5
MEASUREMENT OF LEAF WATER POTENTIAL BY THE DYE METHOD By EDWARD B. KNIPLING Reprinted from ECOLOGY, Vol. 48, No. 6, Autumn 1967

Measure of Water Potential by the Dye Method

Embed Size (px)

Citation preview

Page 1: Measure of Water Potential by the Dye Method

MEASUREMENT OF LEAF WATER POTENTIAL

BY THE DYE METHOD

ByEDWARD B. KNIPLING

Reprinted from ECOLOGY, Vol. 48, No. 6, Autumn 1967

Page 2: Measure of Water Potential by the Dye Method

MEASUREMENT OF LEAF WATER POTENTIAL BY THE DYE METHOD

EDWARD B. KNIPLING!Department of Botany, Duke University, Durham, N. C.

(Accepted for publication August 25, 1967)

Abstract. The dye method for measuring leaf water potential is simple, inexpensive, andsuitable for both laboratory and field work. Leaves are immersed in a graded series of solu-tions, and the solution which neither gains nor loses water is assumed to have a water potentialequal to that of the leaf. Although limited by certain inherent errors, the dye method can beused to measure relative values and changes in potential. Specific procedures are describedfor the successful use of the method.

There are few techniques for measuring leaf waterpotential (Kramer, Knipling, and Miller 1966) that aretruly suitable for field work. The recently describedpressure chamber method appears promising for somespecies (Scholander et al. 1965, Boyer 1967, Waring andCleary 1967). The dye method, also known as theShardakov (1948) or density method, is another usefultechnique because it is simple and requires no elaborateor expensive equipment. The dye method has beendescribed elsewhere, but no comprehensive discussionis readily accessible to many workers. This reportexamines the method and describes specific proceduresand precautions that need to be taken for its successfuluse.

In the dye method (Fig. 1) samples of leaf tissue areimmersed in a graded series of solutions of known waterpotential contained in test tubes. The leaf water poten-tial is assumed to lie between the solutions in which theleaf samples absorb water and those in which the sam-ples lose water. The directions of this water exchangeare determined from the resulting density changes inthe test solutions. Each test solution is colored lightlywith a small amount of a powdered dye such as methy-lene blue or methyl orange. Drops of the colored solu-tions are then introduced with medicine droppers into thecenters of the corresponding members of a parallel seriesof uncolored control solutions. The drops fall if thetest solutions have been concentrated and rise if thesolutions have been diluted. In case the leaf sample hasa water potential equal to that of one of the solutionsused, there theoretically is no water exchange and no

1 Present address: United States Army Cold Re-gions Research and Engineering Laboratory, Hanover,New Hampshire 03755. This work was supported bythe Coweeta Hydrologic Laboratory, SoutheasternForest Experiment Station, U. S. Forest Service.

density change, and hence the colored drop diffuses out-ward in all directions.

The purpose of the dye is to make the test solutiondrops visible when they are placed in the control so-lutions. The small amount of dye used to color the so-lutions lightly does not significantly alter their densities.

The precision of measurement by the dye method de-pends on the size of the water potential increment be-tween test solutions. It is possible to use increments assmall as 0.5 bar, but in order to have a workable numberof solutions (generally 6 to 10) bracketing an unknownleaf water potential, 1- to 5-bar increments often are theonly reasonable choices.

Test solutions generally are made with sucrose ormannitol. When sucrose is used, it is suitable to usethe commercial (trade) type. The solutions of known

« -8.7 BAF-9.0

SIO.O -II. 0

LEAF ABSORBS WATERSOLUTION DENSITY INCREASES LEAF LOSES WATERSOLUTION DENSITY DECREASES

CONTROL

FIG. 1. Diagram representing dye method for measur-ing leaf water potential. The measured water potentialis taken to be between the solutions in which the coloredtest solution drops rise and fall, i.e., —8.5 ± 0.5 bars.

Page 3: Measure of Water Potential by the Dye Method

Autumn 1967 1039water potential can be prepared according to tables ofWalter (1931) or Ursprung and Blum (1916). Theamount of solution used in each test tube should beenough to cover the leaf sample. Test solution volumesof about 2 ml in 13- by 100-mm test tubes are suitablefor most water potential determinations with small piecesof leaf tissue, but larger volumes generally are necessarywhen whole leaves are used. The volume of the controlsolutions should be about 5 ml in order to have asu/hcient depth in which to observe the direction ofdrop movement.

The amount of leaf tissue used per unit volume ofsolution should be as large as is feasible without re-stricting access of solution to the tissue and withoutdamaging the tissue when immersing it into the solution.An exactly equal amount of leaf tissue is not needed ineach solution because the method detects the directionof water exchange, not the amount of exchange. How-ever, it is advisable to use approximately the sameamount of tissue in each solut ion and from one waterpotential determination to the next.

The use of whole leaves in the dye method is con-venient because of the ease with which they can be sam-pled and inserted into the solutions. In addition, thewater potentials measured wi th whole leaves are insome cases more accurate than those measured withcut leaves. This appears to be true for conifer needles,hut the comparisons for broadlcaved species are variable(Knipling and Kramer 1967). In some cases the cut-ting of leaf tissue is desirable because it facilitates waterexchange and permits the variability in water potentialamong a sample of leaves to be represented in eachsolution. Often the use of cut leaves is necessary whenthe leaves are large or the supply limited.

Theoretically, only a short leaf immersion time shouldbe necessary for a water potential measurement by thedye method. This is because the drop movement issensitive to extremely small concentration differences(0.0005M sucrose), and therefore, detectable densitychanges should be established in the solutions long be-fore complete equilibration. The only effect of longerimmersion times should be to change the distinctivenessof drop movement. However, there also is a change

IMMERSION T I M E t h r s )FIG. 2. The effect of progressively longer immersion

times on the leaf water potentials of cut leaf tissuemeasured by the dye method. Species: dogwood, Cormisflorida L.; sourwood, Oxydendnnn arboreum DC.; whiteoak, Quercus alba L.; American elm, Ulmns americanaL.; yellow poplar, Liriodenclron tulipifera L.; tomato,Lycopersicum esculcntmn Mill.; tobacco, Nicotiana taba-cum L.

in the measured water potentials with time (Fig. 2).Characteristically, the values measured after about 30min are 2-5 bars lower (more negative) than the final,stable values reached after 2-8 hr. However, waterpotential measurements of conifer needles continue tochange for about 24 hr. Since the final values providethe best estimates of water potential (Render and Krceb1961, Lemee and Gonzalez 1965), the proper lengthof immersion for a particular species should be deter-mined prior to using the dye method for routine measure-ments.

Numerous explanations have been offered for thechanging values with increasing immersion time. Wal-ter and Ellenberg (1957) suggest that the initial valuesare those of epidermal cells, whereas later readings rep-resent an approach toward the higher water potentialsof interior cells. However, it is questionable if sucha large water potential gradient exists within leaves.Furthermore, with cut leaf tissue, one would expect in-terior cel ls to equilibrate as rapidly as surface cells.

Other workers (Lemee and Gonzalez 1965, Brix 1966)srg ;est that the changing values reflect a gradient ofwater potential created by cutting the leaf tissue, theini t ia l values representing the osmotic potential of cellsalong the cut edges where turgor pressure has beeneliminated. Although this induced gradient may con-tribute to the changing values, it does not fully accountfor the initial values because they generally are lower(more negative) than the leaf osmotic potential. Forexample, the osmotic potential of dogwood leaves wasaccurately measured with a thermocouple psychrometer(Ehlig 1962) to be —15.3 bars. The water potentialmeasured on a parallel sample of leaves by the dyemethod was —24 bars after 10 min immersion and —17bars after 30 min immersion (Fig. 2). Further evi-dence that the initial dye method values are not de-termined by the osmotic potential of cut cells is indi-cated by the fact that the changing values also areobserved with uncut leaves.

The major cause of the erroneous dye-method valuesmeasured after short immersion times appears to be theadsorption of water onto the leaf surfaces (Rehder 1961,Schlafli 1964). This extraction of water from the so-lutions occurs within the first few minutes of immersionand increases the density of all test solutions beforeosmotic water exchange takes place. Thus, drops fromthe test solutions having potentials lower than that ofthe leaf move in the wrong direction when placed inthe control solutions. After progressively longer im-mersion times, the values indicated by the dye methodapproach the true leaf water potential as osmotic waterloss from the tissue compensates for the initial densitychanges. The effect of adsorbent leaf surfaces on solu-tion densities can be simulated by placing filter paperin sucrose solutions.

In most cases leaf surface residues and sap from cutand damaged tissues, which contaminate the test solu-tions (Gaff and Carr 1964, Brix 1966), probably con-tribute to the abnormal density increases of the solutionsduring the early period of immersion. Also, the sub-sequent leakage of low density contaminants from theleaf tissue probably helps compensate for the initialdensity changes.

Once dye-method values reach a stable level after anextended immersion time there still is a residual errorin the measurements. The contamination of the solu-tions appears to be the major source of error (Brix1966, Knipling and Kramer 1967), although the uptakeof solutes by the leaf tissue, which decreases the densities

Page 4: Measure of Water Potential by the Dye Method

1040of the test solutions, contributes to the total error(Goode and Hegarty 1965, Brix 1966). The directionof the contamination error depends on the density ofthe contaminants relative to the density of the solutionhaving a potential equal to that of the leaf. When thecontaminants are more dense than the equipotential so-lution, the dye method indicates a value lower (morenegative) than the true leaf water potential; when thecontaminants are less dense, the method indicates avalue higher than the true potential.

The final size of the contamination error depends onthe extent which osmotic water gain by or loss fromthe leaf tissue compensates for the contamination densitychanges during complete equilibration. In this regard,succulent-leaved species such as tomato, tobacco, andyellow poplar, which undergo relatively large changesin water content per unit change in water potential,generally have smaller contamination errors than leavesof woody species such as white oak and American elm.

To assess the total error in dye-method values, themethod has been compared with the thermocouple psy-chrometer method, which is considered to measure leafwater potentials accurately (Boyer 1966). Valuesmeasured by the two methods on the same leaf materialgenerally have agreed within 3 bars, although the valueshave differed by as much as 5-8 bars for some species(Kramer and Brix 1965, Knipling and Kramer 1967).

Even though the dye method appears not to measuretrue leaf water potentials, it can be used to measure rel-ative values and changes in potential. However, valuesdetermined on different species should be comparedwith caution because of differences in the relationshipsof dye-method values to the true leaf water potentials.

In addition to the procedures already discussed, thereare certain practices that are helpful or necessary inconducting the dye method. It is essential that allglassware be clean and dry to prevent any alteration ofthe solutions. The solutions should be relatively freshso as not to be contaminated with fungi. When the stocksolutions are contained in plastic wash bottles, they canbe rapidly and accurately dispensed to a given level inthe test tubes. The test and control solutions alwaysshould be dispensed at the same time and from the samestocks to insure equal densities at the start of a deter-mination.

To reduce the adverse effects of contamination, Brix(1966) modified the procedure for obtaining the controlsolutions. Since most contamination takes place in thefirst 30 min of immersion, Brix excluded its effect bymeasuring the density changes occurring thereafter. Inpractice, he decanted off portions of the contaminatedtest solutions to use as controls. Similarly, Gaff andCarr (1964) suggested that the effects of contaminationcould be reduced by renewing the test solutions withfresh, uncontaminated solutions after an initial shortperiod of immersion.

In some cases contamination can be reduced by wash-ing the experimental leaf material several or morehours prior to sampling. When cut tissue is used inthe dye method, the cutting should be done with a sharprazor rather than with a cork borer or knife to mini-mize damage to the leaves. When practical, interveinalleaf tissue should be used to minimize leakage of freewater into the solutions from the cut vascular elements.Glass rods aid in pushing the leaf samples into the testsolutions without damaging the tissue.

During the leaf-immersion period the test tubes shouldbe stoppered to prevent density changes in the solutionscaused by evaporation. Density changes also arise

REPORTS Ecology, Vol. 48, No. 6from temperature differences, and the test and controlsohiiions should be kept together during the immersionpci 'Kid . Ex t ra - long medicine droppers (about 5 inches)aid the insert ion of the colored test solution drops intothe con'.rol solutions. A separate dropper should beused for each pair of solutions. The visibility of dropmovement can be enhanced by placing a white back-ground on the test tube rack.

LITERATURE CITEDBoyer, J. S. 1966. Isopiestic technique: measurement

of accurate leaf water potentials. Science 154: 1459-1460.

. 1967. Leaf water potentials measured witha pressure chamber. Plant Physiol. 42: T33-137.

Brix, H. 1966. Errors in measurement of leaf waterpotential of some woody plants with the Schardakowdye method. Canada Dept. Forestry Publ. 1164. 11 p.

Ehlig, C. F. 1962. Measurement of energy status ofwater in plants with a thermocouple psychrometer.Plant Physiol. 37: 288-290.

Gaff, D. F., and D. J. Carr. 1964. An examination ofthe refractometr ic method for determining the waterpotential of p lant tissue. Ann. Bot. N. S. 28: 351-368.

Goode, J. E., and T. W. Hegarty. 1965. Measure-ment of water potential of leaves by methods in-volving immersion in sucrose solutions. Nature206: 109-110.

Knipling, E. B., and P. J. Kramer. 1967. Compari-son of the dye method with the thermocouple psy-chrometer for measuring leaf water potentials.Plant Physiol. 42: 1315-1320.

Kramer, P. J., and H. Brix. 1965. Measurement ofwater stress in plants. In Methodology of plant eco-physiology. 1962 Proc. Montpellier Symp., UNESCO,Arid Zone Res. XXV: 343-351.

Kramer, P. J., E. B. Knipling, and L. N. Miller. 1966.Terminology of cell water relations. Science 153:889-890.

Lemee, G., and G. Gonzalez. 1965. Comparison demethodes de measure du potentiel hydrique (tensionde succion, DPD) dans les feuilles par equilibreosmotique et par equilibre de pression de vapeur.In Methodology of plant eco-physiology. 1962.Proc. Montpellier Symp., UNESCO, Arid Zone Res.XXV: 361-368.

Render, H. 1961. Saugkraftmessungen on: Stachyssilvatica im frischen und welden Zustand. Ber.Deut. Bot. Ges. 74: 84-92.

Render, H., and K. Kreeb. 1961. Vergleichende Un-tersuchungen zur Bestimmung dcr Blattsaugspan-nung- mit der gravimetrischen Methode und derSchardakow-Methode. Ber. Deut. Bot. Ges. 74:95-98.

Schlafli, A. 1964. "Ober die Eignung der Refrakto-meter- und der Schardakowmethode zur Messungosmotischer Zustandgrossen. Protoplasma 58: 75-95.

Scholander, P. F., H. T. Hammel, E. D. Bradstreet,and E. A. Hemmingsen. 1965. Sap pressure in vas-cu la r plants. Science 148: 339-346.

Shardekov, V. S. 1948. New field method for the de-t e r m i n a t i o n of the suction pressure of plants. Dokl.Akad. Nauk SSSR 60: 169-172. (In Russian).

th'sprung, A., and G. Blum. 1916. Zur Kenntnis derSaugkraft. Ber. Deut. Bot. Ges. 34: 525-539.

Walter, H. 1931. Die Hvdratur der Pflanze. Gus-tav Fisher, Jena. 161 p.

Page 5: Measure of Water Potential by the Dye Method

Autumn 1967 REPORTS 1041Walter, H., and H. Ellenberg. 1957. Okologische ture stress: evaluation by pressure bomb. Science

Pflanzengeographie. Fortschr. Bot. 20: 100-117. 155: 1248-1254.Waring, R. H., and B. D. Cleary. 1967. Plant mois-