Unpacking a century‐old mystery: Winter buds and the latitudinal gradient in leaf form

Th is year marks the 100th anniversary of a seminal paper on plant form. In 1916, in the pages of the American Journal of Botany , Irving W. Bailey and Edmund W. Sinnott documented a remarkable observation: in wet tropical forests, the percentage of woody plant species with toothed or lobed leaves is close to zero, but it increases toward 100% moving north into cold-temperate regions ( Bailey and Sinnott, 1916 ). Th is latitudinal gradient has repeatedly been confi rmed (e.g., Little et al., 2010 ; Peppe et al., 2011 ) and is so robust that paleobotanists use the percentage of leaves with entire margins in paleofl oras as a proxy for mean annual temperature ( Wolfe, 1971 ). In the meantime, it has come to light that other aspects of leaf form may be correlated with climate, as temperate leaves also tend to be rounder, while tropical leaves are more elliptical ( Schmerler et al., 2012 ). But, why does leaf form vary so predictably? Th e short answer is that we still don’t know. Here we explore a new angle, focusing attention on changes in the rhythm of growth and leaf development that accompanied evolutionary shift s into strongly seasonal climates. First we must ask: Is this pattern due to many evolutionary shift s in leaf form as lineages moved from tropical into temperate forests (and vice versa)? Or, is it largely driven by just a few successful lineages in northern latitudes that happened to have teeth and lobes (e.g., maples, birches, oaks)? We still don’t have a clear idea of the number of tropical–temperate transitions in plants ( Donoghue and Edwards, 2014 ). Yet, the wide taxonomic distribution of lineages with both tropical and temperate ranges supports the assumption that there were multiple biome shift s accompanied by repeated evolutionary changes in leaf form (e.g., temperate Acer within Sapindaceae, Tilia within Malvaceae, Hamamelis within Hamamelidaceae, Fagus within Fagaceae). And, judging by our experience with Viburnum ( Schmerler et al., 2012 ; Spriggs et al., 2015 ), additional transitions are likely hidden within many of the clades that span these biomes ( Edwards and Donoghue, 2013 ; Donoghue and Edwards, 2014 ). Until now, adaptive explanations for the leaf-form gradient have focused on leaf function either later in development or in mature leaves. For instance, we know that leaf size and shape infl uence boundary layer dynamics; smaller and more dissected leaves facilitate gas exchange and transpirational cooling ( Gates, 1968 ). But, why then should leaves not instead be more dissected in tropical forests, where the air is oft en hot and still? A second explanation points to leaf teeth as sites of early-season gas exchange, arguing that rapid maturation of toothy margins provides a boost in photosynthate production when light and water are more available, before the formation of a full forest canopy ( Baker-Brosh and Peet, 1997 ; Royer and Wilf, 2006 ). Data vary in support of this hypothesis, and there has been no attempt to quantify the total contribution of photosynthesis in teeth of emerging leaves to a plant’s carbon budget, which we imagine is exceedingly small. Another hypothesis is that teeth serve as hydathodes that expel water that might otherwise fl ood developing leaf tissues early in the spring. Th is may be relevant for temperate species that use positive root pressure to remove freeze–thaw embolisms ( Lechowicz, 1984 ; Feild et al., 2005 ), but many species with leaf teeth do not generate positive xylem pressure. A fourth explanation is biomechanical: temperate leaves, it is said, are thinner and rely more heavily on structural support from their vein systems. In such leaves, the optimal tissue confi guration surrounding each major vein is wedge shaped, which in a pinnately veined leaf would result in a toothy margin ( Givnish, 1979 ). It has even been argued that teeth protect leaves against herbivores ( Brown and Lawton, 1991 ). Each of these hypotheses has some merit and might apply in particular cases. But, in our estimation, none of them is terribly well supported, and little attention has been paid to the alternative possibility that selection on other aspects of the organism might indirectly generate certain leaf characteristics, possibly affecting both teeth and shape simultaneously. Here we consider the idea that the repeated emergence of 1 Manuscript received 21 March 2016; revision accepted 22 April 2016. 2 Department of Ecology and Evolutionary Biology, Brown University, 80 Waterman St., Box G-W, Providence, Rhode Island 02912; and 3 Department of Ecology and Evolutionary Biology, Yale University, PO Box 208106, New Haven, Connecticut 06520-8106 4 Author for correspondence (e-mail: erika_edwards@brown.edu), phone: 401.863.2081 5 Present address: Biomedical Imaging Unit, University of Southampton, Southampton, SO16 6YD, UK doi:10.3732/ajb.1600129 O N T H E N AT U R E O F T H I N G S : E S S AY S New Ideas and Directions in Botany