Theevolutionofwatertransportinplants:anintegrated approach · Plants can control climate as well as respond to it. Recent models supported by independent isotopic evidence suggest

The evolution of water transport in plants: an integratedapproachJ . PITTERMANN

Department of Ecology and Evolutionary Biology, University of California, Santa Cruz, CA, USA

ABSTRACT

This review examines the evolution of the plant vascular system from its beginnings in the green algae to modern

arborescent plants, highlighting the recent advances in developmental, organismal, geochemical and climatolog-

ical research that have contributed to our understanding of the evolution of xylem. Hydraulic trade-offs in vascu-

lar structure–function are discussed in the context of canopy support and drought and freeze–thaw stress

resistance. This qualitative and quantitative neontological approach to palaeobotany may be useful for interpret-

ing the water-transport efficiencies and hydraulic limits in fossil plants. Large variations in atmospheric carbon

dioxide levels are recorded in leaf stomatal densities, and may have had profound impacts on the water conser-

vation strategies of ancient plants. A hypothesis that links vascular function with stomatal density is presented

and examined in the context of the evolution of wood and ⁄ or vessels. A discussion of the broader impacts of

plant transport on hydrology and climate concludes this review.

Received 14 September 2009; accepted 26 December 2009

Corresponding author: J. Pittermann. Tel.: +1 831 459 1782; fax: +1 831 459 5353; e-mail: pittermann@

biology.ucsc.edu

INTRODUCTION

Plants can control climate as well as respond to it. Recent

models supported by independent isotopic evidence suggest

that it was the evolution of large, Devonian land plants that

may have contributed to the dramatic reductions in atmo-

spheric CO2 in the Late Devonian (Beerling & Berner, 2002;

Berner, 2005, 2006) while progressively deeper and more

complex root systems accelerated pedogenesis and carbon

sequestration (Retallack, 1997; Algeo & Scheckler, 1998;

Raven & Edwards, 2001; Berner, 2005). Although the contri-

bution of plants to atmospheric and biogeochemical processes

continues to be well explored (Pataki et al., 2003; Bonan,

2008; Malhi et al., 2008), the role of plant water transport in

responding to or acting on both of these components has

been somewhat overlooked.

There is a direct link, however, between plant water

transport and climate. Indeed, important inferences about the

levels of palaeoatmospheric CO2 have been based on measure-

ments of stomatal distributions in fossil plants. Stomata are

microscopic pores on surfaces of green leaves and stems that

allow CO2 to diffuse into photosynthetic tissue, while letting

water escape. Analysis of fossil plant material across the Phan-

erozoic suggests that low stomatal densities reflect elevated

palaeoatmospheric CO2 levels, whereas the opposite is true

for high stomatal densities (McElwain & Chaloner, 1996;

Chaloner & McElwain, 1997; McElwain et al., 1999; Retal-

lack, 2001; Beerling, 2002; Roth-Nebelsick, 2005; Kurschner

et al., 2008). Apparently, a greater number of smaller stomata

are needed to counter CO2 deficits at low ambient CO2,

rather than fewer, but larger stomata (Franks & Beerling,

2009a). However, stomata must also regulate water loss from

the leaf because, for each CO2 molecule fixed by the plant,

200–400 molecules of water may escape through the stomatal

pore. At any point in time, the stomata must optimize plant

water use efficiency, which is the carbon gain for a given

amount of water lost. This trade-off and its implications on

plant water balance, hydraulics and structure–function may

have been apparent early in the evolution of vascular land

plants.

What are the climatic consequences of transpiration for

atmosphere–biosphere interactions and does the answer lie

once again in the stomatal impressions of ancient leaves? How

has the evolution of water transport impacted the overall

design of both ancient and modern flora? The goal of this

review is to address these issues in an interdisciplinary and

accessible manner that draws not only on the fossil record, but

also on the recent advances in molecular, physiological and

evolutionary plant biology. Parts I and II introduce the basics

of the cohesion–tension (C–T) theory of water transport and

examine the evolutionary and developmental transitions that

selected for this ubiquitous and broadly accepted mechanism

112 � 2010 Blackwell Publishing Ltd

Geobiology (2010), 8, 112–139 DOI: 10.1111/j.1472-4669.2010.00232.x

from the earliest vascular land plants to trees. Part III high-

lights recent advances in our understanding of the structure

and function of plant vascular tissue (xylem), and its potential

utility for the reconstruction of water transport in extinct taxa

from climates that were radically different from today. The

review concludes with a discussion of the significance of plant

water transport on ecosystem and global hydrology.

PART I: THE EVOLUTION OF WATERTRANSPORT IN TERRESTRIAL PLANTS

Much of the information in Part I draws from the fossil

record, which is both rich with information yet fundamentally

incomplete. Indeed, the fossil record has limited resolution

with respect to the physiological, biochemical and life-history

traits that are integral to understanding the transition of

photosynthesis from an aquatic to a terrestrial setting. Com-

puter models, systematics and developmental approaches can

fill some of these gaps, but the veracity of these hypotheses is

ultimately tested against fossils.

Thus, fossil specimens are scrutinized with much caution.

The majority of fossils are preserved in riverine depositional

environments, meaning that they may have travelled great

distances from potentially very different habitats before

settling permanently (Stewart & Rothwell, 1993; Taylor

et al., 2009). Furthermore, fragmentation can disassociate

fragile structures such as leaves from the rest of the plant body,

thereby obscuring relationships between various tissue sam-

ples and potentially inflating species numbers. For instance, it

was not until the discovery of an intact specimen bearing both

wood and foliage that Archaeopteris was treated as a single

species rather than two separate genera (Beck, 1960; Meyer-

Berthaud et al., 1999). Lastly, diagenesis, that is the chemical

and physical alteration of fossil and sediment during the

conversion to rock, can also cause substantial changes to the

structure and character of a specimen (Taylor et al., 2009).

Sampling biases may affect approximations of species

diversity in the fossil record leading to errors in estimates of

the rates of evolution, particularly if workers focus on particu-

lar localities or intervals. In general, if diversity is constant,

sampling intensity will not affect results (Raymond & Metz,

1995; Tarver et al., 2007). However, evidence suggests that

earlier estimates of diversity of Mid-Silurian through Late

Devonian land plants may be higher than expected due to sam-

pling bias, usually attributed to workers’ preferential interest

in one era or locality over another (Raymond & Metz, 1995).

Fortunately, plant vasculature is often the most abundant

and best preserved plant tissue. Indeed, some xylem samples

from the earliest land plants appear as though they were

frozen in time, with little distortion to the cell structure.

Remarkably, even the chemical composition of early vascular

tissue can be discerned to the point where we can reasonably

speculate about the biomechanical properties of the stem, as

well physiological ecology of the plant (Kenrick & Crane,

1991; Bateman et al., 1998; Boyce et al., 2003; Wilson et al.,

2008). For this reason, macroevolutionary and neontological

approaches can be fruitfully applied to research addressing

evolution of plant water transport, or the relationships of fossil

plants to their climates.

The mechanism of water transport in plants

Water transport in plants is not difficult to understand, but nei-

ther is it inherently intuitive. In his book ‘Vegetable Staticks’

(1727), the English physiologist Stephen Hales first remarked

that water is released from a plant by ‘perspiration’. He thus

captured an important element of the C–T theory, which

requires that water is lost from the plant by evaporation in

order for bulk flow to occur. The C–T theory was fully devel-

oped by Dixon and Joly (1894), and remains to this day the

most parsimonious and experimentally supported explanation

of plant hydration. The fundamental tenet of the C–T mecha-

nism is that the water column can sustain a very high degree of

tension without ‘breaking’. This phenomenon is entirely con-

sistent with the properties of water, yet so unusual that a few

workers find the C–T mechanism disturbing. Despite recent

controversy, the C–T mechanism has withstood vigorous

experimental scrutiny and remains the most widely accepted

explanation of water movement in land plants (Pockman et al.,

1995; Cochard et al., 2001; Angeles et al., 2004).

Water may be transpired at a high rate from the stomatal

pores, so effective hydraulic transport must efficiently replace

this lost water. Failure to supply water to the shoot as quickly

as it is lost results in the familiar phenomenon of wilting fol-

lowed by damage to cell membranes and ending with death

(McDowell et al., 2008). Roots are a crucial component for

water entry into the plant yet for the most part, it is the aerial

tissues that drive water movement, much like cut flowers in a

vase.

The process of water transport begins when water is lost

from the mesophyll cell walls located in the interior of the leaf.

Plant cell walls are a tight but porous weave of cellulose and

pectin that act much like a wick. This evaporation creates

capillary suction on the menisci within the cell wall pores,

causing tension to be transmitted down the water column in

the xylem (Pickard, 1981). The xylem tension exists because

the water column is engaged in a ‘tug-of-war’ between

adhesion to soil particles and evaporation from the leaf sur-

face. This tension can also be described as a ‘negative pressure’

that is quantitatively referred to as the water potential (W) and

typically measured in Pascals (Pa) such that increasingly nega-

tive water potentials indicate greater tension imposed on the

water column. Because water moves down a water potential

gradient and the water potential of the roots is more negative

than that of the surrounding soil, water enters the roots.

Water movement in plants is referred to as the C–T mecha-

nism because hydrogen bonding holds the water molecules

together and their movement is the result of the pulling action

The evolution of water transport in plants 113

� 2010 Blackwell Publishing Ltd

(i.e. tension) created by evaporation from the cell wall pores

(Sperry, 2000; Tyree & Zimmermann, 2002; Taiz & Zeiger,

2006).

Because transpiration is chiefly driven by sunlight, water

movement through plants is a passive and therefore cheap

process that selected for low resistance, efficient transport

conduits (Sperry, 2003). These are dead, empty, require little

to no metabolic input and can provide a network that allows

water to reach tree canopies over 100 m in height (Koch

et al., 2004; Burgess et al., 2006; Domec et al., 2008).

Underlying this transport capacity is a fundamentally simple

mechanism whose origins required little more than a hydro-

philic cell wall to move water by capillarity. Green algae

possess such a cell wall for the purpose of osmoregulation, so

the framework for the evolutionary transition from wicking

water over the surface of a cell to bulk flow within the special-

ized conduits of vascular plants was realized well before the

colonization of land (Sperry, 2003; Franks & Brodribb,

2005). In addition to the cell wall, the key innovations that

allowed plants to maintain hydration and adequate rates of gas

exchange on land were the cuticle, the stomata and the evolu-

tion of vascular tissue. The functional significance, evolution-

ary history and developmental elements of these traits will be

addressed below.

The green algae: earth’s first land plants

The connection between land plants and green algae (Chloro-

phyta) has been apparent since centuries before Darwin’s

time, and there is no doubt that embryophytic land plants are

descendants of freshwater green algae (Graham, 1993;

Graham et al., 2000; Lewis & McCourt, 2004; McCourt

et al., 2004). Specifically, it is believed that for the Charophy-

tic green algae lacking bicarbonate pumps, competition for

CO2 may have been the driving force behind their venture

onto land (Graham, 1993). Clearly, the photosynthetic

benefits of fourfold increased CO2 diffusivity and higher light

levels outweighed the costs of exposure to UV radiation and

dessication (Denny, 1993; Graham, 1993). A complete dis-

cussion of the morphological, biochemical and reproductive

adaptations predisposing Charophytic algae to life on land is

beyond the scope of this review, but excellent treatments can

be found in Graham (1993), Graham et al. (2009), Gensel

(2008) and Gensel & Edwards (2001). Because the fossil

record of the green algae is fragmentary and incomplete

(Graham, 1993), the discussion will be limited to recent pro-

gress regarding the algal ancestry of land plants.

The green algae display a surprising array of cellular and

colonial morphologies, and some species such as Chara

braunii, may superficially resemble plants (McCourt et al.,

2004). Although green algae are typically associated with

aquatic ecosystems, they can also be indigenous and highly

speciose elements of desert soil microbiotic communities

(Evans & Johansen, 1999). Although the most widely recog-

nized transition of green algae to land has been the one leading

to land plants, recent evidence indicates that the green algae

have colonized terrestrial surfaces on at least a dozen separate

occasions (Lewis & Lewis, 2005). Molecular clock hypotheses

suggest that the first green algae may have appeared in the

Mid-to-Late Mesoproterozoic (1.4–1.0 billion years ago;

Yoon et al., 2004) but the date of their venture onto land is

unclear due to poor preservation (Taylor et al., 2009). The

oldest observed silicaceous or calcium carbonate remains of

green algae are from the Upper Silurian (Feist et al., 2005).

Molecular phylogenies show high support for the two

Chlorophytic and Charophytic lineages of the green algae

(Lewis & McCourt, 2004; Fig. 1), and there is a general

consensus for the placement of the Charophytic lineage as

ancestral to land plants because of the high number of shared

derived molecular, biochemical, reproductive, developmental

and morphological traits between these groups (Graham,

1996; Raven & Edwards, 2004). Nonetheless, the issue of

which Charophytic group is most closely related to terrestrial

plants remains somewhat controversial primarily because of

different approaches to gene and taxon sampling (Qiu et al.,

2006, Qiu, 2008; McCourt et al., 2004; Gensel, 2008; see

also Turmel et al., 2007).

Despite the progress in elucidating the phylogenetic rela-

tionships of the green algae and their affinities to land plants,

the shared physiological and biochemical elements of dessica-

tion tolerance have attracted less attention (Graham & Gray,

2001). This comes despite Raven’s early efforts to provide

such a context for the evolution of Charophytes and his mis-

givings about the suitability of Charales to life on land based

on latter group’s physiology and morphology (Raven, 1977).

Perhaps, the plasticity of algal forms (Bhattacharya & Medlin,

1998), combined with the antiquity of the Charophyte

to embryophyte divergence, hopelessly complicates such a

mapping exercise (Graham & Gray, 2001) but it could

nonetheless inform the plausibility of current evolutionary

scenarios because environmental water stress exerts some of

the strongest selective pressures on terrestrial (and marine)

biota (Yancey et al., 1982).

Graham & Gray (2001) argue that the closest Charophytic

ancestors of land plants would have been dessication-tolerant

and occupied ephemeral, ‘variably unfavourable’ aquatic habitats.

These habitats acted as sites of strong evolutionary pressures

related to the effects of water deficit, and thus provided

the Charophytes with an impetus towards terrestrialization

(Graham & Gray, 2001). Consequently, the water status of

terrestrial Charophyceans would have fluctuated in response

to water availability (so-called poikilohydry), and the evolution

of desiccation tolerance would have been essential in allowing

them to recover from otherwise fatal (80–90%) protoplasmic

water loss. Dessication tolerance is a common phenomenon

in basal plant lineages characterized by small-statured pheno-

types such as the algae, bryophytes and some ferns (Proctor &

Tuba, 2002). This trait has been lost, however, in more

114 J . PITTERMANN


derived plants due to structural constraints that prevent the

collapse and refilling of cells, as well as the high cost of solute

production required to minimize drought injury to cell

membranes and metabolic components (Beck et al., 2007).

The significance of the plant cuticle

Although multicellularity reduces water loss by reducing total

surface to volume ratios, it is unlikely that an algal colony

larger than few hundred cells could have maintained adequate

hydration for long periods of time without a means of restrict-

ing water loss, or efficiently replacing lost water. Additionally,

because a water film represents a significant barrier to CO2

diffusion, such a simple multicellular arrangement may have

significantly constrained photosynthesis (Franks & Brodribb,

2005). All land plants must then reconcile two opposing bio-

physical constraints. On the one hand, photosynthesis

requires large surface areas to facilitate light capture and CO2

diffusion, whereas on the other, water conservation favours

low surface:volume ratios. Three critical innovations – the

cuticle, the stomata and water-conducting tissues – allowed

plants to resolve these conflicting requirements. Of these

three traits, the cuticle is the oldest. We currently know noth-

ing about the evolution of the cuticle, except that it is a syna-

pomorphy of land plants, the earliest fragments of which were

found in the Ordovician (Wellman et al., 2003). Neverthe-

less, one can neither discuss the evolution of plant homoihydry

(that is the ability to maintain hydration over a range of exter-

nal conditions of water availability; Raven & Edwards, 2004;

Raven, 1984) nor the plant fossil record (Willis & McElwain,

2002; Wellman et al., 2003; Taylor et al., 2009) without

mentioning this feature.

The 1 to 15 lm thick cuticle covers most of the aerial

elements of land plants where it protects the shoot from

herbivory, UV-induced damage and water loss (Edwards

et al., 1996; McElwain & Chaloner, 1996; Raven & Edwards,

2004). The cuticle is essentially a waxy layer on the external

cell walls of the epidermis (Bargel et al., 2004, 2006), in

which cutin forms the highest fraction and acts as a matrix for

the deposition of fatty acids, alkanes and waxes. Because the

Fig. 1 The phylogeny of land plants in the context

of palaeoclimate and clade radiations over Phanero-

zoic time, updated from Sperry (2003). The graph

indicates modelled atmospheric CO2 concentrations

according to Berner (2006), as well as correspond-

ing large-scale climatic patterns (Zachos et al.,

2001; Willis & McElwain, 2002; Berner, 2005; Huey

& Ward, 2005; Brezinski et al., 2008; Trotter et al.,

2008). The phylogeny is from Sperry (2003) with

updates from Burleigh & Mathews (2004) and Qiu

(2008). Important morphological features relevant

to the evolution of plant water relations are indi-

cated on the phylogeny according to Sperry (2003)

with several updates (Proctor, 1981; Franks &

Farquhar, 2007; VanAller Hernick et al., 2008).

Presence in the fossil record is denoted by thick

black line. Relative tracheophyte abundances were

estimated or re-drawn from Niklas (1992, 1997),

Willis & McElwain (2002), Singh (2004) and

Montanez et al. (2007). There is good fossil

evidence to suggest that liverworts may have been

present as early as the Late Ordovician (dotted line;

Wellman et al., 2003).



cuticle can significantly restrict water loss, it is unlikely that

the colonization of land by plants would have occurred with-

out the evolution of this layer (Niklas, 1992; Raven &

Edwards, 2004). Waxes are the key constituents of the cuticle

barrier but it is the cutin fraction that is most resilient to

degradation and thus best conserved in the fossil record

(Schreiber & Riederer, 1996). It is noteworthy that some

Charophytic algae can synthesize a wax-free extracellular layer

that may allow a limited amount of water and CO2 diffusion

(Graham, 1993; Edwards et al., 1996).

Historically, the usefulness of fossilized cuticle lay primarily

in the patterns of preserved stomatal impressions whose size,

distribution and abundance are used to infer palaeo-atmo-

spheric CO2 concentrations (McElwain & Chaloner, 1996;

McElwain et al., 1999; Royer et al., 2001; Retallack, 2001;

Kouwenberg et al., 2003). Unfortunately, cuticle thickness

alone is a poor predictor of species’ drought tolerance because

cuticle permeability and thickness may not be related

(Edwards et al., 1996; McElwain & Chaloner, 1996). How-

ever, emerging work has shown that cuticle alkanes record the

deuterium isotope signatures (¶D) of meteoric water and can

serve as a accurate biomarkers for palaeoclimatic reconstruc-

tion, particularly if isotopic models take into account the

relationships between leaf wax ¶D patterns and latitude, plant

form and photosynthetic pathway (Sachse et al., 2004; Liu &

Yang, 2008). Because deuterium enrichment is also partially

controlled by transpiration and soil evaporation, n-alkanes

may also be useful in constraining estimates of transpiration or

evapotranspiration from well-preserved fossil material (Smith

& Freeman, 2006).

The evolution of stomata

The cuticle imposes a high degree of resistance on gaseous dif-

fusion, so it must be perforated in order for CO2 to enter.

These perforations take the form of stomata, the 30- to 80-lm

long epidermal pores that are formed by the aperture of two

adjacent elongated guard cells and found on surfaces of most

photosynthetic tissue (Hetherington & Woodward, 2003;

Taylor et al., 2009). It is the turgor-mediated opening and

closing of the guard cells in response to leaf hydration, light

levels, atmospheric CO2, abscissic acid and soil water availabil-

ity that controls the diffusion of CO2 and water vapour

between the leaf and the atmosphere (Davies & Zhang, 1991;

Assmann & Wang, 2001; Buckley, 2005; Messinger et al.,

2006; Ainsworth & Rogers, 2007). Stomatal conductance is

tightly coupled to hydraulic transport in land plants (Jones &

Sutherland, 1991; Brodribb & Holbrook, 2004), so stomatal

evolution may have occurred closely in time with the vascular

system (Kenrick & Crane, 1997; Sperry, 2003).

The position of stomata on the land plant phylogeny is

decided by the placement of ancestral taxa such as liverworts

and bryophytes, the standing of which has been called into

question (Raven, 2002). Some species of liverworts and

mosses exhibit what appear to be proto- or pseudo-stomata

but as the function of these pores in gas exchange is unclear

(see below; Duckett et al., 2009; Proctor, 1981), they may be

unsuitable candidates for stomatal precursors. Hence, true

stomata are currently accepted to be monophyletic, with their

origins in the hornworts (Raven, 2002).

Unfortunately, the fossil record is silent with regard to

changes in epidermal cells that may have led to the formation

of stomata, but simple perforations are apparent in the cuticles

of taxonomically mysterious taxa such as Spongiophyton

(Gensel et al., 1991) and Nematothallus from the Lower

Devonian (Edwards et al., 1998; Taylor et al., 2009). These

organisms had thick cuticles, so epidermal pores would have

significantly reduced resistance for gaseous diffusion. Similar

perforations found on the thalli of the extant liverwort,

Conocephalum conicum probably function in gas exchange

(Proctor, 1981; Raven & Edwards, 2004), but it is unlikely

that they exert dynamic control over water loss. Without

active regulation courtesy of the guard cells, these plants

passively and continually lose water in response to the atmo-

spheric vapour pressure deficit.

The oldest unequivocal stomatal specimens have been

described from unidentified cuticle and tissue fragments from

the Pridoli epoch (Upper Silurian, 416–418.7 Mya, Upper

Silurian, Edwards et al., 1996, 1998). In these samples, the

stomatal apparatus appears nearly circular, consisting of two

thick guard cells that create a lenticular pore. Perhaps due to

poor preservation and small fragment size, there is no

evidence for a substomatal chamber and in several of these

Silurian samples, epidermal cells appear to underlie the guard

cells. Small substomatal cavities are found, however, in sam-

ples of pro-tracheophytes such as Rhynia gwynne-vaughanii,

Aglaophyton major, Horneophyton lignieri and the lycophyte

Asteroxylon mackiei from the Lower Devonian (397.5–

416 Mya; Edwards et al., 1998). These taxa generally have

low stomatal frequencies, and some such as Asteroxylon have

sunken stomata –features that are consistent with the idea that

water stress was the key abiotic variable acting on early terres-

trial vegetation (Gray et al., 1985).

Based on our understanding of stomatal behaviour in

modern taxa, it has long been accepted that the evolutionary

pressures that drove water conservation strategies favoured

the coupling of the cuticle with stomata (Raven, 2002; Franks

& Brodribb, 2005). However, the underlying assumption

that the regulation of water loss was the impetus for the evolu-

tion of stomata (especially in early land plants) has recently

been challenged by Duckett et al. (2009) who argue that the

‘pseudostomata’ of the Sphagnum moss evolved to facilitate

dehydration in sporophytic capsules, rather than conserve

water. In this plant, it is dehiscence and collapse of guard cells

that opens the stomatal aperture, rather than an increase in

guard cell pressure, as typically observed in higher land plants.

Duckett et al.’s results leave the evolution of stomata open to

interpretation. One could argue that stomata initially

116 J . PITTERMANN


appeared in mosses for the purpose of dehydration and were

subsequently adapted for gas exchange in higher plants, but

alternatively stomata could have evolved to facilitate gas

exchange on land, but were co-opted by the mosses for the

purpose of dehydration. A final option is that, in Sphagnum, the

stomata fulfil both functions over time (Duckett et al., 2009).

Duckett et al.’s study changes our perspective on the evo-

lution of stomata and will undoubtedly motivate future work

on stomatal structure and function in modern taxa. Despite

the fact that stomatal design has remained fundamentally

unchanged, notable differences in size, density, shape and

ornamentation are apparent across the major plant groups

(Ziegler, 1987; Franks & Beerling, 2009a,b). The functional

significance of some of this variation apparently lies in the

tempo of stomatal movements whereby the derived, dumbell-

shaped stomata of grasses open and close much faster than

the classical kidney-shaped stomata of older lineages (Franks

& Farquhar, 2007). The authors argue that faster response

times of graminoid stomata may have increased water-use

efficiency during the global aridification that peaked 45 to

30 Mya, and thus conferred the grasses with a competitive

advantage.

Others have applied finite element analysis to explore the

relationships between stomatal structure and function to show

that sunken stomata, a cuticular lining, small substomatal

chamber size and reduced stomatal aperture, each have the

potential to greatly decrease stomatal conductance (Roth-Ne-

belsick, 2007). How variation in the size of the substomatal

chamber relates to stomatal size is empirically unknown, as are

the consequences of this allometry for photosynthesis, but

Roth-Nebelsick’s model suggests that transpiration estimates

based solely on fossil stomatal density and aperture size may

overestimate stomatal conductance.

Along these lines, the gas-exchange response of Rhyniophy-

tic plants has received considerable attention because the

morphological detail of their tissues is well preserved, and

their terrestial habit represents the beginning of the evolution

of modern plant physiology. It appears that Rhyniophytes col-

onized diverse habitats with variations in temperature

(Edwards et al., 2001) and water availability due to periodic

flooding (Rice et al., 2002; Trewin et al., 2003), thereby pro-

viding the context in which the ecophysiology of these taxa

can be examined. The cuticle of Rhyniophytic taxa is charac-

terized by low stomatal densities suggesting that in the CO2-

rich climate of the Lower Devonian, these plants would have

exhibited high water-use efficiencies (Edwards et al., 1998).

Computer simulations suggest that transpiration and photo-

synthesis were greatest in Nothia aphylla, a ‘leafy’ Rhynio-

phyte with relatively high stomatal densities (El-Saadawy &

Lacey, 1979), when compared with the more conservative

Rhynia and Aglaophyton (Konrad et al., 2000; Roth-Nebel-

sick & Konrad, 2003). The stricter water-use strategies of the

latter two taxa probably reflect the absence of a well-devel-

oped root system (Roth-Nebelsick & Konrad, 2003).

Lastly, increased water use efficiency may have been the

driving force behind selection for presumably ‘optimized’,

i.e. rapid, stomatal responses of angiosperms to short-term

changes in CO2, relative to conifers and ferns (Brodribb

et al., 2009). Considering that water use efficiency positively

affects species’ reproductive output (Farris & Lechowicz,

1990; Dudley, 1996; Ackerly et al., 2000), one could argue

that the evolution of stomata single-handedly provided

ancestral land plants with a competitive advantage, and thus

helped transform the terrestrial landscape into a three-

dimensional mosaic of vegetation, rather than a dense film of

liverworts.

The transition to plant vascular tissue

With the cuticle and the stomata in place, the evolution of

water-conducting cells completed the basic bauplan needed

for plant homoiohydry (Raven & Edwards, 2004). What are

the transitional steps from undifferentiated mesophyll cells to

a bundle of directionally aligned hollow tubes, and what are

the biotic and abiotic signals responsible for this pattern in the

first place? Understanding vascular tissue ontogeny may

inform our interpretation of fossil material and its relevance to

the evolution of xylem.

Our current outlook on the differentiation of unspecified

plant cells into elongated tracheary elements is shaped by

model systems such as Zinnia, Arabidopsis and Populus

(Tuskan et al., 2006; Turner et al., 2007). In plants with

a meristematic, procambial region, cambial cells are pro-

grammed to differentiate into either phloem or vascular tissue.

However, under appropriate culture conditions, mesophyll

cells can also be made to de-differentiate and develop into

tracheary elements (Fig. 2). Auxin, the newly discovered xylo-

gen hormone and other signalling compounds control these

changes and much progress has been made in elucidating the

genetic regulation of cell differentiation (Roberts & McCann,

2000; Chaffey, 2002). What makes this particularly relevant in

an evolutionary context is that the totipotency of plant cells

can be so easily manipulated with the appropriate signalling

compounds, most of which are produced by neighbouring

cells (Nieminen et al., 2004). In other words, the design of

even the most basic of vascular systems belonging to liver-

worts (Ligrone & Duckett, 1996) and bryophytes (Proctor &

Tuba, 2002) required little more than the appropriate signal

to upregulate a suite of genes that control cell differentiation.

One of the critical steps to the formation of vascular tissue

may have been the establishment of a polar auxin gradient

because auxin stimulates the differentiation, patterning and

elongation of tracheary elements (McCann, 1997; Sieburth &

Deyholos, 2006). An auxin gradient can also arise due to

wounding, and generate a response characterized by cell

rotation in the new tissue formed near the injury (Kramer

et al., 2008). Interestingly, evidence for this rotated cell

arrangement was recently discovered in secondary vascular tis-



sue belonging to an Archeopteris (upper Devonian, 375 Mya;

Rothwell & Lev-Yadun, 2005) as well as in wood of tree-sized

fossil equisetophytes and arborescent lycophytes belonging to

different lineages (Rothwell et al., 2008). In these plants, the

independent origins of secondary vascular tissue suggest a role

for the parallel evolution of regulation by auxin in each clade

(Rothwell et al., 2008).

There is evidence that vascular tissue differentiation is a

homologous phenomenon in land plants. Indeed, molecular

data indicate that auxin metabolic pathways are present in

basal plant lineages such as the Charales, liverworts and bryo-

phytes (Cooke et al., 2002). Of further interest are computer

models showing close correspondence between predicted

procambial vascular patterns based on theoretical input auxin

concentrations, and their close correspondence to actual

stelar patterns observed in the Early Devonian Psilophyton

and several members of the pro-gymnosperm group, the

Aneurophytales (Stein, 1993). Altogether, there is growing

evidence for the contribution of auxin to the evolution of vas-

cular plants, and the subsequent diversification of Silurian and

Devonian plant lineages.

Following cell differentiation, the process of expansion is

driven by mechanical forces that are a function of osmotic pres-

surization, which in mature plant cells can generate pressures

ranging from 0.1 to 3 MPa (Somerville et al., 2004; Taiz &

Zeiger, 2006). Brassinosteroid hormones are thought to play

a role in the complex co-ordination of the simultaneous expan-

sion of multiple cells, although several auxin-mediated

enzymes acidify and loosen the fibrous, cellulose cell wall

matrix (Wang & He, 2004; Turner et al., 2007). Importantly,

a class of proteins known as expansins regulates the loosening

of the cell wall (Cosgrove, 2005; Popper, 2008). Expansins

are not only ubiquitous across the land plant taxa, but also

thought to be integral to the development of xylem and evolu-

tion of the plant vascular system (Popper, 2008). The genetic

toolbox required for cell wall expansion is present in algae,

so the only other key innovation in the evolution of plant

vascular tissue would have been the development of the cellu-

losic secondary cell wall for increased rigidity. In plants, this sec-

ondary wall is deposited interior to the primary cell wall

following cell expansion, but excluded from the vessel endwall

and the pitted regions of the conduit that allow water to move

Fig. 2 The differentiation of mesophyll and procambial cells into vascular conduits according to Turner et al. (2007). Different combinations of hormones may act on

the procambial or mesophyll cells to initiate the formation of tracheary elements or conduits. Conduit expansion and elongation is followed by the deposition of the

secondary wall and followed by cell death and digestion of cell endwalls. Lignification completes this developmental sequence.

118 J . PITTERMANN


from one cell to another (Somerville et al., 2004; Choat et al.,

2008).

The rigidification of cell walls with polyphenols probably

occurred synchronously, or nearly so with the evolution of

early plant vascular systems (see next section). Developmen-

tally, the deposition of lignin or its precursors such as lignan

into the cellulose, pectin and protein wall matrix of the sec-

ondary wall is the final step towards increasing the stiffness

and compressive resistance of the conducting cell as well as its

impermeamility to water (Niklas, 1992; Lewis, 1999; Koehler

& Telewski, 2006). For example, Arabidopsis and Nicotiana

mutants with disrupted lignin synthesis not only exhibit mal-

formed vasculature, but also show a high frequency of

crimped or collapsed conduits (Turner & Somerville, 1997;

Piquemal et al., 1998; Jones et al., 2001).

The complex genetic and chemical components of lignifica-

tion have been identified (Boerjan et al., 2003; Peter & Neale,

2004) so a solid grasp of the functional implications of the

process on plant hydraulics is in the near future. Recently, 12

genes were targeted as the probable candidates of vascular lig-

nification in Arabidopsis (Raes et al., 2003) whereas the Black

Cottonwood genome-sequencing project uncovered a high

number of gene families responsible for ligno-cellulosic wall

formation (Tuskan et al., 2006). However, the genetics of

herbaceous model systems such as those of Arabidopsis and

Zinnia are easily manipulated (McCann, 1997; Roberts &

McCann, 2000) so, at this point in time, we know more about

tracheary element formation from these plants than from

woody taxa. Interestingly, evidence from the Zinnia model of

tracheary element formation suggests that lignification follows

cell death, which means that surrounding parenchyma tissue

may be responsible for the synthesis and release of lignin

precursors (Turner et al., 2007). Certainly, once the digestion

of cell contents begins, cell growth has ceased so lignin depo-

sition is the logical, final stage of creating a functioning, but

dead water-filled xylem conduit.

Lignins were generally thought to be absent from the algae,

but the cell walls of the Charophyte Coleochaete were found to

contain lignin-like compounds, adding further support for

their close association with land plants (Delwiche et al.,

1989). True lignin was recently discovered in the flexible,

joint-like tissue, the red algae Calliarthron, prompting the

suggestion that lignin biosynthetic pathways may have had an

early function in biomechanics, in addition to protecting cells

from microbial attack and UV damage (Martone et al., 2009;

see also Niklas, 1992). Both studies indicate that the funda-

mental pathways of lignin biosynthesis are deeply conserved,

so the lignification of xylem was essentially a function of time

and increased tissue specificity.

On a final note, the biosynthesis of lignin is dependent on

oxygen, leading some workers to speculate that elevated O2

produced by the abundant land plants during the Mid-Devo-

nian to the Early Carboniferous, may have spurred the evolu-

tion of xylem lignification (Graham et al., 1995; Berner,

2006; Berner et al., 2007). In the proposed sequence of

events, a rise in O2 levels due to photosynthesis facilitated

lignification, which in turn supported the evolution of

arborescence, further increasing photosynthetic biomass in a

feedforward cycle. Interestingly, some would argue that lignin

is not even a prerequisite for conferring stiffness to the plant

body per se and is merely a secondary requirement for arbores-

cence because non-lignified tissues can impart rigidity to the

plant body (Rowe & Speck, 2004), thereby eliminating the

necessity for the role of oxygen in the first place. However, lig-

nin is essential to preventing cell implosion at the negative

water potentials that can occur at the tops of trees as a result of

gravity (Koch et al., 2004; Domec et al., 2008), so modern

trees would probably have not evolved beyond a height of

5 m without this phenolic (Niklas, 1992). It is unlikely that

selection would favour a combination of weak xylem and non-

lignifed support tissue when the economy of modern wood

yields higher dividends courtesy of simultaneous canopy

support and water transport. Whether O2 facilitated this

streamlining of plant form and function cannot be directly

tested, but a quantitative model, although limited, may

provide us with useful insights.

The transport advantages of conducting tissue

With the basic blueprint for the vascular network in place, the

next step in the evolution of xylem was to maximize the trans-

port gains of this low resistance pathway. Indeed, the need to

reliably move water and solutes throughout the plant body

selected for the evolution of vascular tissue in even the most

primitive plants such as liverworts (Ligrone & Duckett, 1996;

VanAller Hernick et al., 2008). Furthermore, the near simul-

taneous appearance of vascular systems and stomata during

the Mid-Silurian suggests that selection not only favoured the

functional coupling between these traits (Sperry, 2003; Raven

& Edwards, 2004), but also their co-ordinated optimization

to maximize photosynthesis (Brodribb, 2009).

The frictional resistance to water transport is significantly

reduced when water travels through microscopic, hollow

channels rather than the nanoscopic pathways that character-

ize symplastic water movement. Not surprisingly, water

transport through elongated, hollow single cells is 106 times

more efficient than transport through parenchymal tissue via

plasmodesmatal channels (Raven, 1977). Similarly, ectohydric

water transport, which describes the wicking of water through

external capillary structures, is common in both terrestrial

Charophycean algae and bryophytes (Proctor & Tuba, 2002;

Franks & Brodribb, 2005) but is insufficient for taller plants.

Both pathways present such high degrees of resistance that

plants exclusively relying on non-vascular water transport

rarely exceed heights of a few centimetres and are typically

poikilohydric (Niklas, 1992; Proctor & Tuba, 2002).

The hydraulic advantage of wide, hollow conduits over nar-

row channels is evident in the context of the Hagen–Poiseuille



equation, which predicts water-transport efficiency through

smooth-walled capillaries. Here, the flow rate of water, Q,

with a viscosity g, passing through a capillary with a diameter

of D is proportional to the pressure gradient dP ⁄ dx driving

the flow such that,

Q ¼ pD4

128g

� �dP

dx: ð1Þ

Equation 1 demonstrates that hydraulic efficiency is propor-

tional to the conduit diameter raised to the fourth power, so a

modest increase in conduit diameter leads to significant gains

in transport (Tyree et al., 1994; Tyree & Zimmermann,

2002). Single conduit measurements of hydraulic efficiency

agree with the predictions of the Hagen–Poiseuille equation

(Zwieniecki et al., 2001). Functionally, this means that all else

being equal, 16 conduits with a 20-lm diameter are required

to achieve the hydraulic efficiency of one 40-lm vessel

(Fig. 3).

The higher area-specific conductivity of larger conduits

confers several advantages, an important one being that a

plant can invest less carbon in transport tissue, particularly as

the conduits are dead. Secondly, the lower transport resistance

associated with large conduits reduces the pressure drop that

is required to move water from the roots to the leaves. Very

negative water potentials can predispose the xylem to dysfunc-

tion (see Part III) and stress the leaf tissue, both of which can

lead to reductions in gas exchange. Lastly, high transport

efficiencies characteristic of xylem with large conduits can sup-

port increased transpiration rates, which in turn allow higher

rates of photosynthesis (Brodribb & Feild, 2000; Brodribb

et al., 2002; Santiago et al., 2004). Predictably, plants have

exploited the relationship between diameter and efficiency

such that there has been a 60-fold increase in conduit diame-

ter since the evolution of the early 5- to 8-lm wide Cooksonian

tracheids from the upper Silurian (416 Mya; Niklas, 1992;

Fig. 4) to the nearly 500-lm wide vessels of extant tropical

vines and lianas (Ewers, 1985).

PART II: VASCULAR PLANTS IN THE FOSSILRECORD

Conducting tissue in protracheophytes and early vascular

plants

The Late Silurian to the Early Devonian represent a period of

remarkable terrestrial plant speciation as well as exploration of

different forms of vascular structure. Geochemically based

models suggest that elevated atmospheric CO2 coupled with

2–4 �C warmer sea surface temperatures characterized the

climate during this initial phase of plant diversification (Berner,

2006; Came et al., 2007). Some of the extinct taxa from this

period may well be the ancestors of certain bryophytic lineages

or true vascular land plants, while others were early experi-

ments in the homoihydric lifestyle. The most well-studied

(though neither the oldest nor most ancestral) plants of this

period are from the Early Devonian Rhynie Chert (398–

396 Mya). This group has traditionally occupied the position

of the simplest vascular plants but this perspective has changed

as it became clear that Rhyniophytes share features of both

bryophytes and vascular plants (Kenrick & Crane, 1997;

Taylor et al., 2009). Kenrick & Crane (1997) provide the most

widely accepted cladistic treatment of the evolution of early

land plants, using among other character traits, the anatomy of

vascular tissue to decide placement on the phylogeny. What

emerges is a trend towards xylem comprised of progressively

larger tracheids with an increasingly complex and mechanically

robust wall anatomy (Kenrick & Crane, 1991, 1997; Edwards,

2003; Taylor et al., 2009) presumably reflecting selection for

efficient vascular transport and conduit support under negative

xylem pressures. The evolution of angiosperm vessels and

A B

Fig. 3 All else being equal, 16 conduits (vessels)

with a 20-lm diameter (A: Populus sp., diffuse-

porous wood) are required to achieve the hydraulic

efficiency of one 40-lm diameter conduits (B:

Quercus sp., ring-porous wood). Note that the

vessels are embedded in a matrix of fibres.

120 J . PITTERMANN


fibres for water transport and support respectively represents

the pinnacle of vascular specialization in woody plants. Infor-

mative photographs and detailed commentary of the taxa

described in this section can be found in Taylor et al. (2009),

Edwards (2003), Kenrick & Crane (1991), Stewart & Roth-

well (1993) and Kenrick & Crane (1997).

Liverworts occupy the basal position in the land plant phy-

logeny so the nearest analogue to the most primitive form of

vascular tissue probably resides in the gametophyte of the

extant, stomatalless liverwort Symphyogyna brasiliensis and its

relatives in the Pallaviciniineae. Non-lignified, elongate water-

conducting cells with thickened secondary walls are packed

together in the central regions of Symphyogyna’s thallus in

what may ostensibly be the most primitive vascular bundle.

Water travels from one cell to another via plasmodesmatal

pores and perforated, primary wall regions along the length of

the cell (Ligrone & Duckett, 1996; Ligrone et al., 2000).

Close examination of the recently discovered fossil liverwort

Metzgeriothallus sharonae (Mid-Devonian, 390 Mya) could

confirm that vascular tissue may have been present in this

extinct species, although the authors do not discuss this possi-

bility (VanAller Hernick et al., 2008; Fig. 5). It is not unlikely

however, that M. sharonae may represent one of the earliest

experiments in the design of transport tissue as both the

extant Symphyogyna and this fossil liverwort are ascribed to the

order Metzgeriales.

The oldest macrofossil vascular land plants are the cookson-

ioids of the Mid-Silurian (Wenlockian, 428–422 Mya) repre-

sented by Cooksonia (Edwards & Feehan, 1980), and similar

fossils such as Tortilicaulis (Edwards, 1996; Bateman et al.,

1998; Boyce, 2008). Early land plants from the Silurian and

Early Devonian form a diverse group with nebulous evolu-

tionary affinities, but currently accepted hypotheses suggest

that, despite its antiquity in the fossil record, Cooksonia is

derived from the protracheophytes, and belongs between the

rhyniopsids and the lycophytes (Kenrick & Crane, 1997).

Efforts to confidently place Cooksonia on the land plant phy-

logeny are also hampered by limited and partially preserved

Fig. 4 Tracheid diameters increased almost 18-fold from 8 lm during the Upper Silurian, to a maximum of 140 lm in the Upper Devonian (Niklas, 1985; courtesy of

Blackwell Publishing) representing substantial gains in hydraulic efficiency.



fossil material which may or may not actually by Cooksonia in

the strict sense (Boyce, 2008).

Cooksonioids were probably only a few centimetres tall,

with erect, bifurcating axial stems that ranged in diameter

from <100 lm to 1.5 mm (Boyce, 2008). Lacking roots

and leaves, the shoots of these simple plants terminated in

sporangia. Although Cooksonia is accepted as the first

vascular plant (Taylor et al., 2009), the actual contribution

of the limited vascular tissue to water transport is question-

able because the conduits are so narrow (<5 lm; Boyce,

2008). Close examination has shown that the conduits of

Cooksonia pertonii exhibit thick annual and spiral wall rings,

when compared with the thinner primary wall thickenings

found in modern protoxylem (Edwards, 2003). These rings

rigidify an otherwise highly perforated and what appears to

be a mechanically weak conduit wall. Despite their small

size, Cooksonia’s conduits may have adequately delivered

water to axial tissues in order to maintain the necessary

turgor pressure for an erect stature (Bateman et al.,

1998). Cooksonia possessed stomata, but they were low in

frequency and occurred primarily in the vicinity of the

sporangia, prompting the suggestion that their role was to

facilitate nutrient transport via transpiration-driven water

flow rather than increase CO2 diffusion for photosynthesis

(Edwards et al., 1996, 1998; Boyce, 2008).

The younger, more derived flora of the famous Rhynie

Chert has provided an abundance of specimens that reveal the

diversity of conducting tissues in early land plants (Taylor

et al., 2009). The cherts, which were formed in a hot-spring

environment, preserve a terrestrial and freshwater floodplain

habitat containing numerous species of early land plants (Rice

et al., 2002; Trewin et al., 2003). Of these, one the best

understood is Aglaophyton major, an 18-cm tall plant consist-

ing of bifurcating, stomatiferous, cylindrical axes that lie

loosely on the substrate surface and function as rhizomes

(Taylor et al., 2009). Aglaophyton lacked a root system and

leaves, but possessed a central bundle of vascular tissue (proto-

stele), parenchyma with intercellular air spaces, a cuticle and

stomata (Edwards et al., 1998; Taylor et al., 2009). Aglaophy-

ton’s conducting tissue is composed of smooth, thin and

thick-walled cells of variable diameter (18–50 lm; Edwards,

2003) and is understood to be much more primitive than that

of Cooksonia. Indeed, this is one of several reasons why Aglao-

phyton is placed with the protracheophytes, a group more clo-

sely allied with the mosses rather than true vascular plants

(Edwards, 1986; Kenrick & Crane, 1997; Edwards, 2003).

Similarly, Horneophyton lignieri, an abundant and well-docu-

mented 20-cm tall species from the Rhynie Chert, has also

been grouped with the protracheophytes (Kenrick & Crane,

1997; Wellman et al., 2004).

The tracheids of more advanced rhynioids such as Senni-

caulis, Stockmanesella, Huvenia, Sciadophyton and Rhynia

gwynne-vaughanii are typically <25 lm in diameter, and

reveal increasing complexity and ornamentation in the form

of helical thickenings and spongy tissue layers (Kenrick &

Crane, 1991; Edwards, 2003; Taylor et al., 2009). These

plants represent the first tracheophytes (Kenrick & Crane,

1997). The so-called G- and P-type tracheids found in eutra-

cheophytes and euphyllophytes respectively, range from 30 to

80 lm in diameter and may exhibit scalariform (ladder-like)

pitting and a decay-resistant inner layer that covers the pit

apertures (Kenrick & Crane, 1991, 1997; Edwards, 2003). At

first glance, the P-tracheid xylem of the euphyllophyte

Psilophyton resembles that of ferns, thereby representing the

most modern conduit anatomy of Lower Devonian taxa.

Taken together, the general impression from the fossil

record is that vascular tissue in the earliest land plants became

progressively more efficient due to selection for larger lumen

diameters. Simultaneously, tracheids became increasingly

more robust as evidenced by the helical and annular thicken-

ings in S- and G-type tracheids, relative to the smooth, thin-

walled conduits of Aglaophyton. Such fortified conduits would

have withstood the negative xylem pressures associated with

water stress, rather than collapsed. Implosion-resistant, and

A B

Fig. 5 Fossils of the liverwort, Metzgeriothallus

sharonae from the Middle Devonian of eastern

New York, USA (VanAller Hernick et al., 2008;

courtesy of Elsevier Press). A single vein is evident in

each lobe of the gametophyte thallus (A: 10· mag-

nification). The elongate cells present in the vein

closely resemble vascular tissue (B: 120· magnifica-

tion).

122 J . PITTERMANN


perhaps to some extent lignified vasculature may have allowed

Early Devonian taxa to exploit periodically drier environ-

ments. Concurrently, increased perforation of the conduit

walls allowed water to move from one conduit to another,

which not only increased transport efficiency, but also main-

tained transport continuity should one or several conduits

become dysfunctional.

True tracheids are a key synapomorphy of the tracheophytic

land plants, but the first appearance of lignin in the vascular

plant fossil record is equivocal (Friedman & Cook, 2000; Boy-

ce et al., 2002). However, the recent use of novel micro-ana-

lytical techniques to analyse fossil material has lead to

breakthroughs in our understanding of lignification and thus

the structure and function of tracheophytes from both fossil

and extant lineages. Specifically, soft X-ray carbon spectromi-

croscopy combined with isotope ratio mass spectrometry has

proved useful in analysing the wall chemistry of conducting

tissues belonging to fossil plant material (Boyce et al., 2002,

2003). Because lignin has a carbon isotopic signal that is 4&–

8& lighter than cellulose within the same plant, it is possible

to test for tissue lignification by comparing the 13C signal of

taphonomically equivalent fossils that possess lignified tissue

with those whose tissue chemistry is unknown. Concurrent X-

ray spectroscopy can identify the presence of aromatic carbon

compounds that are characteristic of lignin.

These microanalytical techniques allowed Boyce et al.

(2003) to demonstrate that the vascular strand of Aglaophyton

major lacked lignified water-conducting cells unlike the ring

of lignified fibrous cells just beneath the epidermis. These

fibres probably served a mechanical function that rigidified

Aglaophyton’s narrow, cortex-filled stems (Bateman et al.,

1998; Boyce et al., 2003). By contrast, more derived taxa such

as Rhynia gwynne-vaughnii exhibited lightly lignified xylem,

with significant lignification evident in the tracheids of Aste-

roxylon mackiei (see also Kenrick & Crane, 1991). The greater

xylem fraction, and perhaps to some degree, the greater stiff-

ness afforded to taxa with progressively more lignified tissue is

consistent with Rhynia’s and Asteroxylon’s tendency towards

increased height and horizontal branching (Boyce et al.,

2003). New methods, such as X-ray carbon spectromicrosco-

py, are exciting and may offer greater insight into the hydraulic

and mechanical properties of the first vascular plants, particu-

larly those with equivocal phylogenetic affinities.

Vascular tissue in advanced land plants

For nearly 300 million years, tracheid-based xylem remained

the most prevalent transport tissue in the plant kingdom.

The ornate tracheids of the Early Devonian tracheophytes

eventually led to the hydraulically efficient, large-diameter,

scalariform tracheids of more derived groups such as the

horsetails and ferns, and eventually to the thick-walled

tracheids that provided a biomechanical and a transport func-

tion in progymnosperms, gingkos and conifers (Tyree &

Zimmermann, 2002; Pittermann et al., 2006a). According

to the classic scenario of Bailey & Tupper (1918) the subse-

quent evolution of tracheids into the fibres and vessel

elements of angiosperm xylem capitalized on the functional

specialization of each cell type. Fibres are short and thick-

walled, providing the necessary strength for increasingly

lateral canopies while the large vessel elements reduce

resistance to water transport.

Not surprisingly, competition for light and the need for

effective spore dispersal prompted the evolution of arbores-

cence (Niklas, 1985, 1992; Mosbrugger, 1990), well before

the appearance of secondary xylem, that is wood. In fact, the

evolution of secondary xylem is not even a requirement for

elevated canopies assuming that other tissues such as a

hypodermal sterome, thick external periderm or cortex tissue

provide structural support (Niklas, 1997; Rowe & Speck,

2004, 2005). This means that during the Devonian, the tree

habit was free to arise independently and more or less

simultaneously in ferns, lepidophytes, horsetails and progym-

nosperms (Mosbrugger, 1990).

Modern and extinct tree ferns lack secondary xylem so

structural support comes from abundant schlerenchymatous

(i.e. lignifed) fibres, support roots, numerous thin axes and

other plant matter that makes up the false trunk (Mosbrugger,

1990; Taylor et al., 2009). The Mid-Cretaceous fern, Temp-

skya is a good example of this strategy as its main axis looks dis-

proportionately large relative to the size of the emerging

fronds due to the accumulation of roots, fibres and plant deb-

ris (Taylor et al., 2009). Modern ferns, such as those in the

genera Dicksonia and Cyathea rarely exceed heights of 10 m,

but the reasons for this height limit have not been examined.

Buckling is unlikely to occur given the compressive strength

of fern trunks (Mosbrugger, 1990; Niklas, 1997). Although

hydraulic limits in ferns may arise from limited conduit size

and conduit density (Gibson et al., 1984) it may be that there

is little to gain by growing tall in a light-limited environment

already dominated by tall angiosperms with expansive cano-

pies.

The evolution of the vascular cambium (the axial meristem

that gives rise to secondary xylem) relaxed the restrictions on

axial height and girth imposed by primary xylem. Fossil

evidence suggests that the vascular cambium originated early

in the evolution of land plants and may have evolved in

response to strong selective pressure for increased plant size.

The primitive progymnosperm group Aneurophytales were

the first to show limited secondary xylem by the Mid-Devo-

nian, but by the Late Devonian, secondary xylem was present

in representatives of nearly all major plant groups including

Lepidodendrales, horsetails, seed ferns, progymnosperms and

perhaps even in the Cladoxylopsids (fern affiliates), although

the vascular patterns of these fern allies are highly variable even

within a specimen (Niklas, 1997; Taylor et al., 2009).

Of the three types of vascular cambia that evolved in plants,

the bifacial and monocot ‘etagen’ cambia are present in mod-



ern taxa, while the unifacial vascular cambium disappeared

after the Late Devonian (Esau, 1943; Niklas, 1997). The

unifacial vascular cambium was present in Sphenophyllum

(relatives of horsetails), and Sigillaria and Lepidodendron,

both representatives of Carboniferous arborescent Lycopods

(Thomas & Watson, 1976; Taylor et al., 2009). Lepidodon-

dron was a dominant element of the swampy Carboniferous

landscape, whereas Sigillaria may have occupied drier sites

(Phillips & DiMichele, 1992; Stewart & Rothwell, 1993;

Taylor et al., 2009). Because the unifacial cambium was

capable of only producing new xylem cells to the interior of

the main axis (periclinal division), and unable to divide radially

to generate new cambial initial cells (anticlinal division), the

girth of these plants was thought to be determinate and

established early in their development by a large number of

fusiform initials (Eggert, 1961; Cichan, 1985a,b). Conse-

quently, lateral expansion of the secondary xylem as well as the

stem axis in Sphenophyllum and the tree like Lycopods could

only be achieved by the progressive, centrifugal enlargement

of the cambial initials rather than an increase in the number

cambial cells (Cichan, 1985a,b).

Despite the limited cambial expansion of these early trees,

they often reached heights of 40 m and their basal diameters

exceeded 2 m. Reconstructions suggest that Sigillaria and

Lepidodendron were characterized by vertical, monopodial

axes that lacked side-branches but terminated in a large, pro-

fusely branched crown of leafy twigs (Phillips & DiMichele,

1992; Taylor et al., 2009). With greater height, girth and leaf

area presumably became a higher demand for water. The

increased length and diameter of the lateral tracheids (25 mm

long, 120 lm diameter) relative to the central tracheids

(14 mm long, 85 lm diameter) of the main axis (Cichan,

1985a,b) may have adequately supplied this demand, albeit in

a limited capacity. Simple predictions using the Hagen–

Poiseuille equation suggest that hydraulic transport in the

periphery of a Lepidodendron stem may have been over four-

fold higher (on a cell basis) than in the interior of the axis.

However, the trunks of arborescent lycopods contained a

high fraction of cortex tissue (Mosbrugger, 1990), so a deter-

minate meristem producing only a limited amount of xylem

may have been the transport bottleneck failing to deliver

enough water to both maintain turgor and support gas

exchange during drought. This is consistent with the idea

that arborescent lycopods disappeared in response to a drier

climate in the Late Carboniferous (Phillips & DiMichele,

1992). Concurrently, the apparent developmental inflexibility

of the unifacial cambium may have limited structural diversity

and branching, which also would have placed these early trees

at a competitive disadvantage for resources and perhaps con-

tributed to their demise (Rowe & Speck, 2005). Alas, this

‘nascent innovation’ was successful for only 50 million years.

By constrast, the bifacial vascular cambium first character-

ized by the Mid-Devonian progymnosperms Archaeopteris

(Beck, 1960, 1970; Meyer-Berthaud et al., 1999) is found in

all modern trees. As its name implies, the bifacial cambium

produces secondary xylem towards the inside of the stem axis,

and phloem to the outside. Its primary advantage lies in the

capacity of the cambial initials to divide both anticlinally and

periclinally such that each year, new cells are added in a man-

ner that allows for the indeterminate expansion of the stem

axis as well as the addition of branches. Of the two cambial

strategies discussed thus far, selection has favoured the bifacial

vascular cambium because it successfully combines water

transport with mechanical support of the canopy (Rowe &

Speck, 2004, 2005).

The monocot ‘etagen’ cambium is recently derived in the

angiosperm monocots (Lower Cretaceous; Herendeen &

Crane, 1995) and produces lignified fibres in well-ordered,

discrete vascular bundles containing both xylem and phloem

(Esau, 1943; Rudall, 1995; Tomlinson, 2006). There is no

secondary xylem in monocots. Arborescent monocots such as

palms are unusual in the plant world because most of the stem

volume is occupied by primary vascular tissue that is thought

to remain fully functional over the life of the plant (Tyree &

Zimmermann, 2002; Tomlinson, 2006).

On the whole, the evolutionary trajectory of plant vascular

systems has selected for some combination of increasing

hydraulic efficiency and biomechanical support. The evalua-

tion of Murray’s law in the context of different wood types

now serves as a classic example of the support structure versus

hydraulic efficiency trade-off. Murray’s law was originally

developed for cardiovascular design, and states that for a fixed

investment in blood and vessel volume, the optimal transport

arrangement equalizes the sum of conduit radii cubed along

the flow path. Interestingly, this axiom also applies to vascular

plants but only if the support requirement is entirely separate

from the transport function, as in the xylem of leaves, ring-

porous species (see Fig. 3), vines and some ferns (McCulloh

et al., 2003; McCulloh & Sperry, 2005). Significant deviation

from Murray’s law occurs when vascular tissue increasingly

performs a structural function, as in the case of conifers

(McCulloh et al., 2004). Murray’s law has not been evaluated

in extinct taxa such as the Sphenophyllum or Lepidodenron, but

it is unlikely that the unifacial cambium compromised the

water transport in tissue that was largely decoupled from can-

opy support. In other words, transport efficiency is likely to

have selected for adherence to Murray’s law even in morpho-

logically complex, although extinct taxa, but this hypothesis

remains to be tested.

Because the structural and hydraulic performance of fossil

taxa is impossible to examine directly, the remaining option is

to uncover the biophysical priciples that underlie modern

plant structure and function, and to apply them to ancient

taxa, or towards models of new morphologies. Indeed, the

theoretical approach has yielded a vast array of spectacularly

diverse, yet realistic plant ‘morphospace’ that accurately

describes both modern and fossil flora (Niklas, 1997, 2004).

Because the plant vascular system reflects the overall construc-

124 J . PITTERMANN


tion of a broad array of plant life forms including herbaceous,

woody, lianecescent and mococot plants, we can utilize the

structure–function relationships of modern xylem to generate

reasonable hypotheses about the physiology and habitat of

ancient plants.

PART III: THE FUNCTIONAL ANATOMY OFMODERN PLANT VASCULAR TISSUE – THEHYDRAULIC TRADE-OFFS OF RESISTANCE TOABIOTIC STRESS

A good grasp of anatomy is useful to correctly interpret the

relationships between xylem structure and function. This is

because the transport characteristics of the xylem impose

physical constraints on the rate at which water can be supplied

to the leaves, and on the maximum transpiration rates allowed

by stomata (Tyree & Sperry, 1989; Sperry et al., 1998;

Brodribb, 2009). The following discussion will focus on

conifers and woody angiosperms simply because most vascular

structure function work has been performed on woody

species. However, there is no apparent reason why the

principles derived from this research should not apply equally

well to non-woody plants.

The structure of conifer and angiosperm xylem

Conifer wood is homogenous, meaning that it is composed

entirely of overlapping individual cells referred to as tracheids.

Modern conifer tracheids range in size from about 0.5 to

nearly 4 mm in length and 8–80 lm in diameter (Panshin &

de Zeeuw, 1980; Pittermann & Sperry, 2003; Pittermann

et al., 2006a) and their size depends on, among other factors,

their location within the tree and whether or not they function

in supporting the canopy. In general, root tracheids tend to

be longer and wider with thinner conduit walls rather than

stem tracheids (Pittermann et al., 2006b). This is because

root tissue is optimized for the delivery of water to the canopy

whereas stem tissue must compromise between water trans-

port and structural support (Sperry et al., 2006; Pittermann

et al., 2006a,b).

Angiosperm xylem is slightly more complex than that of

conifers and exhibits a greater capacity for variation. Water

transport occurs through multicellular conduits known as ves-

sels. Vessels are composed of two or more single-celled vessel

elements that are stacked one on top of the other to form an

elongated tube that may be up to 0.5 mm in diameter and

2 m in length. Unlike in the tracheids, the distal regions of the

individual vessel elements (known as endwalls) are digested

away either entirely, or in a scalariform pattern with the

end-result resembling an elongated pipe that offers a low resis-

tance pathway for water movement from the root to the can-

opy (Ewers, 1985; Schulte & Castle, 1993; Ellerby & Ennos,

1998). The vessels are embedded in a matrix of short, thick-

walled fibres (Fig. 3). Consequently, angiosperm xylem is

thought to reflect a more optimized vascular strategy that cap-

italizes on the division of labour to efficiently deliver water

through the vessels and while relegating storage and mechani-

cal requirements to the fibres.

In conifers as well as in angiosperms, water moves from one

conduit to another via perforations in the cell wall, also known

as bordered pits (Fig. 6; Choat et al., 2008). These pits are

composed of a water-permeable, cellulosic pit membrane that

is surrounded on either side by the overarching secondary cell

wall, the pit border. The perforated cell wall provides an aper-

ture through which water can pass from one conduit to

another through the pit membrane.

Significant progress has been made in our understanding of

the evolution of xylem structure and function, and the trade-

offs that accompany pit structure, conduit length and conduit

diameter in conifers and angiosperms (Tyree et al., 1994;

Comstock & Sperry, 2000; Sperry et al., 2006). Specifically,

we now have a quantitative grasp of the functional conse-

quences of wood bearing either vessels or tracheids, and are in

better position to consider the vascular performance of fossil

plants.

It is intuitive that conduit lumen diameter and conduit

length highly impact conducting efficiency and that as such,

selection has favoured the evolution of larger conduits

(Fig. 4). Indeed, this perspective has been emphasized in the

literature effectively arguing that angiosperm wood, with its

longer and wider vessels, is inherently more hydraulically

efficient that the homoxylous, tracheid-based wood of coni-

fers (Tyree & Zimmermann, 2002). Broadly speaking, the

hydraulic properties of both plant forms are thought to have

significant implications for their resource competition, fitness

and ecological dominance (Bond, 1989). However, this

perspective has recently shifted because on closer inspection,

the transport issue is not just one of conduit volume but of

overall conduit structure: aside from length, the biggest differ-

ence between vessels and tracheids resides in the endwalls of

these conduits where the interconduit pits are located.

Conifers possess intertracheid pits that are composed of

two distinct regions: the outer margo region is rather porous

with micron-scale openings and thus offers minimal resistance

to water flow whereas the inner, thickened torus region is

impermeable. By contrast, the intervessel pit membranes of

angiosperms are composed of a homogenous, tightly woven

mesh of microfibrils. Pore sizes range from 5 to 420 nm and

the membrane lacks a central thickening (Choat et al., 2008).

The functional ramifications of this differing pit structure on

water transport are profound because the pit resistance to

water flow is almost 60· lower in conifer pit membranes

(6 ± 1 MPa s m)1) than in those of angiosperms

(336 ±81 MPa s m)1). This reduction in pit resistance is

equivalent to an almost eightfold increase in tracheid length,

and thus hydraulically compensates for the otherwise

increased resistance that would be associated with greater

number of endwall crossings due to short tracheids. By



contrast, the greater length of the angiosperm vessel counter-

acts the increased hydraulic resistance offered by the homoge-

nous pit membrane. The end result is that conifer and

angiosperm xylem exhibits similar hydraulic efficiency at an

equivalent conduit diameter (Pittermann et al., 2005; Sperry

et al., 2006).

The ancestral condition of tracheid-based wood relies on

homogenous pit membranes rather than the more efficient

torus–margo arrangement for the interconduit flow of water.

This suggests that these ancestral tracheids may have been up

to 38-fold less efficient than tracheids of equivalent size that

bear the more derived torus–margo membrane arrangement

(Sperry et al., 2006). Furthermore, it has been shown that

across a broad sampling of conifers and angiosperms, the end-

wall contribution to conduit resistance is 64 ± 4% and

56 ± 2% in conifer tracheids and eudicot vessels respectively

(Wheeler et al., 2005; Pittermann et al., 2006b). Since

cycads, ferns and the basal vesselless angiosperms possess

tracheids without a torus–margo pit membrane, how does pit

resistance scale with lumen resistance in these more primitive

taxa? So far, this has only been examined across a broad

sampling of vesselless angiosperms. Interestingly, vesselless

angiosperms exhibit pit resistances that are much closer to

those of conifers than angiosperms (16 ± 2 MPa s m)1)

despite having a homogenous membrane (Hacke et al.,

2007). On average, pit resistance accounted for 77 ± 2% of

the total vessel hydraulic resistance. Scanning electron

microscopy revealed that these pit membranes are much more

porous than those of eudicots, thereby presenting a relatively

lower resistance to water transport. Consequently, the

vesselless angiosperms show hydraulic conductivities that are

much higher than predicted, and surprisingly similar to the

xylem conductivities of conifers (Hacke et al., 2007).

How can we employ these structure–function relationships

in the context of plant palaeo-physiology? Xylem is preserved

extremely well in the fossil record so it is possible with good

sampling and sectioning techniques to re-construct hydraulic

transport in extinct taxa, or at the very least get a sense of it.

However, it is important to acknowledge the implicit assump-

tion here, which is that xylem structure–function relationships

have been conserved over time. This is doubtful because many

extinct taxa combine suites of vascular traits that have no living

analogue. Archaeopteris, the Devonian pro-gymnosperm is

just such a case because its xylem employs tracheids with an

unusual clustered arrangement of intertracheid pits that lack a

torus–margo membrane (Beck, 1970; Meyer-Berthaud et al.,

1999; Taylor et al., 2009). Although it is currently impossible

to directly measure hydraulic efficiency in fossil xylem, it is

possible to construct and test a scaled, physical model much in

the way that Lancashire & Ennos (2002) used glycerine and

toy truck tyres to evaluate the resistance of interconduit pits.

Perhaps the most simplistic way of retrodicting water-

transport efficiency is to use the Hagen–Poiseuille relation-

ship to calculate the lumen conductivity (ks) on a xylem area

(A) basis,

ks ¼pP

D4

A128g; ð2Þ

where A is the area represented by the sum of D4 or the total

xylem area. It is important to keep in mind that scalariform

Fig. 6 A diagram describing the events of freeze–

thaw and air-seeding cavitation in conifers (see

text). Shaded tracheids are water filled and under

negative xylem pressure (PX), clear tracheids are air

filled and at atmospheric pressure (P0). In functional

tracheids, the thickened torus region of the pit

membrane is centrally located within the pit cham-

ber. When a neighbouring tracheid is air-filled, the

pit membrane deflects against the pit border and

the torus is pressed against the pit aperture, thereby

isolating the dysfunctional tracheid. Air seeding

occurs when the torus becomes dislodged from its

sealing position. In angiosperms, air seeding occurs

through the largest pore in the deflected, homoge-

nous pit membrane (see Choat et al., 2008 for

details).

126 J . PITTERMANN


perforation plates in vessels, as well as interconduit pits reduce

the actual conductivity by well over 40% of what the Hagen–

Poiseuille relationship predicts (see discussion above, Sperry

et al., 2006; Schulte & Castle, 1993). Similarly, the helical

and annular thickenings found in the protoxylem of extant

plants, and the tracheids of Cooksonia and Sennicaulis can

reduce the effective diameter of the conduits, and thus reduce

ks (Jeje & Zimmermann, 1979).

The Hagen–Poiseuille method may be most readily applica-

ble in a broadly comparative study of well-classified plants that

fall clearly into either the woody coniferous or woody angio-

sperm category, where we can assume that pit membranes are

likely to constitute a fraction of hydraulic resistance that is

comparable to their living relatives. However, equation 2 can

also be applied to taxa where pit membranes have not broadly

been examined such as in the cycads and ferns, but with much

less certainty. For example, Calkin et al. (1986) determined

that digestion of pit membranes with cellulase increased

hydraulic conductivity by 66% in Pteris vittata. Modelling fern

xylem may be challenging particularly in taxa that harbour ves-

sels in addition to the more commonly observed tracheids

(Carlquist & Schneider, 2007).

However, another interesting exercise is to model fluid flow

in extinct taxa with xylem anatomy that is no longer found in

modern plants. Cichan (1986) was one of the first to recog-

nize the value of this approach, and thus applied the Hagen–

Poiseuille relationship to tracheids in the secondary xylem of

Carboniferous pteridophytic and progymnosperm taxa in

order to estimate the comparative hydraulic performance of

these extinct plants. In his survey, tracheid diameters ranged

from a minimum of 36 lm in the Sphenophyte (i.e. horsetail)

Arthropitys communis to a maximum of 174 lm in the seed

fern Medullosa noei. The tracheid size of Medullosan wood is

exceptional by today’s standards because the largest conduits

in extant plants are found in the lianas, with vessels (not trac-

heids) commonly in excess of 150 lm (Ewers, 1985; Cichan,

1986). As expected, predicted ks was the highest in plants

with the largest tracheid diameters such as M. noei and Spheno-

phyllum plurifoliatum, reaching or exceeding the ks of extant

vines and lianas (Cichan, 1986). Clearly, these extinct plants

exploited a morphospace that was quite different from today’s

forms.

However, given what we know about pit resistance in

woody plants, we can build upon Cichan’s method to achieve

a higher level of precision in our transport models. In practice,

hydraulic conductance (K) quantifies the efficiency of water

transport through xylem, and is measured as the flow rate of

liquid water (Q) for a given pressure gradient (DW) driving

flow:

K ¼ Q

DW: ð3Þ

However, K cannot be measured directly on fossil material,

so an alternative approach for estimating K is to separate

hydraulic resistance (RCA, that is 1 ⁄ K) into its constituent

resistances, those being the conduit lumen resistance (RL)

and the endwall resistance (RW) not unlike resistors in series

(eqn 4; for model details, see Lancashire & Ennos, 2002;

Sperry & Hacke, 2004; Pittermann et al., 2006a,b; Wilson

et al., 2008).

RW ¼ RL þRW ; ð4ÞThe endwall resistance can be broken down into

RW ¼2Rpit

Npits; ð5Þ

where Npits represents the number of pits per endwall. Pit

resistance, Rpit is described by the addition of the pit aperture

and pit membrane resistances, both of which can be separated

into their constituents including the porosity of the pit

membrane, the size of the pit membrane pores, the thickness

of the membrane microfibrils, and the depth and width of the

pit aperture (described in detail in Sperry & Hacke, 2004;

Wilson et al., 2008).

Lastly, the inverse of the Hagen–Poiseuille equation esti-

mates flow across half of a tube’s length,

RL ¼64gL

pD4; ð6Þ

where L represents tracheid length.

Wilson et al. (2008) applied a more complete version of

the above model to investigate the hydraulic resistance of

Medullosa, Cordaites and a Pinus species. This choice of living

and extinct plants allowed the authors to compare the struc-

ture and function of tracheids with and without torus–margo

pitting. Interestingly, Medullosa and Cordaites have tracheids

that have angiosperm-like, homogenous pit membranes,

presumably a suboptimal combination of conduit size and

pitting that is expected to produce a substantial drop in

hydraulic efficiency (on a sapwood area basis) relative to

equivalently sized tracheids with torus–margo pitting

(Pittermann et al., 2005). However, the conduit-specific

conductivity of Medullosa was calculated to be over two

magnitudes higher than that of extant conifers, a reflection of

the unusually wide and long tracheids that constitute this

species’ xylem (Wilson et al., 2008). The authors suggest

that the vasculature and hydraulic performance of these fossil

plants is similar to angiosperm lianas, whose climbing habit

eliminates any need for the xylem to perform canopy support

function. This is consistent with the climbing hooks and

tendrils present on the branches and pinnae of Medullosa

specimens (Krings et al., 2003).

Because water transport sets the upper limit for transpira-

tion, the relatively high hydraulic efficiency of Medullosan

xylem is in line with its large leaves and high leaf stomatal den-

sities (Wilson et al., 2008). This plant probably required large

amounts of water to support photosynthesis, which may have

narrowed its habitat to humid tropical floodplains (DiMichele

et al., 2006).



The approach to partitioning resistances in fossil angio-

sperm xylem can be accomplished in manner similar to that of

Wilson et al. (2008) but it could prove to be more challenging

because vessels may be longer than tracheids by several orders

of magnitude and the vessel network may be tortuous, with

complex pitting patterns (Tyree & Zimmermann, 2002; Kitin

et al., 2004). We can however turn to the structure–function

relationships derived from extant angiosperms to find simple,

anatomically based correlates of hydraulic efficiency. Again,

vessel diameter may prove to be the most useful metric

because it scales well with length and the pitted endwall area,

and thus correlates strongly with xylem transport capacity

(Wheeler et al., 2005; Sperry et al., 2006).

Lastly, the RCA of conifer tracheids with torus–margo

pitting can be inferred from the following summary equation,

RCA ¼128gpD2

þ rPD

FPL2

� �1þ Cð Þ2; ð7Þ

where RL and RW are described by [128g ⁄ (pD2)] and

[rPD ⁄ (FPL2)] respectively (Pittermann et al., 2006b). Essen-

tially, mean RCA is determined by conduit lumen diameter

(D), the length of the conduit (L), intertracheid pit resistance

on an area basis (rP), the fraction of the tracheid wall area

occupied by pits (FP) and the ratio of the tracheid double wall

thickness to lumen diameter (C) (Pittermann et al., 2006b;

Sperry et al., 2006). The average intertracheid rP is

6 ± 1 MPa s m)1 but some degree of variation exists in the

north temperate conifer clades that may reflect tracheid loca-

tion within the plant as well as species’ drought resistance and

habitat preference (Mayr et al., 2002; Burgess et al., 2006;

Pittermann et al., 2006b; Domec et al., 2008).

Relationships between xylem anatomy, and drought and

freeze–thaw induced cavitation

Not only can wood anatomy be used to describe species’

hydraulic efficiency but it can also help identify species’ limits

to water transport. This is because the vascular features that

allow us to model conductivity can also capture the suscepti-

bility of xylem to hydraulic failure caused by drought or

freeze–thaw induced cavitation. In this way, wood anatomy

can potentially serve as an additional indicator of a fossil

plant’s access to water and ⁄ or palaeoclimate.

Water transport by the C–T mechanism works remarkably

well but it is not without its limitations. When leaf transpira-

tion exceeds the transport capacity of the xylem such as during

a drought, the pressure in the water column can become nega-

tive enough to allow the phase change of water from a liquid

to a vapour state. This transition is known as cavitation or

air-seeding, and it is caused by the aspiration of air into the

transpiration stream through the pit membrane. If the pull of

the transpiration stream creates sufficient tension in the sap,

the air bubble expands and the conduit is simultaneously

drained of water. Ultimately, a mixture of air and vapour fills

the entire conduit to create an embolism, which blocks water

transport in that cell (Fig. 6). Because embolism reduces the

number of functional conduits, water must subsequently

travel from the root to the leaf at progressively more negative

xylem pressures through a higher resistance pathway. Under

increasing drought conditions, these higher tensions predis-

pose the xylem to further cavitation events that can potentially

lead to the dangerous situation of runaway embolism and

plant death (Tyree & Sperry, 1988; Hacke & Sperry, 2001;

Tyree & Zimmermann, 2002; Choat et al., 2008).

Species’ cavitation resistance is typically presented in the

form of vulnerability curves, which express the per cent loss of

hydraulic conductivity due to embolism in response to more

negative water potentials (Fig. 7; Hacke & Sperry, 2001).

Vulnerability curves can be generated from almost any portion

of a plant such as roots, stems and leaves. The xylem pressure

at which a segment exhibits 50% loss of hydraulic conductivity

is referred to as the P50, or the ‘cavitation pressure’, and can

be used to compare cavitation resistance between plant organs

or species.

Despite the looming threat of a hydraulic catastrophe,

plants are only as cavitation resistant as their environment

requires them to be. This is because there are several costs

associated with embolism resistance, namely carbon invest-

ment and reduced transport efficiency. Progressively negative

water potentials can threaten xylem conduits with implosion

because the tension exerted on the water column exposes the

conduit walls to a combination of hoop and bending stresses

(Hacke et al., 2001). Hence, xylem conduits must be suffi-

ciently, but not overly reinforced to withstand wall collapse.

Anatomically, this translates to a tight, positive correlation

between the ratio of conduit double wall thickness (t) to

hydraulic mean lumen diameter, that is the empty space con-

fined by the inner walls of the conduit [b; (t ⁄ b)h2] and species’

cavitation pressure (Fig. 8; Kolb & Sperry, 1999; Hacke

Fig. 7 The per cent loss of conductivity (PLC) and cavitation pressure (P50) in

response to either a combination of freeze–thaw and water stress, or water

stress alone at the range of xylem pressures in stems of Utah Juniper (from

Pittermann & Sperry, 2005).

128 J . PITTERMANN


et al., 2001; Hacke & Sperry, 2001; Pittermann et al., 2006a;

Jacobsen et al., 2007). Consequently, the greater investment

in xylem wall content is reflected in higher wood densities as

well as improved mechanical strength, and thus represents the

carbon cost of cavitation resistance (Hacke et al., 2001; Baas

et al., 2004; Jacobsen et al., 2007; Pratt et al., 2007).

However, the thickness:span relationship is driven by variation

in conduit diameter rather than wall thickness, which explains

the general trend for reduced hydraulic conductivity with

increased resistance to cavitation (Hacke et al., 2006; Pitter-

mann et al., 2006a; Sperry et al., 2006). Although conduit

diameter is also coupled with species’ cavitation pressure,

there is much variation in this relationship and a broad

sampling is required to achieve adequate resolution (Fig. 9).

Hence, conduit (t ⁄ b)h2 serves a much better structural

correlate with cavitation pressure than conduit diameter.

Unlike angiosperm xylem which features fibres for struc-

tural support and vessels for water transport, conifer xylem is

homoxylous, meaning that its tracheids serve both a transport

as well as a support role. This means that on average, stem

tracheids exhibit greater (t ⁄ b)h2 than vessels at a given cavita-

tion pressure (Fig. 8; Sperry et al., 2006; Hacke et al., 2001).

To achieve improved cavitation resistance, an increase in

tracheid (t ⁄ b)h2 is accomplished by shrinking the tracheid

lumen diameter rather than increasing tracheid wall thickness.

Consequently, conifer xylem exhibits a 46-fold range in hydrau-

lic conductivity – a significant strength versus transport efficiency

trade-off (Pittermann et al., 2006a; Sperry et al., 2006).

A similar cavitation safety versus efficiency trade-off is

evident in angiosperms but unlike in the conifers, it may be

more tightly coupled to the process of air seeding at the pit

level. The recently proposed ‘pit area hypothesis’ suggests that

vulnerability to air seeding is related to the intervessel pit area

on a conduit. As air seeding in angiosperms is thought to

occur through the largest pore on a pit membrane, the likeli-

hood of an air seeding event rises with increasing pit area

(Wheeler et al., 2005; Sperry et al., 2006; Choat et al., 2008).

Intervessel pit area scales with vessel diameter, thus providing

an explanation for the weak, although consistent relationship

between vessel diameter and vulnerability to cavitation.

In conifers, air seeding is also thought to occur through the

pit membrane but no relationship exists between the pit area

and cavitation pressure (Pittermann et al., 2006b). In water

filled tracheids, water moves from one tracheid to another

through the porous margo region of the pit membrane.

Should one tracheid become embolized, surface tension in

the margo deflects the pit membrane such that the torus is

appressed against the pit aperture and air is prevented from

entering the neighbouring, functional tracheid. Air seeding

occurs when the xylem pressure is negative enough to dis-

lodge the torus from its sealing position (Fig. 6). Conse-

quently, it is the ratio of the torus diameter to the pit aperture

diameter that is thought to control air seeding in conifers,

while the progressive reductions in pit aperture with P50 may

represent the pit-level hydraulic cost of cavitation resistance

(Burgess et al., 2006; Domec et al., 2008; Hacke & Jansen,

2009).

Lastly, hydraulic dysfunction may also be caused by freeze–

thaw cavitation, a common phenomenon in plants inhabiting

seasonal or high elevation climates. When xylem sap freezes,

dissolved gases are coalesced into bubbles during ice forma-

tion. This is because air is insoluble in ice. The bubbles remain

inert while the sap is frozen, but may expand as it melts and

tensions are re-introduced into the xylem. If xylem pressures

are sufficiently negative during thawing, the largest of these

bubbles may expand and nucleate an embolism (Figs 6 and 7).

The likelihood of freeze–thaw induced embolism is described

by a modification of the capillary equation, such that the xylem

pressure (PX) must be more negative than the critical freeze–

thaw cavitation pressure (PXF) in order for the bubble to

expand and embolism to occur,

Fig. 8 Species’ cavitation pressure as a function of the wall thickness to lumen

diameter ratio (data from Hacke et al., 2001; Pittermann et al., 2006a,b; Pratt

et al., 2007, J. Pittermann, unpublished data). Archaeopteris cavitation

pressure was estimated from fossil root and stem material (courtesy of D.

Erwin).

Fig. 9 The relationship between species cavitation pressure and conduit diam-

eter (data from Sperry et al., 2006; J. Pittermann, unpublished data).



PXF ¼�2T

rb; ð8Þ

where T is the surface tension of water and rb is the radius of

the largest bubble. Hence, the larger the bubble, the less

negative the PXF required for bubble expansion (Davis et al.,

1999; Pittermann & Sperry, 2003).

In contrast to drought-induced embolism, conduit diame-

ter is explicitly linked to species’ vulnerability to freezing

induced embolism. This is because rb is a function of radial air

content, which is directly proportional to conduit diameter

(Pittermann & Sperry, 2003, 2005). Practically, this means

that species with an average conduit diameter >30 lm are

highly susceptible to freeze–thaw cavitation, even under

mildly negative xylem tensions. By contrast, plants with

narrow conduit diameters are largely protected from this form

of stress because the smaller bubbles that are produced upon

freezing have a greater tendency to shrink and redissolve

during the thaw (Pittermann & Sperry, 2005). Hence, taxa

with narrow conduits such as alpine conifers are freeze–thaw

cavitation resistant but at the cost of reduced hydraulic

conductance, while the opposite is true of plants with

large-diameter conduits.

The interdependence between xylem pressure and conduit

diameter was examined in conifer wood, and used to roughly

approximate the freeze–thaw vulnerability curves from a

tracheid diameter distribution according to,

PXF ¼ 120D�1:53f ; ð9Þ

where Df is the critical cavitation diameter, which is the

conduit diameter that will embolize due to freezing and thaw-

ing at a pressure PXF (Pittermann & Sperry, 2005). Whether

equation 9 applies to angiosperm wood is currently unknown.

Although the relationships between wood anatomy

and resistance to drought and freeze-thaw stress are well

developed and could potentially be applied to fossil material,

it is important that any inferences about species’ habitat or

microhabitat based on xylem anatomy be approached with

caution. If possible, conclusions should consider the evolu-

tionary lineage, the overall plant form, and leaf venation

patterns, which are coupled to species hydraulic and photo-

synthetic capacity (Sack & Frole, 2006; Brodribb et al., 2007;

Brodribb, 2009), and may also be related to habitat (Royer

et al., 2008). For example, inferences about species’ resistance

to drought-induced cavitation based on diameter alone may

be erroneous, because conduit diameter can be weakly linked

to cavitation in angiosperms (Wheeler et al., 2005; Sperry

et al., 2006) but may be influenced by evolutionary lineage in

both angiosperms and conifers (Cavender-Bares & Holbrook,

2001; Pittermann et al., 2006a,b). We know that despite

having xylem with large diameter conduits, ring-porous spe-

cies such as oaks, are actually resistant to drought-induced

cavitation and exhibit low water potentials in situ (Cochard

et al., 1992) and may thus deviate from the seemingly simple

relationship that should exist between conduit diameter and

P50. However, the relationship between vessel diameter and

vulnerability to freezing-induced cavitation in oaks is generally

consistent with theory (Sperry & Sullivan, 1992; Davis et al.,

1999; Cavender-Bares et al., 2005).

The relationship between stomatal conductance and the

structure and function of xylem

The guiding principle of effective water transport is that the

hydraulic supply of water must meet the transpirational

demands of the leaves (Sperry et al., 1998; Sperry, 2000).

Not surprisingly, the stomata play a key role because should

the water supply become compromised, or transpiration

exceed the plant’s hydraulic capacity, stomatal closure is the

first line of defence against desiccation. However, stomatal

conductance (gs) must also be well regulated because an inap-

propriately low gs will impose a carbon limitation on photo-

synthesis. Hence, photosynthesis and stomatal conductance

are closely coupled in what was probably the first of an

integrated sequence of adaptations that transformed the

gas-exchange characteristics of plants during their coloniza-

tion of land (Hetherington & Woodward, 2003; Franks &

Brodribb, 2005; Franks & Beerling, 2009a,b). Stomatal

distributions have been used to infer palaeoclimates from

fossil leaves but can leaf stomatal distributions lead us to an

improved understanding of plant palaeophysiology?

The close coupling between stomatal conductance and

water transport is well documented and several studies have

shown that alterations in stem water transport can cause either

stomatal closure or stomatal opening. Stomatal behaviour is

governed by the vapour pressure deficit between the leaf inte-

rior and ambient atmosphere, water status, the leaf boundary

layer, soil-to-leaf water transport and chemical signals such as

abscisic acid. However, it is the leaf water status that encapsu-

lates the holistic effect of these factors on stomatal conduc-

tance because leaf water potential regulates the turgor

pressure of both stomatal guard cells as well as the subsidiary

cells surrounding the stomata (Meinzer & Grantz, 1990;

Saliendra et al., 1995; Buckley, 2005). Not surprisingly, the

leaf water potential is to a large extent, controlled by the

transport capacity of the xylem and its ability to recharge water

lost by transpiration. Indeed, an artificial reduction of xylem

transport rapidly induces stomatal closure and reduces transpi-

ration, whereas root pressurization restores leaf water poten-

tial and induces stomatal opening (Saliendra et al., 1995;

Hubbard et al., 2001). Alternatively, partial de-foliation of

the plant canopy increases the supply of water relative to the

demand, thereby inducing an increase in stomatal conduc-

tance (Oren et al., 2001).

A strong relationship exists between stem hydraulic

conductance and photosynthesis suggesting that leaf gas

exchange is proportional to, or constrained by vascular sup-

ply (Brodribb & Feild, 2000; Hubbard et al., 2001; Santi-

130 J . PITTERMANN


ago et al., 2004; Maherali et al., 2008; Brodribb, 2009).

These findings are corroborated by extensive work docu-

menting the coupling between leaf vasculature, stomatal

conductance and gas exchange across a variety of plant life

forms (Brodribb et al., 2003, 2005). Because at least 25% of

the total resistance to plant water flow resides in the leaves,

leaves may exert a disproportionately high degree of control

over plant water relations (Sack et al., 2003). It may also be

that the leaf stomatal and vascular performance may affect

the structure and function of ‘upstream’ stem xylem,

although that possibility has not previously been discussed.

Certainly, such a co-ordinated response would require some

degree of phenotypic plasticity in both the leaves and the

xylem in the short term, but it may also have significant

long-term or even deep-time evolutionary implications on

the structure and function of xylem.

Atmospheric CO2 levels may have constrained the struc-

ture and function relationships of soil-to-leaf hydraulic trans-

port in just such a manner. There is much geological,

biological and theoretical evidence to suggest that atmo-

spheric CO2 levels have fluctuated dramatically over time

(Zachos et al., 2001; Beerling, 2002; Berner, 2006; Fletcher

et al., 2008). Indeed, CO2 levels are thought to have reached

highs well over 4000 ppm during the Devonian and lows

below 500 ppm during the Carboniferous (Fig. 1). Interest-

ingly, much of the evidence for deep-time atmospheric CO2

fluctuations has either been derived from or corroborated by

measures of the leaf stomatal response. Specifically, it has

been shown that across a broad sampling of plants from the

angiosperms and conifers, stomatal density increases in

response to low atmospheric CO2 whereas the opposite is

true under elevated levels of CO2 (Woodward, 1987;

McElwain et al., 1999; Royer et al., 2001; Retallack, 2001;

Beerling, 2002; Kouwenberg et al., 2003). Data derived

from the stomatal index, a more refined measure of the

stomatal response that controls for leaf expansion, also

indicate that in some taxa, CO2 is a developmental driver of

leaf stomatal distributions (Lake et al., 2002). The discovery

that stomatal abundance is related to atmospheric CO2 aug-

mented our understanding of palaeoclimates because CO2

could be retrodicted from leaves much older than ice cores.

In this way, stomatal density in combination with isotope and

palaeosol signals serves as a reasonable CO2 proxy with which

to evaluate the accuracy of climate models.

We know, however, that stomata play a dual role in regulat-

ing water loss as much as CO2 uptake, yet the effect of vacillat-

ing CO2 levels on the palaeo-physiology, especially water

relations of the sampled plants have attracted less attention.

The effect that increasing stomatal densities may have on leaf

water loss was apparent to Woodward (1987) who showed

that the stomatal conductance of Acer pseudoplatanus seed-

lings grown under 225 ppm of CO2 was twice as high as that

of seedlings grown under 340 ppm. Functionally, this meant

that the water-use efficiency of the 225 ppm-grown seedlings

was well over 50% lower than that of the 340 ppm grown

plants. A similar response was observed across eight species of

conifers grown at 280 and 800 ppm (J. Pittermann, unpub-

lished data, Fig. 10). These results imply that plants may have

been predisposed to a high degree of water stress during peri-

ods of low CO2 such as the Pleistocene glacial periods.

Increased susceptibility to drought stress combined with

reduced water use efficiency may have been a contributing fac-

tor to the reduced vegetation cover during the Pleistocene

(Cowling & Sykes, 1999).

If we accept that the apparent changes in stomatal densities

are consistent with the fluctuating CO2 levels of the Mid-to-

Late Phanerozoic, then atmospheric CO2 may have had a pro-

found effect on the whole plant conductance to water. For

example, the ‘topdown’ stomatal response to CO2 may have

at various times selected for plant hydraulic architecture that

was better equipped to mine water via deeper and more com-

plex root systems. Such plants may have also been more effi-

cient at transporting water by allocating a greater proportion

of resources to xylem, or by increasing the transport efficiency

of their xylem tissue. Hence, under periods of good water

availability and low atmospheric CO2, it is tempting to specu-

late that selection may have favoured the presence of more

efficient, large-diameter conduits or a greater fraction of tissue

devoted to water transport, in order to meet the increased

transpirational demands inherent to leaves with high stomatal

densities.

The alternative hypothesis is that reduced atmospheric CO2

levels may have imposed a carbon limitation on growth in a

manner that reduced leaf area, such that whole-plant transpi-

ration scaled with reduced plant size. Consequently, xylem

area ⁄ leaf area, leaf-specific hydraulic efficiency and conduit

diameter would remain unchanged. Indeed, several studies

have shown smaller leaf areas in plants grown under subambi-

ent CO2 (Dippery et al., 1995; Cowling & Sage, 1998).

However, the data do indicate that transpiration rates

continue to be higher in these plants than those grown at

ambient CO2 levels (Ward et al., 1999), suggesting that on a

whole-plant scale, the increased water loss may also impose a

Metasequoia

glyptostroboides

Taxodium

distichumSequoia

Sequoiadendron

sempervirens

giganteum Larix

elliottii

Pseudotsuga

menziesii Pinus

ponderosa

occidentalisPinus

800 ppm grown0.6

0.5

0.4

0.3

0.2

0.1

Sto

mat

al c

on

du

ctan

ce (

g, m

ol m

–2 s

–1)

0

280 ppm grown

Fig. 10 The stomatal conductance of Cupressaceaous and Pinaceous conifer

seedlings grown at 280 and 800 ppm in growth cabinets (mean ± SD; J. Pitter-

mann, unpublished data).



significant physiological limitation. Lastly, an increase or

decrease in the pressure gradients within the xylem may alter

the flow rates such that transpirational demands are effectively

met without a need for anatomical changes in the xylem tissue

(Fig. 11). This response, would however predispose the plants

to a greater possibility of cavitation caused by air seeding.

We can begin to explore the impact of atmospheric CO2 on

plant water relations by returning to some basic relationships

between xylem transport, anatomy and stomatal conductance.

The relationship between gs and leaf-specific hydraulic

conductance (Kl) can be described such that,

gs ¼ K1DW

VPD

� �; ð10Þ

where DW is the water potential difference between the soil

and the leaves (Wsoil ) Wleaf) driving water movement and

VPD represents the vapour pressure deficit between the leaf

interior and the ambient air (Bond & Kavanagh, 1999;

Hubbard et al., 2001). Kl is a measure of hydraulic efficiency

on a whole-plant basis (Bond & Kavanagh, 1999; Sperry,

2000; Hubbard et al., 2001). If we assume that ks scales with

Kl (certainly ks scales with leaf-specific conductivity, Kl; J. Pit-

termann, unpublished data), then substituting equation 2 for

Kl in equation 10 yields the final relationship,

gs1DWð ÞpRD4

A128gVPD: ð11Þ

Equation 11 expresses a theoretical correlation between

stomatal conductance and the theoretical maximal transport

capacity of the xylem, as constrained by conduit diameter.

Simplistically, equation 11 implies that an overall rise in gs

requires an increase in conduit diameter, for a given xylem

area. Such a transport response would require the co-ordina-

tion between the vascular cambium and stomatal patterning

during leaf development, and there is evidence to suggest that

in some species, stomatal development is labile and thus

responsive to atmospheric CO2 (Lake et al., 2002; Beerling,

2005) but whether this couples to vascular development

remains unknown. Alternatively, xylem anatomy may exhibit

limited phenotypic plasticity, and thus impose fundamental

limits on leaf gas exchange as appears to be the case in diffuse

and ring porous angiosperm trees (Bush et al., 2008).

If a link does indeed exist between atmospheric CO2, stoma-

tal conductance, stomatal density and vascular design, it is

tempting to speculate that the 70 million years during the

Mid-Devonian to Early Carboniferous provided atmospheric

conditions that supported the evolution of secondary xylem

(wood). Climate models suggest that this period of the Phan-

erozoic was characterized by CO2 levels that are thought to

have been at least as low or lower than those of today, and

which may have stimulated not only an increase in stomatal

densities but also the appearance of large leaves (Beerling

et al., 2001; Osborne et al., 2004). Furthermore, it is at this

time that elevated atmospheric oxygen levels are thought to

have supported an increase in xylem lignification and possibly

selection for woody taxa (Graham et al., 1995; Berner, 2005;

Berner et al., 2007). A concurrent argument is that low CO2

selected for high stomatal densities, which in turn increased

evaporative cooling thereby releasing leaves from the con-

straints of overheating. This is also consistent with survey

results from the fossil record showing a 25-fold enlargement of

leaf areas during this time (Osborne et al., 2004). In any case,

if atmospheric CO2 precipitated changes in leaf size and stoma-

tal density, then we might expect a co-ordinated response from

the vascular system designed to increase transport efficiency.

A

B

Fig. 11 Some theoretical relationships between

xylem structure and function in response to low and

high ambient CO2.

132 J . PITTERMANN


PART IV: CONCLUDING REMARKS

Plant water transport and climate

Plant water transport irrevocably changed our planet. Even

the earliest transpiring plants would have subtly modified their

local water regime, whereas the Devonian forests were proba-

bly key players in the formation of soils, carbon sequestration

and climate change (Retallack, 1997; Algeo & Scheckler,

1998; Berner, 2005; Berner et al., 2007) Xylem function cer-

tainly impacts atmospheric moisture, and the Amazon rainfor-

est is a dramatic, modern day example of how large-scale

water transport controls continental climates as nearly 50% of

global precipitation is recycled from forests (Malhi et al.,

2008). Deep roots extract and potentially re-distribute the soil

water that is ultimately transpired back to the atmosphere, a

phenomenon that is critical to maintaining hydration when

high evaporative demand and reduced rainfall conspire to

reduce stomatal conductance, such as during the dry season

(Oliveira et al., 2005). Consequently, extensive deforestation

tends to lower evapotranspiration, cloud formation and rain-

fall by a feedforward effect that progressively reduces ecosys-

tem water cycling. Model projections suggest that the

removal of 30–40% of the Amazonian forest could drive the

region into a permanently drier climate regime (Oyama &

Nobre, 2003). Regional meteorological phenomena will also

be affected by alteration of the forest canopy boundary effects

so, taken together, there is indisputable evidence that plant

water transport is a key element of large-scale ecosystem sta-

bility (Betts et al., 2004).

Evapotranspiration is a common metric of area-based

terrestrial water loss but recent work suggests that transpi-

ration alone can modify continental hydrology. Extensive

field studies have shown that elevated CO2 reduces stoma-

tal conductance and transpiration, an effect that generally

improves the water balance of terrestrial vegetation (Korn-

er, 2003). However, the suppression of transpiration leads

to greater water retention in soils and ultimately higher

water discharge into rivers, a phenomenon referred to as

‘physiological forcing’ (Betts et al., 2007). How would this

scenario play out under low CO2 conditions, which typically

give rise to increased stomatal conductance? Given the

uncertainties of climate modelling, a simple answer may

not exist but all things being equal, a reasonable guess may

be that increased transpiration would lower soil moisture,

and thus impact terrestrial water levels.

Future directions

Plant vascular tissue is one of potential drivers of plant diversi-

fication (Donoghue, 2005), so much insight can be gained

into the evolution of plants by applying a neontological and

quantitative approach to fossil wood. Although the Hagen–

Poiseuille method of computing theoretical hydraulic efficien-

cies is useful, incorporating pit resistance into fossil plant

transport models will allow us to broaden our understanding

of xylem function in extinct plants, as well as the significance

of evolutionary transitions from one conduit type to another

such as in the case of the Rhyniophytic plants. As we do this

however, comparative studies of xylem structure and function

in modern taxa need to develop within a broader evolutionary

framework. For example, we know little about the relation-

ships between cavitation resistance, pit resistance and hydrau-

lic efficiency in monocots and Gnetales.

Aside from the extensive descriptive work documenting the

anatomy of fern xylem, not much is currently known about

the comparative hydraulics of ferns, and the homogenous fern

intertracheid pit membrane in particular (Carlquist & Schnei-

der, 2007). This is important information because tracheids

with homogenous pit membranes represent the ancestral state

of xylem in leafy tracheophytes (Euphyllophytes; Kenrick &

Crane, 1997). However, new data collected from across the

fern lineage show that fern xylem is as efficient on an

area basis as that of angiosperms and conifers (J. Pittermann

& E. Limm, unpublished data), suggesting that the large

tracheid size in combination with what appears to be a highly

permeable pit membrane (Carlquist & Schneider, 2007) combine

to ramp up water transport in an otherwise seemingly subopti-

mal combination of vascular traits (Pittermann et al., 2005).

We can also exploit our current understanding of the rela-

tionship between pit area and pit structure in angiosperms and

conifers respectively, to gain more insight into the drought

resistance of early land plants. The pit area hypothesis holds

across angiosperms of different evolutionary lineages,

although membrane porosity may also variably affect resis-

tance to cavitation (Jarbeau et al., 1995; Hacke et al., 2006,

2007; see also Choat et al., 2008). Assuming that pit mem-

branes are well preserved in fossil specimens, perhaps the best

approach may be to examine the pit membrane size, scaling

and porosity in a comparative manner, sampling taxa across

the land plant phylogeny.

Understanding the structure and function of xylem can

inform us not only about the physiological limits of plant

water transport and the constraints it places on photosynthe-

sis, but this information also has potential to augment models

of palaeocommunity assembly. For instance, the Devonian

and Carboniferous are characterized by significantly different

climatic patterns than those of today, as well as changes in the

relative positions of land masses and tremendous diversifica-

tion of the terrestrial flora (DiMichele & Gastaldo, 2008).

Plant distributions in these ancient and unfamiliar landscapes

will, to some extent, be determined by xylem efficiencies and

drought resistance. Altogether, this is an excellent time to

fruitfully study the evolution of plant water transport across a

multitude of scales, ranging from the molecular, organismal

and up to the ecosystem level. Certainly, the intersection of

plant physiology and palaeobotany offer much opportunity

for truly novel discoveries.



ACKNOWLEDGMENTS

I am very grateful to the three anonymous reviewers

whose thoughtful comments and copious advice dramatically

improved this manuscript. Also, many thanks to David

Beerling for the generous opportunity to contribute this

paper. I am also grateful to Lucy Lynn for editorial assistance,

and Brendan Choat, Emily Limm and Jim Wheeler for their

helpful comments. Dr Diane Erwin (UC Berkeley Museum of

Natural History) kindly loaned Archaeopteris fossil material,

and the Miller Institute for Basic Research at UC Berkeley

provided generous support during the preparation of this

manuscript.

REFERENCES

Ackerly DD, Dudley SA, Sultan SE, Schmitt J, Coleman JS, Lin-

der CR, Sandquist DR, Geber MA, Sultan EA, Dawson TE,

Lechowicz MJ (2000) The evolution of plant ecophysiologicaltraits: recent advances and future directions. BioScience 50,

979–995.

Ainsworth EA, Rogers A (2007) The response of photosynthesis

and stomatal conductance to rising [CO2]: mechanisms andenvironmental interactions. Plant, Cell and Environment 30,

258–270.

Algeo TJ, Scheckler SE (1998) Terrestrial-marine teleconnections inthe Devonian: links between the evolution of land plants, weather-

ing processes, and marine anoxic events. Philosophical Transactionsof the Royal Society of London, Series B: Biological Sciences 353,

113–130.Angeles G, Bond B, Boyer J, Brodribb T, Brooks J, Burns M, Cavend-

er-Bares J, Clearwater M, Cochard H, Comstock J (2004) The

cohesion-tension theory. New Phytologist 163, 451–452.

Assmann SH, Wang X-Q (2001) From milliseconds to millions ofyears: guard cells and environmental responses. Current Opinion inPlant Biology 4, 421–428.

Baas P, Ewers FW, Davis SD, Wheeler EA (2004) Evolution of xylemphysiology. In The Evolution of Plant Physiology (eds Hemsley AR,

Poole I). Elsevier, London, pp. 273–295.

Bailey W, Tupper WW (1918) Size variation in tracheary cells: I. A

comparison between the secondary xylems of vascular cryptograms,gymnosperms and angiosperms. Proceedings of the AmericanAcademy of Arts and Sciences 54, 149–204.

Bargel H, Barthlott W, Koch K, Schreiber L, Neinhuis C (2004) Plant

cuticles: multifunctional interfaces between plant and environment.In The Evolution of Plant Physiology (eds Hemsley AR, Poole I).

Elsevier, London, 171–194.

Bargel H, Koch K, Cerman Z, Neinhuis C (2006) Structure-function

relationships of the plant cuticle and cuticular waxes – a smartmaterial? Functional Plant Biology 33, 893–910.

Bateman RM, Crane PR, Dimichele WA, Kenrick PR, Rowe NP,

Speck T, Stein WE (1998) Early evolution of land plants: phylog-eny, physiology and ecology of the primary terrestrial radiation.

Annual Review of Ecology and Systematics 29, 263–292.

Beck CB (1960) Connection between Archaeopteris and Callixylon.

Science 131, 1524–1525.Beck CB (1970) The appearance of gymnospermous structure. Biolog-

ical Reviews 45, 379–400.

Beck EH, Fettig S, Knake C, Hartig K, Bhattarai T (2007) Specific

and unspecific responses of plants to cold and drought stress.Journal of Bioscience 32, 501–510.

Beerling DJ (2002) Low atmospheric CO2 levels during the Permo-

Carboniferous glaciation inferred from fossil lycopsids. Proceedingsof the National Academy of Sciences 99, 12567–12571.

Beerling DJ (2005) Leaf evolution: gases, genes and geochemistry.

Annals of Botany 96, 345–352.

Beerling DJ, Berner RA (2002) Feedbacks and the coevolution ofplants and atmospheric CO2. Proceedings of the National Academyof Sciences 102, 1302–1305.

Beerling DJ, Osborne CP, Chaloner WG (2001) Evolution of leaf-

form in land plants linked to atmospheric CO2 decline in the latePaleozoic era. Nature 410, 352–354.

Berner RA (2005) The rise of trees and how they changed Paleozoic

atmospheric CO2, climate and geology. In A History of AtmosphericCO2 and Its Effects on Plants, Animals and Ecosystems (eds Ehlerin-ger J, Cerling TE, Dearing DM). Springer, New York, pp. 1–7.

Berner RA (2006) GEOCARBSULF: a combined model for Phanero-

zoic atmospheric O2 and CO2. Geochimica et Cosmochimica Acta70, 5653–5664.

Berner RA, VandenBrooks JM, Ward PD (2007) Oxygen and evolu-

tion. Science 316, 557–558.

Betts RA, Cox PM, Collins M, Harris PP, Huntingford C, Jones CD(2004) The role of ecosystem-atmosphere interactions in simulated

Amazonian precipitation decrease and forest dieback under global

climate warming. Theoretical and Applied Climatology 78,

157–175.Betts RA, Boucher O, Collins M, Cox PM, Falloon PD, Gedney

N, Hemming DL, Huntingford C, Jones CD, Sexton DMH,

Webb MJ (2007) Projected increase in continental runoff due to

plant responses to increasing carbon dioxide. Nature 448,1037–1042.

Bhattacharya D, Medlin L (1998) Algal phylogeny and the origin of

land plants. Plant Physiology 116, 9–15.Boerjan W, Ralph J, Baucher M (2003) Lignin biosynthesis. Annual

Review of Plant Biology 54, 519–546.

Bonan GB (2008) Forests and climate change: forcings, feedbacks

and the climate benefits of forests. Science 320, 1444–1449.Bond WJ (1989) The tortoise and the hare: ecology of angiosperm

dominance and gymnosperm persistence. Biological Journal of theLinnean Society 36, 227–249.

Bond BJ, Kavanagh KL (1999) Stomatal behavior of four woodyspecies in relation to leaf-specific hydraulic conductance and

threshold water potential. Tree Physiology 19, 503–510.

Boyce CK (2008) How green was Cooksonia? The importance of sizein understanding the early evolution of physiology in the vascular

plant lineage. Paleobiology 34, 179–194.

Boyce CK, Cody GD, Feser M, Knoll AH (2002) Organic chemical

differentiation within fossil plant cell walls detected with X-rayspectromicroscopy. Geology 30, 1039–1042.

Boyce CK, Cody GD, Fogel ML, Hazen RM (2003) Chemical evi-

dence for cell wall lignification and the evolution of tracheids in

early Devonian plants. International Journal of Plant Sciences 164,691–702.

Brezinski DK, Cecil CB, Skema VW, Stamm R (2008) Late Devonian

glacial deposits from the eastern United States signal an end of the

mid-Paleozoic warm period. Palaeogeography, Palaeoclimatology,Palaeoecology 268, 143–151.

Brodribb TJ (2009) Xylem hydraulic physiology: the functional back-

bone of terrestrial plant productivity. Plant Science 177, 245–251.Brodribb TJ, Feild T (2000) Stem hydraulic supply is linked to leaf

photosynthetic capacity: evidence from New Caledonian and Tas-

manian rainforests. Plant, Cell and Environment 23, 1381–1388.

Brodribb TJ, Holbrook NM (2004) Stomatal protection againsthydraulic failure: a comparison of coexisting ferns and angiosperms.

New Phytologist 162, 663–670.

134 J . PITTERMANN


Brodribb TJ, Holbrook NM, Gutierrez MV (2002) Hydraulic and

photosynthetic coordination in seasonally dry tropical forest trees.

Plant, Cell and Environment 25, 1435–1444.Brodribb TJ, Holbrook NM, Edwards EJ, Gutierrez MV (2003) Rela-

tions between stomatal closure, leaf turgor and xylem vulnerability

in eight tropical dry forest trees. Plant, Cell and Environment 26,443–450.

Brodribb TJ, Holbrook NM, Zwieniecki MA, Palma B (2005) Leaf

hydraulic capacity in ferns, conifers and angiosperms: impacts on

photosynthetic maxima. New Phytologist 165, 839–846.Brodribb TJ, Feild TS, Jordan GJ (2007) Leaf maximum photosyn-

thetic rate and venation are linked by hydraulics. Plant Physiology144, 1890–1898.

Brodribb TJ, McAdam SAM, Jordan GJ, Feild TS (2009) Evolutionof stomatal responsiveness to CO2 and optimization of water-use

efficiency among land plants. New Phytologist 183, 839–847.

Buckley TN (2005) The control of stomata by water balance. NewPhytologist 168, 275–292.

Burgess SO, Pittermann J, Dawson TE (2006) Hydraulic efficiency

and safetyofbranchxylem increases withheight inSequoia sempervi-

rens (D. Don) crowns. Plant, Cell and Environment 29, 229–239.Burleigh JG, Mathews S (2004) Phylogenetic signal in nucleotide data

from seed plants: implications for resolving the seed plant tree of

life. American Journal of Botany 91, 1599–1613.

Bush S, Pataki DE, Hultine K, West A, Sperry J, Ehleringer J (2008)Wood anatomy constrains stomatal responses to atmospheric vapor

pressure deficit in irrigated, urban trees. Oecologia 156, 13–20.

Calkin HW, Gibson AC, Nobel PS (1986) Biophysical model of xylem

conductance in tracheids of the fern Pteris vittata. Journal of Exper-imental Botany 37, 1054–1064.

Came RE, Eiler JM, Veizer J, Azmy K, Brand U, Weidman C (2007)

Coupling of surface temperatures and atmospheric CO2 concentra-tions during the Palaeozoic era. Nature 449, 198–202.

Carlquist S, Schneider EL (2007) Tracheary elements in ferns: new

techniques, observations and concepts. American Fern Journal 97,

199–211.Cavender-Bares J, Holbrook NM (2001) Hydraulic properties and

freezing-induced cavitation in sympatric evergreen and deciduous

oaks with contrasting habitats. Plant, Cell and Environment 24,

1243–1256.Cavender-Bares J, Cortes P, Rambal S, Joffre R, Miles B, Rocheteau A

(2005) Summer and winter sensitivity of leaves and xylem to

minimum freezing temperatures: a comparison of co-occurringMediterranean oaks that differ in leaf span. New Phytologist 168,

597–612.

Chaffey N (2002) Why is there so little research into the cell biology

of the secondary vascular system of trees? New Phytologist 153,213–223.

Chaloner WG, McElwain JC (1997) The fossil plant record and

global climatic change. Review of Palaeobotany and Palynology95, 73–82.

Choat B, Cobb AR, Jansen S (2008) Structure and function of bor-

dered pits: new discoveries and impacts on whole-plant hydraulic

function. New Phytologist 177, 608–626.

Cichan MA (1985a) Vascular cambium and wood development inCarboniferous plants. I. Lepidodendrales. American Journal ofBotany 72, 1163–1176.

Cichan MA (1985b) Vascular cambium and wood development inCarboniferous plants. II. Sphenophyllum. Botanical Gazette 146,

395–403.

Cichan MA (1986) Conductance in the wood of selected Carbonifer-

ous plants. Paleobiology 12, 302–310.Cochard H, Breda N, Granier A, Aussenac G (1992) Vulnerability to

air embolism of three European oak species (Quercus petraea (Matt)

Liebl, Q. pubescens Willd, Q. robur L). Annales Sciences Forestales49, 225–233.

Cochard H, Ameglio T, Cruiziat P (2001) The cohesion theorydebate continues. Trends in Plant Science 6, 456.

Comstock JP, Sperry JS (2000) Some theoretical considerations of

optimal conduit length for water transport in plants. New Phytolo-gist 148, 195–218.

Cooke TJ, Poli D, Sztein AE, Cohen JD (2002) Evolutionary patterns

in auxin action. Plant Molecular Biology 49, 319–338.

Cosgrove DJ (2005) Growth of the plant cell wall. Nature ReviewsMolecular Cell Biology 6, 850–861.

Cowling SA, Sage RF (1998) Interactive effects of low atmospheric

CO2 and elevated temperature on growth, photosynthesis and

respiration in Phaseolus vulgaris. Plant, Cell and Environment 21,427–435.

Cowling SA, Sykes MT (1999) Physiological significance of low

atmospheric CO2 for plant-climate interactions. QuaternaryResearch 52, 237–242.

Davies WJ, Zhang J (1991) Root signals and the regulation of growth

and development of plants in drying soil. Annual Reviews in PlantPhysiology and Molecular Biology 42, 55–76.

Davis SD, Sperry JS, Hacke UG (1999) The relationship between

xylem conduit diameter and cavitation caused by freezing.

American Journal of Botany 86, 1367–1372.

Delwiche CF, Graham LE, Thomson N (1989) Lignin-like com-pounds and sporopollenin in Coleochaete, an algal model for land

plant ancestry. Science 245, 399–401.

Denny MW (1993) Air and Water: The Biology and Physics of Life’sMedia. Princeton University Press, Princeton, NJ.

DiMichele W, Gastaldo RA (2008) Plant paleoecology in deep time.

Annals of the Missouri Botanical Garden 95, 144–198.

DiMichele W, Phillips TL, Pfefferkorn HW (2006) Paleoecology ofLate Paleozoic pteridosperms from tropical Euramerica. Journal ofthe Torrey Botanical Society 133, 83–118.

Dippery JK, Tissue DT, Thomas RB, Strain BR (1995) Effects of low

and elevated CO2 on C3 and C4 annuals 1. Growth and biomassallocation. Oecologia 101, 13–20.

Dixon HH, Joly J (1894) On the ascent of sap. Philosophical Transac-tions of the Royal Society of London, Series B: Biological Sciences 186,

563–576.Domec J-C, Lachenbruch B, Meinzer F, Woodruff D, Warren JM,

Mcculloh K (2008) Maximum height in a conifer is associated with

conflicting requirements for xylem design. Proceedings of theNational Academy of Sciences 105, 12069–12074.

Donoghue MJ (2005) Key innovations, convergence, and success:

macroevolutionary lessons from plant phylogeny. Paleobiology 31,

77–93.Duckett J, Pressel S, P’ng KMY, Renzaglia KS (2009) Exploding

a myth: the capsule dehiscence mechanism and the function of

pseudostomata in Sphagnum. New Phytologist 183, 1053–

1063.Dudley S (1996) Differing selection on plant physiological traits in

response to environmental water availability: a test of adaptive

hypotheses. Evolution 50, 92–102.

Edwards D (1986) Aglaophyton major, a non-vascular land plantfrom the Devonian Rhynie Chert. Botanical Journal of the LinneanSociety 93, 173–204.

Edwards D (1996) New insights into early land ecosystems: a glimpseof a Lilliputian world. Review of Palaeobotany and Palynology 90,

159–174.

Edwards D (2003) Xylem in early tracheophytes. Plant, Cell andEnvironment 26, 57–72.

Edwards D, Feehan J (1980) Records of Cooksonia-type sporangia

from late Wenlock strata in Ireland. Nature 287, 41–42.



Edwards D, Abbott GD, Raven JA (1996) Cuticles in early land

plants: a palaeoecophysiological evaluation. In Plant Cuticles: AnIntegrated Approach (ed. Kerstiens G). Elsevier, Oxford, UK, pp.1–31.

Edwards D, Kerp H, Hass H (1998) Stomata in early land plants: an

anatomical and ecophysiological approach. Journal of ExperimentalBotany 49, 255–278.

Edwards D, Morel E, Poire DG, Cingolani CA (2001) Land plants in

the Devonian Villavicencio Formation, Mendoza Province, Argen-

tina. Review of Palaeobotany and Palynology 116, 1–18.Eggert DA (1961) The ontogeny of Carboniferous arborescent

Lycopsida. Palaeontographica 110B, 99–127.

Ellerby DJ, Ennos AR (1998) Resistances to fluid flow of model xylem

vessels with simple and scalariform perforation plates. Journal ofExperimental Botany 49, 979–985.

El-Saadawy WE, Lacey WS (1979) The sporangia of Horneophyton lig-nieri (Kidston and Lang) Barghoorn and Darrah. Review ofPalaeobotany and Palynology 28, 137–144.

Esau K (1943) Origin and development of primary vascular tissues.

Botanical Review 9, 125–206.

Evans RD, Johansen JR (1999) Microbiotic crust and ecosystemprocesses. Critical Reviews in Plant Sciences 18, 183–225.

Ewers FW (1985) Xylem structure and water conduction in conifer

trees, dicot trees, and lianas. Bulletin of the International Associa-tion of Wood Anatomists 6, 309–317.

Farris MA, Lechowicz MJ (1990) Functional interactions among

traits that determine reproductive success in a native annual plant.

Ecology 71, 548–557.

Feist M, Liu J, Tafforeau P (2005) New insights into Paleozoic charo-phyte morphology and phylogeny. American Journal of Botany 92,

1152–1160.

Fletcher BJ, Brentnall SJ, Anderson CW, Berner RA, Beerling DJ(2008) Atmospheric carbon dioxide linked with Mesozoic and early

cenozoic climate change. Nature Geoscience 1, 43–48.

Franks P, Beerling D (2009a) Maximum leaf conductance driven by

CO2 effects on stomatal size and density over geologic time.Proceedings of the National Academy of Sciences 106,

10343–10347.

Franks P, Beerling D. (2009b) CO2-forced evolution of plant gas

exchange capacity and water-use efficiency over the Phanerozoic.Geobiology 7, 227–236.

Franks P, Brodribb T (2005) Stomatal control and water transport in

stems. In Vascular Transport in Plants (eds Holbrook NM,Zwieniecki MA). Elsevier, Amsterdam, pp. 69–89.

Franks PJ, Farquhar GD (2007) The mechanical diversity of stomata

and its significance in gas-exchange control. Plant Physiology 143,

78–87.Friedman WE, Cook ME (2000) The origin and early evolution of

tracheids in vascular plants: integration of palaeobotanical and

neobotanical data. Philosophical Transactions of the Royal Society ofLondon, Series B: Biological Sciences 355, 857–868.

Gensel PG (2008) The earliest land plants. Annual Review of Ecology,Evolution and Systematics 39, 459–477.

Gensel PG, Edwards D (2001) Plants Invade the Land: Evolutionaryand Environmental Perspectives. Columbia University Press, NewYork.

Gensel PG, Chaloner WG, Forbes WH (1991) Spongiophyton from

the late lower Devonian of New Brunswick and Quebec, Canada.Paleontology 34, 149–168.

Gibson AC, Calkin HW, Nobel PS (1984) Xylem anatomy, water

flow, and hydraulic conductance in the fern Cyrtomium falcatum.

American Journal of Botany 71, 564–574.Graham LE (1993) Origin of Land Plants. John Wiley and Sons, Inc.,

New York.

Graham LE (1996) Green Algae to Land Plants: an evolutionary

transition. Journal of Plant Research 109, 241–251.

Graham LE, Gray J (2001) The origin, morphology and ecophysiol-ogy of early embryophytes: neontological and paleontological

perspectives. In Plants Invade the Land: Evolutionary and Environ-mental Perspectives (eds Gensel Pg, Edwards D). Columbia Univer-sity Press, New York, pp. 140–158.

Graham JB, Dudley R, Aguilar N, Gans C (1995) Implications of the

late Palaeozoic oxygen pulse for physiology and evolution. Nature375, 117–120.

Graham LE, Cook ME, Busse JS (2000) The origin of plants: body

plan changes contributing to a major evolutionary radiation. Pro-ceedings of the National Academy of Sciences 97, 4535–4540.

Graham L, Graham J, Wilcox L (2009) Algae, 2nd edn. BejaminCummings, San Francisco, CA.

Gray J, Chaloner WG, Westoll TS (1985) The microfossil record of

early land plants: advances in understanding of early terrestrializa-

tion, 1970–1984. Philosophical Transactions of the Royal Society ofLondon, Series B: Biological Sciences 309, 167–195.

Hacke UG, Jansen S (2009) Embolism resistance of three boreal coni-

fer species varies with pit structure. New Phytologist 182, 675–686.Hacke UG, Sperry JS (2001) Functional and ecological xylem anat-

omy. Perspectives in Plant Ecology, Evolution, and Systematics 4,

97–115.

Hacke UG, Sperry JS, Pockman WP, Davis SD, Mcculloh KA (2001)Trends in wood density and structure are linked to prevention of

xylem implosion by negative pressure. Oecologia 126, 457–461.

Hacke UG, Sperry JS, Wheeler JK, Castro L (2006) Scaling of angio-

sperm xylem structure with safety and efficiency. Tree Physiology 26,689–701.

Hacke UG, Sperry JS, Feild TS, Sano Y, Sikkema EH, Pittermann J

(2007) Water transport in vesselless angiosperms: conducting effi-ciency and cavitation safety. International Journal of Plant Sciences168, 1113–1126.

Herendeen PS, Crane PR (1995) The fossil history of the monocoty-

ledons. In Monocotyledons: Systematics and Evolution (eds Rudall PJ,Cribb PJ, Cutler DF, Humphries CJ). Royal Botanic Gardens,

Kew, UK, pp. 1–21.

Hetherington AM, Woodward FI (2003) The role of stomata in sens-

ing and driving environmental change. Nature 424, 901–908.Hubbard RM, Stiller V, Ryan MG, Sperry JS (2001) Stomatal con-

ductance and photosynthesis vary linearly with plant hydraulic con-

ductance in ponderosa pine. Plant, Cell and Environment 24,113–121.

Huey RB, Ward PD (2005) Hypoxia, global warming and terrestrial

Late Permian extictions. Science 308, 398–401.

Jacobsen AL, Agenbag L, Esler KJ, Pratt RB, Ewers FW, Davis SD(2007) Xylem density, biomechanics and anatomical traits correlate

with water stress in 17 evergreen shrub species of the Mediterra-

nean-type climate region of South Africa. Journal of Ecology 95,

171–183.Jarbeau JA, Ewers FW, Davis SD (1995) The mechanism of water-

stress-induced embolism in two species of chaparral shrubs. Plant,Cell and Environment 18, 189–196.

Jeje AA, Zimmermann MH (1979) Resistance to water flow in xylemvessels. Journal of Experimental Botany 30, 817–827.

Jones HG, Sutherland R (1991) Stomatal control of xylem embolism.

Plant, Cell and Environment 14, 607–612.Jones L, Ennos AR, Turner SR (2001) Cloning and characterization

of irregular xylem4(irx4): a severely lignin-deficient mutant of Ara-bidopsis. The Plant Journal 26, 205–216.

Kenrick PR, Crane PR (1991) Water-conducting cells in early fossilland plants: implications for the early evolution of tracheophytes.

Botanical Gazette 152, 335–356.

136 J . PITTERMANN


Kenrick P, Crane PR (1997) The origin and early evolution of plants

on land. Nature 389, 33–39.

Kitin PB, Fujii T, Abe H, Funada R (2004) Anatomy of the vessel net-work within and between tree rings of Fraxinus lanuginosa (Olea-

ceae). American Journal of Botany 91, 779–788.

Koch GW, Sillett SC, Jennings GM, Davis SD (2004) The limits totree height. Nature 428, 851–854.

Koehler L, Telewski FW (2006) Biomechanics and transgenic wood.


Kolb K, Sperry JS (1999) Differences in drought adaptation betweensubspecies of sagebrush (Artemisia tridentata). Ecology 80, 2373–

2384.

Konrad W, Roth-Nebelsick A, Kerp H, Hass H (2000) Transpiration

and assimilation of Early Devonian land plants with axially symmet-ric telomes – simulations on the tissue level. Journal of TheoreticalBiology 206, 91–107.

Korner C (2003) Ecological impacts of atmospheric CO2 enrich-

ment on terrestrial ecosystems. Philosophical Transactions of theRoyal Society of London, Series B: Biological Sciences 361,

2023–2041.

Kouwenberg LLR, McElwain JC, Kurschner WM, Wagner F, Beer-ling DJ, Mayle FE, Visscher H (2003) Stomatal frequency adjust-

ment of four conifer species to historical changes in atmospheric

CO2. American Journal of Botany 90, 610–619.

Kramer EM, Lewandowski M, Beri S, Bernard J, Borkowski M,Borkowski MH, Burchfield LA, Mathisen B, Normanly J (2008)

Auxin gradients are associated with polarity changes. Science 320,

1610.

Krings M, Kerp H, Taylor TN, Taylor EL (2003) How Paleozoicvines and lianas got off the ground: on scrambling and climbing

Carboniferous-Early Permian pteridosperms. The Botanical Review69, 204–224.

Kurschner WM, Kvacek Z, Dilcher DL (2008) The impact of Miocene

atmospheric carbon dioxide fluctuations on climate and the evolu-

tion of terrestrial ecosystems. Proceedings of the National Academyof Sciences 105, 449–453.

Lake JA, Woodward FI, Quick WP (2002) Long-distance CO2 signal-

ling in plants. Journal of Experimental Botany 53, 183–193.

Lancashire JR, Ennos AR (2002) Modelling the hydrodynamic resistance

of bordered pits. Journal of Experimental Botany 53, 1485–1493.Lewis NG (1999) A 20th century roller coaster ride: a short account

of lignification. Current Opinions in Plant Biology 2, 153–162.

Lewis LA, Lewis PO (2005) Unearthing the molecular phylodiversity ofdesert soil green algae (Chlorphyta). Systematic Biology 54, 936–947.

Lewis LA, McCourt RM (2004) Green algae and the origin of land

plants. American Journal of Botany 91, 1535–1556.

Ligrone R, Duckett JG (1996) Development of water-conductingcells in the antipodal liverwort Symphyogyna brasiliensis (Metzgeri-

ales). New Phytologist 132, 603–615.

Ligrone R, Duckett JG, Renzaglia KS (2000) Conducting tissues and

phyletic relationships of bryophytes. Philosophical Transactions of theRoyal Society of London, Series B: Biological Sciences 355, 795–813.

Liu W, Yang H (2008) Multiple controls for the variability of hydro-

gen isotopic compositions in higher plant n-alkanes from modern

ecosystems. Global Change Biology 14, 2166–2177.Maherali H, Sherrard ME, Clifford MH, Latta RG (2008) Leaf

hydraulic conductivity are genetically correlated in an annual grass.

New Phytologist 180, 240–247.Malhi Y, Roberts JT, Betts RA, Killeen TJ, Li W, Nobre CA (2008)

Climate change, deforestation and the fate of the Amazon. Science319, 169–172.

Martone PT, Estevez JM, Lu F, Ruel K, Denny MW, Somerville CR,Ralph J (2009) Discovery of lignin in seaweed reveals convergent

evolution of cell wall architecture. Current Biology 19, 169–175.

Mayr S, Wolfschwenger M, Bauer H (2002) Winter-drought induced

embolism in Norwary spruce (Picea abies) at the alpine timberline.

Physiologia Plantarum 115, 74–80.McCann MC (1997) Tracheary element formation: building up to a

dead end. Trends in Plant Science 2, 333–338.

McCourt RM, Delwiche CF, Karol KG (2004) Charophyte algaeand land plant origins. Trends in Ecology and Evolution 19,

661–666.

McCulloh KA, Sperry JS (2005) The evaluation of Murray’s law in

Psilotum nudum (Psilotaceae), an analogue of ancestral vascularplants. American Journal of Botany 92, 985–989.

McCulloh KA, Sperry JS, Adler FR (2003) Water transport in pants

obeys Murray’s law. Nature 421, 939–942.

McCulloh KA, Sperry JS, Adler FR (2004) Murray’s law and thehydraulic versus mechanical functioning of wood. Functional Ecol-ogy 18, 931–938.

McDowell N, Pockman WT, Allen GD, Breshears DD (2008) Mecha-

nisms of plant survival and mortality during drought: why do someplants survive while others succumb to drought? New Phytolgist178, 719–739.

McElwain JC, Chaloner WG (1996) The fossil cuticle as a skeletalrecord of environmental change. Palaios 11, 376–388.

McElwain JC, Beerling DJ, Woodward FI (1999) Fossil Plants and

Global Warming at the Triassic-Jurassic Boundary. Science 285,

1386–1390.Meinzer FC, Grantz DA (1990) Stomatal and hydraulic conductance

in growing sugarcane: stomatal adjustment to water transport

capacity. Plant, Cell and Environment 13, 383–388.

Messinger SM, Buckley TN, Mott KA (2006) Evidence for involve-ment of photosynthetic processes in the stomatal response to CO2.

Plant Physiology 140, 771–778.

Meyer-Berthaud B, Scheckler SE, Wendt J (1999) Archaeoperis is theearliest known modern tree. Nature 398, 700–701.

Montanez IP, Tabor NJ, Niemeier D, DiMichele WA, Frank TD,

Fielding CR, Isbell JL, Biorgenheier LP, Rygel MC (2007) CO2-

forced climate and vegetation instability during Late Paleozoicdeglaciation. Science 315, 87–91.

Mosbrugger V (1990) The tree habit in land plants: a functional com-

parison of trunk construction with a brief introduction into the bio-

mechanics of trees. Lecture Notes in Earth Sciences. Springer-Verlag,Berlin.

Nieminen KM, Kauppinen L, Helariutta Y (2004) A weed for wood?

Arabidopsis as a genetic model for xylem development. Plant Physi-ology 135, 653–659.

Niklas KJ (1985) The evolution of tracheid diameter in early vascular

plants and its implications on the hydraulic conductance of the pri-

mary xylem strand. Evolution 39, 1110–1122.Niklas KJ (1992) Plant Biomechanics. University of Chicago Press,

Chicago.

Niklas KJ (1997) The Evolutionary Biology of Plants. University of

Chicago Press, Chicago.Niklas KJ (2004) Computer models of early land plant evolution.

Annual Review of Earth and Planetary Science 32, 47–66.

Oliveira RS, Dawson TE, Burgess SSO, Nepstad DC (2005) Hydrau-

lic redistribution in three Amazonian trees. Oecologia 145, 354–363.

Oren R, Sperry JS, Ewers BE, Pataki DE, Phillips N, Megonigal JP

(2001) Sensitivity of mean canopy stomatal conductance to vaporpressure deficit in a flooded Taxodium distichum L. Forest: hydrau-

lic and non-hydraulic effects. Oecologia 126, 21–29.

Osborne CP, Beerling DJ, Lomax BH, Chaloner WG (2004) Bio-

physical constraints on the origin of leaves inferred from the fossilrecord. Proceedings of the National Academy of Sciences 101,

10360–10362.



Oyama MD, Nobre CA (2003) A new climate-vegetation equilibrium

state for Tropical South America. Geophysical Research Letters 30,

2199–2202.Panshin AJ, de Zeeuw C. (1980) Textbook of Wood Technology.

McGraw-Hill, New York.

Pataki DE, Ehleringer J, Flanagan LB, Yakir D, Bowling DR, Still CJ,Buchmann N, Kaplan JO, Berry JA (2003) The application and

interpretation of Keeling plots in terrestrial carbon cycle research.

Global Biogeochemical Cycles 17, 1–14.

Peter G, Neale D (2004) Molecular basis for the evolution of xylemlignification. Current Opinions in Plant Biology 7, 737–742.

Phillips TL, DiMichele W (1992) Comparative ecology and life-his-

tory biology of arborescent lycopsids in Late Carboniferous swamps

of Euramerica. Annals of the Missouri Botanical Garden 79, 560–588.

Pickard WF (1981) The ascent of sap in plants. Progress in Biophysicsand Molecular Biology 37, 181–229.

Piquemal J, Lapierre C, Myton K, O’Connel A, Schuch W, Grima-Pettenati J, Boudet AM (1998) Down-regulation of cinnamoyl-

CoA reductase induces significant changes of lignin profiles in trans-

genic tobacco plants. The Plant Journal 13, 71–83.Pittermann J, Sperry JS (2003) Tracheid diameter is the key trait

determining the extent of freezing-induced embolism in conifers.

Tree Physiology 23, 907–914.

Pittermann J, Sperry JS (2005) Analysis of freeze-thaw embolism inconifers: the interaction between cavitation pressure and tracheid

size. Plant Physiology 140, 374–382.

Pittermann J, Sperry JS, Hacke UG, Wheeler JK, Sikkema EH (2005)

The torus-margo pit valve makes conifers hydraulically competitivewith angiosperms. Science 310, 1924.

Pittermann J, Sperry JS, Hacke UG, Wheeler JK, Sikkema EH

(2006a) Mechanical reinforcement against tracheid implosion com-promises the hydraulic efficiency of conifer xylem. Plant, Cell andEnvironment 29, 1618–1628.

Pittermann J, Sperry JS, Wheeler JK, Hacke UG, Sikkema EH

(2006b) Inter-tracheid pitting and the hydraulic efficiency of coni-fer wood: the role of tracheid allometry and cavitation protection.


Pockman WT, Sperry JS, O’leary JW (1995) Sustained and significant

negative water pressure in xylem. Nature 378, 715–716.Popper ZA (2008) Evolution and diversity of green plant cell walls.

Current Opinions in Plant Biology 11, 286–292.

Pratt RB, Jacobsen AL, Ewers FW, Davis SD (2007) Relationshipsamong xylem transport, biomechanics and storage in stems and

roots of nine Rhamnaceae species of the California chaparral. NewPhytologist 174, 787–798.

Proctor MCF (1981) Diffusion resistances in bryophytes. In: Plantsand Their Atmospheric Environment (ed. Jarvis PG). Blackwell

Scientific Publications, London, pp. 219–229.

Proctor MCF, Tuba Z (2002) Poikilohydry and homoihydry: antithe-

sis or spectrum of possibilities? New Phytologist 156, 327–349.Qiu Y-L (2008) Phylogeny and evolution of charophytic algae and

land plants. Journal of Systematics and Evolution 46, 287–306.

Qiu Y-L, Li L, Wang B, many others. (2006) The deepest divergences

in land plants inferred from phylogenomic evidence. Proceedings ofthe National Academy of Sciences 103, 15511–15516.

Raes J, Rohde A, Christensen JH, Van De Peer Y, Boerjan W (2003)

Genome-wide chracterization of the lignification toolbox in Ara-bidopsis. Plant Physiology 133, 1051–1071.

Raven JA (1977) The evolution of vascular land plants in relation to

supracellular transport processes. Advances in Botanical Research5, 153–219.

Raven J (1984) Physiological correlates of the morphology of early vas-

cular plants. Botanical Journal of the Linnean Society 88, 105–126.

Raven JA (2002) Tansley review no. 131. Selection pressures on sto-

matal evolution. New Phytologist 153, 371–386.

Raven JA, Edwards D (2001) Roots: evolutionary origins andbiogeo-chemical significance. Journal of Experimental Botany 52, 381–401.

Raven JA, Edwards D (2004) Physiological evolution of lower em-

bryophytes: adaptations to the terrestrial environment. In The Evo-lution of Plant Physiology (eds Hemsley AR, Poole I). Elsevier,

London, pp. 17–40.

Raymond A, Metz C (1995) Laurussian land-plant diversity during

the Silurian and Devonian: mass extinction, sampling bias, or both?Paleobiology 21, 74–91.

Rice CM, Trewin NH, Anderson LI (2002) Geological setting of the

Early Devonian Rhynie cherts, Aberdeenshire, Scotland: an early

terrestrial hot spring system. Journal of the Geological Society 159,203–214.

Roberts K, McCann MC (2000) Xylogenesis: the birth of a corpse.

Plant Biology 3, 517–522.

Roth-Nebelsick A (2005) Reconstructing atmospheric carbon dioxidewith stomata: possibilities and limitations of a botanical pCO2

sensor. Trees 19, 251–265.

Roth-Nebelsick A (2007) Computer-based studies of diffusionthrough stomata of different architecture. Annals of Botany 23,

23–32.

Roth-Nebelsick A, Konrad W (2003) Assimilation and transpiration

capabilities of rhyniopytic plants from the Lower Devonian andtheir implications for paleoatmospheric CO2 concentration. Palae-ogeography, Palaeoclimatology, Palaeoecology 202, 153–178.

Rothwell GW, Lev-Yadun S (2005) Evidence of polar auxin flow in

375 million-year-old fossil wood. American Journal of Botany 92,903–906.

Rothwell GW, Sanders H, Wyatt SE, Lev-Yadun S (2008) A fossil

record for growth regulation: the role of auxin in wood evolution.Annals of the Missouri Botanical Garden 95, 121–134.

Rowe NP, Speck T (2004) Hydraulics and mechanics of plants: nov-

elty, innovation and evolution. In The Evolution of Plant Physiology(eds Hemsley AR, Poole I). Elsevier, London, pp. 297–325.

Rowe NP, Speck T (2005) Plant growth forms: an ecological and

evolutionary perspective. New Phytologist 166, 61–72.

Royer DL, Wing SL, Beerling DJ, Jolley DW, Koch PL, Hickey LJ,

Berner RA (2001) Paleobotanical evidence for near present-daylevels of atmospheric CO2 during part of the Tertiary. Science 292,

2310–2313.

Royer DL, McElwain JC, Adams JM, Wilf P (2008) Sensitivity of leafsize and shape to climate within Acer rubrum and Quercus kelloggii.New Phytologist 179, 808–817.

Retallack GJ (1997) Early forest soils and their role in Devonian glo-

bal change. Science 276, 583–585.Retallack GJ (2001) A 300-million-year record of atmospheric carbon

dioxide from fossil plant cuticles. Nature 411, 287–290.

Rudall P (1995) New records of secondary thickening in monocotyle-

dons. Journal of the International Association of Wood Anatomists16, 261–268.

Sachse D, Radke J, Gleixner G (2004) Hydrogen isotope ratios of

recent lacustrine sedimentary n-alkanes record modern climate vari-

ability. Geochimica et Cosmochimica Acta 68, 4877–4889.Sack L, Frole K (2006) Leaf structural diversity is related to hydraulic

capacity in tropical rain forest trees. Ecology 87, 483–491.

Sack L, Cowan PD, Jaikumar N, Holbrook NM (2003) The‘hydrology’ of leaves: coordination of structure and function in

temperate woody species. Plant, Cell and Environment 26,

1343–1356.

Saliendra NZ, Sperry JS, Comstock JP (1995) Influence of leaf waterstatus on stomatal response to humidity, hydraulic conductance,

and soil drought in Betula occidentalis. Planta 196, 357–366.

138 J . PITTERMANN


Santiago LS, Goldstein G, Meinzer FC, Fisher JB, Machado K,

Woodruff D, Jones T (2004) Leaf photosynthetic traits scale with

hydraulic conductivity and wood density in Panamanian forest can-opy trees. Oecologia 140, 543–550.

Schreiber L, Riederer M (1996) Ecophysiology of cuticular

transpiration: comparative investigation of cuticular waterpermeability of plant species from different habitats. Oecologia 107,

426–432.

Schulte PJ, Castle AL (1993) Water flow through vessel perforation

plates – a fluid mechanical approach. Journal of ExperimentalBotany 44, 1135–1142.

Sieburth LE, Deyholos MK (2006) Vascular development: the long

and winding road. Current Opinions in Plant Biology 9, 48–54.

Singh G. (2004) Plant Systematics. Science Publishers, Inc., Plym-outh, UK.

Smith FA, Freeman KH (2006) Influence of physiology and climate

on D of leaf wax n-alkanes from C3 and C4 grasses. Geochimica etCosmochimica Acta 70, 1172–1187.

Somerville CR, Bauer S, Brininstool G, Facette M, Hamann T, Milne

J, Osborne E, Paredez A, Persson S, Raab T, Vorwerk S, Youngs H

(2004) Toward a systems approach to understanding plant cellwalls. Science 306, 2206–2211.

Sperry JS (2000) Hydraulic constraints on plant gas exchange. Agri-culture and Forest Meteorology 2831, 1–11.

Sperry JS (2003) Evolution of water transport and xylem structure.International Journal of Plant Sciences 164, S115–S127.

Sperry JS, Hacke UG (2004) Analysis of circular bordered pit func-

tion. I. Angiosperm vessels with homogenous pit membranes.

American Journal of Botany 91, 369–385.Sperry JS, Sullivan JEM (1992) Xylem embolism in response to

freeze-thaw cycles and water stress in ring-porous, diffuse-porous

and conifer species. Plant Physiology 100, 605–613.Sperry JS, Adler FR, Campbell GS, Comstock JP (1998) Limitation of

plant water use by rhizosphere and xylem conductance: results from

a model. Plant Cell and Environment 21, 347–359.

Sperry JS, Hacke UG, Pittermann J (2006) Size and function in coni-fer tracheids and angiosperm vessels. American Journal of Botany93, 1490–1500.

Stein W (1993) Modeling the evolution of stelar architecture in vascu-

lar plants. International Journal of Plant Sciences 154, 229–263.Stewart WN, Rothwell G (1993) Paleobotany and the Evolution of

Plants. Cambridge University Press, Cambridge, UK.

Taiz L, Zeiger E (2006) Plant Physiology, 4th edn. Sinauer Associates,Sunderland, MA.

Tarver JE, Braddy SJ, Benton MJ (2007) The effects of sampling bioas

on palaeozoic faunas and implications for macroevolutionary stud-

ies. Palaeontology 50, 177–184.Taylor TN, Taylor EL, Krings M (2009) Paleobotany: The Biology and

Evolution of Fossil Plants. Academic Press, Oxford, UK.

Thomas BA, Watson J (1976) A rediscovered 114-foot Lepidodendronfrom Bolton, Lancashire. Geological Journal 11, 15–20.

Tomlinson PB (2006) The uniqueness of palms. Botanical Journal ofthe Linnean Society 151, 5–14.

Trewin NH, Fayers SR, Kelman R (2003) Subaqueous silicification of

the contents of small ponds in an Early Devonian hot-stpring com-plex, Rhynie, Scotland. Canadian Journal of Earth Sciences 40,

1697–1712.

Trotter JA, Wiliams IS, Barnes CR, Lecuyer C, Nicoll RS (2008) Didcooling oceans trigger Ordovician biodiversification? Evidence

from conodont thermometry. Science 321, 550–554.

Turmel M, Pombert J-F, Charlebois P, Otis C, Lemieux C

(2007) The green algal ancestry of land plants as revealed bythe chloroplast genome. International Journal of Plant Sciences168, 679–689.

Turner SR, Somerville CR (1997) Collapsed xylem phenotype of

Arabidopsis identifies mutants deficient in cellulose

deposition in the secondary cell wall. The Plant Cell 9, 689–701.

Turner S, Gallois P, Brown D (2007) Tracheary element differentia-

tion. Annual Review of Plant Biology 58, 407–433.Tuskan GA, DiFazio S, many others. (2006) The genome of black

cottonwood, Populus trichocarpa (Torr. & Gray). Science 313,

1596–1604.

Tyree MT, Sperry JS (1988) Do woody plants operate near thepoint of catastrophic xylem dysfunction caused by dynamic

water stress? Answers from a model. Plant Physiology 88, 574–

580.

Tyree MT, Sperry JS (1989) Vulnerability of xylem to cavitation andembolism. Annual Review of Plant Physiology and Plant MolecularBiology 40, 19–38.

Tyree MT, Zimmermann MH (2002) Xylem Structure and the Ascentof Sap. Springer, Berlin.

Tyree M, Davis S, Cochard H (1994) Biophysical perspectives of

xylem evolution - Is there a tradeoff of hydraulic efficiency for vul-

nerability to dysfunction? Journal of the International Associationof Wood Anatomists 15, 335–360.

VanAller Hernick L, Landing E, Bartowski KE (2008) Earth’s oldest

liverworts – Metzeriothalus sharonae sp. nov. from the Middle

Devonian (Givetian) of eastern New York, USA. Review of Palae-obotany and Palynology 148, 154–162.

Wang Z-Y, He J-X (2004) Brassinosteroid signal transduction –

choices of signals and receptors. Trends in Plant Science 9,

91–96.Ward JK, Tissue DT, Thomas RB, Strain BR (1999) Comparative

responses of model C3 and C4 plants to drought in low and ele-

vated CO2. Global Change Biology 5, 857–867.Wellman CH, Kerp H, Hass H (2004) Spores of the Rhynie chert

plant Horneophyton lignieri (Kidston & Lang) Barghoorn &

Darrah, 1938. Transactions of the Royal Society of Edinburgh: EarthSciences 94, 429–443.

Wellman CH, Osterloff PL, Mohluddin U (2003) Fragments of the

earliest land plants. Nature 425, 282–285.

Wheeler JK, Sperry JS, Hacke UG, Hoang N (2005) Inter-vessel

pitting and cavitation in woody Rosaceae and other vesselled plants:a basis for a safety versus efficiency trade-off in xylem transport.

Plant, Cell and Environment 28, 800–812.

Willis KJ, McElwain JC (2002) The Evolution of Plants. Oxford Uni-versity Press, Oxford.

Wilson JP, Knoll AH, Holbrook NM, Marshall CR (2008) Modeling

fluid flow in Medullosa, an anatomically unusual Carboniferous seed

plant. Paleobiology 34, 472–493.Woodward FI (1987) Stomatal numbers are sensitive to increases in

CO2 from pre-industrial levels. Nature 327, 617–618.

Yancey PH, Clark ME, Hand SC, Bowlus RD, Somero GN (1982)

Living with water stress: evolution of osmolyte systems. Science217, 1214–1222.

Yoon HS, Hackett C, Ciniglia C, Pinto G, Bhattacharya D (2004) A

molecular timeline for the origin of photosynthetic eukaryotes.

Molecular Biology and Evolution 21, 809–818.Zachos J, Pagani M, Sloan L, Thomas E, Billups K (2001) Trends,

rhythms, and aberrations in global climate 65 Ma to present. Science292, 686–693.

Ziegler H (1987) The evolution of stomata. In Stomatal Function(eds Zeiger E, Farquhar G, Cowan IR). Stanford University Press,

Palo Alto, pp. 29–57.

Zwieniecki M, Melcher P, Holbrook N (2001) Hydaulic properties ofindividual xylem vessels in Fraxinus americana. Journal ofExperimental Botany 52, 257–264.



Documents

Theevolutionofwatertransportinplants:anintegrated approach · Plants can control climate as well as respond to it. Recent models supported by independent isotopic evidence suggest