vol . 1 8 8 , supplement the amer ican natural i st september 20 16

Sympos ium

Biogeochemistry and Geographical Ecology: Embracing All

Twenty-Five Elements Required to Build Organisms*

Michael Kaspari1,2,† and Jennifer S. Powers2,3,†,‡,§

1. Graduate Program in Ecology and Evolution, Department of Biology, University of Oklahoma, Norman, Oklahoma 73019; 2. SmithsonianTropical Research Institute, Balboa, Republic of Panama; 3. Department of Ecology, Evolution, and Behavior and Department of Plant Biology,University of Minnesota, Saint Paul, Minnesota 55108

abstract: Biogeochemistry is a key but relatively neglected part ofthe abiotic template that underlies ecology. The template has a geog-raphy, one that is increasingly being rearranged in this era of globalchange. Justus von Liebig’s law of the minimum has played a usefulrole in focusing attention on biogeochemical regulation of popula-tions, but given that ∼251 elements are required to build organismsand that these organisms use and deplete nutrients in aggregates ofcommunities and ecosystems, we make the case that it is time to moveon. We review available models that suggest the many different mech-anisms that give rise to multiple elements, or colimitation. We thenreview recent empirical data that show that rates of decompositionand primary productivity may be limited bymultiple elements. In thatlight, given the tropics’ high species diversity and generally moreweathered soils, we predict that colimitation at community and eco-system scales is more prevalent closer to the equator. We conclude withsuggestions for how to move forward with experimental studies ofcolimitation.

Keywords: multiple-element limitation, colimitation, law of the min-imum, decomposition, substrate age hypothesis, soil nutrient avail-ability.

Prelude: Ecology as Rational Field Physiology

Frederic Clements viewed ecology as physiology in serviceof biogeography (Clements 1905; Hagen 1992). In the nine-teenth century,Darwin (1839) andVonHumboldt (VonHum-boldt andBonpland2010)had begun to reveal patterns of spe-cies ranges and the distribution of vegetation. In his ResearchMethods in Ecology, Clements (1905, p. 2) saw these biogeo-graphical patterns as the raw material to develop ecology as

* This issue originated as the 2015 Vice Presidential Symposium presented atthe annual meetings of the American Society of Naturalists.† M.K. and J.S.P. contributed equally to this article.‡ Corresponding author; e-mail: powers@umn.edu.§ ORCIDs: Kaspari, 0000-0002-9717-5768.Correction: This article was reposted on July 19, 2016, with a correction to theDroop equation in the section “Liebig’s Fertilizer and the Origins of NutrientLimitations.”

Am. Nat. 2016. Vol. 188, pp. S62–S73. q 2016 by The University of Chicago.0003-0147/2016/188S1-56522$15.00. All rights reserved.DOI: 10.1086/687576

This content downloaded from 129.0All use subject to University of Chicago Press Term

“rational field physiology,” simultaneously getting the phys-iologist out of the lab and into the field while guiding whathe saw as a largely descriptive natural history (Hagen 1992).Toward that end, he devoted much of the book to measuringwater and temperature, two variables that Krebs (1994, p. 93)would later refer to as ecology’s “master” regulators.Meanwhile,mineralogist Vladimir Vernadsky, summarizing decades ofthinking on the subject, was viewing the totality of living or-ganisms as “expressed as in terms of chemical elements . . .as amoving rock formation endowedwith free energy” (Ver-nadsky 1924; Smil 2000). In doing so, the three parts of ecol-ogy’s abiotic template—water, temperature, and mineralelements—were in place.Vernadsky’s work was given a wider audience by G. Eve-

lyn Hutchinson (1948, p. 393), who noted that in additionto the importance of temperature andwater, “Looking atmanfrom a strictly geochemical standpoint, his most strikingcharacter is that he demands so much—not merely thirtyor forty elements for physiological activity, but nearly allthe others for cultural activity.” Hutchinson elaborated onthe two basic reasons that biogeochemistry was importantto ecology: that the recipe for life is elemental and that hu-mans continue to rearrange the geography of those elementsin the same way we are rearranging water and temperature.More than half a century later, despite pioneering work

(Sterner and Elser 2002; Simpson et al. 2010), we are stillsome distance from fully incorporating biogeochemistry intoClements’s vision of a rational physiology. There are chal-lenges to this synthesis. Water and temperature are singlequantifiable things; biogeochemistry incorporates at least25 chemical elements, all required ingredients to build anorganism (Marschner 1995; Frausto da Silva and Williams2001; Wackett et al. 2004). Data on the organismal chemis-try of individuals and populations still, arguably, lag behinddata on their genomes. Unlike temperature and precipita-tion, we lack the baseline data describing biogeochemistryat scales from square meters to continents.But there are also opportunities to synthesize for gener-

ating deeply integrative biogeochemical ecology. Elemen-

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

The Case for Multiple-Element Limitation S63

tal shortfall creates patterns across scales, resulting in pa-thology and developmental abnormalities in individuals(Marschner 1995; National Research Council 2000, 2005;Snell-Rood et al. 2014), changing growth rates and behavior-al interactions within populations (Jones and Hanson 1985;Elser et al. 1996; McDowell 2003; Simpson et al. 2006) andlimiting ecosystem production and respiration (Elser et al.2007; Kaspari et al. 2008a; Townsend et al. 2011; Wiederet al. 2015). Increasingly economical technologies are aris-ing to quantify elements both within individuals and acrosslandscapes (Asner et al. 2015). A largepart of global change in-volves the global rearranging of elements via nitrogen (N) fer-tilizer manufacture and deposition (Vitousek et al. 1997;Hietz et al. 2011), phosphorus (P) mining (Smil 2000), so-dium (Na) pollution through road salt application (Kaspariet al. 2010), heavy metal pollution (Rauch and Pacyna 2009;Nagajyoti et al. 2010), and increased dust mobility (Fieldet al. 2010; Mahowald et al. 2010). Understanding the ele-mental recipes of organisms, the ecological function of thoseelements, and their geography has considerable potential forunderstanding our past, present, and future earth.

In this article, we discuss efforts so far to come to gripswith biogeochemical limitation in ecology, why we need toembrace the diversity of elements required to build organ-isms and drive ecosystems, and some ways we may do so,all toward achieving Clements’s and Vernadsky’s goal of bet-ter understanding the geography of life. We use this frame-work to speculate on a well-discussed but unresolved issue:the role of biogeochemistry in driving differences betweentemperate and tropical ecosystems (Walker and Syers 1976).Our primary motivation is to convince more young scientistsof a nagging intuition that has grown in the both of us: thatmuch of the pattern in population, community, and ecosys-tem ecology as one moves from place to place arises from ashortage of one or more of the 25 elements required for life.1

Liebig’s Fertilizer and the Originsof Nutrient Limitation

Nutrient limitation2 is a basic concept of integrative biology,beginning inside the cell and scaling up to ecosystems. Con-

1. We offer two caveats before proceeding: we are terrestrial ecologists, andnot particularly good ones. Forgive us if we ignore aquatics. We have also de-cided to pass when it comes to the dark side of biogeochemistry, the role of outand out toxins like Pb, and when too much of an essential element becomes abad thing.

2. At the outset, let us say that nutrient limitation is an imperfect expres-sion that we retain, given its utter ubiquity. To see why, consider opening alecture in principles of ecology with the line, “Nitrogen has long been knownto limit animal populations in grasslands.” The undergraduate audience couldwell be excused if half thought that the lack of nitrogen as a nutrient limits agrasshopper population’s growth rate, while the other half assumes that nitro-gen is keeping populations of soil invertebrates from increasing more by re-ducing the pH of the soil.

This content downloaded from 129.0All use subject to University of Chicago Press Term

sider the Droop equation (Droop 1974), designed to modelgrowth rate of a single cell

m p mm(12 q=Q):

Here, m is the biomass growth rate (day21), mm is the maxi-mum growth rate at theoretically infinite quota, q is themin-imum quota of a nutrient necessary to just support life(moles per cell at zero growth), and Q is the observed quota(moles per cell). As the nutrient supply in the cytosol in-creases, the part of the equation in parentheses convergeson 1, giving the maximum growth rate. On the basis of thesedynamics, nutrient limitation increases (and growth rate de-creases) as nutrient supplies decline. Nutrient limitation thusexists along a range of nutrient availability, where the per-formance of the system is enhanced with the addition ofnutrients or suppressed with their depletion.In the mid-1800s, Justus von Liebig—a polymath chem-

ist and biologist—devised a simple way to identify limitingelements, one not far distant from the cell scenario of theDroop equation (Brock 2002). Liebig aimed to maximize theyield of crop monocultures—one population, one genome,occupying a large area. His model was a metaphor (Liebig1855): he likened the problem to an oaken bucket with ver-tical staves of varying lengths. To increase the capacity of thebucket, it was not necessary to increase the length of each stave;it was only necessary to increase the length of the shorteststave. This metaphor sold lots of fertilizer (Brock 2002).Liebig’s law of the minimum (LLM) became a founda-

tional model for ecology. Codified by Sterner and Elser (2002),LLM has two assumptions. First, an element’s importancecan be ranked by the ratio of biological demand to ecosystemsupply (e.g., table 1). The second is that the physiology oforganisms and ecosystems is organized such that only oneelement—the one with the highest demand to supply ratio—limits metabolism. Placed in the context of Droop’s law, LLMhas two parts. First, elements are essential: any concentra-tion less than m∞ moves performance leftward and down-ward on the Droop curve. Second, energy and resources areexpended to harvest elements, and these costs diminish asavailability increases. Measuring the availability of nutrientsin the environment and in the tissues of the study systemthus informs the working hypothesis for which nutrients aremost likely to limit performance (e.g., Sterner 1997).The application of Liebig logic has since led to a focus

on three elements as candidates for most limiting nutrient,in large part on the basis of their considerable fraction inthe cells and tissues of life. Optimal foraging theory (Ste-phens and Krebs 1986) has successfully predicted how con-sumers—from notonectids to hummingbirds—choose amongfood and habitats simply by assuming that they are max-imizing the intake rate of energy (and ultimately carbon).

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

S64 The American Naturalist

Macroecology uses similar logic to link net primary pro-ductivity—in units of carbon/area/time—to increasing nu-trient abundance at continental scales (McNaughton et al.1989; Kaspari et al. 2000). Nitrogen, the next most commonelement beyond CHO, frequently limits biomass and pro-duction in plants and herbivores (White 1993; LeBauer andTreseder 2008). The growth rate hypothesis (Elser et al. 1996)is an elegant integrative hypothesis, extrapolating from P-richRNA and ribosomes to P-limited growth rates (Elser et al.2003). Clearly, biogeochemistry has a role to play as a domi-nant ecological template.

Why It Is Time to Retire Liebig and EmbraceColimitation across the Periodic Table

One’s success as a scientist can be measured more bythe number of people he or she puts to work on newproblems than by the correctness of specific researchresults. (Raup 1999)

TheDroop equation uses a single cell as its starting point.Nutrient limitation also expresses itself in tissues and or-ganisms as aggregates of cells and in populations and eco-systems as aggregates of individuals (Arrigo 2005; Sperfeldet al. 2015). In these collections of individuals and popula-tions in time and space, the seeds of colimitation are planted.

Liebig’s law has had a good run in introducing ecologiststo the power of biogeochemistry. LLMhas the power of par-simony; a single limiting element is the simplest startingpoint, but at the same time it has an epistemological Achil-les’s heel: it is difficult (and, to our knowledge, unheard of)to rigorously evaluate LLM’s prediction that one element,and one alone, contributes to an ecological process. Anotherreason to retire LLM lies in theH, O, C, N, Ca, P, S, Na, K, Cl,Mg, Si, Fe, Zn, Cu, I, Mn, F, Cr, Se, Bo, Mo, and Co in livingtissue; the patchy distribution of each; and the considerableenergy life invests maintaining them to precise levels of tol-erance (NationalResearchCouncil 2005; Salt et al. 2008).Com-bined, they suggest that the second half of LLM—a stricthierarchy of limitation—is limited to highly simplified sce-narios, like Liebig’s monoculture. LLM is thus wrong most

This content downloaded from 129.0All use subject to University of Chicago Press Term

of the time as we scale up from cells to ecosystems. Put an-other way, the more we look for biogeochemical colimita-tion, the more we are likely to find it.Let us now consider the ways (often complementary; Saito

et al. 2008; Sperfeld et al. 2015) that biogeochemical colimi-tation will be expressed, in which shortfall of more than onechemical element can limit individual performance, popu-lation growth, and ecosystem function.

Elements Serve Different, Essential,Complementary Functions

Studies of nutrient limitation have focused on water, carbo-hydrates, N, and P (roughly, H, O, C, N, and P). But considerhow electrolytes (Na, K, Ca, Cl) govern water balance andmetabolic activity; that Mg is a cofactor throughout glycol-ysis; and that S is essential for protein folding (Frausto daSilva andWilliams 2001).Moreover, while tracemetals com-prise 0.1% of animal mass—Mn, Fe, Co, Ni, Cu, Zn, Mo, B,Si, and Se—they are functional parts of up to one-third of∼4,000 known enzymes that catalyze life (Bairoch 2000; Wal-dron et al. 2009). Each of these elements is essential; short-falls in any cause pathology. And all are patchily distributedin space (Jones and Hanson 1985; John et al. 2007; Smithet al. 2014).Perhaps the single biggest case for ubiquitous colimita-

tion lies in its demonstrated application in nutritional ecol-ogy. Considerable time and energy has been invested in for-mulating optimal diets precisely because they maximize thehealth of model organisms (Smith 1987; Wertz 1987; Cohen2004). Likewise, mineral licks—supplementing structural, elec-trolytic, and trace elements—are placed in the field to enhancethe health and biomass of wildlife and domesticated ani-mals (Jones and Hanson 1985). Geometric optimal diet mod-els are consistently describing how consumers in the lab andfield balance their diets by aiming at mixed nutritional tar-gets (Simpson et al. 2010).

Limitation Cascades

Inmetabolic pathways, what happens in one part of the net-work often ramifies downstream. Limitation cascades arise

Table 1: Leibig logic applied to elemental composition (%) in the tropical forest Barro Colorado Island, Panama

Source

Ca Cu Fe K

15s

Mg

.039.080 on and Condition

Mn

January 04,s (http://ww

N

2017 13:45:1w.journals.u

Na

6 PMchicago.edu/

P

t-and-c).

Zn

Soil

.551 .011 8.556 .034 .230 .316 .429 .021 .038 .008 Litter 1 microbe 2.056 .005 3.155 .127 .317 .096 1.717 .008 .091 .008 Invertebrates .026 .002 .009 1.218 .627 .024 .034 1.377 .024 Ratio:

Soil litter

3.7 3.7 1.4 4.0 2.4 Soil invert 35.3 2.7 1.7 36.1 3.0

Note: Data for soils (n p 10), the litter 1 microbe matrix (n p 10), and decomposer invertebrates (n p 2). Percent N missing for invertebrates. Ratio indi-cates strong bioaccumulation and thus the prospect for nutrient limitation of litter microbes and invertebrates.

The Case for Multiple-Element Limitation S65

when nutrient limitation in one component influences theintensity of nutrient limitation in another. They can resultin a ranking of nutrient effects, not toward identifying a sin-gle limiting factor sensu Liebig but in distinguishing, in Pe-ter Vitousek’s (2010) parlance, proximate limitation (whereadding a nutrient influences a process but does not ramify)from ultimate limitation (where the added nutrient ramifiesand restructures the system, changing both process dynam-ics and standing stocks).

Limitation cascades were powerfully documented whenwhole lake P fertilization transformed oligotrophic lakes intoeutrophic ones (Schindler 1990). In the process, N inputsincreased through biological fixation, as did CO2 diffusion.Likewise, because shortages of N commonly limit popula-tion growth (Vitousek et al. 1993; White 1993), ecosystemrespiration (Horner et al. 1988), and productivity (LeBauerand Treseder 2008), any process increasing ecosystem Nhas the potential to generate a limitation cascade. For exam-ple, the fixation of atmospheric N2 to organic forms usefulfor life requires ample carbohydrates as well as Fe and Moto build the catalyzing enzyme, nitrogenase. In the dark, low-carbohydrate understory of a Panamanian rain forest, fertil-ization with the trace metal Mo generated two- to threefoldincreases in the production of organic N by heterotrophicmicrobes (Barron et al. 2008).Moreover, in an era of increas-ing CO2, the resulting increase in carbohydrate productiondensity should increase plant demand for N—while simulta-neously providing the energy for N-fixing microbes—andthe demand for Mo and Fe (Hungate et al. 2003).

This content downloaded from 129.0All use subject to University of Chicago Press Term

For plant consumers (i.e., herbivores and detritivores), Nais a prime candidate as an ultimate nutrient that can initi-ate limitation cascades. Sodium exists as a trace element inmost plant tissue (Marschner 1995). Herbivores and detriti-vores—particularly fungi and animals—maintain Na tissuelevels 100–1,000-fold higher than plants (Cromack et al. 1977;National Research Council 2005). Consumers confrontingNa shortfall often conserve Na by decreasing activity (Kas-pari et al. 2014). If Na shortfall decreases the physiologicalperformance of plant consumers, one corollary is that thepositive benefits of N and P to herbivores and detritivoreswill vary along a Na gradient. If so, gradients of Na avail-ability may account for the demonstrated variation in con-sumer responses to N and P fertilization (Haddad et al. 2000;Ritchie 2000; Milton and Kaspari 2007; Kaspari et al. 2008b,2009; Bishop et al. 2010; Loaiza et al. 2011; Joern et al. 2012;Lind et al. 2014).

Organisms Deplete Multiple Elements,Equalizing Utility: Availability

Another process promoting colimitation results from life’slong interaction with geochemistry (Frausto da Silva andWilliams 2001; Vernadsky 2012). If we plot the ubiquityof elements in a common vertebrate (us) against the avail-ability of elements in seawater or soil, we find a generallypositive relationship (fig. 1). It is hard, seeing this robust pat-tern, not to conclude that natural selection favors the use ofcommon elements when building ubiquitous biological struc-

Figure 1: The concentration of elements in tissue roughly mirrors its availability in seawater (a) and soil (b), with consequences forcolimitation (from Kaspari 2012).

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

S66 The American Naturalist

tures. One consequence is that the ratio of demand to sup-ply tends to equalize across the range of essential elements.Put another way, if the ratios of demand∶availability con-verge toward a common value, there will never be the clear,enduring maximum. Instead, the elements most in demandrelative to supply may alternate at a timescale proportionalto the physiology of the smallest organisms, among a num-ber of potential candidates (Appling and Heffernan 2014).Any experiment that sums over those cycles will observecolimitation.

Scale: Aggregating Populations over a Matrixof Biogeochemical Availability

A final way of achieving biogeochemical colimitation thatcomplements all the rest is relevant when ecologists mea-sure processes and standing stocks over larger areas, longertimespans, and more populations (Levin 1992). About halfof the globally observed variability in NP stoichiometry existsamong species in a given community (Hillebrand et al. 2009;

This content downloaded from 129.0All use subject to University of Chicago Press Term

Elser et al. 2010). The elemental recipes of taxa differ in waysdirectly relatable to their niche. One of us has found, for ex-ample, that across 26 forest stands in Peru and Panama, soilsricher in calcium supported more Ca-rich isopods; those richerin nitrogen supported more organisms that use N-rich silk(Kaspari and Yanoviak 2009). Any study examining the abun-dance and diversity of soil invertebrates collectively wouldconclude that Ca and N colimit this community.The role of nutrients in limiting decomposition is key to un-

derstanding half the carbon cycle (primary production beingthe other half ). In a square meter of forest floor, thousandsof species and billions of individual bacteria, fungi, and in-vertebrates break down leaf litter (Swift et al. 1979). Scatteredwithin the plot may be numerous kinds of substrate in var-ious states of decay (fig. 2). Suppose that two are a freshlyfallen leaf and a decayed leaf that is little more than ligninand cellulose. A third substrate is a dead grasshopper fallenfrom the forest canopy. If the bacteria and fungal decom-posers exploiting these substrates use different enzymes todo the job, and if a third of these enzymes requires metal

Figure 2: Conceptual model of the processes contributing to colimitation at increasingly coarse spatial scales. Photo of the forest canopy byR. Montgomery.

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

The Case for Multiple-Element Limitation S67

cofactors, it is easy to surmise that different portions ofthat square meter plot will vary in their demand for mineralelements.

Tests for colimitation and importance of elements otherthan N and P in decomposition have thus, not surprisingly,found it. Eleven-week lab incubations of Costa Rican leaflitter with added N, P, Zn, K, Mg, or Ni yielded higher massloss and CO2 efflux with P and Zn (Powers and Salute 2011).In a Panamanian forest, at least three elements—P, K, andat least onemicronutrient (B, Ca, Cu, Fe, Mg,Mn,Mo, S, and/or Zn) when applied to 40# 40-m plots—enhanced decom-position of cellulose wood and/or leaf litter (Kaspari et al.2008a). NaCl in inland forests, applied at dosages mimick-ing urine and coastal rainfall, enhanced decomposition asmuch as P and K (Kaspari et al. 2009, 2014). Ecosystems thatcombine the activity of 10217-g bacteria through 105-g treeswill invariably sum over the activities of a multitude of pop-ulations, each with their own footprint (Waring et al. 2015),most not directly interacting. When we ecologists quantifynutrient limitation, colimitation increasingly seems inevitable.

Is There a Geography of Colimitation?

If colimitation should be the working hypothesis in ecol-ogy, are we equally likely to observe it everywhere, or shouldwe expect geographic variation in which suites of nutrientsare most important? The Walker-Syers model (1976), alsoreferred to as the substrate age hypothesis (Zechmeister-Boltenstern et al. 2015), is a popular paradigm in terrestrialecosystems ecology, and biogeochemistry that provides amodel for where to expect N versus P limitation. In brief, themodel differentiates between elements whose ultimate sourceis fromrocks (e.g., P, cations) versus the atmosphere (primar-ily N). During primary succession (the classic thought exper-iment imagines a freshly formed lava flow condensing andcooling before being colonized by organisms), a terrestrial eco-system has the maximum amount of rock-derived nutrientcapital like P, with negligible quantities of elements like Nderived from the atmosphere. Over geological time as soilsand ecosystems develop, P availability decreases through pro-cesses such as leaching and occlusion to soil minerals, whichrender it unavailable. Concurrently, ecosystem N capital andavailability increase through processes including free-livingand symbiotic nitrogen fixation. Thus, over long periods oftime, nutrient limitation to primary producers is expectedto shift fromN to P, with perhaps an intermediate phase of Nand P colimitation.

Tests of theWalker-Syers hypothesis demonstrate the va-riety of ways ecologists can document the geography of nu-trient limitation. Studies of elemental composition of treesshow higher tissue P and higher P-induced growth rates athigher latitudes (Reich and Oleksyn 2004; Kerkhoff et al.2005; Lovelock et al. 2007). Chronosequence studies allow

This content downloaded from 129.0All use subject to University of Chicago Press Term

one to the test how local patterns of substrate age and weath-ering generate the same transition from N to P limitation(Wardle et al. 2004), often overlaying fertilization experimentswith N and P for good measure (Vitousek et al. 1993, 1997;Herbert et al. 2003). Collectively, these data support the im-portance of soil weathering at multiple scales in generatingtransitions from soil-derived nutrients to air-derived nutrients(Reich and Oleksyn 2004; Huston 2012).But can Walker-Syers’s logic for a latitudinal gradient of

single-element limitation be expanded to predict patternsof colimitation? We suggest two reasons why and then adda big caveat. First, multiple elements like Zn, Cu, and Mgare leached from rock. As one moves in time and space fromrock newly emerged from under a glacier that has been ex-posed to millions of years of weathering, it should result inthe simultaneous decline in a variety of essential elements.This is one way to move into a zone of potential colimita-tion (Saito et al. 2008; Appling andHeffernan 2014). Second,tropical diversity may be more likely to generate colimita-tion if individuals and species differ in their nutrient use anduptake, as has been shown in tropical forest trees (Baribaultet al. 2012; Waring et al. 2015). The increase in diversity ofmany tropical clades combined with the depletion of multi-ple essential elements predicts an increase in the incidenceof colimitation toward the equator.Now the caveat. The more biogeochemists map out the

distribution of elements, the more intriguing variation wefind at multiple spatial scales, including continental ones.For example, geomorphic processes of erosion and deposi-tion and/or nutrient inputs via sea salt inputs or dust depo-sition may disrupt substrate age hypothesis predictions ofwhich nutrients limit productivity (Chadwick et al. 1999;Porder et al. 2007; Weintraub et al. 2015) and, by extension,colimitation, by adding nutrients or exposing unweatheredsubstrate. Consider how a recent interpolated map of zincconcentrations in the top 5 cm of soil across the United States(Smith et al. 2014) demonstrates striking landscape and re-gional variability, with concentrations up to three orders ofmagnitude lower in the highly weathered soils of the South-east (fig. 3). Contemplating this and other maps in Smithet al.’s (2014) opus highlights that biogeochemical gradientsoccur at every spatial scale, arise from a variety of causes (in-cluding anthropogenic), and show just how much potentiallies in a research program that exploits these gradients.That said, we need to build on studies of N and P limita-

tion to reveal the extent of colimitation from populations toecosystems. One model toward this end is NutNet (Fay et al.2015), a distributed fertilization experiment in 42 grasslands,where 5# 5-m plots were fertilized with factorial combi-nations of N, P, or K combined with micronutrients. Acrossthe 42 sites spanning five continents, evidence for colimita-tion of aboveground primary production was pervasive, withNP limitation at 60% of the sites. Colimitation was observed

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

S68 The American Naturalist

on all continents, suggesting that the presence and mag-nitude of colimitation depends locally on climate and back-ground edaphic conditions. Nevertheless, N limitation toaboveground primary production was highest in cool, tem-perate latitudes, although low latitudes were not well repre-sented in the data set. Clearly, much more work is neededto resolve the where, when, and what elements questions re-lated to colimitation.

3. An element that is, with apologies to T. C. R. White, not nitrogen.

Going Forward

We understand that the parsimonious simplicity of LLMhas an attraction, and the first half of Liebig logic—the el-ement with the largest ratio of demand to availability ismost limiting—has gotten us this far. At the same time,as the integration of biogeochemistry into evolutionary andglobal change ecology goes forward, there are ample oppor-tunities for transformative insights. Here we identify a few,but first . . .

This content downloaded from 129.0All use subject to University of Chicago Press Term

Why Assuming Liebig Single-ElementLimitation Can Be Very, Very Bad

There are compelling reasons for embracing colimitationand its corollary: that we investigate the periodic table be-yond C, N, and P. In global change biology, interactions be-tween carbon and other elements are critical for accuratelymodeling the carbon cycle (Hungate et al. 2004; Townsendet al. 2011; Wieder et al. 2015). In studies of insect develop-ment, monarch butterflies reared on milkweeds enrichedwith road salt runoff showed altered muscle mass and neu-ral development compared with those on plants from pris-tine environments (Snell-Rood et al. 2014). Anthropogenicchanges in just one element’s distributions3 have the poten-tial to transform our understanding of such diverse fields ofinquiry as life-history evolution (Snell-Rood et al. 2015)and carbon cycling (Kaspari et al. 2009, 2014).

Figure 3: Zinc concentrations in top soil (0–5 cm) of the continental United States. Reprinted from Smith et al. (2014).

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

The Case for Multiple-Element Limitation S69

Colimitation makes the question of whether nitrogen lim-its X contingent on the availability of others. Thus, under-standing the covariance structure of nutrients in soils is use-ful for two reasons. First, it allows us to refine our answer(i.e., in P-poor soils, N is more limiting than in P-rich soils).Second, it provides us with ways of exploring the interactionof N with other nutrients, using the comparative method (forinspiration, see fig. 3).

Where to begin? Plots of an organism’s elemental com-position versus environmental availability can catalyze a lotof working hypotheses. Figure 1, comparing the compositionof a vertebrate endotherm to seawater, suggests that N, P, Fe,and Znwould be worth a look as limiting elements. Compar-ing that same stoichiometry to the forest soils of Peru, we seethat P, Na, and Zn show LLM’s requisite high ratio of use toavailability. But to make such plots, we need the raw data atscales applicable to the organisms and systems in question.Toward that end,Karasov andMartinezdelRio’s (2007)Phys-iological Ecology is a productive mix of literature and meth-odology toward evaluating dietary requirements.

Describing the Chemical Recipe of Lifeand Its Consequences

We suspect that at the time of this writing, it is easier tofind a population’s genome than its elemental recipe. Docu-menting stoichiometric variation within and between pop-ulations is a critical first step for generating hypotheses nu-trient limitation (Saito et al. 2008; Sterner 2008; Kaspari andYanoviak 2009; Kaspari et al. 2009). Such studies are reveal-ing that about half of the globally observed variability instoichiometry exists among species in a given community(Kraft et al. 2008; Hillebrand et al. 2009; Elser et al. 2010).Some of these differences reflect niche differences linkedto metabolic subunits: high densities of ribosomes (and hightiters of P) are found in bacterial taxa characterized by rapidgrowth (Stevenson and Schmidt 2004). To what extent dogradients of biogeochemistry undergird coexistence in com-munities (Tilman 1982), and what are the functional traitsthat provide a mechanism for these population interactions(McGill et al. 2006)?

Mapping Availability at Multiple Scales

For ecological stoichiometry to be predictive, we need bet-ter maps of biogeochemistry at scales relevant to our studyorganisms (e.g., fig. 3 is more likely to describe the Zn avail-ability experienced by a population of trees, less so a popu-lation of mites). New technologies and approaches have thepotential to transform our understanding of biogeochemi-cal distributions and their consequences. Maps of minerallicks—deposits rich in elements like Na, Mg, and Ca (Jonesand Hanson 1985)—allow for the study of mineral deficiency

This content downloaded from 129.0All use subject to University of Chicago Press Term

using the comparative approach. Networks of in situ solutesensors that collect high-frequency data on nutrient avail-ability may provide information on the temporal uncou-pling of nutrient supply and demand (Appling and Heffer-nan 2014). Last, recent advances in airborne imaging andspectralomics are providing regional-scale snapshots of fo-liar canopy chemistry and revealing strong coupling betweencanopy chemistry and underlying geologic and geomorpho-logical landscape templates (Asner and Martin 2009; Asneret al. 2015).

Testing for Colimitation

Testing for colimitation can seem daunting. How does onesensibly and systematically test which biogeochemicals drivebiology and ecology? As we document above, field factorialexperiments have been key in recognizing the ubiquity ofcolimitation (see also Sperfeld et al. 2015). However, at thesame time, such experiments blow up when more than fourelements—and their interactions—are simultaneously ex-plored. We suggest two expedient routes to effectively ex-plore colimitation. As a test case, we use the hypothesis thatan ecosystem’s decomposition rate is limited by biogeochem-istry. The kitchen sink experiment aims at documentingthe presence of colimitation by fertilizationwithmultiple ele-ments (i.e., everything but the kitchen sink), while the drugdiscovery experiments aims at rapid screening of candidateelements across the periodic table.

Manipulating Fertility: The Kitchen Sink

There are a variety of foundational working hypotheses ofnutrient limitation, which is a fundamentally bottom-upprocess (Power 1992). But beyond the null hypothesis ofno nutrient limitation in the system, a second relativelysimple hypothesis is that increasing the availability of allelements equally serves to maximize decomposition rates.Such experiments could relatively quickly describe what isbiogeochemically possible in the nutrient limitation of de-composition. At the same time, if performed across multi-ple sites, the magnitude of observed increase (or lack thereof )would be informative. If those ecosystems lie along a bio-geochemical, thermal, precipitation, or biodiversity gradi-ent, kitchen sink experiments could shed light on the relativecontribution of these different abiotic versus biotic, bottom-up versus top-down drivers.We envision two basic types of kitchen sink experiments.

The first would compare decomposition on control plotswith those that double the soil availability of N through Zn.The second would raise the nutrient availability of all sitesto that of the richest site. Obviously, the devil is in the details(i.e., what form of nutrients do you add, what concentrations,

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

S70 The American Naturalist

for how long). That said, it is interesting that a micronu-trient kitchen sink in Panama generated higher rates ofdecomposition than P or K fertilization plots (Kaspari et al.2008a).

Drug Discovery as a Model for Decomposingthe Biogeochemistry of Decomposition

Drug prospectors search for the compounds, dosages, andcontexts that control disease (Sterner and Elser 2002; Harvey2008; Kerns and Di 2008). We suggest that Big Pharma’sprogression from in vitro to in vivo—from low-investment,high-throughput tests to study direct effects to long-termstudies with the most promising candidates to study indi-rect effects—is another promising approach for studies ofcolimitation.

We envision a three-step screening process to identifywhich elements affect decomposition, similar to the drugdiscovery model. Much of drug discovery still relies on open(or grind and find) screening: broadly collecting organismsand testing for the efficacy of their constituent compounds(see Adriamycin and Taxol; Subramanian et al. 2006). Can-didate compounds are isolated and tested in vitro over hoursto days in simplified assays against laboratory cell cultures(phase 1). Such tests occur at a massive scale. Consistent suc-cess (e.g., killing cancer cells) moves a candidate drug to thenext phase of the trial. In the case of decomposition, a par-allel experiment would consist of adding a battery of sin-gle nutrient element additions (e.g., N, Ca, P, S, Na, K, Cl,Mg, Si, Fe, Zn, Cu, I, Mn, F, Cr, Se, Mo, Co) to decompos-ing microcosms in vitro for several weeks (following Powersand Salute 2011). Enhanced CO2 fluxes or mass loss rateswould move the candidate element to the next phase.

During phase 2, the fraction of candidate drugs that func-tion as predicted in tissue culture is next tested in model or-ganisms (e.g., mice). The focus is on the direct effect (e.g.,apoptosis, killing cancer cells) in a functioning system. If thecompound fails muster, further testing is discontinued. Suchtests take days to months. In decomposition experiments,candidate elements that increase CO2 efflux in soil shouldbe tested in mesocosms stocked with soil, litter, and macro-invertebrates (e.g., isopods) and monitored for several months.

The!0.1% of candidate compounds that reach therapeu-tic trials are administered in the target organism (typicallyus) under realistic conditions. The goal of phase 3 trials isto evaluate the drug’s function, the chains of ramifying in-direct interactions (side effects), and host acclimation (lossof drug function). Candidate elements advancing to phase 3in decomposition experiments would then be tested in thefield—in large-scale, longer-term (3–5 years) fertilization ex-periments alone and in factorial combinations—to look forthe presence of interactions and track effects throughout tro-phic levels.

This content downloaded from 129.0All use subject to University of Chicago Press Term

Conclusions

Humans are rearranging the geography of biogeochemistryand element availability, with uncertain consequences forecosystem function and evolution. Addressing these chal-lenges requires us to be equipped with robust theories andconceptual models that inform our data collection. We haveargued that it is time to retire Justus von Liebig’s law of theminimum and embrace a more complex view of element lim-itation that acknowledges the interactions of 251 elementsboth inside and outside the cell. We propose the untestedhypothesis that colimitation is more pervasive in the trop-ics than in the temperate or boreal zone. Although movingfrom one square on the periodic table to a much larger num-ber of squares—and from there to the biosphere—may seemdaunting, we think that the ground for such studies is fertile,and there is much to be learned.

Acknowledgments

This article was written with the support of National ScienceFoundation (NSF) EF-1065844 to M.K. and NSF DEB 1053237and the U.S. Department of Energy, Office of Science, Officeof Biological and Environmental Research, Terrestrial Eco-system Science Program award DE-SC0014363 to J.S.P. Wehave both been inspired by the work we have reviewed withinand to J. Hagen and V. Smil, who have written so elegantlyon the origins of many of the ideas we discuss here. Wealso thank M. Zuk for the opportunity to contribute to thissymposium issue and E. Bruna, M. Zuk, and an anonymousreviewer for helpful comments on a previous draft of thismanuscript.

Literature Cited

Appling, A. P., and J. B. Heffernan. 2014. Nutrient limitation andphysiology mediate the fine-scale (de)coupling of biogeochemicalcycles. American Naturalist 184:384–406.

Arrigo, K. R. 2005. Marine microorganisms and global nutrient cycles.Nature 437:349–355.

Asner, G. P., C. B. Anderson, R. E. Martin, R. Tupayachi, D. E. Knapp,and F. Sinca. 2015. Landscape biogeochemistry reflected in shiftingdistributions of chemical traits in the Amazon forest canopy. NatureGeoscience 8:567–573.

Asner, G. P., and R. E. Martin. 2009. Airborne spectranomics: map-ping canopy chemical and taxonomic diversity in tropical forests.Frontiers in Ecology and the Environment 7:269–276.

Bairoch, A. 2000. The ENZYME database in 2000. Nucleic Acids Re-search 28:304–305.

Baribault, T. W., R. K. Kobe, and A. O. Finley. 2012. Tropical treegrowth is correlated with soil phosphorus, potassium, and calcium,though not for legumes. Ecological Monographs 82:189–203.

Barron, A., N. Wurzburger, J. Bellenger, S. Wright, A. Kraepiel, andL. Hedin. 2008. Molybdenum limitation of asymbiotic nitrogen fix-ation in tropical forest soils. Nature Geoscience 2:42–45.

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

The Case for Multiple-Element Limitation S71

Bishop, J. G., N. B. O'Hara, J. H. Titus, J. L. Apple, R. A. Gill, and L.Wynn. 2010. NP co-limitation of primary production and responseof arthropods to N and P in early primary succession on MountSt. Helens volcano. PLOS ONE 5:e13598.

Brock, W. H. 2002. Justus von Liebig: the chemical gatekeeper. Cam-bridge University Press, Cambridge.

Chadwick, O. A., L. A. Derry, P. M. Vitousek, B. J. Huebert, andL. O. Hedin. 1999. Changing sources of nutrients during four mil-lions years of ecosystem development. Nature 397:491–497.

Clements, F. E. 1905. Research methods in ecology. University Pub-lishing, Lincoln, NE.

Cohen, A. C. 2004. Insect diets: science and technology. CRC, BocaRaton, FL.

Cromack, F. J., P. Sollins, R. L. Todd, D. A. J. Crossley, W. M. Fender,R. Fogel, and A. W. Todd. 1977. Soil microorganism-arthropod in-teractions: fungi as major calcium and sodium sources. Pages 78–84in W. J. Mattson, ed. The role of arthropods in forest ecosystems.Springer, New York.

Darwin, C. 1839. Journal of researches into the natural history and ge-ology of the countries visited during the voyage of the H. M. S. Bea-gle round the world, under the command of Capt. Fitz Roy, R. N.Colburn, London.

Droop, M. 1974. The nutrient status of algal cells in continuous cul-ture. Journal of the Marine Biological Association of the UnitedKingdom 54:825–855.

Elser, J., D. Dobberfuhl, N. MacKay, and J. Schampel. 1996. Organismsize, life history, and N∶P stoichiometry. BioScience 46:674–684.

Elser, J., W. Fagan, A. Kerkhoff, N. Swenson, and B. Enquist. 2010. Bi-ological stoichiometry of plant production: metabolism, scaling andecological response to global change. New Phytologist 186:593–608.

Elser, J. J., K. Acharya, M. Kyle, J. Cotner, W. Makino, T. Markow,T. Watts, et al. 2003. Growth rate–stoichiometry couplings in di-verse biota. Ecology Letters 6:936–943.

Elser, J. J., M. E. S. Bracken, E. E. Cleland, D. S. Gruner, W. S. Harpole,H. Hillebrand, J. T. Ngai, et al. 2007. Global analysis of nitrogen andphosphorus limitation of primary producers in freshwater, marineand terrestrial ecosystems. Ecology Letters 10:1135–1142.

Fay, P. A., S. M. Prober, W. S. Harpole, J. M. H. Knops, J. D. Bakker,E. T. Borer, E. M. Lind, et al. 2015. Grassland productivity limitedby multiple nutrients. Nature Plants 1:15080.

Field, J. P., J. Belnap, D. D. Breshears, J. C. Neff, G. S. Okin, J. J.Whicker, T. H. Painter, et al. 2010. The ecology of dust. Frontiersin Ecology and the Environment 8:423–430.

Frausto da Silva, J. J. R., and R. J. P. Williams. 2001. The biologicalchemistry of the elements: the inorganic chemistry of life. OxfordUniversity Press, Oxford.

Haddad, N. M., J. Haarstad, and D. Tilman. 2000. The effects of long-term nitrogen loading on grassland insect communities. Oecologia(Berlin) 124:73–84.

Hagen, J. B. 1992. An entangled bank: the origins of ecosystem ecol-ogy. Rutgers University Press, New Brunswick, NJ.

Harvey, A. L. 2008. Natural products in drug discovery. Drug Dis-covery Today 13:894–901.

Herbert, D. A., M. Williams, and E. B. Rastetter. 2003. A model anal-ysis of N and P limitation on carbon accumulation in Amazoniansecondary forest after alternate land-use abandonment. Biogeochem-istry 65:121–150.

Hietz, H., B. L. Turner, W. Wanek, A. Richter, C. A. Nock, and S. J.Wright. 2011. Long-term change in the nitrogen cycle of tropicalforests. Science 334:664–666.

This content downloaded from 129.0All use subject to University of Chicago Press Term

Hillebrand, H., E. Borer, M. Bracken, B. Cardinale, J. Cebrian, E.Cleland, J. Elser, et al. 2009. Herbivore metabolism and stoichiom-etry each constrain herbivory at different organizational scales acrossecosystems. Ecology Letters 12:516–527.

Horner, J., J. Gosz, and R. Cates. 1988. The role of carbon-based plantsecondary metabolites in decomposition in terrestrial ecosystems.American Naturalist 132:869–883.

Hungate, B. A., J. S. Dukes, M. R. Shaw, Y. Luo, and C. B. Field. 2003.Nitrogen and climate. Science 302:1512–1513.

Hungate, B. A., P. Stiling, P. Dijkstra, D. Johnson, M. Ketterer, G.Hymus, C. Hinkle, et al. 2004. CO2 elicits long-term decline in ni-trogen fixation. Science 304:1291.

Huston, M. A. 2012. Precipitation, soils, NPP, and biodiversity: res-urrection of Albrecht’s curve. Ecological Monographs 82:277–296.

Hutchinson, G. E. 1948. On living in the biosphere. Scientific Monthly67:393–397.

Joern, A., T. Provin, and S. T. Behmer. 2012. Not just the usual sus-pects: insect herbivore populations and communities are associatedwith multiple plant nutrients. Ecology 93:1002–1015.

John, R., J. W. Dalling, K. E. Harms, J. B. Yavitt, R. F. Stallard, M.Mirabello, S. P. Hubbell, et al. 2007. Soil nutrients influence spatialdistributions of tropical tree species. Proceedings of the NationalAcademy of Sciences of the USA 104:864–869.

Jones, R. L., and H. C. Hanson. 1985. Mineral licks, geophagy, andbiogeochemistry of North American ungulates. Iowa State Press,Ames.

Karasov, W. H., and C. Martinez del Rio. 2007. Physiological ecology:how animals process energy, nutrients, and toxins. Princeton Uni-versity Press, Princeton.

Kaspari, M. 2012. Stoichiometry. Pages 34–48 in R. M. Sibly, J. Brown,and A. Kodric-Brown, eds. Metabolic ecology: a scaling approach.Wiley, Chichester.

Kaspari, M., L. Alonso, and S. O'Donnell. 2000. Three energy variablespredict ant abundance at a geographic scale. Proceedings of theRoyal Society B: Biological Sciences 267:485–490.

Kaspari, M., N. A. Clay, D. A. Donoso, and S. P. Yanoviak. 2014. So-dium fertilization increases termites and enhances decompositionin an Amazonian forest. Ecology 95:795–800.

Kaspari, M., M. N. Garcia, K. E. Harms, M. Santana, S. J. Wright,and J. B. Yavitt. 2008a. Multiple nutrients limit litterfall and de-composition in a tropical forest. Ecology Letters 11:35–43.

Kaspari, M., and S. P. Yanoviak. 2009. Biogeochemistry and the struc-ture of tropical brown food webs. Ecology 90:3342–3351.

Kaspari, M., S. P. Yanoviak, and R. Dudley. 2008b. On the biogeog-raphy of salt limitation: a study of ant communities. Proceedingsof the National Academy of Sciences of the USA 105:17848–17851.

Kaspari, M., S. P. Yanoviak, R. Dudley, M. Yuan, and N. A. Clay. 2009.Sodium shortage as a constraint on the carbon cycle in an inlandtropical rainforest. Proceedings of the National Academy of Sciencesof the USA 106:19405–19409.

Kerkhoff, A., B. Enquist, J. Elser, and W. Fagan. 2005. Plant allom-etry, stoichiometry and the temperature-dependence of primaryproductivity. Global Ecology and Biogeography 14:585–598.

Kerns, E. H., and L. Di. 2008. Drug-like properties: concepts, struc-ture design and methods: from ADME to toxicity optimization.Academic Press, Burlington, MA.

Kraft, N. J. B., R. Valencia, and D. D. Ackerly. 2008. Functional traitsand niche-based tree community assembly in an Amazonian for-est. Science 322:580.

Krebs, C. J. 1994. Ecology. 4th ed. HarperCollins, New York.

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

S72 The American Naturalist

LeBauer, D. S., and K. K. Treseder. 2008. Nitrogen limitation of netprimary productivity in terrestrial ecosystems is globally distributed.Ecology 89:371–379.

Levin, S. A. 1992. The problem of pattern and scale in ecology. Ecol-ogy 73:1943–1967.

Liebig, J. V. 1855. Principles of agricultural chemistry with special ref-erence to the late researches made in England. Dowden, Hutch-inson & Ross, London.

Lind, E. M., E. T. Borer, A. Kay, and K. LaPierre. 2014. How changesto grassland landscapes impact insect communities: results from acollaborative distributed experiment. Entomological Society ofAmerica Annual Meeting, Portland, OR.

Loaiza, V., J. L. Jonas, and A. Joern. 2011. Grasshoppers (Orthoptera:Acrididae) select vegetation patches in local-scale responses to fo-liar nitrogen but not phosphorus in native grassland. Insect Sci-ence 18:533–540.

Lovelock, C. E., I. C. Feller, M. C. Ball, J. Ellis, and B. Sorrell. 2007.Testing the growth rate vs. geochemical hypothesis for latitudinalvariation in plant nutrients. Ecology Letters 10:1154–1163.

Mahowald, N. M., S. Kloster, S. Engelstaedter, J. K. Moore, S.Mukhopadhyay, J. R. McConnell, S. Albani, et al. 2010. Observed20th century desert dust variability: impact on climate and biogeo-chemistry. Atmospheric Chemistry and Physics 10:10875–10893.

Marschner, H. 1995. Mineral nutrition of higher plants. AcademicPress, San Diego, CA.

McDowell, L. R. 2003. Minerals in animal and human nutrition. Else-vier, Philadelphia.

McGill, B. J., B. J. Enquist, E. Weiher, and M. Westoby. 2006. Re-building community ecology using functional traits. Trends in Ecol-ogy and Evolution 21:175–185.

McNaughton, S., M. Oesterheld, D. Frank, and K. Williams. 1989.Ecosystem-level patterns of primary productivity and herbivory interrestrial habitats. Nature 341:142–144.

Milton, Y., and M. Kaspari. 2007. Bottom-up and top-down regu-lation of decomposition in a tropical forest. Oecologia (Berlin) 153:163–172.

Nagajyoti, P. C., K. D. Lee, and T. V. M. Sreekanth. 2010. Heavymetals, occurrence and toxicity for plants: a review. Environmen-tal Chemistry Letters 8:199–216.

National Research Council. 2000. Nutrient requirements of beef cat-tle. 7th ed. National Academies Press, Washington, D.C.

———. 2005. Mineral tolerance of animals. National Academies Press,Washington, D.C.

Porder, S., P. M. Vitousek, O. A. Chadwick, C. P. Chamberlain, andG. E. Hilley. 2007. Uplift, erosion, and phosphorus limitation interrestrial ecosystems. Ecosystems 10:158–170.

Power, M. E. 1992. Top-down and bottom-up forces in food webs:do plants have primacy? Ecology 73:733–746.

Powers, J. S., and S. Salute. 2011. Macro- and micronutrient effectson decomposition of leaf litter from two tropical tree species: in-ferences from a short-term laboratory incubation. Plant and Soil346:245–257.

Rauch, J. N., and J. M. Pacyna. 2009. Earth’s global Ag, Al, Cr, Cu,Fe, Ni, Pb, and Zn cycles. Global Biogeochemical Cycles 23:1–16.

Raup, D. M. 1999. The nemesis affair: a story of the death of dino-saurs and the ways of science. Norton, New York.

Reich, P. B., and J. Oleksyn. 2004. Global patterns of plant leaf N andP in relation to temperature and latitude. Proceedings of the Na-tional Academy of Sciences of the USA 101:11001–11006.

This content downloaded from 129.0All use subject to University of Chicago Press Term

Ritchie, M. E. 2000. Nitrogen limitation and trophic vs. abiotic influ-ences on insect herbivores in a temperate grassland. Ecology 81:1601–1612.

Saito, M., T. Goepfert, and J. Ritt. 2008. Some thoughts on the con-cept of colimitation: three definitions and the importance of bio-availability. Limnology and Oceanography 53:276–290.

Salt, D., I. Baxter, and B. Lahner. 2008. Ionomics and the study of theplant ionome. Annual Review of Plant Biology 59:709–733.

Schindler, D. 1990. Experimental perturbations of whole lakes as testsof hypotheses concerning ecosystem structure and function. Oikos57:25–41.

Simpson, S. J., D. Raubenheimer, M. A. Charleston, and F. J. Clis-sold. 2010. Modelling nutritional interactions: from individualsto communities. Trends in Ecology and Evolution 25:53–60.

Simpson, S. J., G. A. Sword, P. D. Lorch, and I. D. Couzin. 2006. Can-nibal crickets on a forced march for protein and salt. Proceedings ofthe National Academy of Sciences of the USA 103:4152–5156.

Smil, V. 2000. Cycles of life. Scientific American Library, New York.Smith, D. B., W. F. Cannon, L. G. Woodruff, F. Solano, and K. J.

Ellefsen. 2014. Geochemical and mineralogical maps for soils ofthe conterminous United States. U.S. Geological Survey Open-FileReport 2014-1082. U.S. Geological Survey, Denver.

Smith, J. C. J. 1987. Methods of trace element research. Pages 21–56in W. Mertz, ed. Trace elements in human and animal nutrition.Academic Press, San Diego, CA.

Snell-Rood, E., R. Cothran, A. Espeset, P. Jeyasingh, S. Hobbie, andN. I. Morehouse. 2015. Life-history evolution in the Anthropo-cene: effects of increasing nutrients on traits and trade-offs. Evo-lutionary Applications 8:635–649.

Snell-Rood, E. C., A. Espeset, C. J. Boser, W. A. White, and R. Smykalski.2014. Anthropogenic changes in sodium affect neural and muscledevelopment in butterflies. Proceedings of the National Academy ofSciences of the USA 111:10221–10226.

Sperfeld, E., D. Raubenheimer, and A. Wacker. 2015. Bridging facto-rial and gradient concepts of resource co-limitation: towards a gen-eral framework applied to consumers. Ecology Letters 19:201–215.

Stephens, D., and J. Krebs. 1986. Foraging theory. Princeton Univer-sity Press, Princeton, NJ.

Sterner, R. W. 1997. Modelling interactions of food quality and quan-tity in homeostatic consumers. Freshwater Biology 38:473–481.

———. 2008. On the phosphorus limitation paradigm for lakes. In-ternational Review of Hydrobiology 93:433–445.

Sterner, R. W., and J. J. Elser. 2002. Ecological stoichiometry: the bi-ology of elements from molecules to the biosphere. Princeton Uni-versity Press, Princeton, NJ.

Stevenson, B. S., and T. M. Schmidt. 2004. Life history implicationsof rRNA gene copy number in Escherichia coli. Applied and En-vironmental Microbiology 70:6670–6677.

Subramanian, B., A. Nakeff, K. Tenney, P. Crews, L. Gunatilaka, andF. Valeriote. 2006. A new paradigm for the development of anti-cancer agents from natural products. Journal of Experimental Ther-apeutics and Oncology 5:195.

Swift, M., O. Heal, and J. Andersen. 1979. Decomposition in terres-trial ecosystems. Studies in Ecology. Vol. 5. University of CaliforniaPress, Berkeley.

Tilman, D. 1982. Resource competition and community structure.Princeton University Press, Princeton, NJ.

Townsend, A. R., C. C. Cleveland, B. Z. Houlton, C. B. Alden, andJ. W. C. White. 2011. Multi-element regulation of the tropical forestcarbon cycle. Frontiers in Ecology and the Environment 9:9–17.

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).

The Case for Multiple-Element Limitation S73

Vernadsky, V. I. 2012. The biosphere. Springer Science & BusinessMedia, New York.

Vernadsky, W. 1924. La géochimie. Alcan, Paris.Vitousek, P. M., J. D. Aber, R. W. Howarth, G. E. Likens, P. A.

Matson, D. W. Schindler, W. H. Schlesinger, et al. 1997. Humanalteration of the global nitrogen cycle: sources and consequences.Ecological Applications 7:737–750.

Vitousek, P. M., S. Porder, B. Z. Houlton, and O. A. Chadwick. 2010.Terrestrial phosphorus limitation: mechanisms, implications, andnitrogen-phosphorus interactions. Ecological Applications 20:5–15.

Vitousek, P. M., L. R. Walker, L. D. Whiteaker, and P. A. Matson.1993. Nutrient limitation to plant growth during primary succes-sion in Hawaii Volcanoes National Park. Biogeochemistry 23:197–215.

Von Humboldt, A., and A. Bonpland. 2010. Essay on the geographyof plants. University of Chicago Press, Chicago.

Wackett, L. P., A. G. Dodge, and L. B. M. Ellis. 2004. Microbial ge-netics and the periodic table. Applied and Environmental Micro-biology 70:647–655.

Waldron, K. J., J. C. Rutherford, D. Ford, and N. J. Robinson. 2009.Metalloproteins and metal sensing. Nature 460:823–830.

Walker, T. W., and J. K. Syers. 1976. The fate of phosphorus duringpedogenesis. Geoderma 15:1–19.

Tropical forest of Barro Colorado Island, Panama. The thousands of sincreasingly seen to respond positively to additions of a variety of elem

This content downloaded from 129.0All use subject to University of Chicago Press Term

Wardle, D. A., L. R. Walker, and R. D. Bardgett. 2004. Ecosystemproperties and forest decline in contrasting long-term chronose-quences. Science 305:509–513.

Waring, B. G., L. Alvarez-Cansino, K. E. Barry, K. K. Becklund, S.Dale, M. G. Gei, A. B. Keller, et al. 2015. Pervasive and strong ef-fects of plants on soil chemistry: a meta-analysis of individual plant‘Zinke’ effects. Proceedings of the Royal Society B: Biological Sci-ences 282:20151001.

Weintraub, S. R., P. G. Taylor, S. Porder, C. C. Cleveland, G. P. Asner,and A. R. Townsend. 2015. Topographic controls on soil nitrogenavailability in a lowland tropical forest. Ecology 96:1561–1574.

Wertz, M. 1987. Trace elements in human and animal nutrition. Ac-ademic Press, San Diego.

White, T. C. R. 1993. The inadequate environment: nitrogen and theabundance of animals. Springer, Berlin.

Wieder, W. R., C. C. Cleveland, W. K. Smith, and K. Todd-Brown.2015. Future productivity and carbon storage limited by terrestrialnutrient availability. Nature Geoscience 8:441–444.

Zechmeister-Boltenstern, S., K. M. Keiblinger, M. Mooshammer,J. Peñuelas, A. Richter, J. Sardans, and W. Wanek. 2015. The ap-plication of ecological stoichiometry to plant-microbial-soil organicmatter transformations. Ecological Monographs 85:133–155.

Symposium Editor: Marlene Zuk

pecies that live here, each with a subtly different chemical recipe, areents. Photo credit: Michael Kaspari.

15.039.080 on January 04, 2017 13:45:16 PMs and Conditions (http://www.journals.uchicago.edu/t-and-c).