Photosynthetic Organs of Desert Plants Structural designs of nonsucculent desert plants cast doubt on the popular view that saving water is the key strategy Arthur C. Gibson

M

an y species of planes and animals are remarkably adapted to live in hot, dry deserts, where annual rain fall is extremely low and organisms must cope wirh lang drought. The natural history and environmental biology of desert organisms-how they man-

age life undee such a dry regime-is of particular interest. Ta explain how organisms survive desert harshness, biologists naturally first assumed that most structural features of desert planes are adaptations to limit water loss-that is, to conserve waterand general textbooks and accounts of desert natural history commonly stress chis basic tenet. However, aver the last quarter-century, plant ecophysiologists have gradually been developing a very different view of desert plant structure, in which saving water is not as central as formerly thought. Instead. there is growing evidence that many of the strucrural and physiological adaptations of photosynthetic leaves and sterns are for maximizing photosynthetic rate and regulating energy budget. Desert plants do, of course, show some water-conserving strategies, but generally not through structural adaptations. Cacti are frequently cited as representative desert plants because they Arthur C. Gibson (e-mai!: agibson@ biology.uda.edu) is a professor in the Department of Organismie Biology, Eeology, and Evolution, University of Califocnia, Los Angeles, and direetor of the Mildred E. Mathias Botanieal Garden, Los Angeles, CA 90095-1606. © 1998 American Institute of Biologieal Scienees.

November 1998

Desert plants do, of course, show some waterconserving strategies, hut generally not through structural adaptations have an exquisite set of adaptations for optimal water management (Gibsan and Nobel 1986). Greatly thickened stern tissues, composed mostly of large, thin-walled parenchyma cells, provide a huge reservoir of water that is stored in vacuoles; a fully hydra ted cactus stern has a water content of 90-94%. The low surface-to-volume ratio of the stern allows the plant to stock pile enough water behind every square millimeter of stern to carry it through several years of drought. Some desert cacti can end ure tissue water content as low as 20% and still resurne growth when stimulated by a single heavy rain. Cactus leaves are small to microscopic and ephemeral, and the green stern is the chief photosynthetic organ. The stem is coated with a thick, waxy cuticle, so that when stomata are closed, transpirational loss is trivial. In addition, cacti have Crassulacean acid metabolism (eAM), a specialized type 01 photosynthesis in which stomata are closed during daylight but open in darkness (Gibson and Nobel 1986, Nobel 1988, 1991a). The great advantage of nocturnal stomatal opening is that it occurs when temperatures tend to

be lower, resulting in substantial savings of water over what would be used with the same amount of stomatal opening in the hot daytime, when transpiration rates would be extremely high due to the sharp drop in water vapor concentration from the internal stern surface to the ambient dry air. As illustrative as cacti and other succulent plants are for water conservation, succulent plants actually constitute a small percentage of plant biomass in most deserts, aside from a few notable regional exceptions. such as the succulent Karroo, Atacama, and Sonoran Deserts. In fact, more than 85% of all desert plant species have no obvious waterstorage tissue to buffer daily dehydration of the shoot. It is on these nonsucculent plants that greater attention needs to foeus for elucidating the full range of desert plant adaptations. Among the many nonsucculent life forms in the desert are drought-deciduous shrubs and subshrubs, evergreen shrubs, phreatophytes (trees and shrubs with roots that tap the water table), green-stemmed shrubs and trees, perennial grasses, and ephemerals, inc1uding annuals and geophytes. T 0 say that these plants eseape, evade, or endure drought is not totally incorrect, but it is somewhat outmoded, as recent highly precise observations on the physiological performance of plant organs indicate. Today's desert plant biologists have been trained to keep in mind that each photosynthetic organ must deal simultaneously with 911

stoma

cutic le

I

-"-

)';rlZ

upper epidermis

ar[,

vascul bund le

substomatal chamber

~ r~

r~

I 11

t..q( ~

lAA'1.

)!.

3 ~

-g. ~

lower epider

lU

I

i( mi~ ' - ....../.......... "-substomatal chamber

r;

11.

I

0-

stoma

Figure 1. Cross-section of a typical desert leaf. Stomata are seen on both leaf surfaces, and adaxial (upper) and abaxial (lower) palisade mesophyll are prescnt throughout the leaf. Each mesophylilayer can have a high density of palisade cells, often exceeding 7000/mm 2, but wirh greater than 90% of the cell wall exposed to intercellular air space. This design pro duces a high Am«/A ratio and facilitates high rates of CO 2 diffusion into the photosynthetic cells.

several functions or problems: harvesting sunlight and performing photosynthesis, regulating its energy budget, managing water relations, and, perhaps, mediating predator resistanee. Emphasizing water relations over the other funetions provides a poor and often misleading interpretation about faetors controlling evolution of desert plant structure.

Maximizing net photosynthesis When analyzing desert plant adaptations, it is important to remember that C0 1 uptake for photosynthesis and water vapor loss via transpiration are both regulated by the guard cells of stomata; therefore, restrieting one gas will also restrietthe other. This situation is often deseribed as a dilemma facing the leaf, in that opening stomata for photosynthesis earries a heavy eost in water vapor loss. Conversely, a typieal plant can keep its stomata closed and thereby lose no water vapor, but during that time it also fixes no atrnospheric C0 1 and henee rnakes na sugar. Describing desert plant life as a defensive strategy that prevents high water loss

912

draws attention away trom an offensive strategy, that is, to maximize photosynthesis as long as environmental eonditions are adequate. For most desert plants, the offensive, opportunistie strategy appears to be the key for interpreting many struetural adaptations of desert plants. Some of the best dues about general photosynthetie strategies of desert plants have been diseovered in the anatomieal design of nonsuceulent leaves. Leaves from the tallest phreatophytie trees to low subshrubs and the tiniest winter annuals share, with rare exception, a common pattern. The standard desert leaf is amphistomatie, that is, stomata are present on both surfaees, at a relatively high density, and the mesophyll is isolateral, with chloroplastrich palisade eells facing eaeh surface and little or no spongy mesophyll in the center (Figure 1). The palisade mesophyll cells are narrow, so that there are many thousands of these cylindrieal eells per square millimeter of leaf surface, generally arranged in four or more layers with minimal contaet between the eells. A leaf with such a high surface density of palisade eells is described

as having a very high Ame'IA ratio (Nobel 1991b). By maximizing internal mesophyll eell surfaee area, that is, by attaining a high Ame'IA ratio, a leafean have more rapid CO 2 diffusion across the cell wall (liquid phase) into the eeUs, and thereby experience a higher rate of photosynthesis. A leaf with more densely paeked mesophyll cells also possesses more chloroplasts per unir area. Hence, if sunlight is not limiting, a relatively thiek, narrow leaf would be capable of the same total photosynthetic output as a much wider, but thinner, leaf. Oceurrence of stomata at high densities, and often in equal numbers on both leaf surfaces, at first seems counterintuitive to anyone accustomed to thinking that water conservation is paramount for xerophytes (plants that oceupy dry habitats). If losing extremely little water vapor was the principal strategy of a desert leaf, one would first expeet to find low densities of stomata and perhaps stomata that are deeply sunken, presumably sheltered from high rates of transpiration. However, arecent survey of nonsueculent leaves in the earth's lowland deserts found that deeply sunken stomata are exceedingly rare (Gibson 1996). What at first seerns counterintuitive is now beeoming obvious: Stomata I patterns are mainly a design to maximize rates of C0 1 uptake, not to prevent water loss. The diffusion pathway of C0 1 under optimal conditions can be improved substantially by splitting stomata on the two surfaees rather than by plaeing thern all on one surface, beeause gas can diffuse rapidly across the two boundary layers in parallel (Mott et al. 1982, Nobel 1991b, Gibson 1996). For the thickest leaves, having stomata on both surfaees may also shorten the diffusion pathway of C0 1 to individual mesophyll eeUs, another means of enhaneing photosynrhetie rate (Parkhurst 1994). Many species living in deserts exhibit the C 4 photosynthetie pathway. These indude most summer annuals (Mulroy and Runde11977) and summer-active bunchgrasses, especially wherever a summer rainfall regime dominates (Hattersley 1983). Plants with C 4 photosynthesis grow rapidly under high temperatures and full sun,

BioScience Val. 48 No. 11

T able 1. Maximum rates of photosynthesis {P maJ for common nonsucculent species in the southern desert regions of the United States. ' P",n' lJmol·m Taxon and growth form Woody plants with drought-deciduous leaves Acamptopappus sphaerocephalus Ambrosia deltoidea F.nce!ia farinosa Encelia frutescens Ellergreen shrubs Atrlplex canescens Atriplex hymenelytra LaTTea fridentata Phreatophytes Acacia greggii Olneya tesota Prosopis glandulosa Perennial grasses Pleuraphis rigida SwaUenia aJexandrae Ephrmerals and hemicryptophytes Abronia viUosa

Amsinckia tessellata Astragalus jaegerianus Astragalus lentiginosus var. {rcmontii Astragalus lentiginosus var. rnicans Camissonia brevipes Camissonia claviformis Chorizanthe rigida Cueurbita palmata Dieoria caneseens Eremalcbe rotundifolia Eriogonum de{lexum Geraea canescens Lotus salsu[?inosus I.upinus arizonicus Lupinus odoratus Mohavea ht:eviflora Oenothera californica ssp. eurekensis Oenothera deltoides PaTafoxia arida Phacelin calthi{olia Phacelia crennlata PhacdllJ fremontii R umex hymenosepalus Salvia columhariae Tidestromia oblongifolia Nonsuceulent aphyllous shrubs al1d rrees Bebhia juncen B. juncea Caesalpinia virgata Cercidium floridum Cereidium mierophyllum Chrysothamnus panieufatus Gutierrez.a microcephala Hymenoclea salsola Porophyllum gracile Psorothamnus spinosus Salazaria mexicana Senn« arma!a Stephanomeria pauäflora Thamnosma m()ntana

Lcaf

2' S -1

Stem b

Reference

25.0

Ehleringer et al. (1987) Szarek and Woodhouse (1977) Monsoll er al. (1992) Ehleringer er al. (1987)

19.9 20.9 21.3

Sharifi et a1. (1997) Pearcy et a1. (1974) Franeo et a1. (1994)

10.0 7.5 25.9

Szarek and Woodhouse (1978) Szarek and Woodhouse (1977) Sosebee and Wan (1989)

38 29

Nobel (1980) Pavlik (1980)

39.8

Sh.:lrifi ;md Gibson (unpublished data) Shanfi and Gibson (unpublished dara) Gibson et al. (1998) Gihson et al. (1998) Pavlik (1980) Seemann er a1. (1980) Monney et al. (1976) Werk et al. (1983) Sharifi and Gihson (unpublished data) Pavlik (1980) Mooney er al. (1981) Werk et al. (1983) Mooney et al. (1981) Werk et a1. (1983) Armond and Mooney (197!l) Sharifi and Gihson (unpublished data) Werk et al. (1983) Pavlik (1980) Sharifi and Gibsnn (unpublished dara) Forseth et al. (1984) Werk et a1. (1983) Werk et aJ. (1983) Werk et a1. (1983) Sharifi .:lud Gibson (unpuhlished dara) Werk et al. (1983) Björkman er al. (1972)

9.2 23.8

28

44.4

9.'

22.4

28 41 59

29.2 42.5 38

43

24.9 35.3 26.5 35 41.6 37.9

36

,')5.9

42 27.3 23.0 24.3 21.!l 28.2

37

20.7

B.S 23.6 21.1

24.!l 37.7 15.1

22.3 21.7

13.2 11.2 10 15.2 17.9

16.6

23.9 11.1 16.7 10.7 23.3 10.5

Ehleringer et a1. (1987) Comstock et a1. (1988) Nilsen et al. (1996) Adams and Srrain (1968) Szarek ami woodhouse (1978) Ehleringer et al. (1987) Ehleringer et al. (1987) F.hleringer et al. (1987) Ehleringer et al. 1987) Nilsen et al. (1989) Ehleringer et al. (1987) Nilsen et al. (1996) Ehleringer et al. (1987) Ehleringer et al. (1987)

'P""x was measured when soil water was not limiting (high leaf water potential) in native habitat under the most favorable, unmanipulated ambient conditions of the growing season, wirh saturated slIlllight and low vapor pressure deficits. "For green sterns, projeeted area was used for calculations exccp! Nilsen et al. (1996), who ll~ed total surfac:e area to calculate Pm .- for Caesalpinia vlrgata.

November 1998

913

and they efficiently convert CO 2 into glucose because the compartmentalization of carbon fixation and photosynthetic carbon reduction into two different cells eliminates wasteful photo respiration (Taiz and Zeiger 1991). Desert C 4 species are amphistomatic, although mesophyll structure differs radically from C 3 leaves because C 4 species exhibit Kranz anatomy, having a highly derived bundle sheath within which glucose is manufactured (Johnson and Brown 1973, Dengier er al. 1985). High rates of net photosynthesis amongnonsucculent C 3 and C 4 desert plants have been measured by ecophysiologists in the field under optimal field conditions of abundant soil water, bright sun, and mild air temperatures (Table 1). For many woody species, maximum net rate of photosynthesis (Pm:l.J.J of leaves in nature can exceed 20 /lmol'm-2 's- 1, which is equivalent to that of many row crops under irrigation; in desert hcrhs, P max ohen exceeds 30 /lmol· m-2 's- 1 , which is at the high end of the range for any C3 plant. Some of the highest rates of net photosynthesis have been measured under optimal field conditions among plants from the deserts of the southwestern United States; for example, Camissonia claviformis, a C 3 winter annual growing in Death Valley, had a maximal photosynthetic rate of 59 /lmol'm- 2 's-1 (Mooney et a1. 1976). Under controlled conditions, the highest photosynthetic capacities ever recorded at ambient C0 1 concentrations are 81 /lmol·m- 2 ·s- l , for the C 4 summer annual Amaranthus palmeri from the Sonoran Desert (Ehleringer 1983), and 67 J..ln10I·m-2 ·s-\ for the springtime-active C 4 bunchgrass Pleuraphis rigida from the Ca lifornia deserts (Nobel 1980). For numerous winter annuals, la bora tory photosynthetic rates of more than 50 IJ,mol'm-1 's-1 have been measured (Forseth and Ehleringer 1983, Werk er al. 1983). As soils dry and leaves are therefore under greater water stress, photosynthetic rates of desert species are reduced but still comparatively high. At levels of water stress at which the leaves of most nondesert plants would experience litde or no net carbon gain, leaves of desert plants still experience substantial rates of 914

CO z uptake. This ability reflects one of the most important adaptations of desert plants: namely, that through physiological-and not structuralmechanisms they can open stomata at much greater water stress than in typical plants and thereby take advantage of leaf architecture designed for high photosynthetic rate. However, not all desert plants have high rates of photosynthesis. Among the woody desert plants, investigators have measured some relatively low rates ofleaf photosynthesis. Most notable is creosote bush, Larrea tridentata, the dominant evergreen shrub of North American warm deserts. The typical values of creosote bush P max under optimal field conditions are 14-16 /lmol·m-2 ·s-1 (Rundei .nd Sh.rifi 1993). Bur despite its structure (i.e., amphistomatic leaves and high Am··'/A ratio), which could enable reasonably high P max' creosote bush typically photosynthesizes at one-fourth that rate during droughr (Sharifi er al. 1997). Creosote bush, and probably other evergreen desert shrubs as well, may have a strategy for gas exchange in which high instantaneous rates of net carbon ga in have been traded for much lower rates of net carbon gain during nearly every day of the year. This slow but steady pattern of gas exchange keeps pace with the continuous cost of maintaining evergreen leaves (Ehleringer 1985).

Leaf size Leaves and leaflets of desert plants are smaller than those of species growing in wetter habitats, a condition that is termed microphylly. For example, 89 % of nonsucculent woody plants from the deserts of southern California have photosynthetic surfaces less than 10 mm wide, and in 76%, mean width is less than 5 mm (Gibson 1996). Winter annuals and grasses from the same desert communities are also mostly narrow or microphyllous. A narrow-Ieaf design has often been seen as a strategy to reduce the total transpiring surface area of the canopy-that is, as an adaptation to limit total water loss. This condusion might be valid if total canopy leaf area of a desert shrub is less than that of shrubs from different habi-

tats. Indeed, a sharp reduction of leaf area of a desert plant's canopy is a function of seasonally declining plant water potential (Srnith et al. 1995). During desert drought, the smallieaf area of woody plants certainly reduces water loss and prevents development of dangerously low shoot water potentials (Orshan 1954, Nilsen er al. 1984). In addition, many drought-deciduous woody plants, called shoot-shedders, experience a major loss of leaf area from the canopy during drought (Zoh.ry .nd Orshan 1954). Even certain evergreens lose as much as 70% of their leaf biomass during droughr (Nilsen er al. 1984), .nd when drought lasts multiple years only the most distal leaves may remain on the shoots. However, during the most rapid growth, a desert shrub may have so rnany narrow leaves that totalleaf area is actually nearly the same as that of equalsized plants with fewer broad leaves. What, then, are the benefits for a desert leaf of being narrow rather than broad? A major explanation for desert microphylly is that narrow leaves can be maintained dose to ambient temperature and below lethai tissue temperature without substantial transpiration, which dissipates heat from the leafby evaporative cooling. Using energy budget equations for a hot summer day in the desert, computer predictions of expected temperature can be made for leaves of three different widths (Figure 2; Gates and Papian 1971). In still air, a broad, sunlit leaf with closed stomates could experience tissue temperatures exceeding 55°C, far above lethai temperature for that leaf, and an extreme1y high rate of transpiration would be required to cool that leaf below the lethai temperature. A leaf of intermediate width would still be likely to exceed lethai temperature without substantial transpirational cooling. Howevcr, the microphyllous leaf would, on the same hot day, remain be10w lethai temperatures and within several degrees of ambient temperature, even without transpiration. If a narrow leaf can be kept cooler than a broader leaf, then the narrow leaf can be expected to have a reduced rate of transpiration. Because

BioScience Val. 48 No. 11

water vapor concentration within the leaf is temperature dependent, a cooler leaf results in a smaller water vapor gradient to the surrounding air and, consequently, in a smaller driving force for transpirational water 1055 (Nobel 1991b). Thus, although early investigators were correet in their hypothesis that narrow leaves are an adaptation to conserve water, the mechanism is hiophysical, rather than simple leaf-area reduction. A cooler narrow leaf also has an advantage on a hot day hecause it is doser to the thermal optimum for photosynthetic enzymes, which tends to be around 25-35 oe. Moreover, lowering leaf temperature also reduces respiratory costs. Having narrow rather than broad leaves mayaiso have major effects on instantaneous and total daily illumination of leaves in inner regions of a shrub canopy. It is possible that total sunlight interception by a microphyllous plant may be improved over one with broader leaves, which produce dense self-shading; therefore, microphylly may sometimes be part of an architectural design to maximize light harvesting for the entire shrub. One prediction suggested by Figure 2 is that broad leaves should exist on a desert plant if the plant taps a constarrt supply of soil water that can be effectively delivered to the leaves. For example, plants such as deep-rooted phreatophytes and shal1ow-rooted dcsert palms at a spring or oasis would have broad leaves; a broad-Ieaved plant mayaiso flourish during a season when ambieot temperatures are relatively low, or whcre surface soil moisture is high enough to meet demands oE trans pirational cooling for a broad-leaved plant. After heavy thunderstorms, fully hydra ted, summer-active broadleaved perennials of the Sonoran Desert have been ahle to keep their leaves more than 10 oe cooler than afternoon summer air temperatures by transpiring at very high rates (Smith 1978). Even broad-Ieaved herbs growing in springtime, such as wild rhubarb, Rumex hymenosepalus, in North American deserts, can be 5 oe cooler than the air temperature when they are experiencing their highest rate of photosynthesis (M. Rasoul Sharifi and Arthur C. Gibson, un-

November 1998

Figure 2. Predicted temperatures of desert plant leaves of various areas. Temperatures for leaves 1 cm, 5 cm, and 20 cm in width were computed using an energy budget equation for a cloudless summer dar in the desert wirh ambient ternperature of 40 oe, relatively humidity of20%, still air (wind ve10city 0.1 rn/s), and direct solar irradiance fm full sun (after Gates andPapian 1971). The broadestleafwould require a substantial rate of evaporation via transpiration to reduce leaf ternperature to bclow lcthallimits, typically less than 47 oe. Conversely, a broad leaf can also experience su bstantial undcr-tempera· tures (less than ambientl when stomata are wide open and the rate of transpiration is exceedingly high.

To.al absorbed enc' 9Y 838W m-~

17.5

Air Temperature 40' C Relative humidity 20% Wind velooity 0. 1 m

c 0

."" .~

$"

7.5

•c

0= 5.0 25

o '-;--"""",~--;;;---'-~~'--;';c-~ 30

35

40

45

50

55

Leaf temperature. oe

published data). Thus, it appears that broad-Iea ved desert plants show little attempt to limit water when soil moisture is available-indeed, to do so would mean c10sing stomata, which would sharply reduce net photosynthesis and, in hot weather, cause the lcaves to die via heating.

Orientation of leaves Further evidence that the strategy of typical desert plants is to maximize photosynthetic rate comes from the orientation strategies that maximize

~\

j

exposure of leaves to sunlight (Wainwright1977,MooneyandEhleringer 1978, Ehleringer 1985). A number of desert dicotyledonous annuals exhibit diaheliotropism (Figure 3), a precise form of solar tracking by Ieaves that allows them to be orierrted perpendicular to direct sunlight at all times throughout the day. Diaheliotropism maximizes interception of solar radiation and, thus, total daily photosynthesis, while maximizing solar heating from infrared radiation. By contrast to the leaves of many desert annuals, leaves of desert shrubs do not track the sun but remain in a fixed position. For example, somc evergreen shrubs have fixed leaves that are approximately vertical.

/~

Figure 3. Diaheliotropic leaves of Eremalche rotundifolia (desert five-spot). Such leaves track the sun's movell1cnt throughout thc day, rcmaining perpcndicular to direct sunlight on a cloudless day and thereby maximizing interception of solar irradiance for cach minute, If a leaf blade were fixed in position to near the horizontal (middle plant), it would receive most solar irradiance during the middle of the dar and sub:ottantially lower amounts during early morning and late afternoon. If a leaf blade were fixed in a ncarly vertical position (plant at the right or Jeft), it would receive maximal solar irradiance during early morning and latc afternoon but would not receive direct illumination at midday.

915

E. farinosa forms densely hairy, sil-

Figure 4. Scanning dectron micrographs of ehe upper leaf surface of creosote bush, Larrea tridentata, from North American deserts. (Ieft) A leaf wirh its restn coaring. (right) A leaf

after resin was dissolved with oeganie solvents, reveaLing scartered trichomes on the sucface and a high densiry of stomata. Stomatal density appears (0 be much lower on the resio-covered specimen, suggesring that many stomata a re sealed by resin, thus lowering p""~. Bar = 2S Ilm . Micrograph ar leEr reprinted from Rundei and Gibson (1996).

Whereas a possible interpretation of vertical orientation of leaves is to reduce summer heating during midday (Gates 1980), ecophysiologieal evidence suggests that vertical orientation is actua lly a strategy to maximize photosynthesis, especially during early morning ho urs, when diu rna l shoo r watet potential and vapot ptessute of the air are highest; even whole evergreen branehes of cteosote bush may show a pattern of fixed otientation fot maxim izing photosy nthesis (Neu feld et al. 19 88). Fot a microphyllous leaf, whieh remains at nearly the same temperature as the air around it, vertieal orien tation should yield 0 0 significa nt relid from midday summer hea t, but being edge-on to the sun ar noon may he\p the leaves of cerrain planrs to reduce deletetious photoinhibition during severe water stress, a proeess in whi ch hi gh levels of light aetually interfere wi th photosynthetic processes (Björkman er a1. 1980, Smith .nd Nobel 1986). Desett leaves tend to have relatively high Jight~saturation values. The few lighr-response curves that have been conducted under field conditions ind icare that in common dese rt perennials, light saturati on oee urs above ha lf full SU ll ; foe some C4 and C 3 herbaceous speeies, satu916

ration may not be achieved even at full sun (Mooney et al. 1976, Ehleringer 1983). This phenomenon raises the que stion of whet her desert planrs can be "light limited" even though they inhabit an environment in whieh it was assumed th ar a rypical leaf receives all of rhe necessary photons of light throughout thc day. The first evidence of light limitation arose when Nobel (1983) demonstrated that re moving spines from desert eacti results in a do ub ling of biomass. This finding revealed th ac the tradeoff fot havi ng spines. presuma bl y for protection again st herbivores, is slower vegetative growth due to seU-shading. Similarly, [here is substantialself-shadingin many desen plant eanopies, so that lower lea ves are mostly operating ar low photon flux density (PFD) . Low interior light rnay be why lower leaves are shed in such plants as creosote bush, in which even the upperrnost canopy leaves appear to be li ght limited. The most thoroug hly documented case of a desert leaf thar shows an eHeet of light limitation on the meso~ phyll is that of the desett brittleb ush, Encelia farinosa , of western North America (Eh leringe r er a l. 1976, Ehleringer and Björkman 1978a, 1978 b, Ehler inger and Mooney 1978, Z han get a l. 1995). Drought-stressed

very leaves. The highly reflective silvery eovering pre vents infrared radiation from causing extra headng of the leaf while at the same time redueing the level of photon flux density reaching the chloroplasts> thereby lowering oet phorosynthesis by greater than one-half. Their thiek hair covering is an additional adaptation to avoid lethai leaf temperature; it results in a cooler leaf and thus in a redueed transpiration rate. as discussed earlier. The thorough understa nding of the hairy leaves of E. farinosa now helps us to reject previous thinking, which suggested that the prcsence of hairs on desert plants is an adaptation that slows water vapor diffusion across the boundaty layer aod, in that way, slows leaf u anspiration, Bur the hairs ha ve a eost in a sha rp reducrion of net phorosynthesis, because the hairs that refleet infra red radiation also refleet some of the visi ble light absorbed by the photosynrhetie pigments.

Water conservation by leaves Since a seminal aceount on xerophytes by Maximov (1931), biologists have known that desert plants ean exhibit high transpiration rates when soil water is plentiful and thar they are more conscrvative in their water use as the soil dties, For a typieal nonsucculent desert plant, stomata open daily at levels of leaf hydration that plants from mesic habitats would not rolerate beca use they would ha ve resulted in leaf abscissioo. Manymicrophyllousdesen shrubsshowsubsta ntial uptake ofC02 even when soils are relatively dry and shoot water potentials drop to - 3 or even -5 MPa, weB below the maxi mum leaf water stress resistance of typical mesophytes. The remarkable evergreen ereosote bush continues to experience positive net photosynthesis when shoot water potentials are below-8 MPa (Odeninger al. 1974). Having the capability to open stomata at such low leaf wa cer potentials is a physiological, not a structural, adapt atio n. This ca pability to eontinue opening stomates during extreme water stress is probably whar predisposes these plant species [Q live in arid habitats. BioScience Vol. 48 No, 11

Nonsueeulenr desert leaves do not co nrain special strucrural dev ices for storing or conser ving water. In gen~ eral, such leaves possess no water reservoirs (that is, cells with large va cuoles to supply water to leaves duri ng periods of peak tra nspira~ rion); insread, rhey have narrow mesophyll ceils with small volumes and, rherefore, the high surfaee~ro~ volurne rarios thar are oprimal for rapid CO 2 uprake. Cutide coars the epidermis, of course, bur this wax layer is not noticeably thicker than that of plants from other ecosystems, although it does appear to be more effective in reducing cuticular tran~ spiration than cutiele on plants from m.n y other habitats (NoheI1991b). The most dramatic speciaJization fo r water conserva ri o n is leaves coared with resin, such as ereosote bush (Meinzer et al. 1990). Resin on leaf surfaces of creosote bush (Fig~ ure 4), 2-4 ~m in thickness and present wirhin the epidermal ceHs, not ooly re duces epicuticular transpiration bur also repels herbivores. Resin can seal stomata c1osed, reducing water loss bur the n also the rate of net phorosynthes is. This effeet may explain why few desert speeies have resin-covered leaves. Water-use efficiency is one measure of desert planrs' water conservati on, because as the shoot experiences greater warer stress, the ratio of moles of CO l gained fO moles of H 2 0 lost per photosynthetic surface area would be expected t O inerease if plants use water more efficiently as water stress increases. There are severa l ways to express water-use effieiency: instanraneous water-use effideney, which is the ratio of assimilation ro transpiration (AlE); intrinsic W3te r-use efficiency. which is the ratio of assimilation to stomata I eooductanee (AIg s ); and inregrated wateruse efficiency, that is, warer-use effi ~ cieney for an entire growing season, which is estimated by using carbonisotope (aUC) ratios (higher values of aUC represent higher water-use effieiency; Rundei er al. 1988). Ooe broad generalil.ati on is that no nsueeuleot C 3 plants ha ve the lowesr water-useeffieiency, nonsucculent C .. plants have imermediate wateruse effieiency! and succ ulent CAM plants have the highest water-use effieieney (Nobel 1991., 1991 b). Some

November 1998

Figure S. C ross~sect ions of the rnicrophyllous leaf and photosynthetic young stern ol cheesebush, Hymenoclea so/soja, horn the California deserts. (Idr) The leal has isolateral meso ph yll wirh strong deve lopmem of palisade layers. (f ight ) The stern also has la yers of palisade cells in the outer cortex, induding abundant intercellular air space between ce lls. Astoma (arrow) is associated with a spacious sub stomata 1air chamber through which C0 1 can diffuse tO the photosynthetic cells. Bars = 25 ~m. srudies indicate that longer~lived Cl desen plants use water more eUi· ciently than shorter-lived species, aod within a species, indi viduals gtOw· ing in drier microhabitats fend [ 0 have significantly higher integrared water-use efficiency than individuals growing e10se to or in washes (Ehleringer 1994, Smith et al. 1997). However, even among deser t ephemerals, relativel y high va lues of insranta neous and inrrinsic wateruse effidency ca n be observed when phorosynt hetic rates are very high and the vapor pressure difference between leaf and air is low. When soi! water reaches field capa city, such as dur ing a rainy sea son, noo succulent desert leaves use water sornewhat more efficientl y t han planrs of mo re mes ic habitats, because P tends to be so high in de se n le~~es. During the course of a day, the warer-useefficiencyof many desert plants characteristically decreases fr om midmorning, when net photosynth esis is typically highest, to lare afternooo, when the photosynthetic rate ca n be subsrantially lower but sto mates are still open. Therefore, not a single paradigm has been fo uod indicating that dese rt species a re necessaril y superior CO pla nts from wetter habitats in Iimiting wate r loss.

Sterns as photosynthetic organs The percentage of noosucculent speeies that have green sterns is not kno wn, but the numbe r of examples is constantly dimbing. In just the warm deserts of the world. rhe list includes woody species from 36 different dicotyledonous families plus the gymnosperm Ephedra (Gibson 1996). Many drought ·deciduous speeies, particularl y in the Asteraceae (sunflower fami ly), have photosynthetic yo ung stems. Vegetative sterns and inflorescence axes of herbs ca n also exhibit signifieant rates of CO 2 uptake. Because aphyll y (the leafless condition ) ha s often been explained as a straregy to recluce transpiration by eliminating leaf surface, photosynrhetic sterns of desert plants-which rend to be leafless o r to possess leaves rhat are eirher minute or vestigialappear at first glanee to be structural adaptations for water conservation (Gibson 198 3) . In realiry, a shrub or tree with large-diameter green sterns may display an impressive amounr of total green surface (Adams and Strain 1968). An alternative hyporhesis is that when sterns are the chief photosynthetic organ of the plant, leaves typicall y have been eliminated because they block sunlight from strik917

ing the stern surface (Gibson and Nobel 1986). A cylindrical stern is less efficient than a planar leaf for intercepting direct sunlight because only half of the stern can be fuHy illuminated at any particular moment. Most surface of astern would therefore be below light saturation, with a high degree of self-shading even in the absence of leaves. Modeling of canopy designs and tests of such models are needed to generalize about light interception by photosynthetie sterns. Although one might assurne that a green stern should lose relatively little water because it is essentially leafIcss, some desert plant sterns aetually have moderate transpiration rates, with stomatal conductance similar to that of leaves on the same plant (Comstoek and Ehleringer 1988). Consequently, these shrubs, with their high total stern surface area, also have relatively high water requirements, which rnay explain why the rnajority of these species either inhabit washes and other water drainage ehannels or have exceedingly deep roots. Many greenstemrned shrubs and trees have relatively wide vessels in the xylem to earry water frorn wet soil to the

transpiring shoots (Gibson 1996). Stomata 1densities are typically lower on green sterns than on desert leaves, but stomatal pores are slightly larger, permitting CO 2 diffusion into the airy photosynthetie cortex of the stcrn (Figure 5). Photosynthetic sterns are often eovcrcd with extremely thiek euticle, whose role has been interpreted as redueingtranspiration. However, the thick cuticle may also help to proteet the stern surface as it inereases in diameter and thereby stretehes or splits during seeondary growth. In many species, wax gives a grayish cast to the stern, suggesting that thick cuticle reflects some infrared radiation, thereby lowering stern temperature. Cuticle contains UV-absorbing phenolies and may prevent radiation darnage to long-lived photosynthetic organs. Nearly all photosynthetic sterns of desert plants have sunken stomata, and scientists have often eoncluded that sunken stomata, or stomata hidden in longitudinal furrows, are an adaptation to reduee transpiration. Any adaptation that lowers water vapor conductanee simultaneously lowers CO 2 conductanee and net photosynthesis, a result that is

Carbon isotope ratio, %0

Bebbiajuncea Chrysothamnus paniculatus Dyssodia porophylloides Gutierrezia microcephala G. sarothrae Hymenoc/ea salsola Lepidium fremontij Porophyllum gracile Psilostrophe cooperi Salazaria mexicana &necio jlaccidus var. douglasii Sphaeralcea parvijolia Stephanomeria parviflora Thamnosma montana

..t..leaf .stem

-28

-27

I

-26

I

I

A

A A A

-25

-24

I

I

•• A

A A

•• A

A. A

I

• • •• •

A A A

-23

• A

•

••

Figure 6. Difference in carbon-isotope fatios (S13C, %0) between teaf and photosynthetic stcm of 14 species of nonsucculent descrt perennials frorn the Sonoran Desert in western Arizona. Photosynrhetic stems consistently ha ve higher integrated wateruse efficiency (i.e., less negative ratios) than leaves on the same plant. Data are frorn Ehleringer er a1. (1987). 918

counterproductivc for astern that has beeome specialized as the chief photosynthetie organ. In addition, if sunken stomata have redueed stomatal conduetanee, then recession has its greatest proportional eHeet when growing eonditions are optimal, that is, when soils are wet. The effects of recessing stomata would, however, be negligible during drought, when the adaptation presumably would have been the most critieal for reducing transpiration. When the probable slowing eHeet of sunken stomata on water vapor loss is ca1culated, most types of sunken stomata yield physiologically insignifieant results beeause the distance of the pathway is substantially increased in only a few extreme examples (Gibson 1983, 1996). There are, in fact, reasons to hypothesize that sunken stomata are an adaptation to proteet the guard cells from direct contact with dry air, thereby preventing them from closing under arid conditions-that is, sunken stomata are an adaptation to improve CO 2 uptake by sterns. Moreover, furrowing of sterns greatly inereases the potential photosynthetic area of a eylinder and therefore may be an adaptation to rnaximize net photosynthesis. A desert plant that has both leaves and photosynthetic sterns has an exeellent overall strategy for annual earbon ga in. Because of its anatomical design, the leaf is optimally suited to intercept maximum solar irradiation, to take up maximum CO 2 with the lowest resistanee, and to regulate temperature better than astern. However, thin leaves are more pro ne to herbivory or damage by wind than are sterns, which are rnechanically strengthened. Integrated water-use eHiciency of green sterns is typically higher than that of the leaves (Figure 6), and stems are better at water conservation and therefore being persistent photosynthetic organs during drought rnonths (Comstoek and EhJeringer 1988) . Returning to the example of the dcscrt eaetus uescribed earlier, it is dear that the succulent green stem i5 another important photosynthetie strategy for desert plant life. Wherea5 CAM caeti are the dominant stern succulents of New World deserts, in Africa and southern Asia a similar

BioScience Vol. 48 No. 11

role is occupied hy succulent CAM euphorbs and stapeliads. Stomata of CAM succulents are closed and highly effective in limiting transpiration during daytime. But these plants have no evaporative cooling during the hottest midday hours and n1Ust tolerate incredibly high temperatures in their sterns, which become he at sinks that require the plant to have biochemical mechanisms for protoplasmic resistance to temperatures above 55 oe (Nobel 1988).

Beyond hot deserts Biologists anticipated that desert plants would provide examples of water conservation useful for showing students important relationships between plant structure and habitat, and, in particular, how plants prevent excessive loss of water to survive drought. However, many adaptations for water conservation in nonsucculent plants are largely physiological in nature. Biologists should be cautious in interpreting nonsueeulent photosynthetie structures as deviees for saving water and should keep in mi nd that a key strategy of desert plants is to maximize photosynthesis whenever soil water is available.1t is thcir high rate of photosynthesis, not a sharply lowered rate of transpiration, that probably explains why desert"plants have relatively high water-use effieicncy. While scientists are still working out the meanings of structural designs in hot-desert plants, it is important to be careful about how structural properties of plants from other dry habitats of the world are interpreted. Neighboring semiarid regions with Mediterranean-type climate have different environmental conditions than hot deserts. Alpine and arctic tundra, which sometimes are compared with deserts, have another climatic pattern, as do deserts that depend on adveetive fog foe moisture, and salt deserts, with dominant halophytes that resist salt stress. Certainly, a great effort is still needed to understand tife forms within each of these habitats before comparisons between habitats can be attempted.

References cited Adams MS, Strain BR. 1968. Phowsvnthesis in sterns and leaves of Cercidium fl~ridum.

November 1998

Spring and summer diurnal field response and relation to temperature. Oecologia Plantarum 3: 285-297. Armond PA, Mooney HA. 1978. Correlation of photosynthetic unit size and density with photosynthetic capacity. Carnegie Institution of Washington Yearbook 77: 234-237. Björkman 0, Pearcy RW, Harrison AT, Mooney HA. 1972. Photosynthetic adaptation 10 high temperatures: A field study in Death Valley, California. Scienee 175: 786-789. Bjärkman 0, Badger MR, Armond PA. 1980. Response and adaptation of photosynthesis to high remperature. Pages 223-249 in Turner NC, Kramer PJ, eds. Adaptations ofPlants to Water and High Temperature Stress. Ncw York: Wiley-Interscience. ComstockJP, Ehleringer JR. 1988. Contrastiog photosynthetic behavior in leave~ and twigs of Hymenoclea salsa/a, a greelltwigged warm desert shrub. American Journal of Botany 75: 1360-1370. ComstockJP, CooperTA, Ehleringer JR. 1988. Seasonal patterns of canopy developmem and carhon gain in nineteen warm descrt shrub species. Oecologia 75: 327-335. DengIer NG, Dengier RF, Hattersley PW. 1985. Differingontogenetic origins ofPCR ("Kranz") sheaths in lcaf blades of C 4 grasses (Poaceae). American Journal of Botany 72: 284-302. Ehleringcr J. 1983. Ecophysiology of Amaranthus palmeri, a Sonoran Desert summer allnllaL Oecologia 57: 107-112. ___ .1985. Annuals and perennials ofwarm deserts. Pages 162-180 in Chabot BF, Mooney HA, eds. Physiological Ecology of North American Plant Communities. Ncw York: Chapman and Hall. Ehleringer JR. 1994. Variation in gas exchange characteriHics among desert plants. Pages 361-392 in Schulze E-D, Caldwell MM, eds. Ecophysiology of Photosynthesis. Berlin: Springer-Verlag. Ehleringer JR, Björkman O. 1978a. A comparison of phorosynthetic charaeteristics of Encelia spccies possessing glabrous and puheseent leaves. Plant Physiology 62: 185-190. ___ . 1978h. Pubescence and leaf spectral eharacteristics in a desert shrub, Encelia farinosa. Oecologia 36: 151-162. Ehleringer JR, Mooney HA. 1978. Leaf hairs: Effects on physiologieal aetivity and adaptive value to a desert shruh. Oecologia 37: 18.1-200. Ehleringer JR, Björkman 0, Mooney HA. 1976. Leaf puhescence: Effects on absorptance and photosynthesis in a desert shrub. Science 192: 376-377. Ehleringer JR, Comstock JP, Cooper TA. 1987. Leaf-twig carbon isotope ratio dif· ferences in photosynthetic-twig desert shrubs. Oecologia 71: 318-320. Forseth IN, EhlcringcI JR. 1983. Ecophysiology of two solar tracking desert winter annual. 111. Gas exchange responses ro light, CO z and VPD in relation to long· term drought. Oecologia 57: 344-351. Forseth IN, Ehleringer JR, Werk KS, Cook es. 1984. Field water relations ofSonoran Desert annuals. Ecology 65: 1436-1444. Franeo AC, de Soyza AG, Virginia RA, Rc)'nolds JF, Whitford WG. 1994. Effcets

of plant size and water relations on gas exchange and growth of the desert shrub Larrea tridentata. Oecologia 97: 171-178. Gates DM. 1980. Biophysical Ecology. New York: Springer-Verlag. Gates DM, Papian LE. 1971. Atlas of Energy Budgets of Plant Leaves. London: Academie Press. Gibson AC. 1983. Anatomy of photosynthetic old stems of nonsuceulent dicotyJedons from North American deserts. Botanical Gazette 144: 347-362. ___ . 1996. Structure-Function Relations of WarmDesert Plants. Berlin: Springer· Verlag. Gibson AC, Nobel PS. 1986. The Cactus Primer. Cambridge (MA): Harvard University Press. Gibson AC, Sharifi MR, Runde! PW. 1998. Ecophysiologieal observations on Lane Mountain milkveteh, Astragafus jaegerianus (Fabaecae), a proposed endangered species of the Mojave Desert. Aliso 17: 77-82. Hattcrsley PW. 1983. The distribution of Cl and C 4 grasses in Australia in relation to climate. Oecologia 57: 113-128. Johnson SC, Brown WV. 1973. Grass leaf ultrastructural variations. AmericanJournal of Botany 60: 727-735. Maximov NA. 1931. The physiological significance of the xeromorphic strueture of plants.Journal of Ecology 19: 273-282. Meinzer FC., Wisdom CS, Gonzalez-Coloma A, RundeI PW, Shultz LM. 1990. Effects of leaf resin on stomatal behaviour and gas exchange of Larrea tridentata (De.) Cov. Funetional Ecology 4: 579-584. Monson RK, Smith SD, Gehring JL, Bowman WD, Szarek SR. 1992. l'hysiological dif· ferentiation within a Encelia farinosa population along a short topographie gradient in thc Sonoran Desert. Functional Ecology 6: 751-759. Moone)' HA, Ehleringer JR. 1978. The carhon gain henefits of solar traeking in a desert annual. Plant, Cell and Environment 1: 307-311. Moone)' HA, Ehleringer JR, Berry JA. 1976. High photosynthetic capa city of a winter desert annual in Death Vallcy. Science 194: 322-324. Mooney HA, Field C, Gulmon SL, Bazzaz FA. 19!H. Photosymhetie capaeity in relation to lcaf positions in descrt versus old-field annuals. Oeeologia 50: 109-112. Mott KA, Gibson AC, O'Leary JW. 1982. The adaptive significance of amphistomatic leaves. Plant, Cell and Environment 5: 455-460. Mulroy TW, RundeI PW. 1977. Annual plants: Adaptations tu desert environments. BioScience 27: 109-114. Neufeld HS, Meinzcr Fe, Wisdom CS, Sharifi MR, Runde! PW, Neufeld MS, Goldring Y, Cunningham GL. 1988. Canopy architecture of Larrea tridentata (DC.) Cov., a desert shrub: Foliage orientation and di· reet heam radiation interception. Oeeologia 75: 54-60. Nilsen ET, Meinzer FC, Runde! PW. 1989. Stern phorosynthesis in Psorothamnus spinosus (smoketree) in the Sonoran Desert of California. Oecologia 79: 193-197. Nilsen ET, RundeI PW, Sharifi MR. 1996. Diurnal gas exchange characteristics of twO ~tem photosynthesizing legumes in relation to the climate at two contrasting sites in the California dcscrt. Flora 191:

919

105-116. Nilsen ET, Sllarifi MR, Rundel PW. 1984. Comparative water relations of phreatf)phytes in the Sonoran Desert 01 California. Ecology 64: 1381-1393. Nohel PS. 1980. Water vapor condllCtanCe and C01 uptake for leaves of a C4 desert grass, Hilaria rigida. Ecology 61: 252-258. _ _ .1983. Spine influences on PAR imerceptian, stern temperature, and nocturnal acid accumulation by cacti. Plant, Cell and Environment 6: 153-159. _ _ .1988. Environmental Biology of Agaves and Cacti. Cambridg(;: (UK): Cambridge University Press. _ _ . 1991a. Achievable pruum:tivitie, of CAM planrs; basis for high vailles compared with C J and C,. Ncw Phytologist 119: 183-205. _ _ . 1991b. Physicochemical and Environmental Plant Physiology. San Diego (CA): Acadcmic Press. Odening WR, Strain BR, Oechel Wc. 1974. The effect of decreasing water potencial on net CO 2 exchange of intact desen shruos. Ecology 55: 1086-1095. Orshan G. 1954. Surface reduction and ICs signifkance as a hydroecologicfll facror. Journal of Ecology 42: 442-444. Parkhur~t DF.1994. Diffusion of CO 2 and other gases inside leaves. Ncw Phytologist 126: 449-479. Pavlik BM. 1980. Patterns of water potential and photosynthesis of desert sand dune plants, Eureka Valley, Calilornia. Oecologia 46: 147-154. Pearcy RW, Harrison AT, Mooney HA, Bjürkman O. 1974. Seasonal changes in net phorosynthesis of Atriplex hymenefy-

tra. shrub~ growing in Deach Valley, California. Oecologia 17: 111-121. Rundel PW, Ginson AC. 1996. F"cological Communities and Processes in a Mojavc Desert Ecosystem. Cambridge (UK): Cambridge University Press. Rundel PW, Sharifi MR. 1993. Carbon isotope discrimination and resource availability in the desert shrub Larrea tridentata.. Pages 173-185 in Ehleringer JR, Hall AE, Farquhar GD, eds. Stable Isotopes and Plant CarbonfWater Rela· tions. San Dicgo {CA}: Academic Press. Rundd PW, Ehleringer JR, Nagy KA. 1988. Stahlt Isotopes ill Ecological Research. Berlin: Springer-Verlag. SeemannJR, Field C, Berry JA. 1980. Photosynthetic capacity of desert winter annuals measured in situ. Carnegie Institution of Washington Yearbook 79: 146-147. Sharifi MR, Gihson AC, Runde! PW. 1997. Surface dust impacts on gas exchange in Mojave Desert shrubs. Journal of Applied Eco[ogy 34: 837-846. Smith SD, Nobel PS. 19S6. Deserts. Pages 13-62 in Baker NR, Long SP, cds. Photosynthesis in Cuntrasting Environments. Arusterdam: Elsevier. Smith SD, Herr C, LeJry K, Piorkowski J. 1995. Soil-plant w,lter relations in a Mojave Desert mixed shrub community: a comparison ofthree geomorphicsllrfaces. Journal of Arid Environments 29: 339351. Smith SD, Monson RK, Anderson JE. 1997. Physiological Ecology ofNorth American Desert rlants. Berlin: Springer-Verlag. Smith WK. 1978. Temperatures of desert plants: Anothcr perspective on the adapt-

ability of kaf sil.e. Science 201: 614-616. Sosebee RE, Wan C. 1989. Plantecophysiology: A case study of honey mesquite. Pages 103-118 in McArthur ED, Haferkamp MR, compilers. Proceedings of the Symposium on Shrub Ecophysiology alld Biotechnology. Ogden (UT): USDA ForeH Service Intermountain Research Station. Szarek SR, Woodhouse RM. 1977. Ecophysiological studies of Sonoran Desert plants. II. Seasonal photosynthesis patterns and primary production of Ambrosia deltoidea and Ollleya tesota. Oecologia 28: 365-375. _ _. 1978. Rcophysiological studies of SOlloran Desen plants. IV. Seasonal photosyntheric capacities of Acada greggjj and Cercidium micruphyllum. Oecologia 37: 221-229. Taiz L, Zeiger E. 1991. Plant Physiology. Redwood City (CA): BenjaminJCummings. Wainwright CM. 1977. Sun-tracking and related leaf movements in a desert lupine (Lupinus arizonicus). American Journal of Bouny 64: 1032-1041. Werk KS, Ehleringer J, Forseth IN, Cook 1983. Photosynthetic chatacteristi(~ of Sonoran Deserc winter annuals. Oecologia 59: 101-105. Zhang H, Sharifi MR, Nobel PS. 1995. Photosynthetic charactcristics of sun versus shade plants of Encelia farinosa as af· fected by photosynthetic photon Hux density, intercellular CO 2 concentration, leaf water potential, and leaf temperature. Australian Journal of Plant Physiology 22: 833-841. Zohary M, Orshan G. 1954. Rcological srudies in the vegetation of the Near East descm 11. Wadi Araba. Vegetatio 7: 15-37.

es.

NATIONAL DIABETES MONTH NOVEMBER

1998

Increase awareness ab out diabetic eye disease and he1p reduce severe vision 10ss and blindness. Of the 16 million people in America with diabetes, almost half will develop diabetic eye disease---a leading cause of blindness. Yet, more than 90 percent of these cases of blindness can be prevented through early detection and timely treatment.

To receive a FREE National Diabetes Month Kit, call toll-free

Please join the National Eye Health Education 1-800-869-2020. Program in a nationwide effort to increase awareness of diabetic eye disease and the importance of receiving a dilated NATIONAL National EYE eye exam at least once a year for . . Eye HEALTH Institute EDUCATION all people with diabetes. PRQGRAM

~E

®

UTiOUl 1ls m nlS GF HlltTM

National Diabetes Month is sponsored by th~ National I!ye Health Education Program Partnership. 1he Partnen;hip represents leading public and private organiLations that arc members of the National Eye Health Education Program, coordinated by the National Eyc Institute, National [mtitutes of Health.

920

BioScience Val. 48 No_ 11