the temporal phenotypic expression TRAJECTORY MODEL OF of genotypes across environmental gradients) GÓMEZ-POMPA & VÁZQUEZ-YANES provides a suitable framework to understand how plants adapt to complex environments and Communities are complex entities. Since how suites of functional traits vary with ontog- organisms of different taxonomic groups eny and environmental conditions. Reaction are assembled along multiple environmental norms serve to analyze the expression of plas- scales, their delimitation in the field can be ticity and integrate complex suites of functional controversial. The fundamental question is traits that shape particular life history strategies “what is a community”? A community is a sys- (Weiner, 2004). tem composed of taxonomic entities interact- Since the seminal paper of Arnold (1983), ing through the exchange of resources across plant ecophysiologists have analyzed adapta- spatial and temporal gradients (Vellend, 2010), tion as the integration of suites of functional which determines the dynamics of species traits responding to environmental changes colonization, establishment, and loss. Resource and determining physiological performance limitation can be represented by light, nutrient (survival, growth, reproduction, and over- availability, or water seasonality. Resource gra- all fitness). Natural selection filters the most dients make communities more than a random advantageous phenotypes under a given set of collection of artifacts, although their limits environmental conditions resulting in adaptive are difficult to determine since they are open evolution. This represents the central mecha- systems and involve a variety of organisms nism linking functional traits with evolution- from different taxonomic groups. The study of ary biology (West-Eberhard, 2003; McGill, communities goes to the heart of ecology by 62 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 observing the factors controlling species diver- deterministic (i.e., Janzen-Connell model and sity, distribution, abundance, and composition the regeneration niche) and stochastic models (Vellend, 2010). (i.e., dispersal limitation and the neutral theory Many variables interact in the analysis of biodiversity and biogeography, Rosindell of community assembly, such as site-specific et al., 2011) to predict species diversity and factors, the type and intensity of previous land community structure. It is more complex than use, the nature and dynamics of the distur- previous models because it considers the wide bance regime, and the nature of the landscape variation in life cycles and successional strate- matrix (Guariguata & Ostertag, 2001; Chaz- gies (see Norden et al., 2015). These ideas are don, 2008). All these factors influence species not new (see Budowski, 1965; Gómez-Pompa composition and abundance, as well as the & Vázquez-Yanes, 1981), but have been incor- distribution of functional traits, variation in porated into more quantitative successional population dynamics, and competitive interac- models (Acevedo, Urban, & Shugart, 1996; tions. The complex interaction between these Ackerly & Cornwell, 2007) and are gain- factors makes it challenging to identify the ing acceptance as the role of succession as a dominant patterns explaining the nature of the selective pressure and species filtering mecha- community, especially through the analysis of nism is better understood (Webb, Ackerly, succession (but see Chazdon, 2008, 2014), and McPeek, & Donoghue, 2002; Letcher et al., makes it difficult to predict the final succes- 2015; Lohbeck et al., 2015). sional trajectory in tropical rain forests (Peter- Gómez-Pompa & Vázquez-Yanes (1981) son & Carson, 2008). However, to understand proposed a model of tropical rain forest suc- the relevance of shade tolerance it is necessary cession based on the characteristics of the to comprehend the fundamental mechanisms of life history strategies of the dominant spe- tropical rain forest succession. cies at different stages of forest regrowth Models explaining tropical rain forest suc- (Fig. 1). In this model of life cycle patterns, a cession are rooted in the first theories of mod- species enters a successional stage depending ern ecology proposed to explain the nature of on the match between environmental condi- plant communities (i.e., Gleason, 1926; Cle- tions and the species’ physiological amplitude ments, 1936). Ecologists trained in temperate as influenced by ontogeny. Species arrive by areas (for instance, Richards, 1952) extended dispersion or advanced regeneration (Martinez- these ideas to the analysis of forest succes- Ramos & Soto-Castro, 1993), stay for variable sion in the tropics. They expected to observe periods, some reproduce and complete their life predictable successional trajectories charac- cycle within a stage or across several stages, or terized by a progressive transition through are excluded without becoming reproductive. identifiable stages towards the climax, with This model follows the Gleasonian view of sets of species replacing each other and parti- forest succession, with continuous trajectories tioning environmental resources (in agreement and many variable scenarios (exclusion, estab- with Clement´s concept of communities as lishment, and permanence until reproduction “superorganisms”). Despite the prevalence and spanning one or several stages). Gómez-Pompa popularity of Clement´s ideas, the Gleasonian & Vázquez-Yanes (1981) illustrate 21 differ- view of tropical plant communities, character- ent regeneration strategies, but there could be ized by continuous variation in species com- many more. This approach emphasizes eco- position over time (Gleason´s “individualistic” logical roles (i.e., functional diversity) more view of continuous changes in community than taxonomic identity and can accommodate composition across environmental gradients) is redundancy since many species could occupy closer to the high dynamism and rapid species similar regeneration niches (Rosindel et al., turnover typical of tropical succession (Hub- 2011). This makes it congruent with mod- bell, 1997). The current synthesis combines els explaining community assembly based on Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 63 Fig. 1. Ontogenetic trajectories reflecting different regeneration niches following the progression of tropical rain forest succession over time. Each line represents a life cycle from seed to reproductive adult (some trajectories do no reach the reproductive stage and die at the seed, seedling or juvenile stage -black lines-). A complete life cycle takes place when the ontogenetic trajectory reaches the adult phase. Trajectories A and B are typical of pioneer species that complete their life cycle within the gap. Trajectory C reaches the juvenile phase but dies before reproduction. Trajectory D regenerates in the gap but persists until the building and mature phases (old-growth pioneer). Trajectory E is typical of shade tolerant species (these species germinate and complete their life cycle in the shade in the mature phase). F and G correspond to cryptic pioneers that start in the gap or building-phases but persist until the canopy closes, reaching reproductive stages in the mature phase. Some old-growth pioneers are functionally equivalent to cryptic pioneers (F and G). Some species reach different stages by dispersion but die before reproduction. Arrows above indicate that forest succession could revert to previous stages. Modified from Gómez-Pompa & Vázquez-Yanes (1981). stochastic factors (i.e., Hubbell, 1997). In fact, the evolutionary consequences of variation in stochastic and deterministic factors interact functional and demographic traits, and how to determine species composition throughout successional habitats drive the evolution of succession and could have different effects niche divergence (Letcher et al., 2015). Similar depending on the successional strategy (pioneer to Gleason´s ideas of continuous and complex vs. shade tolerant or old growth specialists) and community assembly, the life cycle model has the species abundance (rare species vs. second not been given the importance it deserves in growth specialists; Kraft et al. 2008; Norden et understanding forest regrowth. Current ideas al., 2017). Certain stages could vary in duration incorporating the filtering role of succession while being dominated by specific groups, such on functional traits (Kraft et al. 2015) are latent as lianas or palms (Schnitzer et al., 2008). The in this model. model also integrates different filters influenc- In mature tropical rain forests succes- ing species abundance (see Kraft et al., 2015), sion starts with a canopy disturbance (i.e., a including physiological limits to withstanding tree fall), which significantly increases the the physical environment in one point of the understory light levels and creates light envi- trajectory, competitive exclusion, within-site ronments similar in magnitude to those at the heterogeneity favoring species co-existence, surface of the canopy. This is the “initiation and interactions with pathogens, predators, pol- phase” of Oliver & Larson (1996), also called linators, and dispersers. The model integrates the “gap-phase” stage of the successional cycle 64 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 by Denslow (1987). Canopy gaps increase light increased photodamage or competition with and create pulses of nutrients as the subsoil fast-growing pioneers. is exposed by the roots of the fallen tree and As the gap fills, the forest reaches the debris produce an aggregation of decomposing “building, stand-thinning, or stem exclusion matter (Schaetzl, Burns, Johnson, & Small, phase” of pioneers following Oliver and Larson 1988). Species partition the different parts of (1996) and Chazdon (2008). In this stage, the a gap. Brandani, Hartshorn, & Orians (1988) canopy, initially composed by pioneer species, found higher seedling diversity in the root gets denser as pioneers reach their reproduc- section relative to the bole and crown areas tive stage and pre-empt the light conditions of the fallen tree (higher seedling density was for additional pioneer seedlings and favor the concentrated in the crown area). Eighty-seven establishment of shade-tolerant species charac- percent of the tree species at La Selva, Costa terized by large seeds, high wood density, high Rica, were found as seedlings in the 51 gaps leaf specific mass, low photosynthetic rates, examined in this study. Most species started as and slow growth. Shade tolerants increase in colonizers rather than as gap survivors, show- abundance and dominance, leading to a gradual ing that the majority of species require a gap exclusion of pioneers and lianas. As succession to regenerate. Partitioning of microsite condi- progresses, there is a change in the species pool tions also takes place in the understory under as well as in the diversity of functional traits deep shade (Montgomery & Chazdon, 2002) (Boukili & Chazdon, 2017; Plourde, Boukili, demonstrating that species divergence along & Chazdon, 2015). Functional diversity peaks fine resource gradients is not limited to gaps at intermediate stages of succession when pio- (Svenning, 1999). neers still co-occur with shade-tolerants (Mus- In addition to the strong colonization abil- carella et al. 2016). ity by pioneers (small seeds produced and dis- During the “mature or old growth stage” persed in large numbers, low wood density, low of forest succession, shade-tolerants reach the canopy and become dominants in this stra- specific leaf mass), recently dispersed seeds, tum. Shade tolerants transition from the light- and seeds stored in the seed bank, begin to deprived conditions of the forest understory to germinate, creating a carpet of seedlings which the canopy, which implies changes in biomass eventually leads to high rates of seed and seed- allocation, growth strategies, and physiologi- ling mortality (Schupp, Howe, Augspurger, & cal adjustment (i.e., niche shifts, Niinemets, Levey, 1989). High levels of seedling mortality 2010b). The documentation of the mecha- do not impede the dominance of pioneers in the nisms of ontogenetic adjustment at this point first stages of gap colonization. However, over is still scarce (but see Poorter et al., 2008; time, seed and seedling mortality increases in Wright et al., 2010). the shade due to the combined action of limited Cryptic pioneer species present an exam- resources (light and nutrients) and pathogens ple of the transition from pioneers to shade (Augspurger, 1984). Some of the species arriv- tolerants. Seedlings of the palm Euterpe preca- ing first include cryptic pioneers, which start toria are abundant in light gaps and disturbed germination and establishment under disturbed environments (Avalos, Fernández, & Engeln, conditions but can withstand shade once the 2013) but remain suppressed for long periods canopy closes. Here, increased light and nutri- under shade once the canopy closes. This ents lead to a switch in biomass allocation, palm becomes reproductive in 93-158 years architecture and growth (Niinemets, 2010b). In under shade (Peña-Claros & Zuidema, 2000), these species, sudden peaks in resources shift but under semi-open conditions (clearings and the regeneration niche from shade tolerance to forest edges), reproductive stages are reached gap specialization. Some shade-tolerant species in just four years (Avalos, 2016). Variations eventually get excluded from this phase due to in the regeneration trajectories could depart Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 65 from this expected successional sequence. forest understory, and it is usually assumed that Typically, pioneers invade recent gaps, but are canopy plants are not light-limited. In fact, the eventually replaced by shade-tolerants when leaf distribution in the canopy can be highly the canopy closes and when the canopy of heterogeneous, which in turn affects light dis- short-lived pioneers creates increased shade. tribution from the canopy surface down to Long-lived pioneers are finally excluded due to the forest floor (Kitajima, Mulkey, & Wright, shade conditions prevalent during late succes- 2004; Avalos, Mulkey, Kitajima, & Wright, sion and are replaced by long-lived trees that 2007). Since forest canopies are multi-layered, started as shade tolerants (Gómez-Pompa & light distribution follows the complex strati- Vázquez-Yanes, 1981). Long-lived shade toler- fication of leaf mass. Light extinction takes ants dominate the canopy during late stages by place immediately below the canopy surface, positioning their crowns in sun-exposed sites. decreasing to 4-9 % of the light magnitude It is likely that these species express reverse measured at the canopy surface in places domi- ontogenetic niche shifts once light conditions nated by lianas (Avalos, Mulkey, Kitajima, & improve (Clark & Clark, 1992; Dalling et al., Wright, 2007). Liana foliage has different opti- 2001). More long-term data on the growth per- cal properties relative to leaves of supporting formance of life stages intermediate between trees (high leaf absorbance, low transmittance; seedlings and adults is necessary to fully under- Avalos. Mulkey, & Kitajima, 1999) creating stand the physiological mechanisms involved light levels similar to those of the understory in the transition from strictly shade tolerant (Chazdon & Fetcher, 1984; Chazdon, Pearcy, into canopy plants (but see Wright et al., 2010). Lee, & Fetcher, 1996). This light deprivation The life cycle trajectory model of Gómez- could take place over a few days, as the liana Pompa and Vázquez-Yanes (1981) reflects produces a thick monolayer of foliage on top the continuity of the pioneer-shade-tolerant of canopy trees (Avalos, Mulkey, Kitajima, & classification, and the diversity of alternative Wright, 2007). The response of canopy trees regeneration strategies entering into the succes- to very dark microhabitats created by lianas is sion continuum. It also offers a framework to still poorly understood, but it entails decreased understand other aspects of functional ecology, leaf lifespan for the host tree, and changes in such as the distribution and diversity of func- biomass allocation to increase leaf production tional traits, and how the strength of herbivore- in sites of the tree crown with more access plant interactions, competition, and parasitism to light. The distribution of leaf masses from varies during succession. For instance, more different species and life forms follows a tem- diverse plant assemblages give room for more poral trajectory determined by leaf phenology diverse plant herbivores eventually leading to and changes in forest structure (Sapijanskas, more diverse chemical and structural defenses. Paquette, Potvin, Kunert, & Loreau, 2014) favoring niche divergence through functional AREAS FOR FUTURE RESEARCH complementarity and plasticity. Adaptation to changing light conditions within the canopy, Increased representation of microhabi- not to mention the functional and community- tats, ontogenetic stages, and life forms where level consequences of adaptation to different shade tolerance is critical: Analogous to the canopy positions, remains an open area for concentration of research on shade tolerance research (but see Cardelús & Chazdon, 2005). in the rain forest understory and on a limited Other canopy life forms such as epiphytes show range of life stages and life forms (tree seed- little changes in structure and function when lings and saplings), little attention has been exposed to deep shade in comparison with high given to shaded environments in other parts light environments (Benzing, 2008). Research of the forest profile, such as the canopy itself. on epiphyte adaptation to the canopy is biased The canopy controls the light conditions of the towards sites of greater light exposure, whereas 66 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 there should be greater amplitude of research (i.e., photochromic and hormonal control of on epiphytes in different light preferences growth responses). because of their considerable species diversity (20-25 000 species, mostly concentrated in the Refined analyses of the regeneration tropics; Benzing, 2008). Instead, shade toler- niche and ontogenetic niche shifts based ance in epiphytes is poorly understood (Zotz on functional traits: Many functional traits & Hietz, 2001), and its effect tends to be con- contribute to shade tolerance (Valladares & founded with photoinhibition or water stress. Niinemets, 2008) and are partially responsible Epiphytic bromeliads, for instance, evolved for the final expression of a shade-adapted syn- from sun-exposed environments and moved drome. Water and nutrient availability, compet- up colonizing the top of the canopy, whereas itive interactions, and resistance to pathogens shade adaptations evolved secondarily, facili- and herbivores also influence shade adaptation. tating the colonization of understory habitats As mentioned above, traits related to the capac- (Crayn, Winter, & Smith, 2004). The ecological ity to grow under shade are efficient proxies and evolutionary analysis of shade tolerance in to measure the degree of shade tolerance. Our epiphytes is long overdue. understanding of the importance of specific functional traits, and combinations of suites Determination of the molecular and of functional traits, depends on the imple- genetic basis of shade tolerance to under- mentation of long-term demographic studies stand its evolution: Although many functional based on population models having sufficient traits are very labile, and thus, show almost replicates of ontogenetic stages, light condi- immediate adaptation to sudden changes in tions, and life forms, while being able to cor- resource distribution (i.e., stomatal conduc- rect for phylogenetic bias. Such studies face tance and photosynthetic induction), their significant challenges in environments where variation over evolutionary time is very conser- most species are rare, hyper-dispersed, widely vative (Webb, 2000; Donoghue, 2008; Letcher distributed across the forest profile, and where et al., 2015). This explains why species with many ontogenetic stages are not amenable similar ecological requirements are phyloge- for experimental manipulation or have very netically related and tend to select similar habi- long duration. The studies of Clark and Clark tats (Webb, 2000). Phylogenetic analyses are (1992), Poorter et al. (2008) and Wright et al. critical to understand the variation in species (2010) are steps in the right direction. These divergence in functional traits and their role in studies compared multiple sites, analyzed long- facilitating adaptation to heterogeneous condi- term data (including metadata), and determined tions. This can lead to the selection of suitable how functional traits covary within and among study systems to explore the phylogenetic life stages in tropical forest plants. Wright et al. relatedness of plants exploiting shaded envi- (2010) analyzed the survival-growth trade-off ronments. This has been attempted (Letcher et using multiple functional traits in 103 tree spe- al., 2015), but results are still inconclusive due cies, from seedlings to adults, within the 50 Ha to the high lability of functional traits; existent plot of Barro Colorado Island in Panama. As phylogenetic and successional studies often plants traverse different light conditions from consider such traits as static or particular to a the understory to the canopy and transition set of species without accounting for ontog- from seedlings and saplings to adults, the co- eny (Letcher, 2010; Chazdon et al., 2011). variant nature and relative importance of suites Finally, the molecular regulation of shade of functional characters changed, providing tolerance requires more attention (Gommers et evidence for ontogenetic niche shifts in many al., 2013), especially regarding the expression species (congruent with the general results of of physiological pathways influencing plant Clark & Clark, 1992, and Poorter et al., 2008). morphology and overall biomass allocation Wright et al. (2010) reported a strong effect of Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 67 the growth-survival trade-off in saplings when succession (Swenson et al., 2012). Demo- amplifying the differences by including the graphic studies could contribute significantly growth rates of the fastest growing individuals to our understanding of forest succession by and the mortality rates of the slowest grow- integrating functional traits with the carbon ing individuals. For trees, the growth-survival accumulation capacity in critical ontogenetic trade-off was weak. In addition to the changing stages and by explaining the impact of dif- matrix of functional traits through ontogeny, ferences in resource quality on demographic Wright et al. (2010) conclude that most func- parameters throughout ontogeny. Finally, since tional traits function as proxies of the underly- the phylogenies of many plant groups are ing demographic factors that determine carbon particularly well known, we should integrate allocation (such as the intrinsic rate of increase) phylogenetic information in the analysis of which are more likely to affect adaptation to shade tolerance. Despite the ample range of changing light conditions. plant species included in Poorter et al. (2008) The above-mentioned studies stress the and Wright et al. (2010) studies, there was no importance of ontogenetic variation in func- correction for phylogenetic bias. In contrast, tional traits. Traits that are critical in a given the Swenson et al. (2012) study compared the ontogenetic stage may have a negligible effect mature forest within the 50 Ha plot of Barro at another stage. Poorter et al. (2008) and Colorado Island in Panama with the much dis- Wright et al. (2010) report a strong nega- turbed, less diverse, and much smaller Luquillo tive correlation between wood density and Forest Dynamics Plot (16 Ha) incorporating the growth rates across all ontogenetic stages, effect of phylogenetic and functional structure however some functional traits have propor- on community assembly. They found species tionally more importance in seedlings than turnover and functional diversity to be driven in mature trees, such as leaf-level traits (i.e., mainly by environmental filtering, whereas specific leaf mass). In dicotyledonous spe- phylogeny had a weak effect. However, their cies with secondary growth, wood density sampling was limited to plants > 1 cm in diam- can reflect the capacity to resist mechanical eter, and thus, seedlings were not considered. and herbivore damage and the plant´s capac- ity for biomass accumulation (Chave et al., Application of a systems perspective 2009), and thus, the plant´s shade tolerance integrating a broader geographical range (Alvarez-Clare & Kitajima, 2009). Despite its and long-term data to facilitate comparisons clear importance and its wide use as a proxy and the development of a new synthesis: of plant growth, wood density is not necessar- A holistic perspective integrating a systems ily a universal predictor of performance across approach to succession and incorporating all life forms. The variation in wood density is many of the emerging properties of biological a poor predictor of biomass accumulation and systems (self-organization and uncertainty) is carbon sequestration in understory, subcanopy necessary to identify common patterns and and canopy palms (Cambronero, Avalos, & account for significant variability in the succes- Alvarez-Vergnani, 2018) and is unlikely to be sional processes. This requires more interaction an efficient functional character related to the among fields that traditionally have looked at growth and survival trade-off in plants lacking patterns of species diversity and species com- secondary growth. position throughout succession from different We still know relatively little about the link perspectives, methodologies, and spatial and between demographic parameters and func- temporal scales, such as evolutionary biol- tional traits and how they covary among criti- ogy and community ecology (Urban, et al. cal ontogenetic stages (i.e., seedlings) and life 2008). Considering different scales is of utmost forms (but see Santiago & Wright, 2007), or importance, since patterns of trait distribution, the strength of their phylogenetic signal during species composition, and phylogeny change 68 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 with the scale. The core processes determining process in sites of similar environmental con- community assembly are drift (which combines ditions (Guariguata, & Ostertag, 2001). This neutral and stochastic processes), selection result shows the importance of having large (niche filtering), dispersal (combination of reserves to capture a significant component of probabilistic and environmental filtering), and the previous species diversity found in larger speciation (evolution of the species pool). The tracts of tropical rain forests. Long-term data interactions among these four elements are from permanent plots could help to elucidate rooted on organismal performance, but vary the role of stochastic and deterministic factors across levels of organization, from individual affecting species trajectories, species identi- plants to regional and landscape levels, deter- ties, and functional roles (i.e., the increased mining community composition in the long abundance of shade-tolerants as succession run (Vellend, 2010). Chronosequence stud- progresses). Such studies could help to clarify ies (Chazdon, 2008; Craven, Hall, Berlyn, the influence of species diversity (taxonomic Ashton, & van Breugel, 2015), as well as the and functional diversity) on ecosystem func- monitoring of large tropical forest dynamics tions such as carbon sequestration. plots (see Zimmerman, Thompson, & Brokaw, 2008), have the potential of disentangling CONCLUDING REMARKS the relationships among these core processes, and thus to explain the observed patterns of Shade tolerance remains a central concept community assembly at different levels of in tropical rain forest succession critical to complexity (Vellend, 2010; Meiners, Cadotte, understanding the evolutionary and ecologi- Fridley, Pickett, & Walker, 2015). For instance, cal limits of niche diversification. From seed chronosequence studies at different geographi- dispersal and seedling establishment to the cal scales have helped to explain how much development of the reproductive adult, tropi- of the variation in species composition and cal plants experience a complex light gradi- abundance is determined by the site age rela- ent as succession progresses, with dynamic tive to local environmental conditions and ini- interactions among competitors, pathogens, tial species composition (Letcher et al., 2012; herbivores, pollinators, and seed dispersers. Mesquita, Massoca, Jakovac, Bentos, & Wil- Although the initial physical conditions of early liamson, 2015). Craven et al. (2015) reported successional stages are critical in determin- shifts in functional characters during succes- ing their response at later stages (i.e., Poorter, sion. For instance, leaf toughness, wood den- 2007), the environment faced by more mature sity, and adult plant size increased significantly ontogenetic stages also impacts growth, repro- whereas photosynthetic rates decreased in more duction, and survival, the three components of mature stages of secondary forests in Soberanía functional performance (Violle et al., 2007). National Park in Panama. Norden et al. (2015) The paucity of data on ontogenetic niche shifts, found that site identity explained most of the and the characterization of environmental con- variation in species density and basal area in ditions encountered by mature ontogenetic successional plots from Brazil, Costa Rica, stages, continues to be a major research gap. Mexico, and Nicaragua. They conclude that The analysis of shade tolerance should be random factors were more important than stand part of more general, inclusive models, such age or previous land use in predicting species as the life cycle successional trajectory model, composition (there was high among-plot varia- which look for common strategies to explain tion in species composition even for plots with- resource use, encompassing the analysis of in a site). This suggests that local effects, such ecophysiological mechanisms of resource utili- as the landscape and the regional species pool, zation, life history strategies, and demographic could affect species composition within a plot, responses from cells to individuals, popula- and influence the dynamics of the successional tions, communities, and ecosystems (i.e., by Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 69 increasing comparative studies of contrasting claramente en la sección de agradecimientos. ecosystems to find common emerging strate- El respectivo documento legal firmado se gies, see Lebrija-Trejos, Pérez-García, Meave, encuentra en los archivos de la revista. Bongers, & Poorter, 2010; Lohbeck et al., 2015). Novel, integrative approaches, such as the leaf and whole-plant economic spectrum, ACKNOWLEDGEMENTS have great potential to generate a new synthesis Olivia Sylvester, Silvia Alvarez-Clare, beyond community assembly rules (and thus, Robin Chazdon, Pat Heslop-Harrison and go beyond the examination of regeneration two anonymous reviewers provided healthy niches), and are sufficiently general to describe criticisms that improved this manuscript. the consequences of differences in functional Catie Morris corrected the final version of performance on ecosystem processes (Wright the manuscript. et al., 2004; Reich, 2014). Such new synthesis should rest on a robust understanding of the impacts of functional traits on plant fitness. RESUMEN The accumulation of long-term data on forest Tolerancia a la sombra en el contexto del proce- growth and changes in ontogenetic perfor- so de sucesión en los bosques tropicales lluviosos. La mance (Poorter et al., 2008; Wright et al., 2010) tolerancia a la sombra (la capacidad de sobrevivir y crecer could provide the necessary evidence to vali- durante largos períodos bajo sombra profunda) es un com- date the plant-economics spectrum model. This ponente clave del valor adaptativo de la planta y la base de new approach requires a change of paradigms. las teorías actuales de la sucesión forestal de la selva tropi-cal. Sirve como un paradigma para entender la asignación McGill et al. (2006) propose a switch from óptima de recursos limitados bajo regímenes dinámicos more descriptive to more mechanistic views of de luz. En esta revisión analizo cómo la sucesión de los plant performance based on functional traits. bosques tropicales lluviosos influye en la expresión de los The generation of this information has mecanismos ecofisiológicos que conducen a la tolerancia a never been more urgent than now, considering la sombra, e identifico áreas futuras que pueden aumentar nuestra comprensión de las consecuencias ecológicas y that many human impacts on natural systems evolutivas de este fenómeno. La tolerancia a la sombra es are becoming global as reflected in the fast pro- un rasgo funcional continuo y multivariable que refleja el gression of climate change. Climate change is balance de invertir recursos bajo condiciones de luz limi- likely to impact more intensively shade tolerant tada versus crecer más rápidamente en condiciones de luz species, and in general, species exploiting lim- intensa. Propongo el modelo de ciclo de vida a lo largo de la trayectoria de sucesión de Gómez-Pompa y Vázquez- ited resources and having slow response times Yanes como una herramienta integradora para entender la and limited temporal plasticity. Our capacity to sucesión de la selva tropical. Este modelo muestra cómo las conserve, manage and restore tropical diversity especies se distribuyen a lo largo del gradiente ambiental en will benefit from a mechanistic understanding función de su grado de tolerancia a la sombra, y representa of plant adaptation to successional gradients, un paradigma más integrador para comprender la interac-ción entre los diferentes componentes de la diversidad de especially now that light changes are com- especies (diversidad taxonómica y funcional y variación pounded by increased resource scarcity and ontogenética) a lo largo de la sucesión. El modelo propone water stress (Schwalm et al., 2017) triggered que las diferentes combinaciones de caracteres funciona- by more intense climatic fluctuations (Clark, les que determinan la tolerancia a la sombra se expresan Clark, & Oberbauer, 2010). en diferentes etapas del ciclo de vida, y afectan cómo y cuándo las plantas ingresan en el proceso de sucesión. Los modelos que explican la expresión de tolerancia a la Declaración de ética: el autor declara que sombra (disponibilidad de recursos, ganancia de carbono, está de acuerdo con esta publicación; que no CSR, competencia de recursos) se basan en la economía de existe conflicto de interés de ningún tipo; y que toda la planta y no son mutuamente excluyentes. Se están ha cumplido con todos los requisitos y proced- desarrollando explicaciones más integradoras basadas en la distribución de caracteres funcionales entre especies, imientos éticos y legales pertinentes. Todas las etapas ontogenéticas, y micrositios, mediante el uso de fuentes de financiamiento se detallan plena y estudios de cronosecuencia y metadatos colectados a largo 70 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 plazo. El análisis de la tolerancia a la sombra está sesgado Augspurger, C. K. (1984). Light requirements of neotropi- hacia las plántulas de árboles y el sotobosque. Otras formas cal tree seedlings: a comparative study of growth and de vida y microhábitats dentro del perfil del bosque están survival. Journal of Ecology, 72(3), 777-795. casi excluidas de estos análisis. En resumen, la tolerancia Avalos, G. (2016). Growth of the neotropical palm Euterpe a la sombra es un fenómeno complejo, está determinada precatoria Mart. in an agroforestry system in Costa por múltiples caracteres funcionales que cambian onto- Rica. Brenesia, 85-86, 1-8. genéticamente en el espacio y el tiempo, e implica una considerable plasticidad. Los métodos actuales no toman Avalos, G., Fernández, M., & Engeln, J. T. (2013). Succes- en cuenta esta plasticidad. Comprender la naturaleza de la sional stage, fragmentation and exposure to extrac-tion influence the population structure of Euterpe tolerancia a la sombra y su base funcional es fundamental precatoria (Arecaeae). Revista de Biología Tropical, para entender el crecimiento de la planta y mejorar la 61(3), 1415-1424. gestión, restauración, y conservación de los bosques tro- picales, los cuales enfrentan las amenazas combinadas del Avalos, G., & Mulkey, S. S. (2014). Photosynthetic and calentamiento global y la pérdida de hábitat. morphological acclimation of seedlings of tropical lianas to changes in the light environment. American Journal of Botany, 101(12), 2088-2096. Palabras clave: filtrado ambiental; cambios de nicho ontogenéticos; caracteres funcionales; diferenciación de Avalos, G., Mulkey, S. S., & Kitajima, K. (1999). Leaf nicho; espectro económico foliar; espectro de economía optical properties of trees and lianas in the outer de plantas; nicho de regeneración; sucesión secundaria; canopy of a tropical dry forest. Biotropica, 31(3), tolerancia a la sombra. 517-520. Avalos, G., Mulkey, S. S., Kitajima, K, & Wright, S.J. (2007). Colonization strategies of two liana species REFERENCES in a tropical dry forest canopy. Biotropica, 39(3), 393-399. Acevedo, M. F., Urban, D. L., & Shugart H. H. (1996). Models of forest dynamics based on roles of tree spe- Baltzer, J. L., & Thomas, S. C. (2007). Determinants of cies. Ecological Modeling, 87(1-3), 267-284. whole-plant light requirements in Bornean rain forest tree saplings. Journal of Ecology, 95(6), 1208-1221. Ackerly, D. D. (2003). Community assembly, niche con- servatism, and adaptive evolution in changing envi- Bazzaz, F. A., Chiariello, N. R., Coley, P. D., & Pitelka, L. ronments. International Journal of Plant Sciences, F. (1987). Allocating resources to reproduction and K(S3), S165-S184. defense. BioScience, 37(1), 58-67. Ackerly, D. D., & Cornwell, W.K. (2007). A trait-based Benzing, D. H. (2008). Vascular epiphytes: general bio-logy and related biota. Cambridge: Cambridge approach to community assembly: partitioning of University Press. species trait values into within-and among-commu- nity components. Ecology Letters, 10(2), 135-145. Boege, K., & Marquis, R. J. (2005). Facing herbivory as you grow up: the ontogeny of resistance in plants. Álvarez-Clare, S., & Avalos, G. (2007). Light intercep- Trends in Ecology & Evolution, 20(8), 441-448. tion efficiency of the understory palm Calyptrogyne ghiesbreghtiana under deep shade conditions. Eco- Boukili, V. K., & Chazdon, R. L. (2017). Environmental tropica, 13, 1-8. filtering, local site factors and landscape context drive changes in functional trait composition during Álvarez-Clare, S., & Kitajima, K. (2009). Susceptibility tropical forest succession. Perspectives in Plant Eco- of tree seedlings to biotic and abiotic hazards in the logy, Evolution and Systematics, 24, 37-47. understory of a moist tropical forest in Panama. Bio- tropica, 41(1), 47-56. Bradshaw, C. J., Sodhi, N. S., & Brook, B. W. (2009). Tropical turmoil: a biodiversity tragedy in progress. Arnold, S. J. (1983). Morphology, performance and fitness. Frontiers in Ecology and the Environment, 7(2), American Zoologist, 23(2), 347-361. 79-87. Arroyo-Rodríguez, V., Melo, F. P., Martínez-Ramos, M., Brandani, A., Hartshorn, G. S., & Orians, G. H. (1988). Internal heterogeneity of gaps and species richness in Bongers, F., Chazdon, R. L., Meave, J. A., ... & Costa Rican tropical wet forest. Journal of Tropical Tabarelli, M. (2017). Multiple successional pathways Ecology, 4(2), 99-119. in human-modified tropical landscapes: new insights from forest succession, forest fragmentation and Budowski, G. (1965). Distribution of tropical American landscape ecology research. Biological Reviews, rain forest species in the light of successional proces- 92(1), 326-340. ses. Turrialba, 15(1), 40-42. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 71 Cambronero, M., Avalos, G., & Alvarez-Vergnani, C. Clark, D. A., & Clark, D. B. (1992). Life history diversity (2018). Carbon accumulation in seven neotropical of canopy and emergent trees in a neotropical rain palm species from different forest strata. Palms, forest. Ecological monographs, 62(3), 315-344. 62(1), 25-34. Clark, D. B., Clark, D. A., & Oberbauer, S. F. (2010). Campo, J., & Vázquez-Yanes, C. (2004). Effects of nutrient Annual wood production in a tropical rain forest in limitation on aboveground carbon dynamics during NE Costa Rica linked to climatic variation but not tropical dry forest regeneration in Yucatán, Mexico. to increasing CO . Global Change Biology, 16(2), Ecosystems, 7(3), 311-319. 2747-759. Cardelús, C. L., & Chazdon, R. L. (2005). Inner-crown Clements, F. E. (1936). Nature and structure of the climax. Microenvironments of Two Emergent Tree Species Journal of Ecology, 24(1), 252-284. in a Lowland Wet Forest. Biotropica, 37(2), 238-244. Coley, P. D., Bryant, J. P., & Chapin, F. S. (1985). Resource Chave, J., Coomes, D., Jansen, S., Lewis, S. L., Swenson, N. G., & Zanne, A. E. (2009). Towards a worldwide availability and plant antiherbivore defense. Science, wood economics spectrum. Ecology Letters, 12(4), 230(4728), 895-899. 351-366. Condit, R., Hubbell, S. P., & Foster, R. B. (1996). Changes Chazdon, R. L. (1985). Leaf display, canopy structure, and in tree species abundance in a neotropical forest: light interception of two understory palm species. impact of climate change. Journal of tropical ecolo- American Journal of Botany, 72(10), 1493-1502. gy, 12(2), 231-256. Chazdon, R. L. (1986). The costs of leaf support in unders- Craine, J. M. (2005). Reconciling plant strategy theories tory palms: economy versus safety. The American of Grime and Tilman. Journal of Ecology, 93(6), Naturalist, 127(1), 9-30. 1041-1052. Chazdon, R. L. (2008). Chance and determinism in tropi- Craine, J. M. (2007). Plant strategy theories: replies cal forest succession. In W. Carson, & S. Schnitzer to Grime and Tilman. Journal of Ecology, 95(2), (Eds.), Tropical forest community ecology (pp. 384- 235-240. 409). Oxford: Wiley-Blackwell. Craven, D., Hall, J. S., Berlyn, G. P., Ashton, M. S., & van Chazdon, R. L. (2014). Second growth: the promise of tro- Breugel, M. (2015). Changing gears during succes- pical forest regeneration in an age of deforestation. sion: shifting functional strategies in young tropical Chicago: University of Chicago Press. secondary forests. Oecologia, 179(1), 293-305. Chazdon, R. L., Chao, A., Colwell, R. K., Lin, S. Y., Nor- Crayn, D. M., Winter, K., & Smith, J. A. C. (2004). Mul- den, N., Letcher, S. G., ... & Arroyo, J. P. (2011). A tiple origins of crassulacean acid metabolism and novel statistical method for classifying habitat gene- the epiphytic habit in the Neotropical family Bro- ralists and specialists. Ecology, 92(6), 1332-1343. meliaceae. Proceedings of the National Academy of Sciences of the United States of America, 101(10), Chazdon, R. L., & Fetcher, N. (1984). Photosynthetic light 3703-3708. environments in a lowland tropical rain forest in Costa Rica. The Journal of Ecology, 72(2), 553-564. Diaz, S., Hodgson, J. G., Thompson, K., Cabido, M., Cor- Chazdon, R. L., & Pearcy, R. W. (1991). The importance nelissen, J. H. C., Jalili, A., ... & Band, S. R. (2004). of sunflecks for forest understory plants. BioScience, The plant traits that drive ecosystems: evidence 41(11), 760-766. from three continents. Journal of Vegetation Science, 15(3), 295-304. Chazdon, R. L., Pearcy, R. W., Lee, D. W., & Fetcher, N. (1996). Photosynthetic responses of tropical forest Dalling, J. W., Winter, K., Nason, J. D., Hubbell, S. P., plants to contrasting light environments. In S. S. Murawski, D. A., & Hamrick, J. L. (2001). The Mulkey, R. L. Chazdon, & A. P. Smith (Eds.), Tro- unusual life history of Alseis blackiana: a shade- pical forest plant ecophysiology (pp. 5-55). Boston, persistent pioneer tree?. Ecology, 82(4), 933-945. MA: Springer. Denslow, J. S. (1987). Tropical rain forest gaps and tree Clark, D. B., & Clark, D. A. (1991). Herbivores, herbivory, species diversity. Annual Review of Ecology and and plant phenology: patterns and consequences in a Systematics, 18(1), 431-451. tropical rain-forest cycad. In P. W. Price, T. M. Lew- insohn, G. W. Fernandes, & W. W. Benson (Eds.), Donoghue, M. J. (2008). A phylogenetic perspective on Plant-animal interactions: evolutionary ecology in the distribution of plant diversity. Proceedings of tropical and temperate regions (pp. 209-225). New the National Academy of Sciences, 105(Suppl. 1), York: John Wiley & Sons. 11549-11555. 72 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 Gilbert, B., Wright, S. J., Muller-Landau, H. C., Kitajima, Hubbell, S. P., Foster, R. B., O’Brien, S. T., Harms, K. K., & Hernandéz, A. (2006). Life history trade-offs in E., Condit, R., Wechsler, B., ... & De Lao, S. L. tropical trees and lianas. Ecology, 87(5), 1281-1288. (1999). Light-gap disturbances, recruitment limita- tion, and tree diversity in a neotropical forest. Scien- Givnish, T. J. (1988). Adaptation to sun and shade: a ce, 283(5401), 554-557. whole-plant perspective. Functional Plant Biology, 15(2), 63-92. Hubbell, S. P. (2001). The unified neutral theory of bio- diversity and biogeography. Princeton & Oxford: Givnish, T. J. (1999). On the causes of gradients in tropical Princeton University Press. tree diversity. Journal of Ecology, 87(2), 193-210. Huston, M. A. (1999). Local processes and regional pat- Gleason, H. A. (1926). The individualistic concept of the terns: appropriate scales for understanding variation plant association. Bulletin of the Torrey Botanical in the diversity of plants and animals. Oikos, 86(3), Club, 53(1), 7-26. 393-401. Gómez-Pompa, A., & Vázquez-Yanes, C. (1981). Succes- Jansson, R., Rodríguez-Castañeda, G., & Harding, L. E. sional studies of a rain forest in Mexico. In D. C. (2013). What can multiple phylogenies say about the West, H. H. Shugart, & D. B. Botkin (Eds.), Forest latitudinal diversity gradient? A new look at the tro- Succession (pp. 246-266). New York: Springer. pical conservatism, out of the tropics, and diversifi- cation rate hypotheses. Evolution, 67(6), 1741-1755. Gommers, C. M., Visser, E. J., St Onge, K. R., Voesenek, L. A., & Pierik, R. (2013). Shade tolerance: when Janzen, D. H. (1970). Herbivores and the number of tree growing tall is not an option. Trends in Plant Science, species in tropical forests. The American Naturalist, 18(2), 65-71. 104(940), 501-528. Gravel, D., Canham, C. D., Beaudet, M., & Messier, C. Kitajima, K. (1994). Relative importance of photosynthetic (2010). Shade tolerance, canopy gaps and mecha- traits and allocation patterns as correlates of seedling nisms of coexistence of forest trees. Oikos, 119(3), shade tolerance of 13 tropical trees. Oecologia, 98(3- 475-484. 4), 419-428. Greig, N. (1993). Regeneration mode in neotropical Piper: Kitajima, K. (2002). Do shade-tolerant tropical tree seed- habitat and species comparisons. Ecology, 74(7), lings depend longer on seed reserves? Functio- 2125-2135. nal growth analysis of three Bignoniaceae species. Functional Ecology, 16(4), 433-444. Grime, J. P. (1977). Evidence for the existence of three pri- mary strategies in plants and its relevance to ecologi- Kitajima, K., Mulkey, S. S., & Wright, S. J. (2004). The American Naturalist Variation in crown light utilization characteristics cal and evolutionary theory. , among tropical canopy trees. Annals of Botany, 95(3), 111(982), 1169-1194. 535-547. Grime, J. P. (1989). Whole-plant responses to stress in Kitajima, K., & Poorter, L. (2008). Functional basis for natural and agricultural systems. In H. G. Jones, T. J. resource niche partitioning by tropical trees. In Flowers, & M. B. Jones (Eds.), Plants under stress: W. P. Carson, & S. A. Schnitzer (Eds.), Tropical biochemistry, physiology and ecology and their appli- Forest Community Ecology (pp. 172-188). Oxford: cation to plant improvement (pp. 31-46). Cambridge: Blackwell. Cambridge University Press. Kitajima, K., & Poorter, L. (2010). Tissue-level leaf tough- Grime, J. P. (2007). Plant strategy theories: a comment on ness, but not lamina thickness, predicts sapling leaf Craine (2005). Journal of Ecology, 95(2), 227-230. lifespan and shade tolerance of tropical tree species. New Phytologist, 186(3), 708-721. Guariguata, M. R., & Ostertag, R. (2001). Neotropical secondary forest succession: changes in structural Kobe, R. K., & Coates, K. D. (1997). Models of sapling and functional characteristics. Forest Ecology and mortality as a function of growth to characterize Management, 148(1-3), 185-206. interspecific variation in shade tolerance of eight tree species of northwestern British Columbia. Canadian Hillebrand, H. (2004). On the generality of the latitudinal Journal of Forest Research, 27(2), 227-236. diversity gradient. The American Naturalist, 163(2), 192-211. Kozlowski, T. T., & Pallardy, S. G. (2002). Acclimation and adaptive responses of woody plants to environmental Hubbell, S. P. (1997). A unified theory of biogeography stresses. The Botanical Review, 68(2), 270-334. and relative species abundance and its application to tropical rain forests and coral reefs. Coral Reefs, Kraft, N. J., Adler, P. B., Godoy, O., James, E. C., Fuller, 16(1), S9-S21. S., & Levine, J. M. (2015). Community assembly, Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 73 coexistence and the environmental filtering metaphor. Martínez-Ramos, M., Alvarez-Buylla, E., & Sarukhan, Functional Ecology, 29(5), 592-599. J. (1989). Tree demography and gap dynamics in a tropical rain forest. Ecology, 70(3), 555-558. Kraft, N. J., Valencia, R., & Ackerly, D. D. (2008). Functio- nal traits and niche-based tree community assembly Martínez-Ramos, M., & Soto-Castro, A. (1993). Seed rain in an Amazonian forest. Science, 322(5901), 580-582. and advanced regeneration in a tropical rain forest. Vegetatio, 107(1), 299-318. Kursar, T. A., & Coley, P. D. (1992). Delayed greening in tropical leaves: an antiherbivore defense?. Biotropi- McCarthy, M. C., & Enquist, B. J. (2007). Consistency bet- ca, 24(2), 256-262. ween an allometric approach and optimal partitioning theory in global patterns of plant biomass allocation. Lebrija-Trejos, E., Pérez-García, E. A., Meave, J. A., Bon- Functional Ecology, 21(4), 713-720. gers, F., & Poorter, L. (2010). Functional traits and environmental filtering drive community assembly McGill, B. J., Enquist, B. J., Weiher, E., & Westoby, in a species-rich tropical system. Ecology, 91(2), M. (2006). Rebuilding community ecology from 386-398. functional traits. Trends in Ecology & Evolution, 21(4), 178-185. Lebrija-Trejos, E., Pérez-García, E. A., Meave, J. A., Poor- ter, L., & Bongers, F. (2011). Environmental changes Meiners, S. J., Cadotte, M. W., Fridley, J. D., Pickett, S. T., during secondary succession in a tropical dry forest in & Walker, L. R. (2015). Is successional research nea- Mexico. Journal of Tropical Ecology, 27(5), 477-489. ring its climax? New approaches for understanding dynamic communities. Functional Ecology, 29(2), Lee, D. W., Baskaran, K., Mansor, M., Mohamad, H., & 154-164. Yap, S. K. (1996). Irradiance and spectral quality affect Asian tropical rain forest tree seedling develo- Melo, F. P., Arroyo-Rodríguez, V., Fahrig, L., Martínez- pment. Ecology, 77(2), 568-580. Ramos, M., & Tabarelli, M. (2013). On the hope for biodiversity-friendly tropical landscapes. Trends in Letcher, S. G. (2010). Phylogenetic structure of angios- Ecology & Evolution, 28(8), 462-468. perm communities during tropical forest succession. Proceedings of the Royal Society of London B: Biolo- Mesquita, R. D. C. G., Massoca, P. E. D. S., Jakovac, C. C., gical Sciences, 277(1678), 97-104. Bentos, T. V., & Williamson, G. B. (2015). Amazon rain forest succession: stochasticity or land-use lega- Letcher, S. G., Chazdon, R. L., Andrade, A. C., Bongers, cy? BioScience, 65(9), 849-861. F., van Breugel, M., Finegan, B., ... & Williamson, G. B. (2012). Phylogenetic community structure during Mittelbach, G. G., Schemske, D. W., Cornell, H. V., Allen, succession: evidence from three Neotropical forest A. P., Brown, J. M., Bush, M. B., ... & McCain, C. sites. Perspectives in Plant Ecology, Evolution and M. (2007). Evolution and the latitudinal diversity gradient: speciation, extinction and biogeography. Systematics, 14(2), 79-87. Ecology Letters, 10(4), 315-331. Letcher, S. G., Lasky, J. R., Chazdon, R. L., Norden, Montgomery, R., & Chazdon, R. J. (2002). Light gradient N., Wright, S. J., Meave, J. A., ... & Andrade, J. L. partitioning by tropical tree seedlings in the absence (2015). Environmental gradients and the evolution of of canopy gaps. Oecologia, 131(2), 165-174. successional habitat specialization: a test case with 14 Neotropical forest sites. Journal of Ecology, 103(5), Muscarella, R., Uriarte, M., Aide, T. M., Erickson, D. L., 1276-1290. Forero-Montaña, J., Kress, W. J., ... & Zimmerman, J. K. (2016). Functional convergence and phylogenetic Lohbeck, M., Lebrija-Trejos, E., Martínez-Ramos, M., divergence during secondary succession of subtropi- Meave, J. A., Poorter, L., & Bongers, F. (2015). cal wet forests in Puerto Rico. Journal of Vegetation Functional trait strategies of trees in dry and wet Science, 27(2), 283-294. tropical forests are similar but differ in their conse- quences for succession. PloS One, 10(4), e0123741. Niinemets, Ü. (2006). The controversy over traits confe- rring shade-tolerance in trees: ontogenetic changes MacArthur, R., & Levins, R. (1964). Competition, habitat revisited. Journal of Ecology, 94(2), 464-470. selection, and character displacement in a patchy environment. Proceedings of the National Academy Niinemets, Ü. (2010a). A review of light interception in of Sciences, 51(6), 1207-1210. plant stands from leaf to canopy in different plant functional types and in species with varying shade Markesteijn, L., Poorter, L., Bongers, F., Paz, H., & tolerance. Ecological Research, 25(4), 693-714. Sack, L. (2011). Hydraulics and life history of tropical dry forest tree species: coordination of spe- Niinemets, Ü. (2010b). Responses of forest trees to single cies’ drought and shade tolerance. New Phytologist, and multiple environmental stresses from seedlings to 191(2), 480-495. mature plants: past stress history, stress interactions, 74 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 tolerance and acclimation. Forest Ecology and Mana- Poorter, L., & Rose, S. A. (2005). Light-dependent changes gement, 260(10), 1623-1639. in the relationship between seed mass and seedling traits: a meta-analysis for rain forest tree species. Norden, N., Angarita, H. A., Bongers, F., Martínez-Ramos, Oecologia, 142(3), 378-387. M., Granzow-de la Cerda, I., Van Breugel, M., ... & Finegan, B. (2015). Successional dynamics in Poorter, L., Wright, S. J., Paz, H., Ackerly, D. D., Condit, Neotropical forests are as uncertain as they are pre- R., Ibarra-Manríquez, G., ... & Muller-Landau, H. C. dictable. Proceedings of the National Academy of (2008). Are functional traits good predictors of demo- Sciences, 112(26), 8013-8018. graphic rates? Evidence from five neotropical forests. Ecology, 89(7), 1908-1920. Norden, N., Boukili, V., Chao, A., Ma, K. H., Letcher, S. G., & Chazdon, R. L. (2017). Opposing mechanisms Prado, A., Sierra, A., Windsor, D., & Bede, J. C. (2014). affect taxonomic convergence between tree assem- Ecology Leaf traits and herbivory levels in a tropical gym-blages during tropical forest succession. Letters, 20(11), 1448-1458L. nosperm, Zamia stevensonii (Zamiaceae). American Journal of Botany, 101(3), 437-447. Novotny, V., Drozd, P., Miller, S. E., Kulfan, M., Janda, M., Basset, Y., & Weiblen, G. D. (2006). Why are there so Prescott, C. E. (2002). The influence of the forest canopy many species of herbivorous insects in tropical rain on nutrient cycling. Tree physiology, 22(15-16), forests? Science, 313(5790), 1115-1118. 1193-1200. Oliver, C. D., & Larson, B. C. (1996). Forest stand dyna- Putz, F. E. (1984). The natural history of lianas on Barro mics: updated edition. New Jersey: John Wiley. Colorado Island, Panama. Ecology, 65(6), 1713-1724. Pacala, S. W., & Rees, M. (1998). Models suggesting field Rees, M., Condit, R., Crawley, M., Pacala, S., & Tilman, experiments to test two hypotheses explaining suc- D. (2001). Long-term studies of vegetation dynamics. cessional diversity. The American Naturalist, 152(5), Science, 293(5530), 650-655. 729-737. Reich, P. B. (2014). The world-wide ‘fast-slow’ plant Peña-Claros, M., & Zuidema, P. (2000). Demographic economics spectrum: a traits manifesto. Journal of limitations for the sustainable extraction of palm Ecology, 102(2), 275-301. heart from Euterpe precatoria in two forest types in Bolivia. Ecología en Bolivia, 34, 7-25. Reich, P. B., Wright, I. J., Cavender-Bares, J., Craine, J. M., Oleksyn, J., Westoby, M., & Walters, M. B. (2003). Peterson, C. J., & Carson, W. P. (2008). Processes cons- The evolution of plant functional variation: traits, training woody species succession on abandoned spectra, and strategies. International Journal of Plant pastures in the tropics: on the relevance of temperate Sciences, 164(S3), S143-S164. models of succession. In W. Carson, & S. Schnitzer (Eds.), Tropical forest community ecology (pp. 367- Richards, P.W. (1952). The tropical rain forest: an ecolo- 383). Oxford: Wiley-Blackwell. gical study. Cambridge: Cambridge University Press. Plourde, B. T., Boukili, V. K., & Chazdon, R. L. (2015). Rosindell, J., Hubbell, S. P., & Etienne, R. S. (2011). The Radial changes in wood specific gravity of tropical unified neutral theory of biodiversity and biogeogra- trees: inter-and intraspecific variation during secon- phy at age ten. Trends in Ecology &Evolution, 26(7), dary succession. Functional Ecology, 29(1), 111-120. 340-348. Poorter, L. (2007). Are species adapted to their regenera- tion niche, adult niche, or both? The American Natu- Santiago, L. S., & Wright, S. J. (2007). Leaf functional ralist, 169(4), 433-442. traits of tropical forest plants in relation to growth form. Functional Ecology, 21(1), 19-27. Poorter, L. (2009). Leaf traits show different relationships with shade tolerance in moist versus dry tropical Sapijanskas, J., Paquette, A., Potvin, C., Kunert, N., & forests. New Phytologist, 181(4), 890-900. Loreau, M. (2014). Tropical tree diversity enhan- ces light capture through crown plasticity and spa- Poorter, L., & Arets, E. J. (2003). Light environment and tial and temporal niche differences. Ecology, 95(9), tree strategies in a Bolivian tropical moist forest: an 2479-2492. evaluation of the light partitioning hypothesis. Plant Ecology, 166(2), 295-306. Schaetzl, R. J., Burns, S. F., Johnson, D. L., & Small, T. W. (1988). Tree uprooting: review of impacts on forest Poorter, L., Bongers, F., Sterck, F. J., & Wöll, H. (2005). ecology. Vegetatio, 79(3), 165-176. Beyond the regeneration phase: differentiation of height-light trajectories among tropical tree species. Schaller, A. (2008). Induced plant resistance to herbivory. Journal of Ecology, 93(2), 256-267. Berlin: Springer. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 75 Schlichting, C. D., & Pigliucci, M. (1998). Phenotypic Tilman, D. (1994). Competition and biodiversity in spatia- evolution: a reaction norm perspective. Sunderland: lly structured habitats. Ecology, 75(1), 2-16. Sinauer. Tomlinson, P. B. (2006). The uniqueness of palms. Bota- Schnitzer, S. A., Mascaro, J., & Carson, W. P. (2008). nical Journal of the Linnean Society, 151(1), 5-14. Treefall gaps and the maintenance of plant species diversity in tropical forests. In W. P. Carson, & S. A. Urban, M. C., Leibold, M. A., Amarasekare, P., De Mees- Schnitzer (Eds.), Tropical Forest Community Ecology ter, L., Gomulkiewicz, R., Hochberg, M. E., ... & (pp. 196-209). Oxford: Blackwell. Pantel, J. H. (2008). The evolutionary ecology of Schupp, E. W., Howe, H. F., Augspurger, C. K., & Levey, metacommunities. Trends in Ecology & Evolution, D. J. (1989). Arrival and survival in tropical treefall 23(6), 311-317. gaps. Ecology, 70(3), 562-564. Valladares, F., & Niinemets, Ü. (2008). Shade tolerance, Schwalm, C. R., Anderegg, W. R., Michalak, A. M., Fisher, a key plant feature of complex nature and conse- J. B., Biondi, F., Koch, G., ... & Huntzinger, D. N. quences. Annual Review of Ecology, Evolution, and (2017). Global patterns of drought recovery. Nature, Systematics, 39, 237-257. 548(7666), 202. Valladares, F., Laanisto, L., Niinemets, Ü., & Zavala, M. Spasojevic, M. J., Yablon, E. A., Oberle, B., & Myers, J. A. (2016). Shedding light on shade: ecological pers- A. (2014). Ontogenetic trait variation influences tree pectives of understorey plant life. Plant Ecology & community assembly across environmental gradients. Diversity, 9(3), 237-251. Ecosphere, 5(10), 1-20. Valladares, F., Sánchez-Gómez, D., & Zavala, M. A. Sterck, F., Markesteijn, L., Schieving, F., & Poorter, (2006). Quantitative estimation of phenotypic plas- L. (2011). Functional traits determine trade-offs ticity: bridging the gap between the evolutionary and niches in a tropical forest community. Procee- concept and its ecological applications. Journal of dings of the National Academy of Sciences, 108(51), Ecology, 94(6), 1103-1116. 20627-20632. Vellend, M. (2010). Conceptual synthesis in community Strauss-Deberiedetti, S., & Bazzaz, F. A. (1996). Pho- tosynthetic characteristics of tropical trees along suc- ecology. The Quarterly Review of Biology, 85(2), cessional gradients. In S. S. Mulkey, R. L. Chazdon, 183-206. & A. P. Smith (Eds.), Tropical forest plant ecophysio- logy (pp. 162-186). Boston: Springer. Violle, C., Navas, M. L., Vile, D., Kazakou, E., Fortunel, C., Hummel, I., & Garnier, E. (2007). Let the concept Swaine, M. D., & Whitmore, T. C. (1988). On the defi- of trait be functional! Oikos, 116(5), 882-892. nition of ecological species groups in tropical rain forests. Vegetatio, 75(1-2), 81-86. Way, D. A., & Pearcy, R. W. (2012). Sunflecks in trees and forests: from photosynthetic physiology to global Swenson, N. G., Stegen, J. C., Davies, S. J., Erickson, change biology. Tree Physiology, 32(9), 1066-1081. D. L., Forero-Montaña, J., Hurlbert, A. H., ... & Zimmerman, J. K. (2012). Temporal turnover in the Webb, C. O. (2000). Exploring the phylogenetic structure composition of tropical tree communities: functional of ecological communities: an example for rain forest determinism and phylogenetic stochasticity. Ecology, trees. The American Naturalist, 156(2), 145-155. 93(3), 490-499. Webb, C. O., Ackerly, D. D., McPeek, M. A., & Donoghue, Svenning, J. C. (1999). Microhabitat specialization in a M. J. (2002). Phylogenies and community ecology. species-rich palm community in Amazonian Ecuador. Annual Review of Ecology and Systematics, 33(1), Journal of Ecology, 87(1), 55-65. 475-505. Sylvester, O., & Avalos, G. (2013). Influence of light con- Webb, C. O., Cannon, C. H., & Davies, S. J. (2008). Eco- ditions on the allometry and growth of the understory palm Geonoma undata subsp. edulis (Arecaceae) logical organization, biogeography, and the phylo- of neotropical cloud forests. American Journal of genetic structure of tropical forest tree communities. Botany, 100(12), 2357-2363. In W. P. Carson, & S. A. Schnitzer (Eds.), Tropical Forest Community Ecology (pp. 79-97). New Jersey: Tilman, D. (1977). Resource competition between plankton Wiley-Blackwell. algae: an experimental and theoretical approach. Eco- logy, 58(2), 338-348. Weiher, E., Freund, D., Bunton, T., Stefanski, A., Lee, T., & S. Bentivenga. (2013). Advances, challenges Tilman, D. (1990). Constraints and tradeoffs: toward a pre- and developing synthesis of ecological community dictive theory of competition and succession. Oikos, assembly theory. Philosophical Transactions of the 58(1), 3-15. Royal Society B, 366, 2403-2413. 76 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 Weiner, J. (2004). Allocation, plasticity and allometry in vital rates, and size distributions of tropical trees. plants. Perspectives in Plant Ecology, Evolution and Ecology, 84(12), 3174-3185. Systematics, 6(4), 207-215. Wright, I. J., Reich, P. B., Westoby, M., Ackerly, D. D., West-Eberhard, M. J. (2003). Developmental plasticity and Baruch, Z., Bongers, F., ... & Flexas, J. (2004). evolution. Oxford: Oxford University Press. The worldwide leaf economics spectrum. Nature, 428(6985), 821. Wright, J. S. (2002). Plant diversity in tropical forests: a review of mechanisms of species coexistence. Oeco- Zimmerman, J. K., Thompson, J., & Brokaw, N. (2008). logia, 130(1), 1-14. Large tropical forest dynamics plots: testing explana- tions for the maintenance of species diversity. In W. Wright, S. J., Kitajima, K., Kraft, N. J., Reich, P. B., Carson, & S. Schnitzer (Eds.), Tropical forest commu- Wright, I. J., Bunker, D. E., ... & Engelbrecht, B. M. nity ecology (pp. 98-117). Oxford: Wiley-Blackwell. (2010). Functional traits and the growth-mortality tra- de-off in tropical trees. Ecology, 91(12), 3664-3674. Zotz, G., & Hietz, P. (2001). The physiological ecolo- gy of vascular epiphytes: current knowledge, open Wright, S. J., Muller-Landau, H. C., Condit, R., & Hubbell, questions. Journal of Experimental Botany, 52(364), S. P. (2003). Gap-dependent recruitment, realized 2067-2078. Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 67(2) Suppl.: S53-S77, April 2019 77