The University of Chicago How Did the Swiss Cheese Plant Get Its Holes? Author(s): Christopher D. Muir Reviewed work(s): Source: The American Naturalist, Vol. 181, No. 2 (February 2013), pp. 273-281 Published by: The University of Chicago Press for The American Society of Naturalists Stable URL: http://www.jstor.org/stable/10.1086/668819 . Accessed: 28/01/2013 20:45 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. The University of Chicago Press, The American Society of Naturalists, The University of Chicago are collaborating with JSTOR to digitize, preserve and extend access to The American Naturalist. http://www.jstor.org This content downloaded on Mon, 28 Jan 2013 20:45:51 PM All use subject to JSTOR Terms and Conditions vol. 181, no. 2 the american naturalist february 2013 Note How Did the Swiss Cheese Plant Get Its Holes? Christopher D. Muir* Department of Biology, Indiana University, Bloomington, Indiana 47405 Submitted November 30, 2011; Accepted September 3, 2012; Electronically published December 26, 2012 emerged as an important determinant of leaf shape abstract: Adult leaf fenestration in “Swiss cheese” plants (Monstera (Zwieniecki et al. 2004). Because a large fraction of hy- Adans.) is an unusual leaf shape trait lacking a convincing evolu- tionary explanation. Monstera are secondary hemiepiphytes that in- draulic resistance occurs in the leaf (Sack and Holbrook habit the understory of tropical rainforests, where photosynthesis 2006), leaf shape may be important in preventing water from sunflecks often makes up a large proportion of daily carbon stress, especially in hot, sunny conditions. assimilation. Here I present a simple model of leaf-level photosyn- Many shape variations, like lobing or compound leaves, thesis and whole-plant canopy dynamics in a stochastic light envi- have evolved independently many times, in some cases ronment. The model demonstrates that leaf fenestration can reduce convergently in response to similar selective pressures. Un- the variance in plant growth and thereby increase geometric mean usual shapes that have evolved once or a few times are fitness. This growth-variance hypothesis also suggests explanations for conspicuous ontogenetic changes in leaf morphology (hetero- more difficult to study with comparative methods and thus blasty) in Monstera, as well as the absence of leaf fenestration in co- require alternative approaches. Window-like perforations, occurring juvenile tree species. The model provides a testable hy- termed leaf fenestration, are perplexing and found pre- pothesis of the adaptive significance of a unique leaf shape and dominantly in the adult leaves of Monstera Adans. (Ara- illustrates how variance in growth rate could be an important factor ceae). Although leaf fenestration in Monstera was described shaping plant morphology and physiology. by European botanists as early as 1693 (Madison 1977a), Keywords: adaptation, leaf fenestration, leaf shape, Monstera, sun- few hypotheses for its evolutionary origin exist. After re- fleck. viewing these hypotheses, I present and analyze a novel model demonstrating that leaf fenestration may reduce variance in canopy growth rate in understory environ- Introduction ments where a large fraction of carbon gain comes from Compared to other aspects of leaf morphology such as brief, intermittent periods of direct light (sunflecks) that size, little is known about the adaptive significance of leaf are unpredictably distributed in the forest. shape variation in terrestrial plants (Nicotra et al. 2011). Despite the ubiquity of Monstera in tropical understories However, it is generally thought that shape might serve and ornamental settings in the North Atlantic, the adaptive some of the same functions as other aspects of leaf mor- significance of its leaf fenestration has received little at- phology, such as thermoregulation (Parkhurst and Loucks tention from biologists. Madison (1977a) refers to a “fan- 1972; Givnish and Vermeij 1976), light interception (Horn ciful interpretation with no basis in reality” put forth by 1971), and deterring herbivory (Brown and Lawton 1991). early post-Darwinian author H. W. King (1892), who con- Deep lobing, or other shapes that reduce the effective leaf jectured that holes allow water to drip through to the size in hot and dry environments, is the best-studied ex- ground. Madison (1977a) himself suggested that fenestra- ample. Smaller effective leaf size, approximately the di- tion, like other forms of leaf dissection, would be advan- ameter of the largest circle encompassed by the leaf lamina, tageous in portions of the canopy with higher irradiance decreases the boundary layer resistance (Gates 1968). De- by reducing boundary layer resistance and permitting creased boundary layer resistance prevents overheating greater convective leaf cooling. He cited work demonstrat- and increased transpiration in hot, sunny environments, ing reduced leaf temperature in lacerated leaves of Musa. selecting for small and deeply lobed leaves (e.g., McDonald However, this might be an inappropriate comparison, as et al. 2003). More recently, leaf hydraulic architecture has Musa are pioneer species of canopy gaps in full sun, whereas Monstera are found in shade (Madison 1977a). * E-mail: [email protected]. Gunawardena and Dengler (2006) reiterate Madison’s Am. Nat. 2013. Vol. 181, pp. 273–281. ᭧ 2012 by The University of Chicago. thermoregulation argument but also conjecture that fen- 0003-0147/2013/18102-53485$15.00. All rights reserved. estration could act akin to mottling in understory herbs, DOI: 10.1086/668819 which has been ascribed to camouflage from vertebrate This content downloaded on Mon, 28 Jan 2013 20:45:51 PM All use subject to JSTOR Terms and Conditions 274 The American Naturalist herbivores (Givnish 1990). However, the camouflage hy- product of the ground area covered by a leaf and the daily pothesis is applicable for plants very near the forest floor, rate of sunflecks: not climbers several meters above. This hypothesis also p fails to explain why juvenile leaves of the same species lack N lA ground.(2) fenestration. However, more highly dissected leaves will be unable to utilize the fraction of irradiance that falls between the The Growth-Variance Hypothesis for Leaf Fenestration lamina (i.e., into the leaf “holes”). For simplicity, rather This study explores whether leaf fenestration might serve than treat holes as discreet units, I assume that holes in another function altogether, reducing variance in growth leaf lamina are infinitesimally small and uniformly dis- rate and, hence, fitness. I term this the growth-variance persed over the ground area occupied by the leaf. There- hypothesis. In Monstera habitat, tropical rainforest un- fore, the daily carbon assimilation of a leaf (Pleaf)isthe derstories, sunflecks can contribute 150% of carbon gain, product of number of sunflecks intercepted, the assimi- but their distribution is unpredictable in space and time lation per fleck, and the ratio of leaf to ground area: (Chazdon 1988). The patchy distribution of sunflecks may p r lead to high variance in canopy growth rate. Population PleafNP f leck.(3) genetic theory demonstrates that selection maximizes the geometric mean fitness (Gillespie 1973), which is sensitive This assumption is valid if sunflecks are sufficiently large, to both the arithmetic mean and variance. Consequently, such that there is not much variation introduced by some traits, such as leaf shape, that decrease variance in resource falling entirely on lamina or entirely in holes. The as- acquisition and fitness can lead to higher geometric mean sumption is also unaffected if sunflecks are large (on the fitness (Frank 2011 and references therein). In the next order of Aground) because more dissected leaves will partially section, I analyze a model that demonstrates that leaf fen- intercept a large sunfleck that would have been missed by estration and other forms of leaf dissection affect the var- a less dissected leaf. By substitution from equations (1) iance in, but not the mean, canopy growth rate. Assuming and (2), it is evident that daily carbon assimilation in this that canopy growth rate contributes to fitness, the model model is not dependent on r: predicts that leaf fenestration increases fitness when a large A portion of carbon gain depends on stochastic sources of P p (lA )P leaf leaf ground f leck A light (sunflecks). ground (4) p lPAfleck leaf. Model If leaf dissection does not alter mean daily carbon assim- A glossary of symbols used in the model is given in table ilation, how can it affect plant growth and fitness? When 1. The amount of ground area (Aground) covered by a given the light environment is stochastic, and the number of leaf area (Aleaf), assuming constant total leaf mass and leaf sunflecks incident on a leaf is treated as a random variable, mass per area, depends on the amount of leaf dissection. the variance in growth rate is a function of r.Toarrive For leaves with entire margins, the ratio of leaf to ground at this result, I scale up from leaf-level photosynthesis to area is unity. For dissected leaves, this ratio will be less. whole-plant canopy dynamics in a phase of exponential The ratio of leaf area to ground is a dimensionless unit r: growth with a stochastic rate of increase dependent on the number of sunflecks. Let A be the whole plant canopy area at time t. The A t r p leaf ,(1)number of leaves in the canopy at t (L) is the canopy area A ground divided by the leaf area: where r varies from 0 (completely dissected) to 1 (entire) and is analogous to the leaf area index for a single leaf.