Aristotle and adding an evolutionary perspective to models of plant architecture in changing environments.

Abstract

Why do plants grow the way that theydo? According to Aristotle, there are fourkinds of causes, or four fundamentallydifferent ways of answering “why” ques-tionssuchasthis(Aristotle, 1984; Falcon,2012). In reductionist science, answers to“why” questions typically relate to one ofthe first three of Aristotle’s causes, regard-ingchangesinsubstances(materialcause),in form (formal cause) and in the effectsof external influences (efficient cause).This is reflected in much functional-structural pant modeling (FSPM), where“structural” aspects of plant architectureare clearly concerned with formal causesand internal “functional” aspects, such ashormones and transported nutrient areclearly concerned with material causes(Sievanenetal.,2000;Prusinkiewicz,2004;Yan et al., 2004; Godin and Sinoquet,2005; Fourcaud et al., 2008; Hanan andPrusinkiewicz, 2008; Vos et al., 2010). Theenvironmental aspects, such as light, soilwater and nutrients, pests and pathogensthat are also often included in such FSPMandinteractwith bothfunction andstruc-ture are clearly concerned with efficientcauses. However, Aristotle’s fourth kind ofcause, final cause, seems to be less consid-eredinreductionist scienceingeneral,andinFSPM inparticular.Final causes concern the aim or pur-pose being served by the object of interest,a plant in our case. In other words, dis-cussion of final causes concernsansweringthe question of why a plant grows theway it does by reference to the purposeof that growth. Such answers could takethe form of “The plant is growing likethat because it is trying to maximize itslightinterception,”forexample.Inscience,such aresponse maylead to accusations ofanthropomorphism, which can be definedas the attribution of human qualities tothingsotherthanhumans,withaconnota-tionthatsuchattribution iserroneousandproblematic (Horowitz, 2007). If Pavlov(1927)wrotethatanimalsshouldbe“stud-ied as purely physiological facts, withoutanyneed to resort to fantasticspeculationsas to the existence of any possible subjec-tive state in the animalwhich may be con-jecturedonanalogywithourselves,”thenitwould seem an even greater sin to explainthe behavior of plants as “purposeful,” orin terms of what they are trying to achievewiththatbehavior?However, evolutionarytheory provides a clear rationale for thevalue of explanations of behavior in termsof the purpose of that behavior, as long asit can be seen as having an evolutionaryadvantage, and thus having been selectedfor by evolutionary processes. So we canrephrase our “final cause” response morecarefully, “The plant is growing like thatbecause that is an ecological strategy thathasevolvedovertimeduetothefactthatittends to maximize the plant’s light inter-ception.” But how can we know whethera growth strategy has indeed evolved overtime to maximize light interception (oranyotherfunctionthatcontributestoevo-lutionarysuccess)?The dynamic structural developmentof a plant can be seen as a strategy forexploiting the limited resources availablewithin its environment, such as light, soilwater and nutrients, and we would expectthat evolution would lead to efficientgrowth strategies that reduce resourcecosts while maximizing resource acquisi-tion. No one growth strategy will be opti-mal in all environments; which strategiesof structural development are most effec-tive will depend on how the resources onwhich the plant depends are distributedthroughbothtimeandspace.Therelativeadvantage of a plant’s growth strategieswill also depend on how its architec-ture influences factors such as dispersalof seeds and pollen, the impacts of her-bivoury and drought stress, the efficiencyof water transport, biomechanical sup-port, and resistance to wind, along withhow much it costs to produce and main-tain the structures that comprise its archi-tecture (Kuppers, 1989; Gartner, 1995).Therefore, if we are to shed light onAristotle’s final cause and start to under-stand why plants have evolved differentstrategies of structural development, weneed to understand the various costs andbenefits of different growth strategies indifferent environments ( Farnsworth andNiklas,1995; Lynch, 1995).There is a long history of model-ing plants in order to investigate thecosts and benefits of different struc-tural growth strategies (e.g., Shinozakiet al., 1964; Honda and Fisher, 1979;JohnsonandThornley,1987; Niklas,1999;West et al., 1999; Takenaka et al., 2001;Falster and Westoby, 2003; King et al.,2003). However, many potentially impor-tant aspects of plant growth and functionhavenotbeenrepresentedinthesemodels,largely due to computational constraintsand limitations in modeling technology.As simplifications of reality, no modelcan possibly include all aspects of real-ity.Nonetheless, recent yearshaveseen thedevelopment of a new generation of plantmodels that include more of these previ-ouslyneglectedaspects,suchastheexplicittopology andspatialgeometry oftheplantstructure; the way that the plant archi-tecture develops dynamically over timeby changes in existing components andthe addition of new ones; the feedbacksbetween plant structure, function, andenvironment that also change with timeas the plant grows and the environment