Planta (Ber].) 90, 303--322 (1970)
The Role of the Mesophyll Cell Wall in Leaf Transpiration P. G. J A n w s a n d 1%. O. SLAT:Z~n Research School of Biological Sciences, Australian National University, Canberra Received October 28, 1969
Summary. Evidence is presented which suggests that the mesophyll cell walls of cotton leaves may influence observed rates of transpiration. The net diffusive flux of water vapour, from the upper and lower surfaces of a leaf, was compared with the flux of nitrous oxide through a leaf and evidence obtained of an extra resistance in the water-vapour pathway associated with water transport in the mesophyll cell walls. This extra resistance appeared to be insignificant at low transpiration rates and in turgid leaves, but increased with transpiration rate and dehydration. The most likely explanation for its origin appeared to be a reduction in hydraulic conductivity across the internal cuticle which lines the outer surfaces of the mesophyll cell walls. In turn this served to reduce the relative vapour pressure at the sites of evaporation. The experiments were conducted under conditions where stomatM opening was induced by COe-free air. Under normal conditions stomatM closure would tend to reduce the development of this extra resistance. Even so, the results throw doubt on the validity of the long-standing assumption that the water-vapour pressure at the evaporation sites is equM to the saturation vapour pressure under all conditions. Introduetion T r a n s p i r a t i o n from p l a n t s involves the e v a p o r a t i o n of water from sites w i t h i n the leaves a n d the s u b s e q u e n t diffusion of water v a p o u r to the n a t u r a l leaf surface a n d t h e n into the air beyond. Two m a i n p a t h w a y s for water m o v e m e n t are reeognised, one being associated with m o v e m e n t directly across the leaf cuticle, the other associated with m o v e m e n t t h r o u g h the stomatM pores. The cuticular p a t h w a y is relatively short, b u t of high resistance and, when the s t o m a t a are open, carries o n l y a small proportion of the t o t a l t r a n s p i r a t i o n flux. The s t o m a t a l p a t h w a y involves e v a p o r a t i o n of water from the outer surfaces of the mesophyll cells, a n d its diffusion t h r o u g h the intercellular spaces a n d the s t o m a t a l pores. A l t h o u g h m u c h longer, this route n o r m a l l y carries most of the t r a n s p i r a t i o n , b u t the presence of the s t o m a t a introduces a powerful, variable resistance into the p a t h w a y and, when the s t o m a t a are closed, s t o m a t a l t r a n s p i r a t i o n can effectively cease. 21 t?lan~a(Berl.), Bd. 90
304
P.G. Jarvis and R. O. Slatyer:
I t is generally accepted that variation in stomatal aperture is the main mechanism by which the plant exercises control over transpiration. From time to time, however, considerable controversy has arisen as to the degree to which other mechanisms may be of significance and the issue is still an active one (see, for example, Livingston and Brown, ]912; Knight, 1917; Klemm, 1956; Milthorpe and Spencer, 1957; Slavik, 1958; Shimshi, 1963; Rawlins, 1963; Slatyer, 1966, 1967; Fischer, 1968, Gale et al., 1967). Two main mechanisms have been proposed to account for nonstomatal transpiration control, should it occur. One envisages that the evaporation sites may retreat from the outer surfaces of the mesophyll cells, from which most evaporation occurs, into the walls themselves, thereby increasing the length and tortuosity of the vapour pathway, and hence the total diffusive resistance. This has been referred to as "incipient drying" (Livingston and Brown, 1912). The other mechanism proposes that the effective vapour pressure at the liquid-air interfaces may be reduced significantly below that of pure free water, thereby reducing the leaf-air vapour concentration gradient which provides the driving force for water-vapour transport. It is suggested that this may occur as a direct result of the dehydration associated with water stress (Shimshi, 1963; Rawlins, 1963) or by accumulation of solutes carried to the evaporating surfaces in the transpiration stream (Boon-Long, 1951). I t could also arise from the presence of the internal cuticle which lines the mesophyll cell walls (Scott, 1950, 1966). Should either, or both, of these mechanisms operate, they could be expected to be of greatest significance during periods of limiting water supply or excessive evaporative demand. Recently, Slatyer (1966) has argued from a theoretical and physical viewpoint that non-stomatal factors are not likely to be of significance in normal transpiration, and Fischer (1968) and Weatherspoon (personal communication) have obtained experimental evidence for leaves with stripped epidermes which support this view. The present paper contains a study of the problem using a novelapproach, with intact and undamaged leaves.
Theory Transpiration from the upper (adaxial) surface of a leaf can be described by the expression CW__ cU F~
~o-
E r~,o
(1)
where F 1-I20 u is the transpiration rate in units such as g c m -2 see -1, (cw -- CUa) is the water-vapour concentration difference between the sites of evaporation in the cell walls (cw) and the bulk air (CUa)in units of g cm -3
Mesophyll Cell Wall in Leaf Transpiration
305
and ~, r ~ o is the algebraic sum of the resistances encountered by the diffusing vapour (sec cm-1). Because water vapour diffuses from the leaves by both a enticular and a stomatal pathway it can be appreciated that ~ r ul i f O consists of a common external resistance r u, and two groups of internal resistances, connected in parallel, which comprise the resistance to transfer through the cuticle (reu) and that through the stomatal p a t h w a y (r~'). The latter symbol refers to specific resistances connected in series, which are concentrated in the cell wall (rw) and in the zone between the cell-wall surface and the external surface of the stomatM pore (r u) including the stomataJ pore itself. These resistances can be related by the expression F,r~o=r~+
f
r~r~c~ ~__
u
..~-T,-~ - r ~ +
(% +re J
~ (ru + rw) ru
~ ..~;-,_ ~ - u
(r8 + r w + r e ) J
.
(2)
A similar expression describes the resistances encountered by the net efflux of water vapour from the lower (abaxial) surface of the leaf, r Izt ~ O For each flux, r w represents a common resistance. An expression can now be derived for r~, from Eq. (2) by rearranging terms. Firstly, "
~
U
a
so t h a t
(3) rw =
re (2 r~00 - - r D reu -- (r, r u~ o - - rD
u
--r~.
The corresponding expression for the lower surface is: 1
rw=
t
1
f re (E r~,o -- r~) I i --,..-7-~; [ re -- (E rtI~O -- ra)
t
--rs.
(4)
The object of the experiments described in this paper is to determine the value of r w under various conditions. A convenient way of doing this is to compare the resistances to water vapour diffusion with those to some other gas, where the resistance r w does not arise, but where the pathway is in other respects identical, or differs in a manner that can be quantified. For this purpose N20 was used. I t has the advantage of being relatively inert physiologica!ly, at the concentrations used (Slatyer and Jarvis, 1966), and its concentration can be accurately measured with an infra-red gas anMyser. The net diffusive flux of N~O from one side of the leaf to the other can be described by the expression F ~ .-o 21"
~E- ~r s ~ o "
(5)
306
P.G. Jarvis and R. O. Slatyer:
The diffusive resistances to N~O flux can be expressed in terms of the diffusive resistances to water v a p o u r b y m u l t i p l y i n g t h e m b y the ratio of the diffusion coefficients of N20 a n d water v a p o u r i n air [D~c..o/Ds2 o = 0.65 (Slatyer a n d Jarvis, 1966)]. Hence D~O
Z rN20 -- f)~r
l
l
u
u
(ra d- rs d- ri ~- rs d- ra).
(6)
The new symbol ri refers to a n y a d d i t i o n a l resistance, in the intercellular air spaces, e n c o u n t e r e d b y the diffusing 5120 molecules over a n d above t h a t e n c o u n t e r e d b y the water v a p o u r molecules. R e a r r a n g i n g terms i n Eq. (6) therefore gives r s + r ls = 0.65 ~ r ~ 0 - ~za - ~ - r~.
(7)
I n t r o d u c i n g Eq. (7) a n d adding Eq. (3) a n d (4) the following expression can be o b t a i n e d for rw:
(s) - {o.65 ~ ~,~o - d - r :
--r~} l"
All terms i n Eq. (8), except for r w a n d r~, can be i n d e p e n d e n t l y determined. Therefore the m a i n difficulty i n d e t e r m i n i n g r~v lies i n providing a n estimate of r~ u n d e r the various conditions of interest.
Materials and Methods The theory requires tile simultaneous measurement, under varying conditions of stomatal aperture and transpiration rate, of the fluxes of water vapour from each side of a leaf, the flux of N~O through the leaf, the ambient water vapour and NzO concentrations on each side of the leaf, and the leaf temperature. A continuous and simultaneous measure of leaf water potential is also highly desirable. To satisfy these requirements, a leaf chamber of novel design was constructed and the nitrous oxide diffusion porometer developed. Full details are supplied by Slatyer and Jarvis (1966) and Jarvis and Slatyer (1966a, 1966b). Cotton plants were chosen for the experiments, partly because of the large size of the amphistomatous leaves, but also because of the background of knowledge about water-vapour exchange with this material (e.g. Slatyer and Bierhuizen, 1964). The cotton plants (Gossypium hirsutum L. cv. Pope) were grown in solution culture in a modified Itoagland's solution in a growth chamber with illumination from fluorescent tubes (Sylvania, Cool White VHO) supplemented with tungsten globes giving a radiation flux density of ca. 120 W m -2 (400--700 am) for 12 hr each day. The photoperiod was extended to 14 hr with the tungsten lamps alone. Air temperature was 35~ during the photosynthetic period and 24~ in the dark. Relative humidity was ca. 65%. In the experiments the fifth emerged leaf from the apex was used when there were about 10 leaves on the plant and before flowering commenced.
Mesophyll Cell Wall in Leaf Transpiration
307
The leaf chamber was designed so t h a t measurements were made on opposed areas of a leaf (ca. 31 cm 2) which remained attached to the plant. Because of the large area of leaf in the chamber in relation to the perimeter, the results obtained should be virtually independent of chamber dimensions. The direction of flux of NsO , usually from the upper to the lower surface of the leaf, did not affect the result, although there were about twice as m a n y stomata in the lower surface, confirming t h a t the flux through the leaf was essentially one-dimensional and t h a t possible errors arising from lateral flow were insignificant. I n a typical experiment, the attached leaf was placed in the chamber on the evening prior to the experiment and the plant was left overnight in darkness with normal air passing through the chambers. I n the morning, the stomata were opened with light in steps of 50 and 100 W m -2 of a b o u t 0.5 hr each before 150 W m -2 and C O J r e e air were supplied. Experimental treatments were imposed when steady state conditions of stomatal resistance, leaf temperature and leaf water content were obtained. During this warm-up period, the leaf-air vapour-pressure difference was 5 - - 1 0 m m Hg and the leaf temperature 28 ~ Experimental treatments usually consisted of varying the leaf-air vapourpressure difference by changing the h u m i d i t y of air entering the leaf chambers; in some experiments, leaf temperature was also varied and leaf temperatures of 33--35 ~ were used for the highest transpiration rates. Steady-state readings were usually obtained a b o u t an hour after a change in conditions, after which the conditions were again changed. Initially, ambient vapour pressure was changed in step functions from high to low, and vice versa, b u t because of the tendency for stomatal closure in the late afternoon, the treatments were later randomized. The same leaf was never used in more t h a n one experimental r u n or more t h a n 2 days in the chamber. A t the end of the experiment the leaf was sampled for calibration of the/?-gauge according to the procedure developed b y Jarvis and Slatyer (1966). t~elative water content and water potential of the leaves were determined by the methods of Weatherley (1950) and Slatyer (1958), as modified b y Jarvis and Jarvis (1963). IV[ore extreme water stress was induced in certain treatments b y exchanging the n u t r i e n t solution around the roots (water potential a b o u t - - 0 . 5 bars) for a similar n u t r i e n t solution containing either NaC1, mannitol or polyethylene glycol (Carbowax 400) to lower the water potential of the solution to values down to 20 bars. The solution was exchanged while the plant was in the dark, either just prior to illumination on the day oi the experiment, or on the previous evening. From the theory, it can be seen t h a t the most sensitive estimates of r i and r w will be obtained when the other resistances in series with them, notably r s and r a , are as small as possible. The boundary layer resistance, r a , was kept at a low and constant level by the action of the small fans stirring the air in the chambers above a n d below the leaf, and b y the flow of air through the chambers (ca. 40 or ca. 70 ltr h 2) (Jarvis and Slatyer, 1966b). There was no problem in keeping the stomatal resistance, r~, low at small leaf-air vapour-pressure differences, b u t the stomata showed a tendency to close a t vapour-pressure differences of a b o u t 20 m m t t g a n d over, especially when the water potential of the culture solution was lowered by the addition of solutes. There also appeared to be a tendency to close endogenously in the mid-afternoon after a b o u t 8 h r illumination. To keep the stomata open throughout an experiment, the observations were made a t a radiation flux density (400--700 nm) of 150 W m -s and with COs-free air supplied to the leaf chambers. This did not prevent some -
-
308
1). G. Jarvis and R. O. Slatyer:
increase in stomatal resistance at the greater vapour-pressure differences used, especially if these treatments occurred in the late afternoon or evening. I n a n u m b e r of experiments, steady state conditions were never obtained, the stomatal resistances, leaf water content and leaf temperature cycling in a regular manner even though CO~-free air was supplied. I n these cases, all the experimental data were transferred from the chart record to punched tape using an analogue to digital converter (Chapman and Goodspeed, 1967) and the i m p o r t a n t parameters calculated at 30- or 60-sec intervals b y digital computer, There are certain inherent disadvantages in using non-steady-state data. These difficulties are most apparent when the measured parameters are changing rapidly, when the rate of change is of a similar general order to the response characteristics of the measuring instruments, and when the measuring instruments themselves have different response times. Hence these data were used only to support the steady state observations.
Results 1. D e t e r m i n a t i o n
o/Constants
r ua a n d r a. ~ To determine the boundary layer resistance to water-vapour flux, r~, two thicknesses of wet blotting paper, to represent a fl'ee water surface, were placed in the leaf chamber in the leaf position with three thermocouples sandwiched in between. Values of r~ were calculated for each side from the slope of the relationship between the surface-air vapour concentration difference (c s - - Ca) and the steady state evaporation rate, FH~o (g cm -~ sec-1), according to the equation: a)
The
Boundary-Layer
Resistances
ra = [ FH~O /
(9)
(cs was assumed to be equivalent to the saturation vapour pressure at the temperature of the wet paper). The value of r~ was found to be 0.42 sec cm -1 for both the upper and lower surfaces of the blotting paper model when placed in the leaf chamber in the leaf position. This figure was adopted for both surfaces of the cotton leaf, the thorough air circulation being assumed to minimise any differences arising from the rougher lower surfaces of actual leaves. t The cuticular resistance to b ) T h e C u t i c u l a r R e s i s t a n c e s , r cu a n d r c. water transfer from each surface of several leaves was determined from regressions between cuticular transpiration and leaf-air vapour concentration difference in the range (cw - - Ca) = 15 -- 35 • l0 -6 g cm -3 (approximately equivalent to 15--35 mm Hg). The values obtained for r cl were found to be 70.6 & 5.6 sec cm -1, and for r~ to be 191 • sec cm -1. At high resistances the variability is comparatively large because of the difficulties in measuring the very small fluxes with sufficient precision. However, even large variations in r c have little effect on the calculation of r w or r i if r c is large in relation
Mesophyll Cell Wall in Leaf Transpiration
309
to }~ r~, o , a r e q u i r e m e n t t h a t is a l w a y s m e t while the s t o m a t a are open. No influence of (c~,- Ca) on r e was a p p a r e n t . S u m m i n g r~ a n d r~c in p a r a l l e l gives a n average r~ for the leaf of 51 sec cm 1. This is s o m e w h a t larger t h a n t h e figure of 32 sec cm -i given b y S l a t y e r a n d Bierhuizen (1964) for the same v a r i e t y , p o s s i b l y because of t h e different conditions in which t h e p l a n t s were c u l t i v a t e d , a n d p o s s i b l y because of t h e g r e a t e r p r e c a u t i o n s t a k e n to ensure s t o m a t a l closure in these e x p e r i m e n t s . H o w e v e r , even with t h e N~O p o r o m e t e r , it is n o t possible to ensure t h a t complete closure has occurred simult a n e o u s l y in both surfaces a n d some of t h e v a r i a b i l i t y in r e m a y r e s u l t from i n c o m p l e t e s t o m a t a l closure in one leaf surface.
2. Determination o/r~ I t seems p r o b a b l e t h a t t h e r e is a n e x t r a effective p a t h l e n g t h for N 2 0 diffusion t h r o u g h the leaf over a n d a b o v e the sum of t h e intercellular resistances for w a t e r v a p o u r diffusion to each leaf surface; t h a t is, t h a t r i is finite a n d positive. F r o m Eq. (8) it can be seen t h a t t h e lower the value of ri the lower will be the c a l c u l a t e d value of r w. M i n i m u m values for rs are therefore o b t a i n e d on the a s s u m p t i o n t h a t r i : 0. While these values are of i n t e r e s t in setting lower limits for rw, i t is clearly desirable to m a k e realistic e s t i m a t e s of r i so t h a t t h e values assigned to r~s can also be as realistic as possible. F o r this p u r p o s e two procedures were used. The first, a n a n a t o m i c a l m e t h o d , is b a s e d on calculations b y J a r v i s et al. (1966) which g i , e a value for r* - - t h e t o t a l intercellular space resistance b e t w e e n the inner surfaces of t h e s u b - s t o m a t a l cavities on each side of the leaf. F r o m these calculations r* = ~ ~ ;~; D~o
(lO)
where hi, ai a n d n~ are, respectively, the thickness of the mesophyll, the a v e r a g e r a d i u s of t h e i n t e r c e l l u l a r spaces, a n d the n u m b e r of air passages p e r u n i t cross section of leaf p l a n area. A n a t o m i c a l i n f o r m a t i o n a b o u t the leaves r e q u i r e d for the determ i n a t i o n of r~ was p r o v i d e d b y Dr. J. E. Begg (personal communication). There were a b o u t twice as m a n y s t o m a t a in t h e lower surface as in t h e u p p e r surface. Several series of p a r a d e r m a l sections were o b t a i n e d using a freezing m i c r o t o m e a n d 60 cellular a n d inter-cellular i n t e r c e p t s were m e a s u r e d on a set of t r a n s e c t s in different directions across these sections. The a r i t h m e t i c m e a n d i m e n s i o n of the intercellular spaces o b t a i n e d was 1.8 =L0.3 ~. Since t h e t r a n s e c t s were in m a n y different directions, this m e a n dimension was t a k e n as t h e m e a n d i a m e t e r of a cylindrical air passage of l e n g t h e q u a l to t h e thickness of the mesophyll. Because t h e
310
P.G. Jarvis and R. O. Slatyer:
resistance is inversely proportional to the square of the radius [Eq. (10)], and the intercellular spaces cover a range of sizes from 0 to 7.2 ~, the mean of a i~' ( ~ 1.73 ~ ) was used in preference to the square of the arithmetic m e a n of a i which would overestimate the resistance. F r o m the transect data, the n u m b e r of the air passages in the mesophyll per unit leaf plan area (n~) was calculated to be 7.15 • l04 cm -2 (Jarvis et al., 1966). The thickness of the mesophyll was measured in transverse leaf sections and found to be 200 ~= 10 ~. Since the subs t o m a t a ! cavities are included in r~., the p a t h length for diffusion t h r o u g h the intercellular spaces was taken to be of the order of 170 ~. F r o m these data, the estimate of r* from Eq. (10) was found to be 1.65 sec cm -1. This estimate can be expected to exceed the true value of r i unless all the intercellular space resistance to water vapour diffusion is in the sub-stomatal cavities; however, it should provide a reasonable upper figure for r~ and, hence, a basis for a minimal estimate of r w under normal turgid conditions. During stress, leaf water content and leaf volume decline (Kennedy and Booth, 1958). Assuming t h a t shrinkage is uniformly three-dimensional, the reduction in the cross-sectional area of the air passages (a~) should be balanced b y a proportional increase in ni, and h i should be expected to decrease to a value proportional to the cube root of the stressed relative water content. A reduction to a relative water content of 50 % (a figure likely to cause death of cotton leaves) can therefore be expected to reduce r i to 0.8 of its turgid value. A change of this order would n o t affect r w to a noticeable degree. The second, physiological, m e t h o d of estimating r i is based on the assumption that, in turgid plants at low leaf-Mr v a p o u r concentrations and tow transpiration rates, r w ~ O. Under these conditions, Eq. (8) can be used to provide estimates of r i. 20 determinations of r i were therefore made on occasions when (cw - - C a ) ~ 5 m m Hg. These measurements yielded an average value for r~ of 1.79 • 0.25 see cm -~. Although there was a high degree of variability, as the s t a n d a r d error indicates, caused b y the difficulty of measuring low values of FH..O with precision, the error has a relatively small effect on the estimates of rw. I t is also of interest t h a t the values of r i determined in this fashion, agree closely with those obtained from the anatomical method. For calculation purposes, the physiological estimates were used, rounded off to 1.8 sec cm -1. 3. D e t e r m i n a t i o n o / r w
Altogether, over 100 experiments, each leading to determinations of r~, were made under a range of conditions. The following paragraphs
Mesophyll Cell Wall in Leaf Transpiration
311
12 11
FH2o
9 :g
m
8
u. 4 3. 2. 1 0
•
5
10
15
20
25
( C w - C a ) x 10 -~ grn crn -3
Fig. 1. Data from one experiment showing the relation between transpiration from a whole cotton leaf (FIt2O) and from the upper (F~I~O) and lower (F~LO) surfaces, and leaf-air vapour concentration difference (cw -- ca)
describe the effects of rate of transpiration, leaf water content, and the duration of high rates of transpiration on r w .
a) E//ect o/Transpiration Rate. Transpiration d a t a from a typical experiment, conducted under conditions of minimM substrate water stress, are depicted in Fig. 1. The three curves show the p a t t e r n of change in transpiration from the leaf as a whole, and from each surface, as leaf-air v a p o u r concentration difference (zJ c) was increased. I t is a p p a r e n t t h a t there is a progressive divergence from a straight-line relationship as transpiration increases until, at high d c values, a drop in transpiration accompanies a further increase in z] c. D a t a of this type have been reported on a n u m b e r of occasions in the past, indicating d e a r l y t h a t there is a progressive increase in the total diffusive resistance as transpiration rate, a n d A c, increase. I t has frequently been assumed t h a t this departure from linearity is due entirely to progressive stomatal closure; on other occasions it has been assumed t h a t no closure has occurred and t h a t the phenomenon provides evidence of non-stomatal control of transpiration (i.e. an increase in rw).
312
P.G. Jarvis and R. O. Slatyer:
8
~
%0
6u
5-
rw
2~
0
o
~
~ FxIO -6gm
1"s
c m -2 sec -s
Fig. 2. Relationship between the stomatal resistance from the upper (ru) and the lower (rs~)leaf surfaces, cell wall resistance, (rw), and total (upper plus lower surface) transpiration rate. Points shown as circles are from data in Fig. 1, points shown in squares and crosses are from additional experiments I n Fig. 2 the main variable components of the total diffusive resistance, which are associated with this transpiration pattern, are depicted. The d a t a are drawn from the data of Fig. l, plus data from other experiments in which higher and lower transpiration rates occurred. Although these points represent only a small fraction of the data available, t h e y are fully representative. Only r s and rw are plotted since the increase in the total resistance m u s t be due to an increase in either, or both, these parameters; ra, r c and r i being regarded as constant. Fig. 2 brings out clear evidence for the existence of r w and suggests t h a t it is dependent, to some degree, on transpiration rate. The intercept of the rw/F~ o curve appears to pass through, or close to, the origin. I t is a p p a r e n t that, although the values of r w are low, relative to comparable values of rsu or riB, t h e y are high enough to be significant components of the total diffusive resistance, at least at the highest transpiration rates obtained. b) E//ect o / L e a / Water Content. W a t e r deficits were imposed b y the addition o~ osmotic substrates [mannitol and polyethylene glycol 400 (carbowax)] which reduced the substrate water potential to levels down to - - 2 0 bars, and reduced relative leaf water content (O) to values down to about 50 %. Although a n u m b e r ol experiments of this type was conducted, considerable difficulty was experienced in maintaining the s t o m a t a
Mesophyll Cell Wall in Leaf Transpiration
313
,
100
90
80
70
60
50
Fig. 3. Relationships between r w and relative water content (O). Each curve represent a different experiment open and, as r 8 increased, the rehability of the estimate of % decreased. I n Fig. 3 data are presented from three typical experiments, covering a range of relative water contents from turgid to severe stress. Lowmoderate transpiration conditions (Fs~o < 10 x 10 _6 g c m -2 see -1) were used to minimise the development of steep water-potential gradients within the leaf and avoid confounding the effects of transpiration rate with direct effects of dehydration. The figure shows that % increases in a fairly linear fashion over the range of values of 0 that it was possible to obtain. Again, the values % appear to be relatively low, although, at the dry end where they exceed 4, it is apparent that % would be a significant factor in determining leaf transpb'ation if the stomata remained open. If the stomata closed, even partially, to provide ~, r ~ o values of 20 see cm 1 or so, the significance of r w would be much reduced. e) D u r a t i o n o/ H i g h Rates o/ Transpiration. Another possible way in which an apparent increase in rw could occur would be through the accumulation of solutes, carried forward in the transpiration stream, at the sites of evaporation. Should this effect develop, it would be most pronounced under conditions of prolonged transpiration at a high rate. Consequently, transpiration and rw were measured under these conditions over a period of 5 hr. Because the initial accumulation of solutes could be expected to be rapid, a high illumination (200 W m -2) dry air (A c = 2 0 mm Hg) situation was suddenly imposed on a situation of moderate light and A c. The results showed that steady state transpiration was reached within 10 rain and, although some tendency towards cycling of stomatal resistance was observed, rw showed no significant increase over a period of 5 hr.
314
P.G. Jarvis and g. O. Slatyer:
A c c u m u l a t i o n of solutes should also be e n h a n c e d b y t h e a p p l i c a t i o n of r a p i d l y p e r m e a t i n g solutes to the r o o t m e d i u m . H o w e v e r , t h e add i t i o n of NaC1 a t c o n c e n t r a t i o n s of up to 4 b a r s osmotic pressure in six s e p a r a t e e x p e r i m e n t s , h a d no d e t e c t a b l e effect on r w.
Discussion The two most significant results of t h e p r e s e n t s t u d y arc, firstly, t h a t positive values for r w were o b t a i n e d and, secondly, t h a t t h e r w increased w i t h increasing t r a n s p i r a t i o n r a t e a n d with increasing w a t e r stress. B o t h o b s e r v a t i o n s are in accord w i t h t h e recent results of F i s c h e r (1968) in leek mesophyll, b u t the values of r,~ were s u b s t a n t i a l l y g r e a t e r in t h e present s t u d y , exceeding 2 sec cm -1 whereas F i s c h e r d i d n o t observe values greater t h a n 0.5 sec cm -1 even u n d e r severe stress. The p r e s e n t values are more similar in m a g n i t u d e to those r e p o r t e d b y K l e m m (1956) a n d S h i m s h i (1963). Before discussing physiological aspects of r ~ , i t is desirable to e x a m i n e the q u a l i t y of t h e p r e s e n t observations. The q u a l i t y of the e x p e r i m e n t a l m e a s u r e m e n t s themselves a p p e a r s a d e q u a t e to enable e s t i m a t i o n of r w to b e t t e r t h a n :j= 1.0 sec cm -1 as long as t h e t o t a l series sum of the diffusive resistances does n o t exceed a b o u t 20 sec cm -1. As has been i n d i c a t e d before efforts were m a d e to minimise r a a n d rs, a n d t h e o n l y d a t a used were those in which t h e series s u m of resistances o t h e r t h a n r w was less t h a n 10secem -1. The m e a s u r e m e n t which caused m o s t concern was of leaf t e m p e r a t u r e , a n error of 0.5 ~ being sufficient to cause a n error of e s t i m a t e of a b o u t 5 % in ~ r ~ o a t t h e general level of t e m p e r a t u r e s a n d v a p o u r - c o n c c n t r a t i o n differences used. Although e x t r e m e care was t a k e n with l e a f - t e m p e r a t u r e m e a s u r e m e n t s , familiarity w i t h t h e technical p r o b l e m s i n v o l v e d leads us to t h e conclusion t h a t errors of this m a g n i t u d e m a y h a v e occurred. A p a r t from specific deficiencies in the m e a s u r e m e n t s , t h e assumptions concerning the c o n s t a n c y of r e a n d r~, a n d t h e a c t u a l value of r~ a d o p t e d , are open to challenge. I n the case of rc, I t o l m g r e n e t a l . (1965) have shown a d e p e n d e n c e of rc on leaf t e m p e r a t u r e a n d r a d i a t i o n - f l u x density. H o w e v e r , the values of r e are so large in comparison with the other resistances, a n d the r a n g e of leaf t e m p e r a t u r e s utilised was so small, t h a t v a r i a t i o n s in r c could n o t have h a d a significant effect on r w. I n t h e case of ri, a value of 1.8 sec cm -1 was chosen following the a n a t o m i c a l a n d physiological evidence described previously. I t was suggested t h a t , if the leaf s h r a n k t h r e e - d i m e n s i o n a l l y d u r i n g stress, t h e value of r i a t a relative w a t e r c o n t e n t of 50% should be 0.8 of the t u r g i d figure, leading to a slight increase in t h e e s t i m a t e d v a l u e of r w . Unp u b l i s h e d d a t a of one of us (P. G.J.) s u p p o r t the contention t h a t r e l a t i v e l y
Mesophyll Cell Wall in Leaf Transpiration
315
small changes occur in intercellular volume, at least in mature cotton leaves. Although Meidner (1955) found that resistance to bulk air flow under a pressure difference, through the mesophyll of P s y c h o t i a c a p e n s i s and S c o l o p i a m u n d i i , increased as water stress was imposed, his technique does not indicate the response pattern to the transverse diffusive flux of water vapour. Accordingly, the estimates used for the present paper do not appear unreasonable or liable to introduce serious errors to estimates of r w. Turning now to the physiological aspects of rw, it should be remembered t h a t the calculations were made so as to attribute to r,~ any apparent additional resistance in the water-vapour pathway. The decision could equally well have been made to attribute the apparent resistance to a change in Cw, or to divide the effects empirically between % and r w . I t is now important to a t t e m p t to assign it correctly. If r w is a true gas-phase diffusive resistance, the traditional suggestion is t h a t it would most probably arise as a result of a withdrawal, into the interfibrillar wall spaces, of the liquid-air interfaces which constitute the evaporation surfaces, thereby creating an extra, gaseous segment for the evaporation pathway. Should this be the correct explanation, rw would be expected to increase as water stress increases, and possibly also increase with increasing transpiration rate should water supply be inadequate to maintain transpiration rate without the development of steep local gradients of water potential. Although the data of Figs. 1, 2 and 3 provide supporting evidence for this view, these data can also support other explanations, and a consideration of the cell-wall morphology throws doubt on this simple explanation. For example, Slatyer (1966) has calculated that if the wall spaces have the characteristics of capillaries and maintain hemispherical menisci at the liquid-ah ~interfaces, they would not be expected to drain until local leaf water potentials reached extremely low levels. In fact, voids as large as 5 nm radius would not drain until water potentials fell to - - 3 0 0 bars and much lower potentials, below --1,000 bars, would be required for smaller voids, even though the degree to which capillary theory can be applied at such small values is in doubt. The cotton leaves used in this study were examined by electron microscopy, through the courtesy of Dr. D. J. Goodehild 1. No pore structure was detectable when wall sections were stained either with potassium permanganate and Reynolds lead (Reynolds, 1963) or when stained with glutaraldehyde, osmie acid and Reynolds lead. However, the preparations did indicate the presence of interfibrillar spaces of the order of 5--10 nm. I n addition, in the outer layers, a zone of about 1 C.S.I.R.O. Division of Plant Industry, Canberra.
316
P.G. Jarvis and R. O. Slatyer:
50 nm across was noticeably denser with a laminar structure and much smaller spaces. This zone corresponds to the internal cuticle of Scott (1950, 1966). From this examination it can be appreciated that extremely low water potentials would be required in order to cause a retreat of the liquid-air menisci. Such values were not approached by the general level of leaf-water potential used in this study; relative water contents of the order of 50% (as in the data of Fig. 3), being associated with leaf water potentials of the order of - - 4 0 bars. I n consequence it appears unlikely t h a t "incipient drying", in the context used in the literature, is a real phenomenon, at least with respect to cotton leaves. If the gas phase resistance component of r w is very small, or nonexistent, the implication is that there is liquid phase continuity right up to the surfaces of the mesophyll cells. I n this case, the effect attributed to r w must be assigned instead to a reduction in c w below the value equivalent to the saturation vapour pressure at the leaf temperature Such a reduction could have three obvious origins, all of which could act in combination: (1) The general level of leaf water potential (~eaf) could depress %, according to the relationship between water potential and relative vapour pressure (or relative vapour concentration), viz.
~leaf-- R_T In %/Csat,
(11)
where the new symbols R, Vw and csat refer, respectively, to the ideal gas constant, the partial molal volume of water (cm 3 mole-l), and the saturation vapour concentration (g em-a). (2) In addition to the effects of general leaf water potential described in (1), an additional reduction in % could be caused by a steep water potential gradient across the outer cell wall layers. This could arise if there is a significant source of hydraulic resistance to liquid flow across the outer, denser, layers of the cell wall (3) There could be an accumulation of osmotically active solutes transported to the liquid-air interface by the transpiration stream, which would depress the relative vapour pressure and hence %. If there is no vapour phase resistance in the cell wall (i.e. if r w = 0 ) , the m a x i m u m degree of reduction in the relative vapour pressure which occurs at the liquid-air interfaces can be calculated as follows: Eq. (1) is first re-written, for stomatal transpiration (st) only, as F ~H
The Role of the Mesophyll Cell Wall in Leaf Transpiration P. G. J A n w s a n d 1%. O. SLAT:Z~n Research School of Biological Sciences, Australian National University, Canberra Received October 28, 1969
Summary. Evidence is presented which suggests that the mesophyll cell walls of cotton leaves may influence observed rates of transpiration. The net diffusive flux of water vapour, from the upper and lower surfaces of a leaf, was compared with the flux of nitrous oxide through a leaf and evidence obtained of an extra resistance in the water-vapour pathway associated with water transport in the mesophyll cell walls. This extra resistance appeared to be insignificant at low transpiration rates and in turgid leaves, but increased with transpiration rate and dehydration. The most likely explanation for its origin appeared to be a reduction in hydraulic conductivity across the internal cuticle which lines the outer surfaces of the mesophyll cell walls. In turn this served to reduce the relative vapour pressure at the sites of evaporation. The experiments were conducted under conditions where stomatM opening was induced by COe-free air. Under normal conditions stomatM closure would tend to reduce the development of this extra resistance. Even so, the results throw doubt on the validity of the long-standing assumption that the water-vapour pressure at the evaporation sites is equM to the saturation vapour pressure under all conditions. Introduetion T r a n s p i r a t i o n from p l a n t s involves the e v a p o r a t i o n of water from sites w i t h i n the leaves a n d the s u b s e q u e n t diffusion of water v a p o u r to the n a t u r a l leaf surface a n d t h e n into the air beyond. Two m a i n p a t h w a y s for water m o v e m e n t are reeognised, one being associated with m o v e m e n t directly across the leaf cuticle, the other associated with m o v e m e n t t h r o u g h the stomatM pores. The cuticular p a t h w a y is relatively short, b u t of high resistance and, when the s t o m a t a are open, carries o n l y a small proportion of the t o t a l t r a n s p i r a t i o n flux. The s t o m a t a l p a t h w a y involves e v a p o r a t i o n of water from the outer surfaces of the mesophyll cells, a n d its diffusion t h r o u g h the intercellular spaces a n d the s t o m a t a l pores. A l t h o u g h m u c h longer, this route n o r m a l l y carries most of the t r a n s p i r a t i o n , b u t the presence of the s t o m a t a introduces a powerful, variable resistance into the p a t h w a y and, when the s t o m a t a are closed, s t o m a t a l t r a n s p i r a t i o n can effectively cease. 21 t?lan~a(Berl.), Bd. 90
304
P.G. Jarvis and R. O. Slatyer:
I t is generally accepted that variation in stomatal aperture is the main mechanism by which the plant exercises control over transpiration. From time to time, however, considerable controversy has arisen as to the degree to which other mechanisms may be of significance and the issue is still an active one (see, for example, Livingston and Brown, ]912; Knight, 1917; Klemm, 1956; Milthorpe and Spencer, 1957; Slavik, 1958; Shimshi, 1963; Rawlins, 1963; Slatyer, 1966, 1967; Fischer, 1968, Gale et al., 1967). Two main mechanisms have been proposed to account for nonstomatal transpiration control, should it occur. One envisages that the evaporation sites may retreat from the outer surfaces of the mesophyll cells, from which most evaporation occurs, into the walls themselves, thereby increasing the length and tortuosity of the vapour pathway, and hence the total diffusive resistance. This has been referred to as "incipient drying" (Livingston and Brown, 1912). The other mechanism proposes that the effective vapour pressure at the liquid-air interfaces may be reduced significantly below that of pure free water, thereby reducing the leaf-air vapour concentration gradient which provides the driving force for water-vapour transport. It is suggested that this may occur as a direct result of the dehydration associated with water stress (Shimshi, 1963; Rawlins, 1963) or by accumulation of solutes carried to the evaporating surfaces in the transpiration stream (Boon-Long, 1951). I t could also arise from the presence of the internal cuticle which lines the mesophyll cell walls (Scott, 1950, 1966). Should either, or both, of these mechanisms operate, they could be expected to be of greatest significance during periods of limiting water supply or excessive evaporative demand. Recently, Slatyer (1966) has argued from a theoretical and physical viewpoint that non-stomatal factors are not likely to be of significance in normal transpiration, and Fischer (1968) and Weatherspoon (personal communication) have obtained experimental evidence for leaves with stripped epidermes which support this view. The present paper contains a study of the problem using a novelapproach, with intact and undamaged leaves.
Theory Transpiration from the upper (adaxial) surface of a leaf can be described by the expression CW__ cU F~
~o-
E r~,o
(1)
where F 1-I20 u is the transpiration rate in units such as g c m -2 see -1, (cw -- CUa) is the water-vapour concentration difference between the sites of evaporation in the cell walls (cw) and the bulk air (CUa)in units of g cm -3
Mesophyll Cell Wall in Leaf Transpiration
305
and ~, r ~ o is the algebraic sum of the resistances encountered by the diffusing vapour (sec cm-1). Because water vapour diffuses from the leaves by both a enticular and a stomatal pathway it can be appreciated that ~ r ul i f O consists of a common external resistance r u, and two groups of internal resistances, connected in parallel, which comprise the resistance to transfer through the cuticle (reu) and that through the stomatal p a t h w a y (r~'). The latter symbol refers to specific resistances connected in series, which are concentrated in the cell wall (rw) and in the zone between the cell-wall surface and the external surface of the stomatM pore (r u) including the stomataJ pore itself. These resistances can be related by the expression F,r~o=r~+
f
r~r~c~ ~__
u
..~-T,-~ - r ~ +
(% +re J
~ (ru + rw) ru
~ ..~;-,_ ~ - u
(r8 + r w + r e ) J
.
(2)
A similar expression describes the resistances encountered by the net efflux of water vapour from the lower (abaxial) surface of the leaf, r Izt ~ O For each flux, r w represents a common resistance. An expression can now be derived for r~, from Eq. (2) by rearranging terms. Firstly, "
~
U
a
so t h a t
(3) rw =
re (2 r~00 - - r D reu -- (r, r u~ o - - rD
u
--r~.
The corresponding expression for the lower surface is: 1
rw=
t
1
f re (E r~,o -- r~) I i --,..-7-~; [ re -- (E rtI~O -- ra)
t
--rs.
(4)
The object of the experiments described in this paper is to determine the value of r w under various conditions. A convenient way of doing this is to compare the resistances to water vapour diffusion with those to some other gas, where the resistance r w does not arise, but where the pathway is in other respects identical, or differs in a manner that can be quantified. For this purpose N20 was used. I t has the advantage of being relatively inert physiologica!ly, at the concentrations used (Slatyer and Jarvis, 1966), and its concentration can be accurately measured with an infra-red gas anMyser. The net diffusive flux of N~O from one side of the leaf to the other can be described by the expression F ~ .-o 21"
~E- ~r s ~ o "
(5)
306
P.G. Jarvis and R. O. Slatyer:
The diffusive resistances to N~O flux can be expressed in terms of the diffusive resistances to water v a p o u r b y m u l t i p l y i n g t h e m b y the ratio of the diffusion coefficients of N20 a n d water v a p o u r i n air [D~c..o/Ds2 o = 0.65 (Slatyer a n d Jarvis, 1966)]. Hence D~O
Z rN20 -- f)~r
l
l
u
u
(ra d- rs d- ri ~- rs d- ra).
(6)
The new symbol ri refers to a n y a d d i t i o n a l resistance, in the intercellular air spaces, e n c o u n t e r e d b y the diffusing 5120 molecules over a n d above t h a t e n c o u n t e r e d b y the water v a p o u r molecules. R e a r r a n g i n g terms i n Eq. (6) therefore gives r s + r ls = 0.65 ~ r ~ 0 - ~za - ~ - r~.
(7)
I n t r o d u c i n g Eq. (7) a n d adding Eq. (3) a n d (4) the following expression can be o b t a i n e d for rw:
(s) - {o.65 ~ ~,~o - d - r :
--r~} l"
All terms i n Eq. (8), except for r w a n d r~, can be i n d e p e n d e n t l y determined. Therefore the m a i n difficulty i n d e t e r m i n i n g r~v lies i n providing a n estimate of r~ u n d e r the various conditions of interest.
Materials and Methods The theory requires tile simultaneous measurement, under varying conditions of stomatal aperture and transpiration rate, of the fluxes of water vapour from each side of a leaf, the flux of N~O through the leaf, the ambient water vapour and NzO concentrations on each side of the leaf, and the leaf temperature. A continuous and simultaneous measure of leaf water potential is also highly desirable. To satisfy these requirements, a leaf chamber of novel design was constructed and the nitrous oxide diffusion porometer developed. Full details are supplied by Slatyer and Jarvis (1966) and Jarvis and Slatyer (1966a, 1966b). Cotton plants were chosen for the experiments, partly because of the large size of the amphistomatous leaves, but also because of the background of knowledge about water-vapour exchange with this material (e.g. Slatyer and Bierhuizen, 1964). The cotton plants (Gossypium hirsutum L. cv. Pope) were grown in solution culture in a modified Itoagland's solution in a growth chamber with illumination from fluorescent tubes (Sylvania, Cool White VHO) supplemented with tungsten globes giving a radiation flux density of ca. 120 W m -2 (400--700 am) for 12 hr each day. The photoperiod was extended to 14 hr with the tungsten lamps alone. Air temperature was 35~ during the photosynthetic period and 24~ in the dark. Relative humidity was ca. 65%. In the experiments the fifth emerged leaf from the apex was used when there were about 10 leaves on the plant and before flowering commenced.
Mesophyll Cell Wall in Leaf Transpiration
307
The leaf chamber was designed so t h a t measurements were made on opposed areas of a leaf (ca. 31 cm 2) which remained attached to the plant. Because of the large area of leaf in the chamber in relation to the perimeter, the results obtained should be virtually independent of chamber dimensions. The direction of flux of NsO , usually from the upper to the lower surface of the leaf, did not affect the result, although there were about twice as m a n y stomata in the lower surface, confirming t h a t the flux through the leaf was essentially one-dimensional and t h a t possible errors arising from lateral flow were insignificant. I n a typical experiment, the attached leaf was placed in the chamber on the evening prior to the experiment and the plant was left overnight in darkness with normal air passing through the chambers. I n the morning, the stomata were opened with light in steps of 50 and 100 W m -2 of a b o u t 0.5 hr each before 150 W m -2 and C O J r e e air were supplied. Experimental treatments were imposed when steady state conditions of stomatal resistance, leaf temperature and leaf water content were obtained. During this warm-up period, the leaf-air vapour-pressure difference was 5 - - 1 0 m m Hg and the leaf temperature 28 ~ Experimental treatments usually consisted of varying the leaf-air vapourpressure difference by changing the h u m i d i t y of air entering the leaf chambers; in some experiments, leaf temperature was also varied and leaf temperatures of 33--35 ~ were used for the highest transpiration rates. Steady-state readings were usually obtained a b o u t an hour after a change in conditions, after which the conditions were again changed. Initially, ambient vapour pressure was changed in step functions from high to low, and vice versa, b u t because of the tendency for stomatal closure in the late afternoon, the treatments were later randomized. The same leaf was never used in more t h a n one experimental r u n or more t h a n 2 days in the chamber. A t the end of the experiment the leaf was sampled for calibration of the/?-gauge according to the procedure developed b y Jarvis and Slatyer (1966). t~elative water content and water potential of the leaves were determined by the methods of Weatherley (1950) and Slatyer (1958), as modified b y Jarvis and Jarvis (1963). IV[ore extreme water stress was induced in certain treatments b y exchanging the n u t r i e n t solution around the roots (water potential a b o u t - - 0 . 5 bars) for a similar n u t r i e n t solution containing either NaC1, mannitol or polyethylene glycol (Carbowax 400) to lower the water potential of the solution to values down to 20 bars. The solution was exchanged while the plant was in the dark, either just prior to illumination on the day oi the experiment, or on the previous evening. From the theory, it can be seen t h a t the most sensitive estimates of r i and r w will be obtained when the other resistances in series with them, notably r s and r a , are as small as possible. The boundary layer resistance, r a , was kept at a low and constant level by the action of the small fans stirring the air in the chambers above a n d below the leaf, and b y the flow of air through the chambers (ca. 40 or ca. 70 ltr h 2) (Jarvis and Slatyer, 1966b). There was no problem in keeping the stomatal resistance, r~, low at small leaf-air vapour-pressure differences, b u t the stomata showed a tendency to close a t vapour-pressure differences of a b o u t 20 m m t t g a n d over, especially when the water potential of the culture solution was lowered by the addition of solutes. There also appeared to be a tendency to close endogenously in the mid-afternoon after a b o u t 8 h r illumination. To keep the stomata open throughout an experiment, the observations were made a t a radiation flux density (400--700 nm) of 150 W m -s and with COs-free air supplied to the leaf chambers. This did not prevent some -
-
308
1). G. Jarvis and R. O. Slatyer:
increase in stomatal resistance at the greater vapour-pressure differences used, especially if these treatments occurred in the late afternoon or evening. I n a n u m b e r of experiments, steady state conditions were never obtained, the stomatal resistances, leaf water content and leaf temperature cycling in a regular manner even though CO~-free air was supplied. I n these cases, all the experimental data were transferred from the chart record to punched tape using an analogue to digital converter (Chapman and Goodspeed, 1967) and the i m p o r t a n t parameters calculated at 30- or 60-sec intervals b y digital computer, There are certain inherent disadvantages in using non-steady-state data. These difficulties are most apparent when the measured parameters are changing rapidly, when the rate of change is of a similar general order to the response characteristics of the measuring instruments, and when the measuring instruments themselves have different response times. Hence these data were used only to support the steady state observations.
Results 1. D e t e r m i n a t i o n
o/Constants
r ua a n d r a. ~ To determine the boundary layer resistance to water-vapour flux, r~, two thicknesses of wet blotting paper, to represent a fl'ee water surface, were placed in the leaf chamber in the leaf position with three thermocouples sandwiched in between. Values of r~ were calculated for each side from the slope of the relationship between the surface-air vapour concentration difference (c s - - Ca) and the steady state evaporation rate, FH~o (g cm -~ sec-1), according to the equation: a)
The
Boundary-Layer
Resistances
ra = [ FH~O /
(9)
(cs was assumed to be equivalent to the saturation vapour pressure at the temperature of the wet paper). The value of r~ was found to be 0.42 sec cm -1 for both the upper and lower surfaces of the blotting paper model when placed in the leaf chamber in the leaf position. This figure was adopted for both surfaces of the cotton leaf, the thorough air circulation being assumed to minimise any differences arising from the rougher lower surfaces of actual leaves. t The cuticular resistance to b ) T h e C u t i c u l a r R e s i s t a n c e s , r cu a n d r c. water transfer from each surface of several leaves was determined from regressions between cuticular transpiration and leaf-air vapour concentration difference in the range (cw - - Ca) = 15 -- 35 • l0 -6 g cm -3 (approximately equivalent to 15--35 mm Hg). The values obtained for r cl were found to be 70.6 & 5.6 sec cm -1, and for r~ to be 191 • sec cm -1. At high resistances the variability is comparatively large because of the difficulties in measuring the very small fluxes with sufficient precision. However, even large variations in r c have little effect on the calculation of r w or r i if r c is large in relation
Mesophyll Cell Wall in Leaf Transpiration
309
to }~ r~, o , a r e q u i r e m e n t t h a t is a l w a y s m e t while the s t o m a t a are open. No influence of (c~,- Ca) on r e was a p p a r e n t . S u m m i n g r~ a n d r~c in p a r a l l e l gives a n average r~ for the leaf of 51 sec cm 1. This is s o m e w h a t larger t h a n t h e figure of 32 sec cm -i given b y S l a t y e r a n d Bierhuizen (1964) for the same v a r i e t y , p o s s i b l y because of t h e different conditions in which t h e p l a n t s were c u l t i v a t e d , a n d p o s s i b l y because of t h e g r e a t e r p r e c a u t i o n s t a k e n to ensure s t o m a t a l closure in these e x p e r i m e n t s . H o w e v e r , even with t h e N~O p o r o m e t e r , it is n o t possible to ensure t h a t complete closure has occurred simult a n e o u s l y in both surfaces a n d some of t h e v a r i a b i l i t y in r e m a y r e s u l t from i n c o m p l e t e s t o m a t a l closure in one leaf surface.
2. Determination o/r~ I t seems p r o b a b l e t h a t t h e r e is a n e x t r a effective p a t h l e n g t h for N 2 0 diffusion t h r o u g h the leaf over a n d a b o v e the sum of t h e intercellular resistances for w a t e r v a p o u r diffusion to each leaf surface; t h a t is, t h a t r i is finite a n d positive. F r o m Eq. (8) it can be seen t h a t t h e lower the value of ri the lower will be the c a l c u l a t e d value of r w. M i n i m u m values for rs are therefore o b t a i n e d on the a s s u m p t i o n t h a t r i : 0. While these values are of i n t e r e s t in setting lower limits for rw, i t is clearly desirable to m a k e realistic e s t i m a t e s of r i so t h a t t h e values assigned to r~s can also be as realistic as possible. F o r this p u r p o s e two procedures were used. The first, a n a n a t o m i c a l m e t h o d , is b a s e d on calculations b y J a r v i s et al. (1966) which g i , e a value for r* - - t h e t o t a l intercellular space resistance b e t w e e n the inner surfaces of t h e s u b - s t o m a t a l cavities on each side of the leaf. F r o m these calculations r* = ~ ~ ;~; D~o
(lO)
where hi, ai a n d n~ are, respectively, the thickness of the mesophyll, the a v e r a g e r a d i u s of t h e i n t e r c e l l u l a r spaces, a n d the n u m b e r of air passages p e r u n i t cross section of leaf p l a n area. A n a t o m i c a l i n f o r m a t i o n a b o u t the leaves r e q u i r e d for the determ i n a t i o n of r~ was p r o v i d e d b y Dr. J. E. Begg (personal communication). There were a b o u t twice as m a n y s t o m a t a in t h e lower surface as in t h e u p p e r surface. Several series of p a r a d e r m a l sections were o b t a i n e d using a freezing m i c r o t o m e a n d 60 cellular a n d inter-cellular i n t e r c e p t s were m e a s u r e d on a set of t r a n s e c t s in different directions across these sections. The a r i t h m e t i c m e a n d i m e n s i o n of the intercellular spaces o b t a i n e d was 1.8 =L0.3 ~. Since t h e t r a n s e c t s were in m a n y different directions, this m e a n dimension was t a k e n as t h e m e a n d i a m e t e r of a cylindrical air passage of l e n g t h e q u a l to t h e thickness of the mesophyll. Because t h e
310
P.G. Jarvis and R. O. Slatyer:
resistance is inversely proportional to the square of the radius [Eq. (10)], and the intercellular spaces cover a range of sizes from 0 to 7.2 ~, the mean of a i~' ( ~ 1.73 ~ ) was used in preference to the square of the arithmetic m e a n of a i which would overestimate the resistance. F r o m the transect data, the n u m b e r of the air passages in the mesophyll per unit leaf plan area (n~) was calculated to be 7.15 • l04 cm -2 (Jarvis et al., 1966). The thickness of the mesophyll was measured in transverse leaf sections and found to be 200 ~= 10 ~. Since the subs t o m a t a ! cavities are included in r~., the p a t h length for diffusion t h r o u g h the intercellular spaces was taken to be of the order of 170 ~. F r o m these data, the estimate of r* from Eq. (10) was found to be 1.65 sec cm -1. This estimate can be expected to exceed the true value of r i unless all the intercellular space resistance to water vapour diffusion is in the sub-stomatal cavities; however, it should provide a reasonable upper figure for r~ and, hence, a basis for a minimal estimate of r w under normal turgid conditions. During stress, leaf water content and leaf volume decline (Kennedy and Booth, 1958). Assuming t h a t shrinkage is uniformly three-dimensional, the reduction in the cross-sectional area of the air passages (a~) should be balanced b y a proportional increase in ni, and h i should be expected to decrease to a value proportional to the cube root of the stressed relative water content. A reduction to a relative water content of 50 % (a figure likely to cause death of cotton leaves) can therefore be expected to reduce r i to 0.8 of its turgid value. A change of this order would n o t affect r w to a noticeable degree. The second, physiological, m e t h o d of estimating r i is based on the assumption that, in turgid plants at low leaf-Mr v a p o u r concentrations and tow transpiration rates, r w ~ O. Under these conditions, Eq. (8) can be used to provide estimates of r i. 20 determinations of r i were therefore made on occasions when (cw - - C a ) ~ 5 m m Hg. These measurements yielded an average value for r~ of 1.79 • 0.25 see cm -~. Although there was a high degree of variability, as the s t a n d a r d error indicates, caused b y the difficulty of measuring low values of FH..O with precision, the error has a relatively small effect on the estimates of rw. I t is also of interest t h a t the values of r i determined in this fashion, agree closely with those obtained from the anatomical method. For calculation purposes, the physiological estimates were used, rounded off to 1.8 sec cm -1. 3. D e t e r m i n a t i o n o / r w
Altogether, over 100 experiments, each leading to determinations of r~, were made under a range of conditions. The following paragraphs
Mesophyll Cell Wall in Leaf Transpiration
311
12 11
FH2o
9 :g
m
8
u. 4 3. 2. 1 0
•
5
10
15
20
25
( C w - C a ) x 10 -~ grn crn -3
Fig. 1. Data from one experiment showing the relation between transpiration from a whole cotton leaf (FIt2O) and from the upper (F~I~O) and lower (F~LO) surfaces, and leaf-air vapour concentration difference (cw -- ca)
describe the effects of rate of transpiration, leaf water content, and the duration of high rates of transpiration on r w .
a) E//ect o/Transpiration Rate. Transpiration d a t a from a typical experiment, conducted under conditions of minimM substrate water stress, are depicted in Fig. 1. The three curves show the p a t t e r n of change in transpiration from the leaf as a whole, and from each surface, as leaf-air v a p o u r concentration difference (zJ c) was increased. I t is a p p a r e n t t h a t there is a progressive divergence from a straight-line relationship as transpiration increases until, at high d c values, a drop in transpiration accompanies a further increase in z] c. D a t a of this type have been reported on a n u m b e r of occasions in the past, indicating d e a r l y t h a t there is a progressive increase in the total diffusive resistance as transpiration rate, a n d A c, increase. I t has frequently been assumed t h a t this departure from linearity is due entirely to progressive stomatal closure; on other occasions it has been assumed t h a t no closure has occurred and t h a t the phenomenon provides evidence of non-stomatal control of transpiration (i.e. an increase in rw).
312
P.G. Jarvis and R. O. Slatyer:
8
~
%0
6u
5-
rw
2~
0
o
~
~ FxIO -6gm
1"s
c m -2 sec -s
Fig. 2. Relationship between the stomatal resistance from the upper (ru) and the lower (rs~)leaf surfaces, cell wall resistance, (rw), and total (upper plus lower surface) transpiration rate. Points shown as circles are from data in Fig. 1, points shown in squares and crosses are from additional experiments I n Fig. 2 the main variable components of the total diffusive resistance, which are associated with this transpiration pattern, are depicted. The d a t a are drawn from the data of Fig. l, plus data from other experiments in which higher and lower transpiration rates occurred. Although these points represent only a small fraction of the data available, t h e y are fully representative. Only r s and rw are plotted since the increase in the total resistance m u s t be due to an increase in either, or both, these parameters; ra, r c and r i being regarded as constant. Fig. 2 brings out clear evidence for the existence of r w and suggests t h a t it is dependent, to some degree, on transpiration rate. The intercept of the rw/F~ o curve appears to pass through, or close to, the origin. I t is a p p a r e n t that, although the values of r w are low, relative to comparable values of rsu or riB, t h e y are high enough to be significant components of the total diffusive resistance, at least at the highest transpiration rates obtained. b) E//ect o / L e a / Water Content. W a t e r deficits were imposed b y the addition o~ osmotic substrates [mannitol and polyethylene glycol 400 (carbowax)] which reduced the substrate water potential to levels down to - - 2 0 bars, and reduced relative leaf water content (O) to values down to about 50 %. Although a n u m b e r ol experiments of this type was conducted, considerable difficulty was experienced in maintaining the s t o m a t a
Mesophyll Cell Wall in Leaf Transpiration
313
,
100
90
80
70
60
50
Fig. 3. Relationships between r w and relative water content (O). Each curve represent a different experiment open and, as r 8 increased, the rehability of the estimate of % decreased. I n Fig. 3 data are presented from three typical experiments, covering a range of relative water contents from turgid to severe stress. Lowmoderate transpiration conditions (Fs~o < 10 x 10 _6 g c m -2 see -1) were used to minimise the development of steep water-potential gradients within the leaf and avoid confounding the effects of transpiration rate with direct effects of dehydration. The figure shows that % increases in a fairly linear fashion over the range of values of 0 that it was possible to obtain. Again, the values % appear to be relatively low, although, at the dry end where they exceed 4, it is apparent that % would be a significant factor in determining leaf transpb'ation if the stomata remained open. If the stomata closed, even partially, to provide ~, r ~ o values of 20 see cm 1 or so, the significance of r w would be much reduced. e) D u r a t i o n o/ H i g h Rates o/ Transpiration. Another possible way in which an apparent increase in rw could occur would be through the accumulation of solutes, carried forward in the transpiration stream, at the sites of evaporation. Should this effect develop, it would be most pronounced under conditions of prolonged transpiration at a high rate. Consequently, transpiration and rw were measured under these conditions over a period of 5 hr. Because the initial accumulation of solutes could be expected to be rapid, a high illumination (200 W m -2) dry air (A c = 2 0 mm Hg) situation was suddenly imposed on a situation of moderate light and A c. The results showed that steady state transpiration was reached within 10 rain and, although some tendency towards cycling of stomatal resistance was observed, rw showed no significant increase over a period of 5 hr.
314
P.G. Jarvis and g. O. Slatyer:
A c c u m u l a t i o n of solutes should also be e n h a n c e d b y t h e a p p l i c a t i o n of r a p i d l y p e r m e a t i n g solutes to the r o o t m e d i u m . H o w e v e r , t h e add i t i o n of NaC1 a t c o n c e n t r a t i o n s of up to 4 b a r s osmotic pressure in six s e p a r a t e e x p e r i m e n t s , h a d no d e t e c t a b l e effect on r w.
Discussion The two most significant results of t h e p r e s e n t s t u d y arc, firstly, t h a t positive values for r w were o b t a i n e d and, secondly, t h a t t h e r w increased w i t h increasing t r a n s p i r a t i o n r a t e a n d with increasing w a t e r stress. B o t h o b s e r v a t i o n s are in accord w i t h t h e recent results of F i s c h e r (1968) in leek mesophyll, b u t the values of r,~ were s u b s t a n t i a l l y g r e a t e r in t h e present s t u d y , exceeding 2 sec cm -1 whereas F i s c h e r d i d n o t observe values greater t h a n 0.5 sec cm -1 even u n d e r severe stress. The p r e s e n t values are more similar in m a g n i t u d e to those r e p o r t e d b y K l e m m (1956) a n d S h i m s h i (1963). Before discussing physiological aspects of r ~ , i t is desirable to e x a m i n e the q u a l i t y of t h e p r e s e n t observations. The q u a l i t y of the e x p e r i m e n t a l m e a s u r e m e n t s themselves a p p e a r s a d e q u a t e to enable e s t i m a t i o n of r w to b e t t e r t h a n :j= 1.0 sec cm -1 as long as t h e t o t a l series sum of the diffusive resistances does n o t exceed a b o u t 20 sec cm -1. As has been i n d i c a t e d before efforts were m a d e to minimise r a a n d rs, a n d t h e o n l y d a t a used were those in which t h e series s u m of resistances o t h e r t h a n r w was less t h a n 10secem -1. The m e a s u r e m e n t which caused m o s t concern was of leaf t e m p e r a t u r e , a n error of 0.5 ~ being sufficient to cause a n error of e s t i m a t e of a b o u t 5 % in ~ r ~ o a t t h e general level of t e m p e r a t u r e s a n d v a p o u r - c o n c c n t r a t i o n differences used. Although e x t r e m e care was t a k e n with l e a f - t e m p e r a t u r e m e a s u r e m e n t s , familiarity w i t h t h e technical p r o b l e m s i n v o l v e d leads us to t h e conclusion t h a t errors of this m a g n i t u d e m a y h a v e occurred. A p a r t from specific deficiencies in the m e a s u r e m e n t s , t h e assumptions concerning the c o n s t a n c y of r e a n d r~, a n d t h e a c t u a l value of r~ a d o p t e d , are open to challenge. I n the case of rc, I t o l m g r e n e t a l . (1965) have shown a d e p e n d e n c e of rc on leaf t e m p e r a t u r e a n d r a d i a t i o n - f l u x density. H o w e v e r , the values of r e are so large in comparison with the other resistances, a n d the r a n g e of leaf t e m p e r a t u r e s utilised was so small, t h a t v a r i a t i o n s in r c could n o t have h a d a significant effect on r w. I n t h e case of ri, a value of 1.8 sec cm -1 was chosen following the a n a t o m i c a l a n d physiological evidence described previously. I t was suggested t h a t , if the leaf s h r a n k t h r e e - d i m e n s i o n a l l y d u r i n g stress, t h e value of r i a t a relative w a t e r c o n t e n t of 50% should be 0.8 of the t u r g i d figure, leading to a slight increase in t h e e s t i m a t e d v a l u e of r w . Unp u b l i s h e d d a t a of one of us (P. G.J.) s u p p o r t the contention t h a t r e l a t i v e l y
Mesophyll Cell Wall in Leaf Transpiration
315
small changes occur in intercellular volume, at least in mature cotton leaves. Although Meidner (1955) found that resistance to bulk air flow under a pressure difference, through the mesophyll of P s y c h o t i a c a p e n s i s and S c o l o p i a m u n d i i , increased as water stress was imposed, his technique does not indicate the response pattern to the transverse diffusive flux of water vapour. Accordingly, the estimates used for the present paper do not appear unreasonable or liable to introduce serious errors to estimates of r w. Turning now to the physiological aspects of rw, it should be remembered t h a t the calculations were made so as to attribute to r,~ any apparent additional resistance in the water-vapour pathway. The decision could equally well have been made to attribute the apparent resistance to a change in Cw, or to divide the effects empirically between % and r w . I t is now important to a t t e m p t to assign it correctly. If r w is a true gas-phase diffusive resistance, the traditional suggestion is t h a t it would most probably arise as a result of a withdrawal, into the interfibrillar wall spaces, of the liquid-air interfaces which constitute the evaporation surfaces, thereby creating an extra, gaseous segment for the evaporation pathway. Should this be the correct explanation, rw would be expected to increase as water stress increases, and possibly also increase with increasing transpiration rate should water supply be inadequate to maintain transpiration rate without the development of steep local gradients of water potential. Although the data of Figs. 1, 2 and 3 provide supporting evidence for this view, these data can also support other explanations, and a consideration of the cell-wall morphology throws doubt on this simple explanation. For example, Slatyer (1966) has calculated that if the wall spaces have the characteristics of capillaries and maintain hemispherical menisci at the liquid-ah ~interfaces, they would not be expected to drain until local leaf water potentials reached extremely low levels. In fact, voids as large as 5 nm radius would not drain until water potentials fell to - - 3 0 0 bars and much lower potentials, below --1,000 bars, would be required for smaller voids, even though the degree to which capillary theory can be applied at such small values is in doubt. The cotton leaves used in this study were examined by electron microscopy, through the courtesy of Dr. D. J. Goodehild 1. No pore structure was detectable when wall sections were stained either with potassium permanganate and Reynolds lead (Reynolds, 1963) or when stained with glutaraldehyde, osmie acid and Reynolds lead. However, the preparations did indicate the presence of interfibrillar spaces of the order of 5--10 nm. I n addition, in the outer layers, a zone of about 1 C.S.I.R.O. Division of Plant Industry, Canberra.
316
P.G. Jarvis and R. O. Slatyer:
50 nm across was noticeably denser with a laminar structure and much smaller spaces. This zone corresponds to the internal cuticle of Scott (1950, 1966). From this examination it can be appreciated that extremely low water potentials would be required in order to cause a retreat of the liquid-air menisci. Such values were not approached by the general level of leaf-water potential used in this study; relative water contents of the order of 50% (as in the data of Fig. 3), being associated with leaf water potentials of the order of - - 4 0 bars. I n consequence it appears unlikely t h a t "incipient drying", in the context used in the literature, is a real phenomenon, at least with respect to cotton leaves. If the gas phase resistance component of r w is very small, or nonexistent, the implication is that there is liquid phase continuity right up to the surfaces of the mesophyll cells. I n this case, the effect attributed to r w must be assigned instead to a reduction in c w below the value equivalent to the saturation vapour pressure at the leaf temperature Such a reduction could have three obvious origins, all of which could act in combination: (1) The general level of leaf water potential (~eaf) could depress %, according to the relationship between water potential and relative vapour pressure (or relative vapour concentration), viz.
~leaf-- R_T In %/Csat,
(11)
where the new symbols R, Vw and csat refer, respectively, to the ideal gas constant, the partial molal volume of water (cm 3 mole-l), and the saturation vapour concentration (g em-a). (2) In addition to the effects of general leaf water potential described in (1), an additional reduction in % could be caused by a steep water potential gradient across the outer cell wall layers. This could arise if there is a significant source of hydraulic resistance to liquid flow across the outer, denser, layers of the cell wall (3) There could be an accumulation of osmotically active solutes transported to the liquid-air interface by the transpiration stream, which would depress the relative vapour pressure and hence %. If there is no vapour phase resistance in the cell wall (i.e. if r w = 0 ) , the m a x i m u m degree of reduction in the relative vapour pressure which occurs at the liquid-air interfaces can be calculated as follows: Eq. (1) is first re-written, for stomatal transpiration (st) only, as F ~H
