Cactus forests: are giant crassulacean acid metabolism plants trees?

Volkmar Vareschi (1980) has spoken of succulents or cactus forests describing an arid plant community in northern Venezuela dominated by giant columnar cacti. He discusses this referring to a definition of forest by Dengler (1935) and argues that by branching in their upper parts these cacti may also develop kind of a canopy, although walking around in such “forests” I often got the impression that the sensation of being under a canopy is much more determined by Mimosaceae, Capparidaceae and Caesalpiniaceae trees mixing with the columnar cacti (see also Vareschi 1980; Fig. 1a,b). Similar life forms as these columnar cacti are the candelabrous giant euphorbias in Africa (Fig. 1c) and the Didieraceae of Madagascar. They are all performing crassulacean acid metabolism (CAM) but none of them has secondary growth based on a circumferential stem cambium. If this were the morphological definition of a tree, none of them is a tree. Then the only bona fide dicotyledonous trees having CAM are species of the genus Clusia (Clusiaceae, Guttiferae) but they perform CAM with their leaves (Lüttge 2006, 2007). Conversely, if we accept a physiognomic definition as Vareschi (1980) and also Menninger (1967) apply it we might well justify treatment of the giant stem succulent plants in the context of stem photosynthesis of trees.

Fig. 1
figure 1

a, b Aspects of a cactus forest (sensu Vareschi 1980) near Carora, Venezuela. c Woodland savanna with Euphorbia candelabrum Tremant ex Kotschy, Great Rift Valley, Ethiopia

Full size image

Phylogenetic relations

Phylogenetically we may ask the question if all stem succulent plants are performing CAM. Woody dicotyledonous bottle or barrel trees, such as those listed in Table 1 as well as Brachychiton (Sterculiaceae), Colpothrinax (Arecaceae, Palmae) and Phytolacca (Phytolaccaceae) are species with C3 photosynthesis. Their water storage serves adaptation to seasonally dry climates especially in savannas and cerrados. However, as far as I know, all fleshy stem succulents are CAM plants.

Table 1 Families and species of tropical woody stem-succulent plants. (Borchert and Rivera 2001, and personal observations)
Full size table

A key for understanding the evolution of stem CAM may lie in the family of the Cactaceae. The family has threes subfamilies, the Pereskioideae, the Opuntioideae and the Cactoideae. While the latter two are always fleshy stem succulents, the Pereskias have shrub or tree like physiognomy (Fig. 2). Moreover, the Pereskias and some of the Opuntioideae are developing leaves on their stems. Extant Pereskia is the most basal taxon of the family (Barthlott and Hunt 1993) and the ancestral forms of the family are thought to have been similar to the leaf-bearing Pereskias (Winter and Smith 1996b; Edwards and Diaz 2006). Leaves of Pereskias perform C3-photosynthesis but may develop a weak CAM activity under the stress of drought. This has the features of CAM cycling, where stomata are kept closed in the night but respiratory CO2 is re-fixed via phosphoenolpyruvate carboxylase (PEPC) and stored in the form of organic acid (mainly malic acid) as it is typical for the CO2 dark fixation of CAM, and organic acid decarboxylation provides CO2 as a substrate for assimilation in the subsequent day (Rayder and Ting 1981; Diaz and Medina 1984; Nobel and Hartsock 1987; Edwards and Diaz 2006). Such CAM-cycling has been suggested to be an initial step in the evolution of full CAM (Guralnick et al. 1986; Guralnick and Jackson 2001), where stomata are open during the night and CO2 is fixed by PEPC and stored in the form of organic acid in the cell vacuoles, and where the organic acid is remobilized again, decarboxylated and the CO2 regained fixed via ribulosebisophosphate carboxylase/oxygenase (RubisCO) behind closed stomata in the subsequent day. However, among the 17 species of Pereskia only one, i.e. P. horrida Kunth. (D.C.), displays a little stem photosynthesis (Edwards and Diaz 2006). It is a stimulating suggestion that a delayed and reduced formation of bark together with the evolution of stem stomata was an important evolutionary step towards the development of stem CAM in the Cactaceae (Edwards and Diaz 2006).

Fig. 2
figure 2

a Pereskia aculeata Mill. State of Rio de Janeiro, Brazil, in the flowering and leaf bearing phenological stage. b Pereskia guamacho F.A.C. Weber in the leafless phenological state

Full size image

Among the Opuntioideae there are also leafy species, e.g. Pereskiopsis porteri Britton et Rose, Quiabentia chacoensis Backeb., Austrocylindropuntia subulata (Muehlenpfordt) Backeb. They show drought induced shifts from C3-photosynthesis to CAM. Interestingly in the leaf-less Opuntia ficus-indica (L.) Miller young 2-week-old cladodes (flat succulent stems) show C3 patterns of photosynthesis. CAM capacity as indicated by the amplitude of diurnal oscillations of malic acid levels strongly increases as young cladodes mature from being carbon sinks to performing as carbon sources in the entire plant where only the mature cladodes exhibit full CAM. All the Cactoideae are leafless and perform stem CAM (Nobel and Hartsock 1986a, 1987; Wang et al. 1998).

These CAM performing stems of the Opuntioideae and Cactoideae of the Cactcaceae family are succulent and fleshy. They have an external photosynthetically active chlorenchyma and an internal water storing hydrenchyma of mainly non-green cells built up from cortex and pith of the stems (Haberlandt 1918). This type of stem succulence has emerged polyphyletically in several taxa, i.e. besides the Cactaceae also in the Asclepiadaceae, Apocynaceae, Asteraceae, Didieraceae, Euphorbiaceae and Vitaceae. All these stems are performing CAM while in the leaves if present as in some Opuntioideae there is C3-photosynthesis and only very weak CAM. Hence, we can conclude that stem CAM is a very typical trait of this particular life form.

This is corroborated by looking at some other taxa. Some of the stem succulent Euphorbiaceae performing CAM also are seasonally producing leaves (Batanouny et al. 1991) and the Didieraceae as well have leaves and photosynthesis is performed by both leaves and stems. CAM activity is related to carbon isotope ratios (δ13C) of the plant biomass, because primary CO2 fixation via PEPC (CAM) has a much lower discrimination against 13CO2 than RubisCO (C3-photosynthesis). δ13C values of leaves and stems of several Didieraceae species ranging from −16.6 to −22.7‰ and −15.8 to −21.4‰, respectively, suggest that on average leaves perhaps may make a little less use of a CAM-like primary CO2 fixation via PEPC than stems, but clearly both organs are performing CAM since with pure C3-photosynthesis much more negative values of δ13C would be obtained (Ziegler 1996). In the Asclepiadaceae Frerea indica Dalzell we also find leaves and stems, and here the leaves clearly perform C3-photosynthesis and the succulent stems perform CAM (Lange and Zuber 1977). A similar situation is observed in the Vitaceae Cissus quadrangularis L. (Ting et al. 1983) for which we have much information from excellent detailed studies (Virzo DeSanto and Bartoli 1996). The leaves largely perform C3-photosynthesis but have some intrinsic capacity for CAM with features of CAM cycling. A CAM-like gas exchange pattern is elicited by drought in the leaves of C. quadrangularis, however, the rates are very low. The stems always perform CAM (Fig. 3).

Fig. 3
figure 3

Time courses of net CO2 exchange, \( {\text{J}}_{{{\text{co}}_{{\text{2}}} }} , \) of stem (St) and leaves (L) of Cissus quadrangularis in well watered control plants and in leaves of plants droughted for 8 days (L*). CAM is clearly expressed in the stem with phase I of CAM, i.e. nocturnal CO2-uptake, phase III, i.e. day time stomatal closure and zero CO2 exchange when organic acid stored in the night is decarboxylated and the CO2 refixed via RubisCO behind the closed stomata, and phase IV, i.e. stomatal opening, CO2 uptake and fixation in a C3 mode via RubisCO in the afternoon. The transiently increased CO2 uptake often seen in the early light period (phase II of CAM) is not expressed in the stem here. Dark bar, dark period; open bar, light period. (Data of Virzo DeSanto and Bartoli 1996)

Full size image

Stem CAM in stem succulents

In leafy Opuntioideae at least 88% of the CO2 uptake over 24 h of day and night was by the leaves (Nobel and Hartsock 1986a). In the Vitaceae C. quadrangularis the photosynthetic surface was 19% leaves and 81% stem, and depending on availability of water to the plants night time CO2 fixation by the stem was 97–100% and by the leaves 3–0% and day time CO2 fixation was 34–27% and 66–73% by stem and leaves, respectively. Total CO2 uptake over 24 h was 71–75% and 29–25% in stems and leaves, respectively (Virzo DeSanto and Bartoli 1996). Here we can see that the relative contribution of stem CAM can be substantial, and of course, in leafless fleshy stem succulents stem CAM is the only income of the plants in terms of energy and carbon.

The succulent CAM stems are differentiated in a peripheral green photosynthetically active chorenchyma and a largely translucent water storing hydrenchyma. Stem CAM in the chlorenchyma of these succulents displays all the typical and well known aspects of CAM (Winter and Smith 1996a; Lüttge 2004). Nocturnal CO2 uptake and dark fixation is mediated by PEPC and the organic acid produced is accumulated in the vacuole of green photosynthetically active cells (phase I). In the early morning there is a transition phase II when gradually PEPC is down regulated and RubisCO is up regulated as it was also well characterized in the cactus Opuntia erinacea (Griffiths) L. Benson (Littlejohn and Ku 1984). Then stomata close, organic acid is remobiloized from the vacuoles, decarboxylated and the resulting CO2 assimilated via RubisCO in phase III. Eventually if water availability is permitting stomata can open again in the later afternoon for direct uptake and assimilation of atmospheric CO2 via RubisCO (Fig. 3). Quantitative analyses of light use in stems of cacti are given by Nobel and Hartsock (1983). In the cladodes of O. ficus-indica the transmittance of the epidermis and cuticle for photosynthetically active radiation was 0.57 and the transmittance through the chlorenchyma was 0.01 (Nobel and Hartsock 1983). Physiological activity occurred and carboxylating enzymes were present over a great depth of the chlorenchyma including the inner half where red and blue light were attenuated by 99% (Nobel et al. 1994b). Photosynthetic functions in the stems include photoinhibition (Adams et al. 1987) as well as protective mechanisms like photoprotective energy dissipation via the violaxanthin/zeaxanthin or xanthophylls cycle, e.g. in the stems of Cissus quinquangularis Chiov. (D’Ambrosio et al. 1994). In C. quinquangularis exposed to high irradiance a much stronger decrease of photochemical quenching of chlorophyll fluoresecence of chlorophyll a of photosytem II was observed than in the stems (D’Ambrosio et al. 1996).

Division of labour between the peripheral stem chlorenchyma and the central hydrenchyma

Succulence is very important for stem CAM. In six epiphytic cacti a close correlation was observed between nocturnal CO2 uptake and succulence parameters, such as stem thickness, fresh weight per unit area and H2O mass per unit area (Nobel and Hartsock 1990). The non-photosynthetically active central hydrenchyma which does not participate in the diurnal oscillations of organic acid levels of CAM (Lüttge et al. 1989) protects the chlorenchyma when water availability is limited under drought stress. In C. quadrangularis scanning electron microscopy showed that while hydrenchyma cells were shrinking during drought chlorenchyma cells kept their structural integrity and were able to maintain their hydration state (Virzo DeSanto et al. 1984). Also in the epiphytic cactus Hylocereus undatus (Haworth) Britton and Rose cell walls of the hydrenchyma were more flexible having a lower volumetric cell wall elasticity modulus, ε, than those of the chlorenchyma (Nobel 2006).

In the peripheral stem chlorenchyma of 9 months old seedlings of the giant columnar cactus Cereus validus Haworth nocturnal malate accumulation of CAM was accompanied by an increase in the osmotic pressure, п, and the turgor pressure, P (Lüttge and Nobel 1984, Fig. 4). Such diurnal changes in water relation parameters also determine diel timing of growth cycles in cladodes of Opuntias with a midday maximum of growth when turgor is still high but malate mobilisation also provides a source for production of carbohydrates (Gouws et al. 2005).

Fig. 4
figure 4

Diurnal changes of malate levels and osmotic pressure in the stem of young seedlings of Cereus validus and of turgor pressure (cellular hydrostatic pressure) in chlorenchyma cells (Lüttge and Nobel 1984)

Full size image

In the outer chlorenchyma of O. ficus-indica nocturnal organic acid accumulation was also shown to have osmotic consequences increasing п of the cells so that they can take up water from the hydrenchyma. The volumetric elasticity modulus, ε, of the cell walls is lower in the hydrenchyma cells so that they are more elastic and can take up and release, respectively, more water with smaller changes in turgor pressure, P, than the chlorenchyma cells. Diurnal dynamic cycles of radial internal water distribution in the stem succulent cacti are such that water moves more readily towards the water storage tissue at dusk and towards the chlorenchyma at dawn as supported by detailed quantitative assessments of water relation parameters п and P and using hydrogen isotopes 3H and 2H (tritium and deuterium, respectively) to assess mixing of water between the two tissues (Goldstein et al. 1991; Tissue et al. 1991).

Physiological ecology and factors determining stem CAM performance at the community level

Global distribution

The only stem succulent types in the Negev desert are a few species of Caralluma. The CAM plant Caralluma negevensis Zohary is restricted to shaded niches between rocks where dew produces after cold nights and CAM obviously allows only a low carbon gain (Lange et al. 1975). The global phytogeographic distribution of giant stem succulents in the arid zones of the globe, the Cactaceae on the American continent, the Euphorbiaceae in Africa and the Didieraceae in Madagascar has been assessed by Ellenberg (1981). These large succulents do require seasonal precipitation. They grow best with an average annual precipitation between 75 and 500 mm, i.e. they can live with low precipitation in arid environments but precipitation must be regular to allow them to refill their water storage parenchyma (see below in the section on drought). This explains the absence of succulents with stem CAM in the severe deserts of Asia, Australia and the inner Sahara with highly variable precipitation over many years (Ellenberg 1981).Thus, as also shown by the example of Caralluma in the Negev, very harsh deserts are not suitable habitats for stem CAM succulents and arido-active dwarf shrubs with C3 photosynthesis are more successful. Conversely, for example, the entire physiognomy of semi-deserts on the American continent with their regular winter rains is determined by giant cacti. An interesting advantage of stem CAM in drought deciduous dry forests is that the arborescent cacti, e.g. Opuntia excelsea Sanchez-Mejorda in a deciduous forest in western Mexico, can continue carbon acquisition and benefit from increased light availability when the C3-trees have shed their leaves (Lerdau et al. 1992).

Rascher et al. (2006) have established a cactus mesocosm in a sizeable glasshouse growth chamber approximating that of natural Sonoran Desert (western USA) ecosystems. Net CO2 exchange by the mesocosm was modulated by plant assimilation and soil respiration processes and reflected the typical CAM phases described above. However, the mean carbon budget of this CAM model desert system was negative with a release of 22.5 mmol CO2 m−2 day−1 and such a model-community exclusively with CAM did not develop the capacity to recycle CO2 from plant and soil respiration.

Drought

CAM is often considered as an ecophysiological adaptation to drought because the nocturnal opening of stomata with carbon acquisition in phase I when evaporative demand is low saves transpiratory loss of water. In the stem succulents the hydrenchyma plays a large role as an internal water reservoir (see above). This is particularly important during the so called CAM-idling. When drought stress temporarily becomes very severe in semi-deserts, stomata may also be kept closed during the nights. The plants then internally recycle respiratory CO2 via PEPC which prevents loss of carbon and minimises loss of water to cuticular transpiration when stomata are closed continuously day and night. Such CAM-idling has been well documented for cacti of the American semi deserts (Szarek et al. 1973; Szarek and Ting 1974 a,b; Nobel 1977; Ting and Rayder 1982; Nobel and Hartsock 1983; Holthe and Szarek 1985; Brulfert et al. 1987). The water lost by cuticular transpiration can be replaced to the chlorenchyma by the reserves in the hydrenchyma. Thus, plants with stem CAM can overcome many weeks and even months by CAM idling. However, when the cacti loose more than 50% of their total water the CAM-cycle ceases and the plants die (Holthe and Szarek 1985) if they can not refill the water reserves in their hydrenchyma which, however, may even be effective in response to rather low amounts of rain following a dry period (Szarek and Ting 1975). This is also a particular problem for seedling establishment at the community level, because with their larger surface : volume ratio young seedlings and small plants can only survive shorter periods of drought as their surface with cuticular transpiration is relatively larger and their water reserves are smaller than in large adult plants (Lüttge et al. 1989). This is also reflected in a census of plants relating age classes to climate records and showing that seedlings do not come up in all years and only survive when rainy seasons are well pronounced (Jordan and Nobel 1982).

Water reserves in the hydrenchyma can not only be used within the same stem organ but also over larger distances in the system of the whole plant via movements in the xylem and phloem. This is dominated by nocturnal changes in chlorenchyma osmotic pressure and not by transpiration (Schulte et al. 1989). Younger developing cladodes of O. ficus-indica, which initially may be C3-like performers, can be provided with water from the underlying mother cladodes (Pimienta-Barrios et al. 2005). They have lower water potentials than the mature underlying cladodes maintaining a water potential gradient for supply of water to the young cladodes via the xylem. Conversely the developing fruits which are supplied via the phloem have higher water potentials than the underlying cladodes (Nobel et al. 1994a).

Salinity

In relation to the fact that succulence and CAM are good adaptations to drought and osmotic stress it might be expected that the stem CAM species are also fit for stress due to salinity. However, there are no stem CAM halophytes, and cacti are quite sensitive to NaCl (Nobel 1983a). This observation is intriguing. However, where this has been studied in cacti it became evident that these stem CAM plants have different strategies to deal with the problem when they are subject to salinity stress. They are stress avoiders as they exclude salt from the stem chlorenchyma and hydrenchyma at the root level (Nobel et al. 1984; Lüttge 2002b). They also may sacrifice their absorptive fine roots during periods of stress totally avoiding the salt stress and rapidly grow new roots when conditions improve (Lüttge et al. 1989).

Temperature

In the semi-deserts often large diurnal temperature differences are prevailing. It is noteworthy therefore, that cacti have a remarkable range of temperature tolerance. Spines of the cacti may play a role in temperature regulation as they moderate the diurnal temperature extremes of the stem surface. This could help to extend the ranges of certain cacti with appreciable spine coverage into colder and perhaps also warmer regions. However, the spines also have a disadvantage in reducing photosynthetic productivity by shading the stem chlorenchyma (Nobel 1983b).

The highest soil surface temperature tolerated by some cacti is 74°C on the bare ground where they grow, the highest temperature tolerance known for higher vascular plants (Nobel et al. 1986). The photosynthetic electron transport apparatus also tolerates longer periods at higher temperatures than any other vascular plant (Chetti and Nobel 1987). Acclimation to high temperatures is associated with protein synthesis (Kee and Nobel 1986; Nobel and de la Barrera 2003).

Nights in the semi-deserts can become quite cold and frequently subfreezing temperatures have to be overcome by cacti (Nobel and Loik 1990). Temperature optima for nocturnal CO2 acquitision of CAM plants are generally low. Depending on environmental conditions such as water supply, irradiance and especially temperature regimes during growth nocturnal temperature optima of between 11 and 23°C have been observed (Nobel and Hartsock 1984). The phytohormone abscisic acid (ABA) in relation to drought was shown to modulate freezing tolerance of cacti (Loik and Nobel 1993).

Nevertheless, the effective performance of cacti at sites with regularly occurring low night temperatures remains quite remarkable. The lowest temperatures tolerated are −10°C by O. ficus-indica and O. streptacantha Lem. and −24°C by O. humifusa (Raf.) Raf. (syn. O. compressa (Salisb.) J.F. Macbr. (Goldstein and Nobel 1994). Acid fluctuations of CAM are closely correlated with seasonal temperature variations (Littlejohn and Williams 1983). Cold tolerance is important for cacti that reach northern latitudes in southern Canada and the eastern United States of America where there is seasonal cold-hardening and acclimation to sub-freezing temperatures in the winter (Nobel and Smith 1983). Temperature and water regulation interact and water stress overrides temperature regulation in O. erinacea and Opuntia polyacantha Haw. (Gerwick and Williams 1978; Littlejohn and Willimans 1983). At the cellular level these low-temperature adaptations involve the production of osmolytes and compatible solutes, such as the sugars glucose, fructose, sucrose and mannitol (Goldstein and Nobel 1991,1994). Cacti do not perform supercooling. The only alternative to supercooling for remaining viable is external ice formation. In low-temperature acclimation there is formation of mucilage in the apoplast. Extracellular mucopolysaccharides serve as an extracellular ice nucleation agent. Moreover, they are hygroscopic and a passive H2O capacitor (Nobel et al. 1992a) reducing H2O efflux from the cells and hence damage due to too rapid dehydration during extracellular freezing. Extracellular nucleation of ice closer to the equilibrium freezing temperature makes cellular dehydration more gradual and therefore less damaging (Goldstein and Nobel 1991,1994; Loik and Nobel 1991). In the light in the early morning at temperatures below freezing photosynthetic quantum use efficiency is very low and during the day effects of exposition and solar irradiance are correlated with protective functions of the xanthophyll cycle in Opuntia stricta Haw. (Barker et al. 1998).

The performance of CAM under situations with regular sub-freezing temperatures during the nights is noteworthy because the complex metabolism of glycolysis providing phosphoenolpyruvate as CO2 acceptor, CO2 fixation via PEPC and active transport processes leading to vacuolar organic acid accumulation must take part in the night. This is particularly intriguing at very high altitudes in the tropics and their typical diurnal climate with “summer every day and winter every night” (Hedberg 1964). Keeley and Keeley (1989) studied the cacti Oroya peruviana (K. Schum.) Britton et Rose and Tephrocactus floccosus (Salm-Dyck) Backeb. (syn. Australocylindropuntia floccosa (Salm-Dyck) F. Ritter) in the Puna vegetation of the Andes of Central Peru at an elevation of 4,000–4,700 m a.s.l. Nocturnal malate accumulation was still observed in nights with an air temperature of −8°C and subepidermal temperatures in the chlorenchyma of −3°C. For T. floccosus there are a hairy and a hair-less form. The isolation by the hairs has the advantage of protecting from cooling while at the same time it has the disadvantage of shading and thus reducing photosynthetic activity. The plants optimize this dilemma in that hairy and hairless forms dominate populations at sites, where either of the factors is the more limiting one. Thus, in the observations of Keeley and Keeley (1989) where the hairless type was rare, the hairy morph had higher nocturnal malate accumulation, and where the glabrous type was frequent, the hairy type had lower malate accumulation.

Productivity

Productivity of CAM plants in natural vegetation is generally much lower than for C3 and C4 species. However, productivity in stem succulent CAM species may also be quite large and this is mainly due to daytime photosynthesis in phase IV (Green and Williams 1982; Acevedo et al. 1983; Luo and Nobel 1993; Nobel 1996). Well irrigated and fertilized Opuntia amiclea Ten. and O. ficus-indica had an average productivity of 46,000 kg ha−1 year−1, which is among the highest productivities reported for any plant species (Nobel et al. 1992b). Nobel (1991) (also Nobel and Hartsock 1986b) has developed an environmental productivity index based on data on soil water status, air temperature and photosynthetically active radiation and the respective responses of net CO2 uptake to model and predict the global perspectives for cultivation and also fathoming potential increases of arable land using such crops. Crops from cacti are the fruits and the cladodes as vegetables for human consumption as well as animal forage and fodder (Kluge and Ting 1978; Russel and Felker 1987; Nobel 1988,1996). A more exotic appliation is that of O. ficus-indica for an indirect source for the carmine like red pigment cochineal which is used in cosmetics and produced by a shieldlouse Dactylopius coccus feeding on the cacti (Kluge and Ting 1978).

Due to the much higher CO2 affinity of PEPC acting in primary CO2 fixation as compared to RubisCO CAM basically is a CO2 concentrating mechanism (Lüttge 2002a) where in phase III behind closed stomata in cacti such as Opuntia monacantha Haw., O. ficus-indica and Opuntia basilaris Engelm. et Bigelow internal CO2 concentrations of 0.12–2.50% may build up (Cockburn et al. 1979; Spalding et al. 1979). Therefore, one might expect that stem CAM plants are not much affected by the current man-made increase in atmospheric CO2 concentrations. However, most studies performed indicate a stimulation of growth by elevated CO2 concentrations (Nobel et al. 1994c; Drennan and Nobel 2000). Dry weight production of O. ficus-indica was on average 32% higher during experiments over 12 weeks in which ambient CO2 concentration was doubled (Cui and Nobel 1994). There was no downward acclimation during long-term exposure to elevated CO2 concentrations (Wang and Nobel 1996). Activity of various enzymes was increased including higher substrate affinity of PEPC and RubisCO activation (Nobel et al. 1996; Wang and Nobel 1996). Thus, elevated CO2 increases CO2 uptake and biomass production not only due to daytime CO2 fixation (RubisCO) but in spite of the basically much higher CO2-affinity of PEPC also due to night-time fixation (Cui et al. 1993). This relative benefit of phase I CO2 uptake is accelerated by conditions of stress, e.g. drought and extreme day/night temperatures (Raveh et al. 1995).