Pot size matters: a meta-analysis of the effects of rootingvolume on plant growth

Hendrik PoorterA,C, Jonas BhlerA, Dagmar van DusschotenA, Jos ClimentB

and Johannes A. PostmaA

AIBG-2 Plant Sciences, Forschungszentrum Jlich, D-52425, Germany.BINIA, Forest Research Centre, Department of Forest Ecology and Genetics, Avda A Corua Km 7.5.,28040 Madrid, Spain.

CCorresponding author. Email: h.poorter@fz-juelich.de

Abstract. Themajority of experiments in plant biology use plants grown in some kind of container or pot.We conducted ameta-analysis on 65 studies that analysed the effect of pot size on growth and underlying variables.On average, a doubling ofthe pot size increased biomass production by 43%. Further analysis of pot size effects on the underlying components ofgrowth suggests that reduced growth in smaller pots is caused mainly by a reduction in photosynthesis per unit leaf area,rather than by changes in leafmorphology or biomass allocation. The appropriate pot sizewill logically depend on the size oftheplants growing in them.Basedonvarious lines of evidencewe suggest that an appropriate pot size is one inwhich theplantbiomass does not exceed 1 g L1. In current research practice ~65% of the experiments exceed that threshold. We suggestthat researchers need to carefully consider thepot size in their experiments, as small potsmaychange experimental results anddefy the purpose of the experiment.

Additional keywords: container volume, experimental setup, meta-analysis, pot size, plant growth, rooting volume.

Received 16 February 2012, accepted 11 May 2012, published online 15 June 2012

Introduction

A large number of studies in plant biology focus on geneexpression, physiology or biomass production of individually-grown plants. To this end, experiments are often conducted onplants grown in somekindof container, fromhere on referred to aspots. Pot size has received special attention in forestry (Carlsonand Endean 1976) and horticulture (Kharkina et al. 1999), wherecommercial companies can prot from choosing the smallest potvolume that still delivers a product with an appropriate quality.Occasionally, the issue of pot size has received attention in otherelds of plant biology. Arp (1991), for example, emphasisedthat pot size might be an important issue in experiments thatconsidered the effect of elevated CO2 on plants; and recently, adiscussion in the eld of ecophysiology has emerged, wherestudies on the recognition of roots of neighbours are thought tobe confounded by pot size (Hess and De Kroon 2007).

Apart from the elds mentioned above, pot size seems tohave received little consideration in the scientic literature andis regulary not reported in the materials and methods section ofpublications. Nonetheless, it is an important issue. In mostlaboratories there is large demand for growth chamber facilitiesand the use of small pots generally implies more experiments orincreased replication. Space is probably less of an issue in mostglasshouses. However, the plant biology community currentlymakes a great effort to develop automated systems for plant

phenotyping (Granier et al. 2006; Nagel et al. 2012). Thesehigh-throughput systems allow for the handling of many plants,which automatically implies that spatial demands on growthfacilities will increase. The use of small pots has the additionaladvantage that it does not exceed the capabilities of the transportrobots to move the weight of pot plus plant.

The use of small pots for research purposes may also havedisadvantages, which are more related to biological constraints.A small container implies a small quantity of soil or othersubstrate and thereby, almost invariably, a reduction in theavailability of water and nutrients to the plant. In addition toreduced resource availability, pots generally impede root growth.Many species easily produce roots of more than 1m in length(Jackson et al. 1996), even at a relative young age (Drew 1975;Fusseder 1987), thereby exceeding the dimensions of mostcontainers used. Large plants in small pots may have a largefraction of roots pot-bound, with all kind of secondaryconsequences (Herold and McNeil 1979). In this paper weexplore the consequences of the choice of pot size for plantsstudied for experimental purposes. We rst analysed in aquantitative way to what extent pot size affects plant growth inthe studies that explicitly considered this factor. We did so by ameta-analysis of 65 studies reported in the literature. Second,we reported on some of the morphological and physiologicalcomponents that might explain the observed growth patterns.

CSIRO PUBLISHINGFunctional Plant Biology Reviewhttp://dx.doi.org/10.1071/FP12049

Journal compilation CSIRO 2012 www.publish.csiro.au/journals/fpb

Third, we analysed whether there is a threshold in the plantsize : pot size relationship above which growth is affected andcompared the data for the current pot-size experiments with amore general database on plant research and to plants grown inelds. Based on these comparisons we suggest what might be anappropriate pot size for a given experiment.

1. Effect of pot size on plant biomass

To analyse the effect of rooting volume on plant growth, wescreened the literature of the last 100 years and arrived at a totalof 65 publications where this factor was studied in pots and~10 where rooting volume was constrained in hydroponically-grown plants (see Appendix 1). For this analysis we will focusmainly on the results for the pot-grown plants. A description ofthe methodological approach is given in Appendix 2. The effectof pot size has been studied in a wide range of pot volumes, withvalues ranging from 5mL for a herbaceous greenhouse crop(Bar-Tal and Pressman 1996), to 1700L for trees growing over aperiod of several years (Hsu et al. 1996). The range of pot sizesused within the various experiments was also large, varying by afactor 2 at least and a factor 35 at most. A clear example in ourcompilation is the experiment by Endean and Carlson(1975), who followed the growth of Pinus contorta Douglasover time. In the very beginning of the experiment, plants grewpresumably well in all pot sizes, but after 4 weeks of growthbiomass was already reduced in the smallest pot volume(Fig. 1a), and by the last harvest, none of the pots seemedlarge enough to ensure unrestricted growth.

In the full compiled data set the picture is similar: in almostall of the experiments considered plants increase in weight withevery larger pot that is used (Fig. 1b). Most experimentsfall between the dotted lines, which indicate a xed ratio ofbiomass to pot volume of 2 and 100 gL1, respectively, for thelower and higher line. Only a few experiments with trees in largecontainers exceed the 100 gL1. As all experiments start withsmall seedlings, plants move over time from the bottom part ofthe graph upwards (cf. Fig. 1a). Clearly, the experiments at theupper right end of Fig. 1b are also the ones that were plannedto last for the longest time, 2 years in case of the experiment withthe largest pots (Hsu et al. 1996), 5 years in case of the experimentwith the highest ratio (Bar-Yosef et al. 1988).

Theoretically, doseresponse curves should level off athigher pot sizes, as in Fig. 1a. However, this is not the casefor many of the studies plotted in Fig. 1b, which mostly showlinear responses. We discuss this further in section 4. To scale alldata to an equal change in pot size, we calculated for eachspecies in each of the experiments the average slope of thelog-transformed mass versus pot size relationship and derivedfrom those values an easy to understand expression that indicatesthe percentage increase in biomass for a 100% increase in potsize (Appendix 3). On average, plants increased 43% in mass forevery doubling in pot size (Fig. 2, P < 0.001), with no signicantdifferences in response between herbaceous and woody species.These response values are substantial and imply that plantsare likely to be over 3 times larger in 2 L pots than in pots of0.2 L. Given these differences, it is clear that pot size shouldbe given careful consideration during the planning phase ofan experiment.

2. Effect on components of the carbon-budget

What is the cause of the growth retardation in smaller pot sizes?As clearly shown by Endean and Carlson (1975), the effect of potsize on biomass gradually increases throughout (the later part of)the experimental period. The rate bywhich biomass of individualplants is accumulated is proportional to the size of the plant andconveniently described by the relative growth rate (RGR, therate of increase in biomass per unit biomass present; Evans1972). The differences in RGR of plants growing in differentpot sizes are always smaller than the differences in biomass at theend of the experiment. This implies that the physiological andmorphological factors that underlie the variation in biomass willalso be affected to a smaller extent than biomass itself (Poorteret al. 2012a). We analysed growth in terms of the plants carboneconomy, using a top-down approach where the RGR isfactorised into three components:

RGR SLA LMF ULR; 1

0.01 0.10 1.000.001

0.01

0.1

1

10

0.01 0.10 1.00 10.00 100.00 1000.000.01

0.1

1

10

100

1000

10 000

100 000

Tota

l dry

mas

s (g)

4

8

14

20

0

Pot size (L)

Herb.Woody

(a)

(b)

Fig. 1. (a) Doseresponse curves of total plant dry mass of Pinus contortaas dependent on pot size. Plants were grown in pots with six different sizesand harvested at several times during a 20-week period. Numbers indicatethe time of harvest after sowing, in weeks. Data from Endean and Carlson(1975), except for seedmass (week 0), which was taken fromMcGinley et al.(1990). (b) Summary of total biomass data of plants grown in pots ofvarious sizes, as reported in the literature (65 experiments on 69 species,see Appendix 1). Each line connects the observations of one species orgenotype in one experiment. Values in red indicate woody species, in blueherbaceous species. Dotted lines indicate a total plant biomass per unit potvolume of 2 (lower line) and 100 (upper one) g L1. Additional, unpublisheddata were obtained for Barrett and Gifford (1995) and Liu and Latimer(1995), as well as from L. Mommer and H. de Kroon, and R. Pierik.

B Functional Plant Biology H. Poorter et al.

according to Evans (1972). Here, SLA denotes specic leaf area(leaf area per unit leaf dry mass; m2 kg1), LMF the relativeallocationof biomass to leaves (leafmass fraction, g leaf g1plant)and ULR is a parameter that indicates the growth rate per unitleaf area (unit leaf rate, gm2 day1). ULR is basically the netresult of carbon gain through photosynthesis, corrected for therate of respiration in the whole plant and the C-content of thenewly added biomass. ULR and photosynthesis per unit leafarea are often strongly positively correlated (Poorter 2002).

A characteristic of RGR is that an absolute difference in RGRbetween treatments causes a relative difference in biomass over

time (Poorter and Navas 2003). However, in the case where wewant to understand the reason for the difference in growth andonly have fragmented information, it ismore amenable to analysethe proportional differences in RGR relative to that of the threegrowth parameters that underlie RGR (Eqn 1). Only a fewexperiments report data on these underlying components, butthe information we could gather points into the followingdirection (see Fig. 2 and Fig. S1, available as SupplementaryMaterial to this paper): a doubling in pot size increases RGR by~5%. Consequently, each of the growth parameters at the right-hand side of Eqn 1 may change by a very small proportion, orone of them by a somewhat larger proportion in the range of5%. At the nal harvest, SLA increased somewhat in someexperiments and decreased in others. We could quantitativelycompare the various experiments by calculating the percentageincrease in SLA with a 100% increase in pot size and found thattaken over all experiments, this variable did not deviatesignicantly from zero (Fig. 2). Most, but not all experimentswith root restriction in hydroponics conrm this response (e.g.Carmi et al. 1983; Kharkina et al. 1999; but see Tschaplinski andBlake 1985).

Allocation patterns are more frequently reported, generallyas the biomass ratio between shoot and root. We prefer aclassication in at least three plant organs (Poorter et al.2012b) and therefore use LMF, SMF and RMF, which are thefractions of total vegetative biomass invested to leaves, stemsand roots respectively.The relatively scarce information indicatesthat LMF is not affected (Fig. 2a), whereas SMF increasedslightly but non-signicantly (0.05

limited and fragmentary evidence yet available, netphotosynthesis is likely to be the process that is strongestaffected by pot size and may explain best the observed potsize effect on biomass (Fig. 1b). Additional support comesfrom experiments where photosynthesis recovered quicklyafter plants were repotted in larger rooting volumes (Heroldand McNeil 1979).

3. What mechanism could explain a reducedphotosynthesis in smaller pots?

Several factors could explain the reduced rate of photosynthesis and thereby growth in smaller pots. A rst possible explanationis that containers of smaller dimensions can be placed at a higherdensity, with less light available for each shoot and hence, alower rate of photosynthesis. Although this could be the case insome of the compiled experiments, strong pot size effects onbiomass and photosynthesis are generally also observed whendensity is specically controlled for (e.g. Endean and Carlson1975; Robbins and Pharr 1988; NeSmith et al. 1992; Climentet al. 2011).

In the type of experiments that we included in our meta-analysis, a smaller pot size will inadvertently decrease the totalnutrient content in the pot. Low nitrogen and phosphorusavailability are known to decrease photosynthesis (e.g.,Sinclair and Horie 1989; Lynch et al. 1991) and growth andincrease the root mass fraction (Poorter et al. 2012b). Thus,lower resource supply could form a plausible explanation. Wecalculated the response of leaf nitrogen concentrations to changesin pot size, expecting to see an increase if nutrient availabilitywould explain the pot size effect. On average, there was a slight,but non-signicant increase in leaf nitrogen concentrations withpot size (Fig. 2), suggesting that this factor cannot completelyexplain the observed differences in photosynthesis or growth.Similar results were found for phosphorus (Krizek et al. 1985).This conclusion is to a certain extent supported by observationson hydroponically-grown plants, which have decreasedphotosynthesis and growth (Fig. 2b) when the root volumewas restricted, despite a continuous high supply of nutrients.However, unlike plants grown in pots, hydroponically-grownplants do not show an increased RMF when restricted (Fig. 2b).As increased RMF is a good indicator for nutrient stress, wepresume that nutrient limitation in small pots is still a factor,although we cannot exclude possible allometric effects whichcould explain a largerRMF in smaller plants aswell (Poorter et al.2012b).

Water is the other commodity that may be in short supply.Small pots could negatively impact the water status of plantsas they have a reduced total water holding capacity andwill, therefore, dry out more quickly (Tschaplinski and Blake1985) and at severe stress levels increase RMF (Poorter et al.2012b). Ray and Sinclair (1998) demonstrated with theirdrought experiment that soil in small pots dries out faster andthereby caused more severe drought stress in plants. However,pot size does not necessarily affect stomatal conductance orleaf water potential (Ronchi et al. 2006) and as fornutrients with plants growing in hydroponics there is still aclear effect of root connement, even though water availability isnot restricted.

Besides resource availability, the temperature of therooting volume could be affected by pot size (de Vries 1980).Pots can intercept a substantial amount of solar radiationespecially in experimental gardens and glasshouses, whichmay increase the soil temperature at the edge and eventuallyin the middle of the pot if no precautions are made (Martiniet al. 1991; Xu et al. 2001). Small pots have greater surfaceareas relative to their volume and thereby heat up morequickly. Townend and Dickinson (1995) measured 5Chigher day temperatures in 0.19 L pots compared with 1.9 Lpots. Keever et al. (1986) suggest that the greater temperatureuctuations in small pots may explain the reduced growth ofthe plants. High temperatures in the pot may have several direct(respiration, root growth) and indirect (through increasedmicrobial activity) effects on plant growth. Pot temperaturesare rarely reported so it is difcult to evaluate how oftentemperature differences between pots contribute to reducedgrowth. However, given that growth reductions also have beenobserved in hydroponically-grown plants suggests thattemperature differences alone cannot explain the observed poteffects either.

If neither nutrient orwater availability nor higher temperaturescan (fully) explain the pot size effects on photosynthesis andgrowth, it could be that root connement per semay cause growthretardation, with reduced photosynthesis as a consequence.Root growth is known to respond directly to impedance.Impeded roots stay shorter whereas the initiation and growthof side branches increases (Bengough and Mullins 1991; Faliket al. 2005). Furthermore, Young et al. (1997) showed that within10minof increasing the impedance to root growth, leaf expansionrate is reduced. This suggests that some kind of signal mayregulate shoot growth when a large proportion of the roots areimpeded. The actual signal for such a response remains as yetunknown. Possibly a reduced sink strength of the root systemcould cause a direct negative feedback on photosynthesis (Pauland Pellny 2003). Alternatively, a specic rootshoot signal isinvolved (Jackson 1993).

A crucial point in the evaluation of the lastly discussedoption is knowledge on the actual distribution of roots withinthe pots. Although the vertical root distribution is relativelyeasily measured (Price et al. 2002; Suriyagoda et al. 2010),analysis of the horizontal distributions is more complicated.Using non-destructive magnetic resonance imaging (MRI), wefollowed the root development of Hordeum vulgare L. andBeta vulgaris L. plants over time in three dimensional space.Representative nuclear magnetic resonance (NMR) images ofroot systems at the end of the experiment are shown in Fig. 3.We calculated the percentage of roots that was located inthe inner half of the soil volume, furthest away from wall andbottom and the percentage of roots present in the outer 4mm ofthe pot. Only 2025% of the root biomass was in the inner partof the pot (Table 1), whereas ~50% was found in the outer 4mm(20% of the total volume). The proportional distributionremained remarkably constant over time. Hence, if theseobservations have wider validity, we conclude that a relativelylarge fraction of roots is close to the edge of the pot, whereunfavourable environmental conditions, for example, largetemperature uctuations and impedance of the pot wall maynegatively impact growth.

D Functional Plant Biology H. Poorter et al.

4. When does pot size limitation starts?

In sections 1 and 2 we considered for each experiment theoverall effect on plant growth, morphology and physiologywhen pot size was doubled. However, it is to be expected thata plant of a given size will be constrainedmore in a small than in alarge pot. That is, young plants are initially not affected by potsize, but as plants grow older, the pot size effect becomes morepronounced, even in medium-sized pot volumes (Fig. 1a). When

experiments last for sufciently long time, even the largest potsize might not be large enough for unrestricted growth, i.e. thesaturating part of the curve extends beyond the largest pots used.For the experimental data this implies that the relationshipbecomes close to linear again. In fact, many of the curves inFig. 1b show a linear relationship. For experiments where onlytwo pot sizes were used, it is impossible to deduce whether theresponse of the plants is indeed linear or not. But even in many ofthe other experiments in Fig. 1b no clear saturation is shown.What is the reason for that?

One objective way to relate plant and pot size acrossexperiments is to calculate the plant biomass that is present ata given volume of rooting space. This variable, for which we useBVR as an acronym (total plant biomass : rooting volume ratio;g L1), has, to our knowledge, been used only by Kerstiens andHawes (1994). BVR values vary widely and ranged in ourdatabase from as low as 0.01 in work by Climent et al. (2008)to over 300 g L1 in work reported by Biran and Eliassaf (1980).The median value in the pot size experiments is around 9.5 forexperiments both with herbaceous and woody species (Fig. 4b).We tested whether the BVR could explain the form of thedoseresponse curves in the 65 experiments shown in Fig. 1b,by calculating for each point what the slope of the doseresponsecurves was, as well as the BVR. In order to be consistent withFig. 2, we derived the percentage increase in biomass with potsize doubling for these data as well. We found that very fewexperiments hadBVRvalues lower than 2 g L1 (Fig. 4b). Hence,for this part of the analysis we included not only data of the lastharvest, but alsodata of earlier harvestswhere available.Thismayimply that not all data points in the analysis are formallyindependent, but it increases the power of our analysis in thiscrucial range.We binned all values in ve BVR ranges and showthe resulting distribution in Fig. 5a. Estimation of slopes isalways more challenging than determining the absolute valuesper point and this may be one of the reasons that there isconsiderable variation within each category. In the categorywith a BVR between 1 and 2 g L1 the effect of pot size isclearly noticeable. Pot size effects are saturating when BVRvalues exceed the 2 g L1 (P < 0.01).

Another way to obtain a greater insight into the relationshipbetween plant biomass and pot size is to express the biomass of

(a) (b)

Fig. 3. (a) NMR image of a Hordeum vulgare plant grown in a pot with avolume of 1.3 L for 44 days. Roots in the inner 50% of the soil volume(furthest away from wall and bottom) are colour-coded yellow, roots in theouter 50% blue. The stem part that was masked from the analysis is shown inred. (b) Idem for a Beta vulgaris plant 48 days after sowing. The developingstorage root is colour-coded red and was not included in this case.

Table 1. The proportion of the root mass that is present in the inner 50% of the pot volume (more than 12.5mm fromthe wall or bottom) and in the space less than 4mm from wall or bottom of the pot, for Hordeum spontaneum and

Beta vulgaris plants growing in 1.3 L potsValues are based on six plants followed over time. Time is days after sowing. Standard error of the mean was on average 1.6

and 2.9 percentage points for the inner half and outer 4mm respectively

Hordeum vulgare Beta vulgarisTime(days)

Root mass ininner half (%)

Root mass inouter 4mm (%)

Time(days)

Root mass ininner half (%)

Root mass inouter 4mm (%)

26 32.0 43.2 31 20.3 53.528 28.9 50.0 33 20.8 53.131 17.7 55.3 34 21.5 54.133 21.3 52.7 39 22.9 52.634 23.1 51.7 41 23.0 52.339 25.2 48.7 48 22.3 52.041 28.2 48.0 44 32.4 44.7

Pot size matters Functional Plant Biology E

plants grown at various pot sizes relative to the biomass atthe largest pot size and plot these values against the BVR(Fig. 5b). Experiments where pot sizes are limiting growththroughout the full range of pot sizes are characterised by linesthat decline linearly with BVR. However, as long as biomass isnot affected, the linewill remain around 1 and only drop at greaterBVR values when pot size starts to reduce growth. For the fewexperiments where this was the case, we could show that thisinection point occurred somewhere between 0.2 and 2 gL1

(Fig. 5b). Thus, from both the full sets of experiments compiled,as for the more detailed analyses over time, we derive that potsize effects are particularly strong when BVR values are greaterthan 2 g L1.

5. How do these BVR values relate to otherexperiments and the eld?

As mentioned in section 1, the range of pot sizes used in thiscompilation is large. The median value is around 0.9 L (Fig. 4a).How does that compare to common practice in ecophysiologicalexperiments? This will partly depend on the species studied.

Arabidopsis, for example, is generally grown in much smallerpots (with an interquartile range of 0.080.21 L) than Zeamays L. (1.85.0 L). An overall impression of used pot sizescan be obtained from the metaphenomics database describedby Poorter et al. (2010), where the response of ~900 differentspecies to 12 different environmental factors is compiled for atotal of ~800 experiments from the literature. The median potsize used in that compilation of experiments turns out to be~23 times larger than those in the pot size studies (Fig. 4a),whereas themedianBVRvalue is ~4-fold lower (Fig. 4b). Hence,we conclude that most studies on pot sizes have focussed onrelatively small pots and have grown plants to larger sizesthan is common in ecophysiological experiments. In contrast,experiments in plant biology generally use relatively larger potsand harvest plants at younger stages, when the BVR value is stillbelow 8 gL1.

0.01

0.1

1

10

100(a)

(b)

9.3

0.5

1.6

2.5

3.0

2.0

10.5

Pot s

ize

(L)

1.2

0.01

0.1

1

10

100

Herb. Woody

WoodyWoodyMetaphenomics

database

BVR

(g L

1 )

Pot sizeexperiments

Herb. Herb.

Fig. 4. (a) Distribution of pot volumes as represented in the current meta-analysis of pot size studies and in a compilation of ~800 studies on the effectof 12 environmental factors on growth and related ecophysiological traits(meta-phenomics database, Poorter et al. 2010). (b) Total plant biomass : potvolume ratios (BVR) in the current meta-analysis and the meta-phenomicsdatabase. The distribution is characterised by boxplots (see legend Fig. 2).Blue boxes indicate the values for herbaceous species, red ones for woodyspecies. Numbers above/below each box show the median values.

0.1 1.0 10.00.0

0.2

0.4

0.6

0.8

1.0

1.2

Scal

ed to

tal d

ry m

ass

(g g

1 )

BVR (g L1)

20

0

20

40

60

80

100

120(a)

(b)

>2012 25 520

Most experiments with pots are conducted to eventuallyunderstand how (agro-)ecosystems function. It may thereforebe relevant to consider what normal BVR values are in the eld.Maximum dry matter yield of major crops varies between10003000 gm2 (Unkovich et al. 2010; assuming 20%biomass in roots). If we assume a rooting depth of 1m, thiswould correspond toBVRvalues in the range of 13 gL1. Giventhat at least half of the DM production takes place duringgrain lling, we can expect that during the vegetative state, theBVR value will not exceed 1.5 g L1. Similar calculations fornatural ecosystems are more difcult as large variation existsin root depth and the standing biomass. Given a rooting depthof 0.35m (Nagel et al. 2012) and a density of 500 plants per m2,a eld of Arabidopsis thaliana Heynh. plants of 0.1 g dry mass,would have a BVR of 0.15 g L1. Although we realise that plantsin the eld experience conditions that are very different fromthose where plants are grown singly in pots in controlledconditions, we conclude from these rough estimates that BVRvalues around 1 are of the same order of magnitude as those ofvegetative plants in the eld.

6. Does pot size affect experimental conclusions?

Up to now we have considered the effect of pot size per se. Mostresearchers are also interested in whether the outcome of theirexperiments is affected by the choice of the pot size. Arp (1991)was one of the rst to draw attention to the fact that pot sizemight restrict the response to elevated CO2. This would limit thepossibilities to drawconclusions fromexperiments that have beenconducted in this eld.

The analysis by Arp (1991) was a compilation of differentstudies that worked with different pot sizes. Kerstiens and Hawes(1994), however, published a meta-analysis of the results of arange of studies with trees in which they show that the biomassresponses to elevated CO2 did not correlate with pot size, or evendecreased with a BVR over 18 g L1. This suggests that small

pot sizes do not reduce responses to elevated CO2 universally.However, in both meta-analyses the evidence could only becircumstantial, as pot size was not an experimental factor inthe compilations. Here we analysed experiments where pot sizewas specically included in the experimental design, not onlyfor interaction with CO2 but also for nutrients, water andirradiance (Fig. 6). Given that nutrient and water availabilityalready increase when pot size is increased, we would expectless additional effect on plant growth if more nutrients or waterwere supplied. However, we would expect increasingly strongergrowth responses with larger pot size when light or CO2 wouldbe increased, simply because of the higher demand for nutrientsand water in larger plants. Although some of the experiments doindeed follow the expected trend, results are not equivocal. Asresults depend on the variability in at least four different harvests,the number of experiments is likely too small to draw any strongconclusion.

Besides possible interactions with abiotic factors, interactionswith biotic factors have been shown. For example, vesiculararbuscular mycorhizae (VAM) infection rates, which wouldnormally increase with reduced nutrient availability, arereduced in small pots (Bth and Hayman 1984; Koide 1991).As a consequence, VAM colonisation is less benecial fornutrient uptake and results in smaller growth increases whensmall pots are used (Kucey and Janzen 1987; Koide 1991).Similarly, Baldwin (1988) showed that pot-bound Nicotianaplants do not respond to leaf damage, whereas repotted plantsdo. Thus, although we cannot draw a rm conclusion here wesuggest that the use of small pots with crowded roots carries asubstantial risk of inuencing the experimental results.

7. Other considerations

This analysis focussed on the effect of pot volume on plantgrowth. However, choosing a pot for an experiment not onlyincludes choosing the right volume, but also the right shape.Although shape is less important than volume (McConnaughayet al. 1993), shallow and deep-rooting species may responddifferently to the actual diameter and height of the pots, atequal pot volume (von Felten and Schmid 2008). Pot height isalso an important factor in determining the free-draining watercontent of pots and thereby the water potential as well as theoxygen availability in the pots (Passioura 2006).

An alternative to standard plastic pots are containers that haveribbed inner sides and small air holes. Such containers promoteself-pruning of roots close to the holes, which avoids rootspiralling and promotes development of lateral roots (Rune2003). Not only shape, but also the material (Bunt andKulwiec 1970) and the colour of the pot (Markham et al.2011) may affect plant growth, mainly through their effect onsoil and root temperature. For a broader discussion on the use ofpots for growing plants in the context of experimental setup seePoorter et al. (2012a).

Conclusions

A meta-analysis of the effects of pot size on growth showsthat on average a doubling of the pot size results in 43% morebiomass. In most cases reduced growth in small pots will becaused by a reduction in net photosynthesis. It is the plant

0.1 1 10 1000.7

1

2

3

4

5

Rat

io b

iom

ass

high

vs.

low

(g g

1 )

Pot size (L)

Irradiance CO2 Nutrients Water

Fig. 6. Interaction of pot size with various abiotic factors. Results aregiven for total plant biomass and are expressed as the ratio between thebiomass at a high level of that factor and the lower level. Data are fromBilderback (1985), Thomas and Strain (1991), McConnaughay et al. (1993),NeSmith (1993), Ismail et al. (1994), Nobel et al. (1994), Barrett and Gifford(1995), Will and Teskey (1997), Houle and Babeux (1998), Centritto (2000)and R. Pierik (unpublished data).

Pot size matters Functional Plant Biology G

mass per unit rooting volume that is relevant rather than pot sizeper se. Large plant mass per pot volume not only reduces growthof plants but also carries the risk of inuencing therelative differences between treatments. We conclude that it isimportant for researchers to minimise such effects by choosingpots that are large enough for their plants, even at later stages ofgrowth. Our advice is to avoid plant biomass to pot volumeratios larger than 2 g L1 and preferably work with plant and potsizes where this ratio is

Kerstiens G, Hawes C (1994) Response of growth and carbon allocation toelevated CO2 in young cherry (Prunus avium L.) saplings in relation toroot environment. New Phytologist 128, 607614. doi:10.1111/j.1469-8137.1994.tb04024.x

KharkinaT,OttosenCO,RosenqvistE (1999)Effects of root restrictionon thegrowth and physiology of cucumber plants. Physiologia Plantarum 105,434441. doi:10.1034/j.1399-3054.1999.105307.x

Koide RT (1991) Density-dependent response to mycorrhizal infection inAbutilon theophrasti Medic. Oecologia 85, 389395. doi:10.1007/BF00320615

KrizekDT,CarmiA,MireckiRM,SnyderFW,Bunce JA (1985)Comparativeeffects of soil moisture stress and restricted root zone volume onmorphogenetic and physiological responses of soybean (Glycine max(L.) Merr.). Journal of Experimental Botany 36, 2538. doi:10.1093/jxb/36.1.25

Kucey RMN, Janzen H (1987) Effects of VAM and reduced nutrientavailability on growth and phosphorus and micronutrient uptake ofwheat and eld beans under greenhouse conditions. Plant and Soil104, 7178. doi:10.1007/BF02370627

Liu A, Latimer JG (1995) Water relations and abscisic acid levels ofwatermelon as affected by rooting volume restriction. Journal ofExperimental Botany 46, 10111015. doi:10.1093/jxb/46.8.1011

Lynch JP,LauchliA,EpsteinE (1991)Vegetativegrowthof the commonbeanin response to phosphorus nutrition. Crop Science 31, 380387.doi:10.2135/cropsci1991.0011183X003100020031x

Markham JW, Bremer DJ, Boyer CR, Schroeder KR (2011) Effect ofcontainer color on substrate temperatures and growth of red maple andredbud. HortScience 46, 721726.

Martini CA, Ingram DL, Nell TA (1991) Growth and photosynthesis ofMagnolia grandiora St Mary in response to constant and increasedcontainer volume. Journal of the American Society for HorticulturalScience 116, 439445.

McConnaughay KDM, Berntson G, Bazzaz F (1993) Limitations to CO2-induced growth enhancement in pot studies. Oecologia 94, 550557.doi:10.1007/BF00566971

McGinleyM, Smith C, Elliott P, Higgins J (1990) Morphological constraintson seed mass in lodgepole pine. Functional Ecology 4, 183192.doi:10.2307/2389337

Nagel KA, Putz A, Gilmer F, Heinz K, Fischbach A, Pfeifer J, Faget M,Blofeld S, Ernst M, Dimaki C, Kastenholz B, Kleinert AK, Galinski A,Scharr H, Fiorani F, Schurr U (2012) GROWSCREEN-Rhizo is a novelphenotyping robot enabling simultaneous measurements of root andshoot growth for plants grown in soil-lled rhizotrons. FunctionalPlant Biology, in press.

NeSmith DS (1993) Summer squash response to root restriction underdifferent light regimes 1. Journal of Plant Nutrition 16, 765780.doi:10.1080/01904169309364573

NeSmith DS, Duval JR (1998) The effect of container size. HortTechnology8, 495498.

NeSmith DS, Bridges DC, Barbour JC (1992) Bell pepper responses to rootrestriction. Journal of Plant Nutrition 15, 27632776. doi:10.1080/01904169209364507

Nobel PS, Cui M, Miller PM, Luo Y (1994) Inuences of soil volume and anelevatedCO2 level on growth andCO2 exchange for the crassulacean acidmetabolism plant Opuntia cus-indica. Physiologia Plantarum 90,173180. doi:10.1111/j.1399-3054.1994.tb02208.x

Passioura JB (2006) The peril of pot experiments. Functional Plant Biology33, 10751079. doi:10.1071/FP06223

Paul MJ, Pellny TK (2003) Carbon metabolite feedback regulation of leafphotosynthesis and development. Journal of Experimental Botany 54,539547. doi:10.1093/jxb/erg052

Poorter H (2002) Plant growth and carbon economy. In Encyclopedia oflife sciences. (Nature Publishing Group: London) Available at:http://www.els.net

Poorter H, Navas ML (2003) Plant growth and competition at elevated CO2:on winners, losers and functional groups.New Phytologist 157, 175198.doi:10.1046/j.1469-8137.2003.00680.x

Poorter H, Niinemets , Walter A, Fiorani F, Schurr U (2010) A methodto construct doseresponse curves for a wide range of environmentalfactors and plant traits by means of a meta-analysis of phenotypic data.Journal of Experimental Botany 61, 20432055. doi:10.1093/jxb/erp358

PoorterH, FioraniF, StittM,SchurrU,FinckA,GibonY,UsadelB,MunnsR,Atkin OK, Tardieu F, Pons TL (2012a) The art of growing plantsfor experimental purposes; a practical guide for the plant biologist.Functional Plant Biology, in press.

Poorter H, Niklas KJ, Reich PB, Oleksyn J, Poot P, Mommer L (2012b)Biomass allocation to leaves, stems and roots: meta-analyses ofinterspecic variation and environmental control. Tansley Review.New Phytologist 193, 3050. doi:10.1111/j.1469-8137.2011.03952.x

PriceAH,SteeleKA,GorhamJ,Bridges JM,MooreBJ,Evans JL,RichardsonP, Jones RGW (2002) Upland rice grown in soil-lled chambers andexposed to contrasting water-decit regimes. I. Root distribution, wateruse and plant water status. Field Crops Research 76, 1124. doi:10.1016/S0378-4290(02)00012-6

R Development Core Team (2011) R: A language and environment forstatistical computing. (R Foundation for Statistical Computing: Vienna,Austria) Available at: http://www.R-project.org/

Ray JD, Sinclair TR (1998) The effect of pot size on growth andtranspiration of maize and soybean during water decit stress. Journalof Experimental Botany 49, 13811386.

Robbins NS, Pharr DM (1988) Effect of restricted root growth oncarbohydrate metabolism and whole plant growth of Cucumis sativusL. Plant Physiology 87, 409413.

Ronchi CP, DaMatta FM, Batista KD, Moraes GABK, LoureiroME, DucattiC (2006) Growth and photosynthetic down-regulation in Coffea arabicain response to restricted root volume. Functional Plant Biology 33,10131023. doi:10.1071/FP06147

Rune G (2003) Slits in container wall improve root structure and stemstraightness of outplanted scots pine seedlings. Silva Fennica 37,333342.

Sinclair TR, Horie T (1989) Leaf nitrogen, photosynthesis, and crop radiationuse efciency: a review. Crop Science 29, 9098. doi:10.2135/cropsci1989.0011183X002900010023x

Suriyagoda LDB, RyanMH, RentonM, LambersH (2010)Multiple adaptiveresponses of Australian native perennial legumes with pasture potentialto grow in phosphorus- and moisture-limited environments. Annals ofBotany 105, 755767. doi:10.1093/aob/mcq040

Thomas RB, Strain BR (1991) Root restriction as a factor in photosyntheticacclimation of cotton seedlings grown in elevated carbon dioxide. PlantPhysiology 96, 627634. doi:10.1104/pp.96.2.627

Townend J, Dickinson AL (1995) A comparison of rooting environments incontainers of different sizes. Plant and Soil 175, 139146. doi:10.1007/BF02413019

Tschaplinski TJ, Blake TJ (1985) Effects of root restriction on growthcorrelations, water relations and senescence of alder seedlings.Physiologia Plantarum 64, 167176. doi:10.1111/j.1399-3054.1985.tb02331.x

Unkovich M, Baldock J, Forbes M (2010) Variability in harvest index ofgrain crops and potential signicance for carbon accounting: examplesfrom Australian agriculture. Advances in Agronomy 105, 173219.doi:10.1016/S0065-2113(10)05005-4

von Felten S, SchmidB (2008) Complementarity among species in horizontalversus vertical rooting space. Journal of Plant Ecology 1, 3341.doi:10.1093/jpe/rtm006

Will R, Teskey R (1997) Effect of elevated carbon dioxide concentrationand root restriction on net photosynthesis, water relations and foliarcarbohydrate status of loblolly pine seedlings. Tree Physiology 17,655661. doi:10.1093/treephys/17.10.655

Pot size matters Functional Plant Biology I

XuG,WolfS,KafkaU(2001) Interactiveeffect of nutrient concentrationandcontainer volume on owering, fruiting, and nutrient uptake of sweetpepper. Journal of Plant Nutrition 24, 479501. doi:10.1081/PLN-100104974

Young IM, Montagu K, Conroy J, Bengough AG (1997) Mechanicalimpedance of root growth directly reduces leaf elongation rates ofcereals. New Phytologist 135, 613619. doi:10.1046/j.1469-8137.1997.00693.x

Appendix 1. List of references used for the meta-analysis

A. Pot studies

Cris and Stout (1929) Agric. Exp. Stat. MSC, No 6.; Stevenson (1967) Can. J. Soil Sci. 47, 163174; Endean and Carlson (1975) Can.J. For. Res. 5, 5560; Hocking and Mitchell (1975) Can. J. For. Res. 5, 440451; Carlson and Endean (1976) Can. J. For. Res. 6,221224; Herold and McNeil (1979) J. Exp. Bot. 30, 11871194; Biran and Eliassaf (1980) Sci. Hortic. 12, 385394; Peterson et al.(1984) Agron. J. 76, 861863; Bilderback (1985) J. Env. Hortic. 3, 132135; Krizek et al. (1985) J. Exp. Bot. 36, 2538; Carmi(1986) Field Crops Res. 13, 2532; Hanson et al. (1987) HortSci. 22, 12931295; Kucey and Janzen (1987) Plant Soil 104, 7178;Ruff et al. (1987) J. Amer. Soc. Hort. Sci. 112, 763769; Tilt et al. (1987) J. Amer. Soc. Hort. Sci. 112, 981984; Bar-Yosef et al.(1988) Plant and Soil 107, 4956; Robbins and Pharr (1988) Plant Physiol. 87, 409413; Bar-Tal et al. (1990) Agron. J. 82, 989995;Gurevitch et al. (1990) J. Ecol. 78, 727744; Koide (1991) Oecologia 85, 389395; Latimer (1991) HortScience 26, 124126;Martini et al. (1991) J. Am. Soc. Hort. Sci. 116, 439445; Simpson (1991) North. J. Appl. For. 8, 160165; Thomas and Strain (1991)Plant Physiol. 96, 627634; Dubik et al. (1992) J. Plant Nutr. 15, 469486; NeSmith et al. (1992) J. Plant Nutr. 15, 27632776;Samuelson and Seiler (1992) Env. Exp. Bot. 32, 351356; Beeson (1993) J. Amer. Soc. Hort. Sci. 118, 752756; McConnaughayet al. (1993) Oecologia 94, 550557; NeSmith (1993) J. Plant Nutr. 16, 765780; Ismail et al. (1994) Aust. J. Plant Physiol. 21,2335; Menzel et al. (1994) J. Hortic. Sci. 69, 553564; Nobel et al. (1994) Physiol. Plant. 90, 173180; Ran et al. (1994) Agron. J.86, 530534; Barrett and Gifford (1995) Aust. J. Plant Physiol. 22, 955963; Liu and Latimer (1995) HortSci. 30, 242246; Mandreet al. (1995) J. Amer. Soc. Hort. Sci. 120, 228234; Agyeman et al. (1996) Ghana J. For. 2, 1424; Hsu et al. (1996) HortSci. 31,11391142; Huang et al. (1996) Plant and Soil 178, 205208; Ismail and Noor (1996) Sci. Hortic. 66, 5158; McConnaughay et al.(1996) Ecol. Appl. 6, 619627; Giannina et al. (1997) Acta Hortic. 463, 135140; Van Iersel (1997) HortSci. 32, 11861192; Willand Teskey (1997) Tree Physiol. 17, 655661; Houle and Babeux (1998) Can. J. Bot. 76, 16871692; Nishizawa and Saito (1998)J. Amer. Soc. Hort. Sci. 123, 581585; Ray and Sinclair (1998) J. Exp. Bot. 49, 13811386; Boland et al. (2000) J. Amer. Soc. Hort.Sci. 125, 135142; Centritto (2000) Plant Biosystem. 134, 3137; Haver and Shuch (2001) Plant Growth Reg. 35, 187196; Yeh andChiang (2001) Sci. Hortic. 91, 123132; Aphalo and Rikala (2003) New Forests 25, 93108; Loh et al. (2003) Urban For. UrbanGreen 2, 5362; Ronchi et al. (2006) Funct. Plant Biol. 33, 10131023; Arizaleta and Pire (2008) Agrociencia 42, 4754; Chirinoet al. (2008) For. Ecol. Man. 256, 779785; Climent et al. (2008) Silvae genetica 57, 187193; Goreta et al. (2008) HortTechnology18, 122129; Kurunc and Unlakara (2009) New Zeal. J. Crop and Hortic. Sci. 37, 201210; Oztekin et al. (2009) J. Food Agric. Env.7, 364368; Climent et al. (2011) Eur. J. Forest Res. 130, 841850; Nord et al. (2011) Funct. Plant Biol. 38, 941952; Mommer andDe Kroon (pers. comm.); Pierik (pers. comm.).

B. Hydroponically grown plants

Richards and Rowe (1977) Ann. Bot. 41, 729740; Carmi et al. (1983) Photosynthetica 17, 240245; Tschaplinski and Blake (1985)Physiol. Plant. 64, 167176; Hameed et al. (1987) Ann. Bot. 59, 685692; Peterson et al. (1991) J. Exp. Bot. 42, 12331240; Thomas(1993) Plant Growth Reg. 13, 95101; Ternesi et al. (1994) Plant Soil 166, 3136; Bar-Tal et al. (1995) Sci. Hortic. 63, 195208; Barand Pressman (1996) J. Amer. Soc. Hort. Sci. 121, 649655; Kharkina et al. (1999) Physiol. Plant. 105, 434441; Xu et al. (2001)J. Plant Nutr. 24, 479501.

J Functional Plant Biology H. Poorter et al.

Appendix 2

For this meta-analysis we screened the literature of the last 100 years. A total of 63 publications plus two additional unpublishedexperiments dealt with plants grown in pots of various sizes, 11 with plants grown in hydroponics with different levels of rootconnement. The references are listed in Appendix 1. As we were interested in not only the physical aspect of container volume, butalso in the resources that come with it, we compared pot size treatments based on size, including the possible additional benets ofincreased nutrient and water availability. The experiments with hydroponically-grown plants are not included in the main analysis.Differences in the shape of the pots were not independently analysed either. In several publications only pot diameter was reported.For a sample of 30 different pots ranging in diameter () from 7 to 40 cm we derived an estimate for pot volume (V) based on theempirical equation V =p(/2)2 (0.46 + 0.8397 0.0023072). If additional data were missing, we assumed pots to be lledwith substrate up to the rim.

For each species or genotype in a given experiment, we determined the biomass at the last harvest and separated this variable inbiomass of leaves, stems, roots and reproductive mass as far as data were provided. To capture the importance of pot size for growth wecalculated for each experiment the proportional increase in total plant dry mass relative to the proportional increase in pot size, bycalculating the slope of the lines that were tted through the observed log-transformed plant masses and pot volumes, separately foreach species in each experiment. Because experiments differed in the range of pot size used, we scaled these slopes in such a way thatthe number reects the percent change in biomass (or another variable) given a doubling in pot size (see Appendix 3 for more details).For a more detailed analysis, log-transformed data from experiments which included three or more pot sizes were also tted with asaturating equation, as described in Appendix 3 and calculated with the nls procedure in R (R Development Core Team 2011).

Appendix 3

Let V1 and V2 be pot volume 1 and 2 and B1 and B2 the total plant dry biomass that is observed at the respective pot volumes:because proportional differences are the focus of interest, we calculate the slope of the line that connects these points as:

S logB2 logB1logV 2 logV 1 ; 1

Suppose we want to know the fraction f by which plant mass increases when pot size doubles. Then

B2 f B1; 2and

V 2 2 V 1: 3Assuming a common slope s over the whole trajectory of pot masses considered, this results then in

S log f B1 logB1log2 V 1 logV 1

logf B1B1 log2V 1V 1

log f log2 ; 4

and, after rearrangement

f 10slog2 loglog2S 2S : 5The same procedure applies if slope s is determined by linear regression through more than two points.To relate the slope of the line to the observed values of total plant mass per unit pot volume (BVR), we took the following approach:

In cases where an experiment consisted of two pot sizes, a linear regression as above was calculated and the resulting slope attributedto both points. In case an experiment consisted of more than two pot sizes and a second order polynomial showed no signicantsaturating trend, we tted a straight line through all points. Otherwise, a saturating curve was tted through the points, of the form

B a P bP c ; 6

with the constraint that a, b and c should be positive. The slope of the line at each pot size is then given by the derivative:

S ac bV c : 7

After the slope and f were calculated for each pot volume in each experiment, data were binned in ve categories of BVR.

Pot size matters Functional Plant Biology K

Appendix 4.

We measured the root distribution of Hordeum vulgare (barley) and Beta vulgaris (sugar beet), grown in a glasshouse in cylindricalcontainers with a volume of 1.3 L (length 26 cm, inner diameter 8 cm). Measurements were done by non-destructive imaging of rootsusing nuclear magnetic resonance imaging (MRI) as described extensively by Jahnke et al. (2009). This method is able to detect rootswith a diameter down to 300mM, which implies that ne roots go unnoticed. Image segmentation of the root system was done bythresholding the image and removing the stem and the sugar beet, with the RegionGrowingMacro module in MeVisLab (ver. 2.2.1;MeVisLab, Bremen, Germany). After segmentation, the pixel values were integrated for the inner and outer regions of the soil core anddivided by the integral over the whole pot.

L Functional Plant Biology H. Poorter et al.

www.publish.csiro.au/journals/fpb