CAM photosynthesis in submerged aquatic plants.
Crassulacean acid metabolism--or CAM, as it is commonly known--is one of three recognized photosynthetic pathways. It involves nighttime fixation of carbon, largely into malic acid, which is temporarily stored, followed by daytime incorporation of [CO.sub.2]--derived from decarboxylation of malate--into the Calvin cycle. The name derives from the substantial diel change in organic acid content of photosynthetic organs and the fact that the pathway was originally studied in plants of the family Crassulaceae. In terrestrial species CAM is best represented in arid land floras, a fact generally understood to result from the greater water-use efficiency conferred upon plants with this photosynthetic pathway (Kluge & Ting, 1978). Thus, the report of CAM in a submerged aquatic plant (Keeley, 1981) was initially met with some skepticism.
The diel cycle of overnight acidification, followed by daytime deacidification (here denoted [Delta][H.sup.+]) of photosynthetic tissues is considered an essential and defining feature of CAM photosynthesis (Fig. 1). While [sup.14]C-labeling studies show that several dicarboxylic acids are produced during dark [CO.sub.2] fixation, malate(malic acid) is considered the primary acid involved in autotrophism (Luttge, 1995). Therefore, I begin with a survey of [Delta][H.sup.+] and Amalate reports for aquatic algae and macrophytes. This will be followed by a review of evidence for CAM in aquatic species with diel acid fluxes and associated ecological and physiological characteristics, and will conclude with a discussion of the distribution and evolution of aquatic CAM plants.
[Figure 1 ILLUSTRATION OMITTED]
III. Diel Acid Changes ([Delta][H.sup.+]) in Submerged Aquatic Plants
The first suggestion of CAM in an aquatic macrophyte was the report of weak acid accumulation and dark [CO.sub.2] fixation in Hydrilla verticillata (Holaday & Bowes, 1980), soon followed by a report of substantial [Delta][H.sup.+] and dark [CO.sub.2] fixation in Isoetes howellii (Keeley, 1981) [John Raven pointed out that Allsopp (1951) earlier reported high acid levels in Isoetes, although Allsopp did not observe diel changes]. Over the past 15 years there has been a plethora of published and unpublished reports on presence and absence of [Delta][H.sup.+] in aquatic plants (Table I). To date, 180 aquatic species have been tested; 69 species, distributed in 14 genera, have significant overnight accumulation of acids, ranging from 5 to 290 mmol [H.sup.+] [kg.sup.-1] fresh mass (FM). For comparison, terrestrial CAM plants commonly have [Delta][H.sup.+] levels [is less than] 100 and seldom [is greater than] 200 mmol [H.sup.+] [kg.sup.-1] FM (Kluge & Ting, 1978; Winter & Smith, 1995a).
[TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII]
Aquatic species in five genera stand out as having acid accumulation that is substantially higher than others and within the range of terrestrial CAM plants. These include the sporebearing Isoetes (Lycophyta: Isoetaceae) and flowering plants (Anthophyta), both monocots, Sagittaria (Alismataceae) and Vallisneria (Hydrocharitaceae), and dicots, Crassula (Crassulaceae), and Littorella (Plantaginaceae). In these genera there is further evidence, beyond just the [Delta][H.sup.+] reports, that points to CAM photosynthesis (Section V). The extent to which CAM is implicated in aquatic species with more limited [Delta][H.sup.+] (Table I), will be discussed in Section X.
Isoetes (Fig. 2) is the largest genus of aquatic CAM plants, with all 38 aquatic species tested showing substantial [Delta][H.sup.+] (Table I), with some species exhibiting [Delta][H.sup.+] levels comparable to the highest levels for terrestrial CAM plants; [Delta][H.sup.+] = 290 mmol [kg.sup.-1] FM or 62 mmol [m.sup.-2] total leaf area. The Isoetes tested represent a quarter of this worldwide genus (Tryon & Tryon, 1982) and include much of the geographical range and most all aquatic habitats occupied by the group (Section VII). These data suggest that all aquatic species in the genus may prove to be CAM; there are a few terrestrial species, some of which are not CAM (Section XII.A.1).
[Figure 2 ILLUSTRATION OMITTED]
Sagittaria comprises about 20 species, largely in the Americas. All are aquatic and four of six species tested have substantial [Delta][H.sup.+] and other characteristics of CAM and two species have low-level acid accumulation. Vallisneria is a genus of approximately six species, two of which have significant, although not consistent, [Delta][H.sup.+]. Crassula is a genus of more than 200 species. The vast majority are succulent terrestrial perennials with CAM, and are mostly endemic to South Africa. A small number of Crassula are diminutive annuals, which are distributed worldwide and include both aquatics with CAM and terrestrials, which are not CAM (Section XII.B). Littorella includes only three aquatic taxa distributed at high latitudes in Europe, North America, and South America. I agree with those who consider them to be subspecific varieties of L. uniflora, and in the remainder of this review I will refer to them simply as "Littorella."
IV. Criteria for CAM Photosynthesis
Biochemically, CAM requires nighttime fixation of inorganic carbon catalyzed by the cytoplasmic phosphoenolpyruvate carboxylase (PEPC). In order to be considered an autotrophic process this must be coupled with net uptake of [CO.sub.2]. The first stable product, malate, is transported across the vacuolar tonoplast as malic acid. During the day it is transported out of the vacuole and [CO.sub.2] is released by cytoplasmic and/or mitochondrial decarboxylases, followed immediately by refixation of [CO.sub.2] with the chloroplastic ribulose 1,5-biphosphate carboxylase, oxygenase (RUBISCO). All reactions occur within a single photosynthetic cell (Winter, 1985). Criteria for CAM include:
1. Dark fixation of [CO.sub.2] via [Beta]-carboxylation with malate(malic acid) the first stable product.
2. Overnight storage of malic acid with little metabolism of this product in the dark.
3. Daytime decarboxylation of malic acid, resulting in substantial diel changes in both acidity and malate concentrations.
4. Opposite diel pattern of overnight starch (or sugar) depletion.
5. Refixation of the [CO.sup.2] resulting from decarboxylation of malate into products of the Calvin or PCR (photosynthetic carbon reduction) cycle.
6. Sufficient PEPC activity to account for overnight acidification.
7. Sufficient decarboxylase activity to account for daytime deacidification.
8. Net uptake of [CO.sup.2] in the dark.
Other characteristics often associated with CAM--such as preference for arid habitats, leaf succulence, diel pattern of high stomatal conductance at night and low daytime conductance, stoichiometry of (1:2:1) for (dark-[CO.sup.2] uptake: [Delta][H.sup.+]:[Delta]malate), the daytime suppression of [Beta]-carboxylation, pyruvate [P.sub.i] dikinase activity, among others--are not strictly associated with the CAM pathway, in either terrestrial or aquatic floras.
V. Evidence of the CAM Pathway in Aquatic Plants
A. DARK FIXATION
Steady-state [sup.14]C-labeling in the dark shows that all five of the genera Isoetes Sagittaria, Vallisneria, Crassula, and Littorella exhibit substantial dark fixation into malate (Table II). Presumably this is via [Beta]-carboxylation by the [C.sub.4] enzyme PEPC [as demonstrated for Vallisneria spiralis by Helder and van Harmelen (1982)], although detailed studies of C-atom position of the [sup.14]C-label have not been done for other aquatics (as is true of most terrestrial CAM species).
Table II Dark fixation products following a 3 h [sup.14][CO.sub.2]-pulse and after a 9 h [sup.14][CO.sub.2]-free chase in the dark (from Keeley, unpubl. data)
Percentage distribution of [sup.14]C-label(a) Malate Other Insoluble soluble Taxa 3h 3h+9h 3h 3h+9h 3h 3h+9h Isoetes bolanderi 80 72 20 26 0 2 I. howellii 89 78 11 22 0 0 L orcuttii 88 82 12 17 0 1 Sagittaria subulata 66 70 29 27 5 3 Vallisneria americana Spring 61 66 36 29 3 5 Autumn 39 27 47 65 14 8 V. spiralis 54 53 43 42 3 5 Crassula aquatica 79 75 21 24 0 1 Littorella uniflora 83 79 15 20 2 1
(a) Average of 2 or more replicates.
In all of these aquatic species, malate produced by dark-fixation is stored overnight and largely not metabolized in the dark, as is evident from the pulse-chase studies in the dark (Table II). The bulk of the remaining dark-fixed label is in citrate (or isocitrate). Malate comprises the storage carbon utilized in CAM photosynthesis, a role apparently not ascribed to the other dicarboxylic acids, which apparently are labeled in the dark by transfer of [sup.14]C-label from malate, and serve other metabolic functions (Luttge, 1995). Seasonal changes in labeling patterns have been observed for Vallisneria americana (Table II), indicating greater CAM activity in the spring than in the autumn. This accounts for conflicting reports on acid accumulation in the related K spiralis (Table I); significant [Delta][H.sup.+] occurred in a summer study, whereas two other winter studies failed to find significant [Delta][H.sup.+]. Seasonal changes in level of CAM activity have been reported for several aquatic species and are discussed in Sections VIII and IX.
These labeling studies are incapable of distinguishing between malate and malic acid. However, consistent with the conclusion that dark-fixed label is transported in the protonated form malic acid is the highly significant correlation between [Delta][H.sup.+] and Amalate, evident across species of Isoetes (Fig. 3). If malate were the only acid accumulating, a 2:1 stoichiometry for [Delta][H.sup.+]:Amalate would give a regression line slope of 0.5. The observed deviation (Fig. 3) from that expectation is consistent with 10-20% dark-fixed label in citrate(citric acid) (Keeley, 1981, 1996), assuming a stoichiometry of [2H.sup.+] per malate and [3H.sup.+] per citrate. The slope of this regression line for Isoetes is close to the slope of 0.42 reported for pineapple (Medina et al., 1993). Littorella, on the other hand did not deviate from a 2:1 stoichiometry for [Delta][H.sup.+]:[Delta]malate (Madsen, 1987a), indicating either that the ~20% citrate produced by dark fixation (Table II; Keeley, unpubl, data) is stored as the anion or that citric acid generation is variable between studies. Patterns similar to Isoetes are evident in Sagittaria subulata and species of Crassula, where the molar ratio of [Delta][H.sup.+]:[Delta]malate (x [+ or -] S.D.) = 2.3 [+ or -] 0.3 and 2.0 [+ or -] 0.2, respectively (Table I).
[Figure 3 ILLUSTRATION OMITTED]
These acid changes are restricted to photosynthetic organs and are absent from roots and corms of I. howellii (Keeley, 1981) and I. setacea (Gacia & Ballestros, 1993).
Consistent with glycolytic production of the [CO.sup.2]-accepter molecule PEP, is the overnight depletion of starch observed in I. bolanderi (Keeley et al., 1983a) and I. howellii (Keeley, 1983a). In mid-season, diel changes in I. howellii leaf starch were 144 mol glucose-equivalents [kg.sup.-1] Chl, comparable to the 122 mol malic acid [kg.sup.-1] Chl (Keeley, 1987). Early in the season, however, diel changes in starch in the leaves were insufficient to account for levels of [Delta][H.sup.+], suggesting either that there was a dependence upon starch stored in corms or that PEP was generated at this time from sugars (Black et al., 1995).
B. DAYTIME DEACIDIFICATION
During daytime deacidification (Fig. 1, Table I) there is substantial evidence that the released [CO.sup.2] is refixed via the [C.sub.3] pathway (Fig. 4). Isoetes orcuttii and Littorella also show a mover of [sup.14]C-labeled malate, with label initially in phosphorylated compounds (not shown), followed by transfer of label to other soluble and insoluble compounds. Other aquatic CAM species demonstrate a similar pattern during the light deacidification phase (Keeley, unpubl. data).
[Figure 4 ILLUSTRATION OMITTED]
C. CAM ENZYMES
Carboxylase activities (Table III) show that RUBISCO activities are similar between aquatic and terrestrial CAM plants, perhaps reflecting broadly similar photosynthetic rates (Section V.D). However, PEPC activities are substantially lower for aquatic CAM species than for terrestrial CAM plants (Dittrich et al., 1973), which is surprising since rates of acid production are similar. Nonetheless, PEPC activities in aquatic CAM plants are sufficient to account for the rates of nighttime malate production (10-20 mmol [kg.sup.-1] FM [hr.sup.-1]). Even though ratios of RUBISCO/PEPC are higher in aquatic CAM plants, they nonetheless are still much lower than for a typical [C.sub.3] plant such as spinach (Table III). Also, when aquatic CAM plants are exposed to the atmosphere, the RUBISCO/PEPC increases to levels comparable to terrestrial C3 plants (Table III), which is consistent with the concomitant switch from CAM to [C.sub.3] (Section IX).
Table III Activity of carboxylating enzymes, RUBISCO and PEPC, and other photosynthetic enzymes in submerged aquatic foliage or emergent aerial leaves and selected terrestrial species included for comparison(a)
Data Taxa source(b) RUBISCO PEPC Aquatic CAM species Isoetes howellii Submerged 8 256 36 Aerial 8 553 18 I. lacustris Submerged 4 75 22 Aerial 1 141 -- I. orcutii Submerged 8 225 46 Aerial 8 480 15 Crassula aquatica Submerged 8 392 178 Aerial 8 854 45 Littorella uniflora Submerged 4 187 95 Submerged 7 -- 165 Submerged 5 -- 819 Aerial 5 -- 65 Terrestrial CAM species Ananas comosus 8 -- -- Crassula argenta 2 59 270 Kalanchoe daigremontiana 8 -- -- Mesembryanthemum crystallinum CAM mode 6 306 1074 [C.sub.3] mode 6 438 24 Other terrestrial species Spinacea oleracea [C.sub.3] 8 865 54 Zea mays [C.sub.4] 8 462 842 Zea mays [C.sub.4] 1 184 -- Taxa RUBISCO/PEPC [ME-NAD.sup.+] Aquatic CAM species Isoetes howellii Submerged 7.1 2 Aerial 30.7 nd I. lacustris Submerged 3.4 -- Aerial -- -- I. orcutii Submerged 4.9 -- Aerial 32.0 -- Crassula aquatica Submerged 2.2 2 Aerial 19.0 4 Littorella uniflora Submerged 2.0 -- Submerged -- -- Submerged -- nd Aerial -- nd Terrestrial CAM species Ananas comosus -- -- Crassula argenta 0.2 192 Kalanchoe daigremontiana -- 96 Mesembryanthemum crystallinum CAM mode 0.3 -- [C.sub.3] mode 183 -- Other terrestrial species Spinacea oleracea [C.sub.3] 16.0 -- Zea mays [C.sub.4] 0.5 -- Zea mays [C.sub.4] -- -- Pyruvate Taxa ME-NADP PEPCK P-dikinase Aquatic CAM species Isoetes howellii Submerged 37 nd 110 Aerial 42 nd 186 I. lacustris Submerged -- -- -- Aerial -- -- -- I. orcutii Submerged -- -- -- Aerial -- -- -- Crassula aquatica Submerged 78 nd 208 Aerial 156 nd nd Littorella uniflora Submerged -- -- -- Submerged -- -- -- Submerged 42 -- -- Aerial nd -- -- Terrestrial CAM species Ananas comosus -- 83 908 Crassula argenta -- -- -- Kalanchoe daigremontiana 73 -- -- Mesembryanthemum crystallinum CAM mode -- -- -- [C.sub.3] mode -- -- -- Other terrestrial species Spinacea oleracea [C.sub.3] -- -- nd Zea mays [C.sub.4] -- -- 289 Zea mays [C.sub.4] -- -- --
(a) nd, not detectable; --, not assayed.
(b) 1, Beer et al., 1991; 2, Dittrich et al., 1973; 3, Farmer, 1987; 4, Farmer et al., 1986; 5, Groenhof et al., 1988; 6, Holtum & Winter, 1982; 7, Hostrup & Wiegleb, 1991a; 8, Keeley, 1997a, and unpubl. data.
Thus, relative to terrestrial CAM plants, aquatic CAM species are capable of similar magnitudes of acid accumulation with a lower investment of energy and nutrients in PEPC. I hypothesize that the basis for this stems from differences in water and carbon availability. In aquatic CAM plants there is no obvious selective advantage to rapid dark fixation, whereas in terrestrial species higher PEPC activity may translate into a shorter duration of stomatal opening, and thus higher water use efficiency. Also, aquatic habitats have substantially higher [CO.sup.2] levels than air (Section VII). Under elevated carbon conditions, the naturally high substrate affinity of PEPC may result in vacuolar storage capacity for malic acid being a greater limiting factor to carbon gain, thus favoring reduced investment in PEPC. This explanation is supported by the increase in RUBISCO/PEPC observed for terrestrial CAM plants in response to elevated [CO.sup.2], despite showing little change in [Delta][H.sup.+] (Nobel et al., 1996). Also, the aquatic CAM Littorella exhibits a threefold drop in PEPC activity under elevated [CO.sup.2], without any drop in [Delta][H.sup.+] (Hostrup & Wiegleb, 1991 a).
Kinetic studies show many similarities between the PEPC from the aquatic CAM Littorella and terrestrial CAM plants (Groenhof et al., 1988); e.g., increased [V.sub.max] and decreased [K.sub.m] in the dark or in response to glucose-6-phosphate, and the opposite pattern in response to malate.
Decarboxylase activities are sufficient to account for rates of daytime deacidification and in three species studied, NADP malic enzyme is the primary decarboxylase (Table III). Another potential decarboxylase, PEP carboxykinase, has not been detected in I. howellii or C. aquatica (Keeley, 1998b), and, like terrestrial CAM plants lacking this enzyme (Kelly et al., 1989; Black et al., 1995), these two aquatics have significant pyruvate, [P.sub.i] dikinase activity. Also, consistent with lack of PEP carboxykinase (Winter & Smith, 1995a; cf. Christopher & Holtum, 1996), I. howellii utilizes starch as the source of the [CO.sup.2] acceptor PEP (Keeley, 1983a).
D. GAS EXCHANGE
Gas exchange patterns for aquatic CAM plants are more complex than for terrestrial CAM plants due to multiple carbon sources and dynamic diel changes in availability. In this section, gas exchange characteristics under steady-state conditions (pH 5.5 with vigorous agitation) will be described, and in Section VIII these patterns will be contrasted with patterns under field conditions.
For solutions equilibrated near atmospheric levels of [CO.sup.2] (~0.011 mol [m.sup.-3]), Isoetes howellii exhibits no net [CO.sup.2] uptake in the dark (Keeley & Bowes, 1982), but at higher [CO.sup.2] levels, more typical of its natural environment, dark uptake rates were ~27 mol [kg.sup.-1] Chl [hr.sup.-1] (Fig. 5), or, based on allometric values in Keeley & Sandquist, 1991, 210 mmol [kg.sup.-1] dry mass [hr.sup.-1] or 2.8 mmol [m.sup.-2] total leaf area [hr.sup.-1]. These rates are comparable to dark [CO.sup.2] uptake in terrestrial CAM plants (Kluge & Ting, 1978)--a surprising conclusion since, collectively, aquatic plants have substantially lower photosynthetic rates than terrestrial plants (Bowes & Salvucci, 1989). This seeming paradox may be explained as follows. Differences in daytime photosynthetic rate between aquatic and terrestrial plants are largely a function of transport processes, which are very different between land and water (Raven, 1984). Dark fixation, on the other hand, is more a function of vacuolar storage capacity (Kluge & Ting, 1978), which is more equitably distributed between aquatic and terrestrial CAM plants.
[Figure 5 ILLUSTRATION OMITTED]
In contrast to many, but not all, terrestrial CAM plants, under steady-state [CO.sup.2] conditions, the aquatic CAM I. howellii shows no daytime suppression of [CO.sup.2] uptake (Keeley & Bowes, 1982). In terrestrial CAM plants, suppression results from stomatal closure but does not occur in aquatic plants under steady-state conditions because they lack functional stomata (Section VI.A). In these aquatics, [CO.sup.2] uptake is controlled by ambient [CO.sup.2] concentration and diffusive resistances, factors that, under field conditions (Section VIII), produce more dynamic patterns of [CO.sup.2] uptake than observed in steady-state (Fig. 5). This explanation is supported by the fact that terrestrial CAM plants exhibit [CO.sup.2] uptake in the light if stomatal resistance is overcome, either by removal of the epidermis or with isolated protoplasts (Chellappan et al., 1980; Winter & Smith, 1995a).
Under steady-state conditions (Fig. 5), [CO.sup.2] uptake in the light may be 2-3 times greater than uptake in the dark, across a wide range of naturally occurring [CO.sup.2] concentrations. As with terrestrial CAM plants, [CO.sup.2] uptake in the light is assimilated directly through the C3 pathway--as demonstrated (for Crassula aquatica and Isoetes spp.) by the initial fixation of [sup.14]C. label in PGA and transfer to other phosphorylated compounds, coupled with lack of label in dicarboxylic acids (Keeley, 1998b).
VI. Other Attributes of Aquatic CAM Plants
A. STRUCTURAL CHARACTERISTICS
Three of the five genera with CAM have the "isoetid" growth form, so named because of the resemblance to Isoetes (e.g., Fig. 2), although not all isoetids have CAM (Richardson et al., 1984).
1. Morphological Variation in Isoetes
Despite the rather large number of Isoetes, there is remarkable morphological similarity. All but three species (Hickey, 1990) have the isoetid rosette of stiff terete leaves attached to a small rounded corm. Isoetids have a relatively low surface:volume ratio (1-2 vs. 10-20 for other aquatic macrophytes) and high root:shoot ratio ([is greater than] 1 vs. <0.2 for other macrophytes) (Raven et al., 1988; Boston et al., 1989; Keeley, 1991; Madsen et al., 1993). All isoetids have lacunal air chambers, and in Isoetes species, both aquatic and terrestrial, there are always four lacunae, which, depending on species and habitat, represent 20-90% cross-sectional airspace. A common feature is the concentration of chloroplasts in mesophyll cells surrounding lacunae and, unlike other aquatic macrophytes, few if any epidermal chloroplasts. Both aquatic and terrestrial species have a relatively substantial-appearing cuticle, although little is known about permeability characteristics (but see Keeley et al., 1984). Leaves are attached to a modified stem-rhizophore with traces from the central vascular core connecting leaves and roots (Sculthorpe, 1967).
The most obvious variation in the genus lies in size, which ranges from a centimeter in some rock-outcrop seasonal pool species to large robust species with leaves nearly half a meter long and roots several times longer in some tropical alpine lacustrine species. On rich floodplain sites in the eastern United States, specimens up to 90 cm have been reported (Musselman & Knepper, 1994).
Variation in vegetative structure is apparent in stomatal distribution and root architecture and is closely tied to habitat. Amphibious or seasonal pool species are all drought deciduous and have nonfunctional stomata on submerged foliage. Upon exposure to the atmosphere, stomata become functional and there is a greater density on leaves produced under aerial conditions. Lacustrine species are largely evergreen, although those in lakes subject to thick snowpack are winter deciduous (Keeley, 1987). These lake species exhibit two patterns, apparently tied to latitude. In the Temperate Zone, species such as Isoetes bolanderi, I. macrospora, and I. lacustris produce astomatous leaves underwater but, if exposed, will initiate leaves with functional stomata (Keeley, unpubl, data). In tropical alpine species such as I. palmeri, I. lechleri, and I. karsteni, submerged leaves are astomatous, and stomata are rarely produced under aerial conditions (Keeley, unpubl, data). Terrestrial species, comprising about 10% of the genus, exhibit a similar latitudinal pattern; Temperate Zone species are low-elevation, summer-deciduous plants with functional stomata whereas tropical alpine Isogtes are evergreen plants lacking stomata.
Roots are remarkably variable. Amphibious species from seasonal pools, commonly on fine clay sediments, have relatively thin, highly branched roots with extensive root-hair development. In contrast, many lacustrine species, particularly in tropical alpine lakes with sandy substrates, have thick, unbranched roots, lacking root hairs (Keeley, unpubl, data). In at least some Isoetes these differences are plastic responses to sediment (Karrfalt, 1984). All Isoetes have a single large lacunal chamber that fills the center of the root and varies in cross-sectional area. Also, all species have a mechanism for burying corms that is analogous to "contractile roots" (Karrfalt, 1977).
2. Other Aquatic CAM Plants
Littorella resembles Isoetes in the isoetid growth form, although the corm is replaced by a stolon or rhizome. Linorella leaves have extensive lacunal airspace, lack of epidermal chloroplasts and concentration of chloroplasts in cells surrounding lacunae (Hostrup & Wiegleb, 1991b). This species can alter the extent of lacunal surface area in response to sediment characteristics (Robe & Griffiths, 1988) or upon emergence (Hostrup & Wiegleb, 1991b). Leaf orientation varies from stiffly erect terete leaves in submerged plants to reflexed flattened leaves in terrestrial plants, a character shared with Isogtes.
Some Sagittaria are also isoetids, with rosettes of stiff semi-terete to subulate phyllodes in the aquatic stage. Depending on environmental conditions, these cylindrical leaves are replaced by elongated ribbon-shaped submerged leaves (pseudo-lamina) or broadened sagittate semi-floating leaves (Sculthorpe, 1967). Some, e.g., S. cuneatus and S. graminea (with limited [Delta][H.sup.+], Table I) apparently lack the isoetid stage.
Two aquatic CAM genera are not isoetids: Vallisneria spp. have ribbon-shaped leaves and Crassula spp. are diminutive caulescent annuals, with short semi-cylindrical leaves and often prostrate stems, which constitute much of the photosynthetic surface area.
Succulence is a characteristic typical of a great many terrestrial CAM plants but is not characteristic of aquatic CAM plants. For terrestrial species, mesophyll succulence (kg [H.sub.2]O [g.sup.-1] Chl) is [is less than] 1 for non-CAM plants but up to an order of magnitude higher for most terrestrial CAM plants (Kluge & Ting, 1978). Aquatic CAM plants commonly have mesophyll succulence ratios [is greater than] 1, but as a group are indistinguishable in this character from non-CAM aquatic plants
Two aquatic CAM genera are not isoetids: allisneria spp. have ribbon-shaped leaves and Crassula spp. are diminutive caulescent annuals, with short semi-cylindrical leaves and often prostrate stems, which constitute much of the photosynthetic surface area.
Succulence is a characteristic typical of a great many terrestrial CAM plants but is not characteristic of aquatic CAM plants. For terrestrial species, mesophyll succulence (kg [H.sub.2]O [g.sup.-1] Chl) is I for non-CAM plants but up to an order of magnitude higher for most terrestrial CAM plants (Kluge & Ting, 1978). Aquatic CAM plants commonly have mesophyll succulence ratios 1, but as a group are indistinguishable in this character from non-CAM aquatic plants (Keeley, unpubl, data). Succulence, however, leads not only to a higher water content but also to a low surface area:volume ratio (Gibson & Nobel, 1986), a feature shared by both aquatic and terrestrial CAM plants.
B. INORGANIC CARBON SOURCE
Aquatic plants have access to carbon sources not available to terrestrial plants. Bathed in solution, these plants are exposed to dissolved [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], or [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], with [CO.sub.2] predominating at acidic pH but nil above pH 8. Aquatic plants are often described as "preferring" [CO.sub.2], meaning the apparent [K.sub.m] is substantially lower for [CO.sub.2] uptake, even in species with the capacity for [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] uptake. Despite the fact that bicarbonate is the active form assimilated by PEPC, aquatic CAM species lack the capacity for bicarbonate uptake. In Isoetes spp., at constant [CO.sub.2] concentration, photosynthetic rates at pH 5 are higher than rates at pH 8, despite the substantially higher inorganic carbon present at the higher pH (Keeley, unpubl, data). Of course, this could reflect inhibition due to high pH or alkalinity.
The pH-drift technique, where final pH is a function of alkalinity plus carbon-extracting ability of the plant (Allen & Spence, 1981), shows Elodea canadensis (a known bicarbonate user) has much greater carbon extracting ability than the non-bicarbonate user Isoetes howellii (Fig. 6). While species such as E. canadensis may drive up the pH during such experiments to above pH 10, non-bicarbonate users such as I. howellii seldom raise the pH much beyond 8. A useful comparative parameter is the final total carbon ([C.sub.t]):alkalinity ratio, which is 0.73-0.79 for E. canadensis and 0.97-1.00 for I. howellii (Gearhart & Keeley, unpubl. data), values characteristic of bicarbonate and non-bicarbonate users, respectively. Using similar techniques, Sand-Jensen (1987) demonstrated a lack of bicarbonate uptake also for the CAM species Isoetes lacustris, and also for I. macrospora and Littorella (Boston et al., 1987; Maberly & Spence, 1983, 1989), Crassula aquatica (Keeley, unpubl, data) and C. helmsii (Newman & Raven, 1995). Capacity for bicarbonate uptake is widespread in aquatic plants but is likely missing from many species because ions such as [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] must be actively transported across the epidermal membrane, which makes it energetically more expensive than passive uptake of [CO.sub.2]. Bicarbonate uptake is a [CO.sub.2]-concentrating mechanism best viewed as an alternative to CAM.
[Figure 6 ILLUSTRATION OMITTED]
C. ISOTOPE FRACTIONATION
Keeley and Sandquist's (1992) review of [sup.13]C: [sup.12]C ratios in aquatic species can be summarized as follows. Consistent with the pattern in terrestrial CAM plants, [Delta] [sup.13]C values for Isoetes species are substantially lower for submerged leaves in the CAM mode than for aerial leaves in the [C.sub.3] mode (see Section IX). Also, in Isoetes, [Delta] [sup.13]C is lower for aquatic CAM species than for terrestrial [C.sub.3] species (Richardson et al., 1984; Keeley & Sandquist, 1992). However, aquatic CAM species often have ratios indistinguishable from aquatic [C.sub.3] species (Keeley & Sandquist, 1992; cf. Richardson et al., 1984). This derives from additional factors that determine ratios in
VII. Habitat Distribution
Enhanced water use efficiency is an important selective force in the evolution and maintenance of CAM in terrestrial plants and is reflected in the abundance of CAM in many arid land floras (Kluge & Ting, 1978). Even in tropical rain forest CAM epiphytes, water use efficiency is considered an important selective factor (Griffiths, 1989). Clearly, such is not the case with aquatic CAM plants; rather, this pathway is strongly correlated with habitats imposing severe carbon-limitation. These habitats include shallow rain-fed seasonal pools and oligotrophic lacustrine habitats.
A. SEASONAL POOLS
Shallow seasonal pools form in many parts of the world and commonly have species of Isoetes and/or Crassula (Keeley & Zedler, 1998). Many fill during winter and spring, when precipitation exceeds evapotranspiration, and because they are rain-fed, such "vernal pools" typically have low conductance, with pH controlled by the weak buffer system of [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. They are generally shallow with high levels of photosynthetically active radiation (PAR) (Keeley et al., 1983b). Plant biomass is high, and thus early morning photosynthetic consumption of [CO.sub.2] drives pH up and by mid-day free-[CO.sub.2] in the bulk water is nil (Fig. 7A). This leaves bicarbonate as the primary source of carbon, and most communities have some species capable of utilizing this source and thus driving up the pH to 9-10 (Keeley & Busch, 1984). Since these pools are densely vegetated and relatively stagnant, [CO.sub.2] depletion in the leaf boundary layer is likely to occur rapidly (Smith & Walker, 1980), suggesting that plants are subject to a considerably longer period of [CO.sub.2] starvation than is evident in the bulk water (Fig. 7A). At night, release of respiratory carbon drives up the ambient [CO.sub.2] levels, resulting in a largely biogenically driven diel pattern of [CO.sub.2] availability, or what Raven and Spicer (1995) refer to as a landscape-level "[CO.sub.2] pump." Dynamic fluctuations in pool chemistry, similar to those illustrated for California (Fig. 7A), have been demonstrated for seasonal pools in Spain (Gacia & Ballestros, 1993), Chile, and South Africa (Keeley, unpubl, data). As a matter of speculation, forest understories exhibit similar diel changes in [CO.sub.2] availability (Broadmeadow & Griffiths, 1993), which may account for the odd occurrence of terrestrial CAM plants in these habitats.
[Figure 7 ILLUSTRATION OMITTED]
Seasonal pools develop under many circumstances, but not all are suitable CAM plant habitats (Keeley & Zedler, 1998). Alkaline pools generally lack CAM species, as the high pH results in little diel change in pH and [CO.sub.2] availability. Pools that develop along temporary stream courses or within large drainage basins also seldom are dominated by CAM plants. This is because the enriched nutrient content, due to allochthonous input of inorganic and organic nutrients (Wetzel, 1975), buffer the water against sharp diel changes in carbon as well as favoring faster-growing competitors.
Lacustrine habitats dominated by CAM plants are generally softwater oligotrophic lakes, which are common at high latitudes or, in lower latitudes, only at high elevations. Oftentimes such lakes are completely dominated by CAM plants. For example, in Lake Kalgaard (Table IV) 99% of the biomass is contributed by two CAM species, Littorella in a zone 0-2 m deep and Isoetes lacustris at 2-4.5 m (Sand-Jensen & Sondergaard, 1979)--a pattern repeated elsewhere in Europe (Szmeja, 1994). In North America, CAM species such as I. macrospora reach peak biomass at depths below 7 m (Collins et al., 1987). Depth distribution patterns in general vary in accordance with water transparency (Middelboe & Markager, 1997). In shallow neotropical alpine lakes, Isoetes and Crassula often cover three-fourths or more of the lake bottom, with few other species present (Keeley, pers. obs.). Although Isoetes are commonly distributed in lakes with circumneutral pH (Jackson & Charles, 1988; Gacia et al., 1994), they often dominate under more acidic conditions (Moyle, 1945; Pietsch, 1991; Voge, 1997).
Table IV Comparison of typical water and sediment chemistry characteristics of selected lakes dominated by CAM macrophytes and lakes dominated by non-cAM macrophytes(a)
Lake (country) Latitude Elev. (m) Kalgaard (Denmark) 56 [degrees] N 75 Esthwaite (Denmark) 56 [degrees] N -- Weber (U.S.A.) 43 [degrees] N -- Ellery (U.S.A.) 38 [degrees] N 2900 "Km 31" (Colombia) 4 [degrees] N 3650 "Larga" (Colombia) 4 [degrees] N 3650 "Temprano" (Ecuador) 0 [degrees] 4050 Searsville (U.S.A.) 37 [degrees] N 110 Photosynthetic Lake (country) Dominant macrophytes pathway Kalgaard (Denmark) Isoetes lacustris CAM Littorella uniflora CAM Esthwaite (Denmark) Isoetes lacustris CAM Littorella uniflora CAM Weber (U.S.A.) Isoetes macrospora CAM Littorella uniflora CAM Ellery (U.S.A.) Isoetes bolanderi CAM Eleocharis acicularis non-CAM "Km 31" (Colombia) Isoetes karstenii CAM Crassula paludosa CAM "Larga" (Colombia) Isoetes palmeri CAM Crassula paludosa CAM "Temprano" (Ecuador) Isoetes peruvianum CAM Crassula paludosa CAM Searsville (U.S.A.) Myriophyllum brasiliense non-CAM Potamogeton spp. non-CAM Ceratophyllum demersum non-CAM Water column Data Free-[CO.sub.2] Lake (country) source(b) pH (mol [m.sup.-3]) Kalgaard (Denmark) 1 7.4 0.03 Esthwaite (Denmark) 2 6.0 0.06 Weber (U.S.A.) 3 6.1 0.10 Ellery (U.S.A.) 4 6.8 0.12 "Km 31" (Colombia) 5 5.1 0.23 "Larga" (Colombia) 5 5.3 0.12 "Temprano" (Ecuador) 5 6.3 0.14 Searsville (U.S.A.) 5 7.7 4.86 Water column Sediment water Conductivity ([micro]S Free-[CO.sub.2] Lake (country) [cm.sup.-3]) pH (mol [m.sup.-3]) Kalgaard (Denmark) 66 5.5 3.00 Esthwaite (Denmark) -- 6.5 1.01 Weber (U.S.A.) -- 5.8 0.80 Ellery (U.S.A.) 22 6.5 1.70 "Km 31" (Colombia) 10 4.8 1.51 "Larga" (Colombia) 15 4.9 1.79 "Temprano" (Ecuador) -- 5.9 0.75 Searsville (U.S.A.) 750 -- --
(a) --, data not available.
(b) 1, Sand-Jensen & Sondergaard, 1979b; 2, Robe & Griffiths, 1988; 3, Boston & Adams, 1985; 4, Keeley et al., 1983a, and Sandquist & Keeley, 1990; 5, Keeley, unpubl, data.
Diel changes in [CO.sub.2] and [O.sub.2] are a function of metabolic and physical processes and in poorly buffered water are controlled by the ratio of biomass:water-volume. Because this ratio is very low in oligotrophic lakes, these habitats do not exhibit predictable diel patterns of [CO.sub.2] availability (Sand-Jensen et al., 1982; Keeley et al., 1983a; Sand-Jensen, 1989; Sandquist & Keeley, 1990). These habitats, however, have inorganic carbon levels one to two orders of magnitude lower than for seasonal pools or for mesotrophic lakes dominated by non-CAM plants (e.g., Searsville Lake, Table IV). Although [CO.sub.2] levels in oligotrophic lakes are still greater than the levels expected from equilibrium with the atmosphere (~0.01 mol [m.sup.-3]), the diffusive resistance of water ([10.sup.4] times greater than air) limits the availability of [CO.sub.2] in unstirred layers around leaves. These infertile habitats are also low in other inorganic nutrients, in particular nitrate and phosphate (Sondergaard & Sand-Jensen, 1979b; Pietsch, 1991). Irradiance levels are higher than in mesotrophic lakes (due to low phytoplankton biomass) but substantially lower than in shallow seasonal pools (Kirk, 1983). In addition to the irradiance attenuation with depth, some high-elevation lakes experience abbreviated day length due to shading by adjacent forests and rugged terrain (Sandquist & Keeley, 1990).
One noteworthy characteristic of lacustrine habitats dominated by CAM species is the substantially higher sediment [CO.sub.2] level (Table IV), an important factor in the carbon balance of isoetids (Section VIII.B.1). It is of some interest that Isoetes distributed in acidic infertile lakes in tropical Andean sites have a tendency to grow in extremely dense clumps of [10.sup.3]-[10.sup.4] plants [m.sup.-2], due in part to vegetative reproduction by axillary gemmae (Hickey, 1986; Keeley, pers. obs.). As a consequence, organic matter is concentrated beneath the clumps and thus sediment [CO.sub.2] levels are substantially greater than in the interstitial spaces between clumps (Keeley, unpubl, data), perhaps facilitating [CO.sub.2] uptake from the sediment.
In general, CAM species are poorly represented in mesotrophic lakes and are seldom found under eutrophic conditions (Seddon, 1965, 1972; Rorslett & Brettum, 1989; Gacia et al., 1994). Eutrophication often leads to the disappearance of CAM species (Kurimo & Kurimo, 1981; Farmer & Spence, 1986). Numerous authors have suggested that the restriction of isoetids to infertile sites is because they are competitively displaced in more fertile habitats--a hypothesis with some experimental support (Lee & Belknap, 1970). Preference for oligotrophic conditions by aquatic CAM plants is similar to the pattern observed for terrestrial CAM plants.
C. OTHER HABITATS
There is some overlap between oligotrophic lake and seasonal pool habitats--e.g., Littorella often is distributed in the eulittoral zone that periodically dries. These habitats are shallow enough to potentially experience diel changes similar to seasonal pools, and populations persist in this amphibious state (Nielsen et al., 1991). Isoetes asiatica is a species of shallow lakes where only a portion of the population is amphibious (Pietsch, 1991). Also, some tropical alpine ephemeral pools dominated by CAM species (Isoetes and Crassula) are very oligotrophic and, because of this state and the low temperatures, fail to generate significant diel changes in [CO.sub.2] (Keeley, unpubl, data).
Other CAM habitats include slow-moving shallow streams (Isoetes flaccida), shaded sections of relatively fast-moving irrigation canals (I. malinverniana), and the eulittoral zone of freshwater tidal rivers (I. riparia and Sagittaria subulata) (Keeley, 1987). These require further study to elucidate the relevant selective factors favoring CAM.
In summary, aquatic CAM distribution is a function of two factors: inorganic carbon and irradiance. CAM plants dominate under carbon-limited conditions, and as trophic conditions improve and free [CO.sub.2] levels go up, CAM plants dominate only under conditions that generate marked diel patterns of availability. Within oligotrophic habitats, irradiance may play a role by limiting the length of time available for light-requiring reactions, and here CAM may play a role in extending the depth to which certain Isoetes can colonize.
VIII. CAM and the Carbon Budget
Although enhanced water use efficiency is the ultimate selective force in terrestrial CAM plant evolution, the proximal selective factor is enhanced daytime intercellular [CO.sub.2] partial pressure ([p.sub.i]). High [CO.sub.2]([p.sub.i]) on the order of 40 mPa [Pa.sup.-l] or 4% v/v results from high stomatal resistance, coupled with decarboxylation of malate stores (Winter & Smith, 1995a). In effect, CAM is a [CO.sub.2]-concentrating mechanism and thus requires a physical setting in which a disequilibrium is created between exogenous and endogenous [CO.sub.2] pools.
In aquatic plants, several factors inhibit [CO.sub.2] leakage during daytime decarboxylation of malate, thus creating a disequilibrium in [CO.sub.2] pools. The primary factor is the high diffusive resistance of water ([10.sup.4] times greater than air). Also, water per se has an ameliorating effect on gas exchange, which, relative to leaves in air, inhibits outward diffusion of [CO.sub.2] (Steinberg, 1996). The cuticle, a feature uncommon in aquatic plants (Sculthorpe, 1967), is quite apparent in many aquatic CAM plants and may be an important resistance factor. Additionally, anatomical features play a role because chloroplasts are concentrated in mesophyll cells surrounding the lacunae, and consequently, sites of decarboxylation are several cell layers removed from the ambient environment, which constitutes a substantial diffusional resistance (Raven, 1977) and further contributes to disequilibrium. The standard to which these resistances are measured is the RUBISCO activity. For decarboxylation to be effective, [CO.sub.2] leakage must not be greater than the rate at which it can be fixed. Also, daytime PEPC activity may, through its substantially lower [K.sub.m], capture carbon and thus inhibit leakage (Osmond, 1984; Winter, 1985). Estimates of leakage rates for Littorella and Isoetes lacustris indicate that only 1-2% inorganic carbon is lost, and leakage rate is not sensitive to [CO.sub.2] concentration (Sondergaard & Sand-Jensen, 1979a; Madsen, 1987b).
Habitats differ in the factors contributing to disequilibrium between ambient and endogenous [CO.sub.2] sources.
A. SEASONAL POOL CAM PLANTS
CAM plants in seasonal pools show diel patterns of carbon uptake in the light and dark that are correlated with changes in ambient [CO.sub.2]. An example of one spring day for Isoetes howellii shows that as available carbon declines during early morning (Fig. 7A), [CO.sub.2] uptake is suppressed (Fig. 7B). Tracking this decline is a rapid decarboxylation of vacuolar malic acid stores (Fig. 7B), as photosynthesis switches to increasing dependence upon this endogenous carbon source. Three of the four phases of [CO.sub.2] exchange recognized by Osmond (1978) for a "well-irrigated CAM plant" are evident in this aquatic (Fig. 7B).
Phase 1, the period of dark [CO.sub.2] uptake and assimilation, matches well with terrestrial CAM plants, including the suppressed uptake late in the dark phase (Fig. 7B). This depression is also observed under steady-state conditions in the lab (Keeley & Bowes, 1982) and may reflect feedback inhibition of malic acid on PEPC activity (Groenhof et al., 1988; Kluge & Brulfert, 1995).
Phase 2 shows an acceleration in uptake due to the light-induced switch to direct assimilation of carbon by the [C.sub.3] pathway, a pattern also seen in terrestrial CAM plants. It is not known how much of this initial burst in [CO.sub.2] uptake in the light results from a combination of both PEPC and RUBISCO activity. In Osmond's prototype CAM plant, Phase 2 is characterized by a rapid suppression of [CO.sub.2] uptake, resulting from stomatal closure, although there is much species-specific variation in rate of stomatal closure (Kluge & Ting, 1978; Borland & Griffiths, 1995; Winter & Smith, 1995a). Since functional stomata are lacking in aquatic plants, the drop in [CO.sub.2] uptake during Phase 2 is obviously not related to stomatal behavior; rather, it is due to the depletion of ambient [CO.sub.2] (Fig. 7A).
Phase 3 is a period of limited [CO.sub.2] uptake, controlled in terrestrial CAM plants by stomatal closure, which is a response to high internal [CO.sub.2]([p.sub.i]), generated by malate decarboxylation. Phase 3 in this aquatic CAM plant is controlled by the depletion of ambient [CO.sub.2].
Phase 4 in terrestrial plants is a period in which the Phase 3 suppression of [CO.sub.2] uptake is overcome because malate is depleted; as a consequence, [CO.sub.2]([p.sub.i]) decreases and this induces stomatal opening. Phase 4 is missing in this aquatic CAM plant because ambient [CO.sub.2] remains depleted, due to slow gas exchange with the atmosphere (Smith, 1985) and high pH resulting from bicarbonate uptake by other species in the community.
In I. howellii the pattern of acidification (Phase 1) and deacidification (Phases 2 & 3) track ambient [CO.sub.2] (Fig. 7). Deacidification is insignificant during the first three hours of Phase 2 and appears to be controlled by high ambient [CO.sub.2], as suggested by the fact that percentage deacidification is correlated with percentage [CO.sub.2] depletion of the water. Also, deacidification can be experimentally slowed by incubation under elevated [CO.sub.2] levels (Keeley, 1983a). A similar suppression of deacidification by elevated [CO.sub.2] is also observed in terrestrial CAM plants (Fischer & Kluge, 1985). In the aquatic habitat, I. howellii deacidification is correlated with irradiance, such that on cloudy days, decarboxylation of malate slows and [Delta][H.sup.+] is suppressed. This may be tied to the fact that lower PAR reduces photosynthetic demand for [CO.sub.2] by the pool flora, causing [CO.sub.2] in the water to remain high through mid-day (Keeley & Busch, 1984).
Integrating the area under the [CO.sub.2] uptake curve (Fig. 7B) shows that on this particular date, [CO.sub.2] uptake contributed 49% of the total 24 hr gross carbon gain. Under shorter day lengths and cooler temperatures earlier in the season, both total gross carbon uptake and the dark contribution are lower (Keeley & Busch, 1984).
A comparison of total [CO.sub.2] uptake in the dark and total [CO.sub.2] fixation in the dark (predicted by [Delta][H.sup.+]) indicates that carbon uptake never matches carbon assimilation. This is because dark fixation utilizes both ambient [CO.sub.2] and an endogenous source arising from respiration. Refixation of respiratory [CO.sub.2] is illustrated by the substantial overnight acid accumulation possible under [CO.sub.2]-free conditions (Fig. 8). It is estimated that throughout the season this may account for 50-75% of the dark carbon fixation in I. howellii (Keeley & Busch, 1984) and in Crassula helmsii (Newman & Raven, 1995).
[Figure 8 ILLUSTRATION OMITTED]
In summary, dark fixation affects carbon balance both by extending the period of [CO.sub.2] uptake and by recycling [CO.sub.2]. Terrestrial CAM plants are similar, in that a portion of overnight acid accumulation is due to refixation of respiratory carbon and this can be up to 100% in what is referred to as "CAM-cycling" or "CAM-idling" (Griffiths, 1988; Martin, 1995).
Root uptake of [CO.sub.2] from interstitial water in the sediment may be substantial in many lacustrine isoetids (Section VIII.B.1) but is less significant in amphibious seasonal pool species. Although [CO.sub.2] concentration in these sediments is about one order of magnitude higher than the peak water column levels (Keeley & Sandquist, 1991), soils are commonly fine clay sediments with small interstitial spaces. Also, seasonal pool Isoetes have less intercellular airspace than do lacustrine species. Laboratory studies with leaves and roots in separate compartments show that for I. howellii, under [CO.sub.2] levels matching field conditions around leaves and roots, uptake by leaves is about 5-10 times greater than by roots, and this is under conditions in which the solution surrounding the roots is stirred (Keeley, unpubl.data). When one considers the diffusive resistances in these sediments, it is apparent they are not likely a major carbon source for these plants.
B. LACUSTRINE CAM PLANTS
The absence of diel changes in ambient [CO.sub.2] availability (Section VII.B) means that the evolution of CAM in these environments has been driven by factors distinct from those effective in seasonal pools. There is evidence that both carbon and light may be limiting. In addition, other nutrients are scarce in these infertile habitats, and the CAM pathway potentially could enhance nitrogen-use efficiency (Griffiths, 1989; Robe & Griffiths, 1994). Evaluating these factors is complicated by [CO.sub.2] uptake from both the water column and sediment.
1. Sediment [CO.sub.2] Uptake
In Littorella, the permeability for [CO.sub.2] transport across the root surface is 0.6-0.8 mm [hr.sup.-1] and across the leaf surface is 3.8-5.8 mm [hr.sup.-1] (Madsen, 1987a). This, coupled with the substantially shorter source-to-sink path length in leaves, makes it no surprise that, under equal [CO.sub.2] concentrations, leaves exhibit greater [CO.sub.2] uptake (per unit surface area) than roots (Sondergaard & Sand-Jensen, 1979a). However, oligotrophic lakes typically have carbon-rich sediments that may contain one to two orders of magnitude more free-[CO.sub.2] than the water column (Table IV). Macrophytes with the isoetid growth form, including both CAM and non-CAM species, capitalize on this rich carbon source and derive a substantial portion of their car- bon from the sediment.
Under ambient [CO.sub.2] levels in the water column (0.015 mol [m.sup.-3]) and sediment ([is greater than] 1 mol [m.sup.-3]), for both Littorella and Isoetes species, more than 95% of [CO.sub.2] uptake in the light is through the roots (Sondergaard & Sand-Jensen, 1979a; Boston et al., 1987). However, as the water column [CO.sub.2] level rises, root uptake may decline to [is less than] 50% of the total uptake (Richardson et al., 1984; Sandquist & Keeley, 1990).
Dark [CO.sub.2] uptake shows a similar pattern where, under natural levels of [CO.sub.2] in the water column and sediment, all [CO.sub.2] uptake is through the roots (Fig. 9B). As root medium [CO.sub.2] level goes down, uptake from the water column increases (Fig. 9A), and when root medium levels are higher, there is net [CO.sub.2] evolution from the foliage (Fig. 9C). It is of some interest that the overnight acid accumulation in Littorella, which matches very closely the estimated total dark [CO.sub.2] fixation (= direct uptake from the water + root uptake from the sediment + re-fixation of respiratory carbon), does not differ significantly across the range from 0.7 to 3.1 mol [m.sup.-3] sediment [CO.sub.2]; rather, all that changes is the path of [CO.sub.2] uptake (Madsen, 1987a).
[Figure 9 ILLUSTRATION OMITTED]
Root uptake results in a substantial increase in [CO.sub.2]([p.sub.i]) in the lacunae (Fig. 9 caption), and this endogenous [CO.sub.2] is an important source for carbon assimilation in both the light and the dark. In addition to being a rich carbon source, transport to chlorenchymous cells surrounding the lacunae is through the gas phase, and thus substantially faster than aqueous phase transport from the water column (Raven, 1984). This internal [CO.sub.2] supply can exceed demand at night, as evidenced by inorganic carbon leakage (Sandergaard, 1981), but may be limiting during the day (Sondergaard & Sand-Jensen, 1979a; Madsen, 1987b). As this [CO.sub.2] source becomes limiting, CAM--through decarboxylation of malate stores--enhances internal [CO.sub.2] concentration (Robe & Griffiths, 1988). Under natural substrate levels of [CO.sub.2], it appears that CAM is capable of maintaining endogenous [CO.sub.2] levels sufficient to suppress photorespiration and make PAR the limiting factor to photosynthesis (Robe & Griffiths, 1990).
Root uptake of [CO.sub.2] is by passive diffusion through airspaces in the roots, stems, and leaves (Raven et al., 1988; Keeley et al., 1994). There is also a net flow of [O.sub.2] into hypoxic sediments which has beneficial effects on nutrient uptake (Tessenow & Baynes, 1978; Sand-Jensen et al., 1982; Smits et al., 1990; Pedersen et al., 1995).
Characteristics associated with the isoetid growth form which enhance carbon uptake from the roots are 1) high root:shoot ratio, 2) short pathway from roots to leaves, 3) extensive air space, and 4) chloroplasts in cells surrounding the lacunae. Species with other growth forms, such as the non-CAM Myriophyllurn spicatum, obtain very little carbon from the sediment (Loczy et al., 1983; Raven et at., 1988). It appears that isoetids can alter their root permeability in response to sediment characteristics--e.g., highest lacunal [CO.sub.2] concentrations were observed in Littorella grown on the lowest [CO.sub.2] sediments (Robe & Griffiths, 1988)
Although Crassula species lack the isoetid growth form conducive to root uptake, they are generally prostrate and therefore may benefit from enhanced sediment [CO.sub.2]; for instance, water column [CO.sub.2] concentration a few centimeters above the sediment may be more than one order of magnitude greater than the level in bulk water (Robe & Griffiths, 1992).
2. Factors Affecting Acidification and Deacidification Patterns
The decline in [CO.sub.2] uptake late in the dark period observed for Littorella (Fig. 9A-C) is similar to that observed for the seasonal pool species Isoetes howellii (Fig. 7B). Also in common with that seasonal pool species is the substantial role of nighttime refixation of respiratory [CO.sub.2] in Littorella and I. lacustris: from 1/3 to 2/3 of the total acid accumulation (Madsen, 1987a; Robe & Griffiths, 1990; Richardson et al., 1984; Smith et al., 1985).
In Littorella, incubation for several weeks under a 12 hr photoperiod of low photosynthetically active radiation (PAR = 40-50 [micro]mol [m.sup.-2] [s.sup.-1]) greatly reduces overnight acid accumulation (Madsen, 1987c; Robe & Griffiths, 1990). This damping effect of low light also has been reported for Isoetes kirkii (Rattray et al., 1992). Perhaps this is due to low stores of starch for glycolytic PEP production or the extra ATP required to drive the tonoplast transfer of malate (Smith et al., 1995; Luttge, 1987) and is consistent with the high photon costs of net [CO.sub.2] fixation by CAM plants (Raven & Spicer, 1995). A similar effect of low daytime PAR inhibiting [Delta][H.sup.+] is observed in terrestrial CAM plants (Osmond, 1978). Seasonal changes in light and temperature also contribute to lower levels of CAM in autumn and winter for the aquatic I. macrospora (Boston & Adams, 1985) and I. lacustri (Gacia & Ballestros, 1993).
When light is less limiting (450-500[micro]mol [m.sup.-2] [s.sup.-1]), CAM activity is maintained at [CO.sub.2] levels between 0.01 and 1.5 mol [m.sup.-3) but reduced or eliminated at 5.5 mM free-[CO.sub.2] (Madsen, 1987b, 1987c; Robe & Griffiths, 1990). In Littorella, a [CO.sub.2] level sufficient to suppress CAM is 3.0 mol [m.sup.-3] around the leaves, but 5.4 mol [m.sup.-3] is required around the roots, reflecting the substantially greater resistances, less surface area, and longer path length from roots to the site of carboxylation (Madsen, 1987b). Inhibition of CAM by elevated [CO.sub.2] operates by suppressing daytime decarboxylation, as indicated by the fact that high ([is greater than] 1 mol [m.sup.-3]) [CO.sub.2] in the dark phase produces high [Delta][H.sup.+] but the same [CO.sub.2] level in the light phase causes an immediate suppression of CAM (Madsen, 1987b; Hostrup & Wiegleb, 1991 a).
3. Contribution of CAM
Calculation of a carbon budget is complicated by the necessity to include carbon uptake from both leaves and roots, and carbon fixation in the light and dark, as well as refixation of respiratory carbon. Light is potentially limiting, and its effect is likely to differ between species. Littorella, which occupies shallow water, typically experiences mid-day photosynthetically active radiation (PAR) levels of 100-200 [micro]mol [m.sup.-2] [s.sup.-1]) at the leaf tips and receives an annual photon flux density (PFD) estimated at 1760 mol [m.sup.-2] [yr.sup.-1] (Sand-Jensen & Madsen, 1991). Isoetes lacustris is distributed more deeply (PFD = 455 mol [m.sup.-2] [yr.sup.-1]) and, in response to these zonation differences, has higher chlorophyll levels, lower light-saturated net photosynthesis, and higher photosynthetic rates under low irradiance than Littorella (Sand-Jensen, 1978). The extent to which these factors affect differences in expression of CAM (e.g., stoichiometry of uptake: fixation in both the dark and light) has not been explored.
Field studies of I. bolanderi showed that daytime carbon uptake tracked irradiance and that substantial uptake was restricted to about a 6 hr period around mid-day (Sandquist & Keeley, 1990). In this study dark [CO.sub.2] uptake contributed about 30% of the gross carbon uptake, which approximates the 28% calculated for the contribution of dark [CO.sub.2] uptake by I. lacustris (Richardson et al., 1984).
A reasonably complete carbon budget for Littorella has been provided by Robe and Griffiths (1990), under natural carbon conditions and little or no light limitation (Fig. 10):
[Figure 10 ILLUSTRATION OMITTED]
1. 55% of the total carbon gain is derived from dark [CO.sub.2] uptake
2. [CO.sub.2] uptake accounts for only 30% of the dark fixation (i.e., there is substantial refixation of respiratory [CO.sub.2])
3. 81% of the [CO.sub.2] supply for daytime photosynthesis is derived from decarboxylation of malate.
The importance of CAM is further demonstrated by the lack of congruence in [O.sub.2] evolution and [CO.sub.2] uptake (Fig. 11); during the day, Littorella exhibits substantial [O.sub.2] evolution but minimal [CO.sub.2] uptake. This seeming disconnection of the light reactions and carbon reduction reactions is because carbon assimilation is utilizing endogenous [CO.sub.2] sources, such as that derived from decarboxylation of malate. A consequence of using this endogenous [CO.sub.2] source is a reduction in the [CO.sub.2] compensation point and increase in carboxylation efficiency (Madsen, 1987b, 1987c).
[Figure 11 ILLUSTRATION OMITTED]
Limitations of nutrients other than carbon appear to play a relatively minor role in controlling CAM activity (Madsen, 1987c; Robe & Griffiths, 1994). However, evolution of carbon-concentrating mechanisms such as CAM, in plants on infertile sites, potentially makes nutrients other than carbon the limiting resource in primary productivity (Raven, 1995). Even though nutrient limitations may have minimal proximal effect, ultimately the infertility of oligotrophic lakes has likely been a strong selective influence on growth rates (Boston, 1986; Boston & Adams, 1987). Reflective of these CAM plants' adaptation to nutrient-poor habitats is the observation that Littorella plants grown on the lowest sediment [CO.sub.2] concentrations maintained the highest levels of lacunal [CO.sub.2], [Delta][H.sup.+], and photosynthesis (Robe & Griffiths, 1988).
Most studies of aquatic CAM production concern lacustrine species from infertile carbon-poor habitats. Standing above-ground biomass of macrophytes in oligotrophic lakes is commonly one to three orders of magnitude lower than in eutrophic lakes lacking CAM species (Sculthorpe, 1967; Wetzel, 1975). Within the littoral zone dominated by macrophytes, standing crops often are 0.1-2.0 mg oven-dry mass [ha.sup.-2] (Sand-Jensen & Sondergaard, 1979; Toivonen & Lappalainen, 1980; Keeley et al., 1983a; Boston & Adams, 1987; Gacia & Ballestros, 1994). Growth rates are generally low and, even when placed under enriched carbon conditions, species (both CAM and non-CAM) from such oligotrophic lakes have rates lower, by an order of one magnitude or more, than species from more meso- or eutrophic habitats (Boston et al., 1989). In the lacustrine habitat, the CAM pathway contributes about 50% of the total annual carbon gain, largely through the extension of the carbon assimilation period (Boston & Adams, 1986). This nocturnal carbon contribution was equivalent to the total 24 hr dark respiration and a critical component to success in these lakes.
Seasonal pools are densely vegetated with as much as 10 mg dry mass [ha.sup.-2] [yr.sup.-1] production each growing season (Keeley & Sandquist, 1991). While not a record for CAM plant productivity (Nobel, 1995), it is significantly higher than the productivity of many arid CAM habitats. In one study, gross [CO.sub.2] uptake was about 10% higher for Isoetes than associated non-CAM species (Keeley & Sandquist, 1991). Gross measures of productivity (i.e., biomass changes during the growing season) showed I. howellii production at 9.9 [+ or -] 0.1 g dry mass [m.sup.-2] [day.sup.-1]; this species represented 37% of the biomass early in the season and 53% late in the season. These seasonal pools are mesotrophic habitats, and under the right conditions certain CAM plants are capable of considerable productivity, potentially outcompeting other species, as evidenced by the aggressive invasive ability of the aquatic CAM Crassula helmsii (Dawson & Warman, 1987; Newman & Raven, 1995).
IX. Aquatic CAM Plants in an Aerial Environment
Seedlings of terrestrial CAM species commonly are [C.sub.3] and switch to CAM later in development (Raven & Spicer, 1995), whereas amphibious CAM species exhibit an opposite pattern. During early stages of development underwater they exhibit CAM, but upon exposure to an aerial environment amphibious species switch off CAM and rely strictly on the [C.sub.3] pathway. This has been demonstrated both by diminished [Delta][H.sup.+] (Table V) and [.sup.14]C-labeling studies (Keeley, 1998b). This switch occurs on a cell-by-cell basis as the emergent tips of leaves will reduce overnight acid accumulation, whereas submerged bases retain CAM (Keeley, 1988). As the dry season approaches, and these aerial plants are exposed to increasing aridity, they do not regain the CAM pathway. Many eulittoral lacustrine species also will switch off CAM upon exposure (Table V).
Table V Diel changes in titratable acidity ([Delta][H.sup.+) under submerged and aerial conditions for aquatic and terrestrial species
Latitudinal Taxa Habitat(a) Habit(b) zone Isoetes howellii Seas. pool Sum. decid Temperate Crassula aquatica Seas. pool Sum. decid. Temperate C. natans Seas. pool Sum. decid. Temperate Isoetes bolanderi Lacustrine Win. decid. Temperate I. macrospora Lacustrine Evergreen Temperate Littorella uniflora Lacustrine Evergreen Temperate Isoetes palmeri Lacustrine Evergreen Tropical I. karstenii Lacustrine Evergreen Tropical I. nuttallii Terrestrial Sum. decid. Temperate I. butleri Terrestrial Sum. decid. Temperate I. stellenbosensis Terrestrial Sum. decid. Temperate Crassula erecta Terrestrial Sum. decid. Temperate C. oblanceolata Terrestrial Sum. decid. Temperate Isoetes andicola Terrestrial Evergreen Tropical I. andina Terrestrial Evergreen Tropical I. novo-granadensis Terrestrial Evergxeen Tropical Data Taxa source(c) Country Latitude Isoetes howellii 3 U.S.A. 34 [degrees] N Crassula aquatica 4 U.S.A. 34 [degrees] N C. natans 7 S. Africa 33 [degrees] N Isoetes bolanderi 5 U.S.A. 38 [degrees] N I. macrospora 7 U.S.A. 47 [degrees] N Littorella uniflora 1 Finland 61 [degrees] N Isoetes palmeri 7 Colombia 4 [degrees] N I. karstenii 7 Colombia 4 [degrees] N I. nuttallii 2 U.S.A. 38 [degrees] N I. butleri 2 U.S.A. 35 [degrees] N I. stellenbosensis 7 S. Africa 33 [degrees] S Crassula erecta 4 U.S.A. 34 [degrees] N C. oblanceolata 7 S. Africa 33 [degrees] S Isoetes andicola 6 Peru 11 [degrees] S I. andina 6 Colombia 4 [degrees] N I. novo-granadensis 6 Ecuador 0 [degrees] ([Delta][H.sup.+])(d) (mmol [kg.sup.-1] FM 24- Submerged Aerial Taxa Elev. (m) ([bar]x + SD) ([bar]x + SD) Isoetes howellii 610 294 + 22 14 + 4 Crassula aquatica 610 103 + 9 28 + 1 C. natans 200 100 4 Isoetes bolanderi 2900 187 + 9 32 + 3 I. macrospora 100 182 + 10 4 + 2 Littorella uniflora -- 141 + 12 1 + 6 Isoetes palmeri 3650 68 + 13 80 + 21 I. karstenii 3650 98 + 5 85 + 6 I. nuttallii 500 2 + 1 1 + 1 I. butleri 500 1 + 1 1 + 1 I. stellenbosensis 1200 1 + 1 2 + 1 Crassula erecta 610 3 + 1 2 + 1 C. oblanceolata 1200 3 + 1 2 + 1 Isoetes andicola 4135 -- 90 + 15 I. andina 3650 -- 182 + 22 I. novo-granadensis 4050 -- 142 + 25
(a) Seas. pool, seasonal pool.
(b) Sum. decid., summer deciduous; Win. decid., winter deciduous.
(c) 1, Aulio, 1985; 2, Keeley, 1983b; 3, Keeley & Busch, 1984; 4, Keeley & Morton, 1982; 5, Keeley et al., 1983a; 6, Keeley et al., 1994; 7, Keeley, unpubl, data.
(d) N [is greater than or equal to] 3
Aquatic CAM plants exhibit further plasticity in their adaptation to a terrestrial existence; stomata become functional or are initiated de novo, and there are increases in protein, total chlorophyll, percentage chlorophyll a, RUBISCO/PEPC ratio, and photosynthetic rate (Table III; Groenhof et al., 1988; Keeley, 1990, 1998b). Coupled with these physiological changes are subtle changes in leaf anatomy, such as increased stomatal density, thicker cuticle, and smaller lacunae (Keeley, 1990; Hostrup & Wiegleb, 1991 b). It is apparent that water potential changes at the leaf surface are involved in switching off CAM, as I. howellii maintained at [is greater than] 90% relative humidity will retain CAM in the aerial environment (Keeley, 1988), as does I. setacea (Gacia & Ballestros, 1993) and Littorella (Aulio, 1986b). These structural and functional changes are likely mediated by hormonal changes induced by lower water potentials (e.g., Schmitt et al., 1995).
Switching off CAM in the aerial environment is ultimately a response to enhanced availability of [CO.sub.2]. Despite the fact that atmospheric partial pressure of [CO.sub.2] is lower than in most aquatic habitats, substantially lower diffusional resistances in air dramatically reduce carbon limitation in the leaf boundary layer. As with terrestrial species exhibiting similar photosynthetic flexibility (e.g., Bloom & Troughton, 1979), the shift from CAM to [C.sub.3] is potentially tied to enhanced productivity in these amphibious species as well.
Some aquatic characteristics are retained in the terrestrial environment--e.g., sediment-based [CO.sub.2] uptake continues in terrestrial populations of Littorella (Nielsen et al., 1991) as well as in the non-CAM Lobelia dortmanna (Pedersen & Sand-Jensen, 1992). High cuticular resistance of the terrestrial leaves was noted by these authors as reason for hypothesizing a terrestrial origin for this mode of nutrition. However, all aquatic plants possess cuticles (Raven, 1984), and it is particularly prominent in many lacustrine Isoetes, although thickness is not a reliable indicator of permeability (Kerstiens, 1996). With respect to both sediment-based nutrition and CAM, there are clear selective advantages to cuticular development in aquatic plants.
Not all lacustrine Isoetes switch off CAM upon emergence. Some tropical alpine species, for instance, retain CAM for at least six months in an aerial environment with low humidity (Table V), and leaves initiated under terrestrial conditions fail to produce stomata.
X. Diel Acid Changes in Other Aquatic Species
Not all 69 species demonstrating significant [Delta][H.sup.+] (Table I) have been included in this discussion of aquatic CAM. In addition to the five genera already discussed, others may deserve this designation. For example, Lilaeopsis lacustris (Apiaceae) was reported to have substantial overnight accumulation of acidity and malate (Table I), but was not included due to the lack of other supporting data and absence of [Delta][H.sup.+] in other aquatic species of Lilaeopsis. Scirpus subterminalis likewise has not been included for lack of further data and the low amplitude of [Delta][H.sup.+] (Table I), which, of course, does not preclude presence of the CAM pathway.
Prudence is justified, as some species with significant [Delta][H.sup.+] clearly are not CAM. For example, Orcuttia spp. (Poaceae) have a low but consistent [Delta][H.sup.+] (Table I; Keeley, 1998a), and labeling studies indicate that malate is the first stable product of dark fixation. However, dark pulse-dark chase studies show nearly all label fixed in the dark is transferred out of the malate pool in the dark, and a substantial proportion ends up in insoluble compounds (Fig. 12A). By the end of the dark period, over 50% of the label is in citrate (not shown), suggesting that dark- fixed carbon has been transported to the mitochondria (Kalt et al., 1990; Olivares et al., 1993). Eleocharis acicularis (Cyperaeeae) exhibits a similar pattern of malate turnover in the dark (Keeley, unpubl. data). Hydrilla verticillata was early documented as exhibiting dark fixation and slight acid accumulation (Holaday & Bowes, 1980). It, too, metabolizes a substantial portion of the dark-fixed carbon in the dark in apparently non-autotrophic metabolism (Fig. 12B). These observations do not conclusively demonstrate absence of the CAM pathway, as even well-recognized terrestrial CAM plants utilize some portion of dark-fixed carbon for non-autotrophic metabolism (Luttge, 1988). However, when coupled with data on rates of uptake, it appears that dark [CO.sub.2] fixation in these species may not contribute significantly to autotrophism. Typological designations such as CAM are always problematic when dealing with phenomena that vary quantitatively.
[Figure 12 ILLUSTRATION OMITTED]
Downingia bella has [CO.sub.2] fixation in the dark, and the fact that malate accumulates (Fig. 12C) suggests it may contribute to autotrophism, but this species lacks certain CAM criteria: It exhibits a highly significant Amalate, but, despite repeated sampling, there is no indication of [Delta][H.sup.+] (Table I). It is comparable to Isoetes in the RUBISCO/PEPC ratio, and activity of NADP Malic Enzyme and pyruvate, [P.sub.i]-dikinase (Keeley, 1998b). This plant deserves further study, as it is a prime candidate for the scheme proposed by Raven et al. (1988) for a CAM mechanism that would couple [H.sup.+] disposal with [K.sup.+] uptake. They envisioned an autotrophic pathway that would simulate CAM in most details, except [malate.sup.2-] + 2[K.sup.+] would be stored in the vacuole, resulting in significant [Delta]malate but no [Delta][H.sup.+], as is observed in D. bella (Table I).
Some marine algae in all three of the major phyla have long been noted for their dark [CO.sub.2] fixation (e.g., Joshi et al., 1962; Akagawa et al., 1972b; Willenbrink et al., 1979; Church et al., 1983), and certain of the brown algae (Phaeophyta) have significant [Delta][H.sup.+] ((Table I). This, coupled with evidence of photosynthetic use of endogenous [CO.sub.2] (Ryberg et al., 1990), has evoked labels of CAM and CAM-like for several brown algae (Johnston & Raven, 1986; Raven & Samuelsson, 1988; Axelsson et al., 1989; Raven et al., 1989; Raven & Osmond, 1992). One such species is the well-studied Ascophyllum nodosum, which has been reported to accumulate 10-20 mmol [H.sup.+] kg FM (Surif & Raven, 1983; Johnston & Raven, 1986). Deviations from CAM are evident in the type of carboxylating enzyme (PEP carboxykinase: Kremer, 1979; Kerby & Evans, 1983) and lack of carbon storage in malate; only 5% of dark-fixed carbon remains in malate at the end of the 12 hr dark period (Fig. 13). Products labeled in the dark include glutamate, aspartate, succinate, and various amino acids, but during the dark period chase, most label accumulates in fumarate and citrate (Keeley, unpubl. data), which are organic acids not likely to act as carbon storage compounds for autotrophism (Luttge, 1988). These labeling patterns are not markedly different from those observed for other brown algae (Akagawa et al. 1972a; Kremer, 1979; Coudret et al., 1992).
[Figure 13 ILLUSTRATION OMITTED]
Documenting the potential non-autotrophic uses of dark-fixed carbon is beyond the scope of this review. However, it is worth noting that dark [CO.sub.2] fixation may contribute carbon to several pathways, though not necessarily tied to acid accumulation. Non-autotrophic uses of dark-fixed carbon include involvement as a pH stat mechanism for reducing cytoplasmic ionic disequilibrium (Raven, 1986) or in anaplerotic reactions related to nitrogen assimilation (Turpin et al., 1991). Relevant to the latter mechanism, dark [CO.sub.2] fixation in the macrophytic brown algae Ascophyllum nodosum can be stimulated under enhanced nitrogen conditions (Keeley, unpubl.data).
XI. Systematic Distribution
Significant [Delta][H.sup.+] has not been detected in either the Chlorophyta or Rhodophyta, and the acidification cycle in the brown algae (Phaeophyta) may not represent CAM (Section X). Apparent restriction of CAM to the Tracheophyta may be explained in part by the greater carbon allocation to cell wall material in these macrophytes, resulting in C acquisition being a more rate-limiting step than N, P, or Fe acquisition (Raven & Spicer, 1995).
Within the vascular plant flora, aquatic CAM plants are from widely unrelated taxa, such as lycopods, monocots, and dicots. Of the 134 vascular plant species reported here, 37% had CAM but more could be added with additional information. Estimating the proportion of the world's aquatic flora with CAM is problematic due to incomplete information on the total number of amphibious species. If we restrict our attention to just those 33 characteristic aquatic families listed by Sculthorpe (1967), thus removing species of Crassula and Littorella from our analysis, and assuming all aquatic Isoetes are CAM, it is calculated that 6% of the aquatic flora is CAM, which compares exactly with the 6% reported for terrestrial floras (Winter & Smith, 1995a).
Of course, such comparisons are phylogenetically biased because of the potential linkage of CAM and the aquatic habitat in certain linkages. While lacking a precise phylogenetically corrected comparison (Eggleton & Vane-Wright, 1993), we can obtain a less biased view of aquatic and terrestrial CAM distribution by focusing at the family level. Of the 33 aquatic families (representatives in about one-half have been tested), three (Isoetaceae, Alismataceae, and Hydrocharitaceae) have evolved CAM, or 9% of the aquatic plant families. For comparison with the distribution of terrestrial CAM, most attention has been focused on flowering plants, where there are 26 families with CAM (Smith & Winter, 1995). Based on an estimated 321 "terrestrial" families [349 flowering plant families reported by Stebbins (1974), minus the aquatic families considered above], gives an estimate that 8% of the terrestrial plant families have CAM, quite comparable to the aquatic flora. This suggests that CAM has had an equal likelihood of evolving in water as on land.
XII. Evolution of Aquatic CAM Plants
Being restricted to the Tracheophyta means that CAM is found only in secondarily aquatic plants. Did CAM originate in an aquatic milieu or was it present in terrestrial ancestors?
In Isoetes, the earliest aquatic CAM plants, the view that they represent recent herbaceous descendants of a long linear reduction sequence from the arborescent Lepidodendrales (Stewart, 1983), could be interpreted as suggesting a terrestrial or at least emergent-aquatic origin for the group. However, recent evidence disputes this view and suggests that Isoetes's origins are tied to similar aquatic corm-bearing plants well developed in the Carboniferous, which coexisted with arborescent Lycophyta (Taylor, 1981; Skog & Hill, 1992; Kovach & Batten, 1993; DiMichele & Bateman, 1996). Several recent studies have shown complete Isogtes specimens in early Triassic ([is greater than] 230 Ma) sediments, apparently forming dense monocultures in ephemeral pools (Wang, 1996; Retallack, 1997). Throughout the Triassic, these Isoetes coexisted with other herbaceous Lycophyta, such as the extinct Tomiostrobus (Retallack, 1997) and Isoetites Munster (Ash & Pigg, 1991; Pigg, 1992), both of which were amphibious, and remarkably similar to extant Isoetes. Indeed, Hickey (1986, 1990) suggests that the three neotropical Isoetes that form subgenus Euphyllum are basal to the genus and "represent relictual morphotypes" of the extinct Isoetites. These species (and perhaps I. worrnwaldii from South Africa) have in common a laminate leaf, which clearly separates them from the rest of Isoetes. Although Isoetites were cosmopolitan, these primitive Isoetes have populations that are highly restricted (and mostly extirpated), but like Isoetites they are aquatic.
Based on a cost-benefit evaluation of atmospheric conditions, Raven and Spicer (1995) speculated that terrestrial environments conducive to CAM were unlikely during geological periods relevant to the early evolution of the Isoetaceae. Their arguments, however, apply less to aquatic habitats, where biogenic processes buffer the system from the impact of atmospheric changes in [CO.sub.2]. Carbon-limiting factors conducive to aquatic CAM evolution, such as diel changes in [CO.sub.2] availability in shallow seasonal pools, could have been present since the early Triassic history of the Isoetaceae. In addition, the rising temperature of the early Triassic (Spicer, 1993) would have exacerbated the tendency for sharp diel changes in [CO.sub.2] availability in shallow pools.
This scenario is supported by other observations. Based on the widespread distribution of derived traits, it is apparent that the initial morphological divergence from ancestral aquatic Isoetites, giving rise to modem Isoetes, was in traits conducive to surviving dry dormant periods, indicative of an amphibious origin for the group (Hickey, 1986; Taylor & Hickey, 1992). Such an amphibious lifecycle is also supported by the presence of stomata in the earliest known Isoetes and in other paleoecological characteristics (Retallack, 1997). Possibly the origin of Isoetes was in amphibious habitats at the edges of Triassic swamps. Such habitats would have had diel changes in carbon limitation, which would have favored the evolution of CAM. High organic matter in these swamp sediments may also have favored [CO.sub.2] uptake from the sediment, as suggested by the similarity in lacunal volume between Isogtes roots and fossil roots of the extinct Stigmaria (Karrfalt, 1980) and Pleuromeia (Munster) Corda (Grauvogel-Stamm, 1993), and this in mm would have favored the evolution of CAM (Osmond, 1984).
The Cretaceous radiation of modem Isogtes (Pigg, 1992), into less fertile lacustrine habitats (Hickey, 1986), may reflect increasing competition from faster-growing aquatic flowering plants (Section VII). If so, CAM would have been an important pre-adaptation to colonizing these oligotrophic lacustrine habitats.
An amphibious origin for CAM keeps alive Cockburn's (1981) "stomatal-hypothesis," but other biochemical origins are equally reasonable (Osmond, 1984; Winter, 1985). Griffiths's (1989) suggestion that CAM evolution proceeded from dark refixation of respiratory carbon to dark uptake, would not apply to aquatic CAM plants, since [CO.sub.2] uptake in these plants is not dependent on evolution of unique stomatal behavior. The near-ubiquitous presence of CAM photosynthesis in Isoetes suggests that CAM has had a long and monophyletic relationship with the group and therefore Isoetes represents the oldest clade of CAM plants (Winter & Smith, 1995b). Thus, the evolution of CAM photosynthesis dates back to the Paleozoic or shortly thereafter.
Despite this apparently very early origin for CAM, its widespread and highly disjunct phylogenetic distribution leads to the inescapable conclusion that, within the Tracheophyta, it is not a homologous trait (Luttge, 1987; Monson, 1989; Ehleringer & Monson, 1993). Further insights into the evolution of aquatic CAM photosynthesis are possible through comparative studies of certain taxa. Particularly promising are Isogtes and Crassula, which are large genera (100-200 species), dominated by CAM species but also having non-CAM species. Comparison of these genera is of interest because Isogtes comprises mostly aquatics with a few terrestrial species, whereas Crassula is mostly terrestrials, with very few aquatic species.
A. PATTERNS OF RADIATION IN ISOETES
Cladistic analysis indicates that radiation of modem Isoetes has been from seasonal pools into both terrestrial habitats and infertile lacustrine habitats (Hickey, 1986; Taylor & Hickey, 1992).
1. Putative Amphibious-to-Terrestrial Transitions
Evolutionary changes in photosynthetic biology occurred in the transition from water to land. Strictly terrestrial(1) species I. nuttallii and I. butleri, of western and eastern North America, respectively, and L stellenbossiensis, from the Cape Province of South Africa, lack CAM even when artificially submerged (Table V); possibly the terrestrial I. durieui of Europe is similar (Richardson et al., 1984). Lack of CAM, high [Delta] [sup.13]C values, and the absence of Kranz anatomy indicates that these terrestrials are [C.sub.3], which is consistent with their summer-deciduous nature, as there are few, if any, examples of [C.sub.4] or CAM terrestrial geophytes. These Temperate Zone terrestrial species are summer-deciduous plants restricted to vernally moist sites with relatively short growing seasons. They have functional stomata and develop rapidly until dormancy is imposed by drought, even in summer-rain climates (Baskin & Baskin, 1979). Normal growing conditions are similar to those experienced by amphibious species following dry-down of the seasonal pool habitat. An aquatic ancestry is supported by the presence of four lacunal chambers, structures that are atypical for terrestrial plants and missing from terrestrial outgroups in the Lycophyta (Hickey, 1986). Consistent with this model is the placement of terrestrial I. butleri as an offshoot of a clade that has radiated into various amphibious habitats (Hickey et al., 1989). On the other side of the continent, a similar origin applies to the terrestrial I. nuttallii, which would appear to be a recent derivative of the amphibious I. orcuttii; these species are so close that they have been synonymized in some taxonomic treatments.
In summary, I. nuttallii, I. butleri, I. stellenbossiensis, and I. durieui--plus an unnamed species from Chile (Keeley & Hickey, unpubl, data) and probably species from Australia (Keeley, unpubl, data)--are secondarily terrestrial and secondarily [C.sub.3]. Systematic (Pfeiffer, 1922) and cladistic (Hickey, 1986; Hickey et al., 1989; Taylor & Hickey, 1992) analyses suggest a polyphyletic origin for this terrestrial syndrome.
2. Putative Amphibious-to-Lacustrine-to-Terrestrial Transitions
Given the absence of many plesiomorphic traits, it appears that lacustrine species of Isoetes are more recently derived from amphibious ancestors (Hickey, 1986; Taylor & Hickey, 1992). CAM would have assisted in the invasion of these infertile lakes, and these sites would have enhanced further development of sediment-based [CO.sub.2] uptake. Many of these aquatic species have retained the facultative responses to emergence so that, under terrestrial conditions, they develop stomata and switch off CAM (Section IX).
However, in some neotropical alpine lacustrine species, adaptations to the aquatic environment appear to be genetically fixed; when grown in air, they retain CAM and fail to produce stomata (Section IX). This constitutive response could reflect a much earlier origin, an idea consistent with the neotropical distribution of the most primitive Isogtes (Hickey, 1990). These neotropical alpine species often grow in relatively flat lake basins subject to siltation, and as a consequence many have very long leaves, with the lower % buried in the sediment.
Adjacent to many lakes, from Peru to Colombia, are terrestrial Isogtes that are likewise "buried" in the sediment. They are evergreen with astomatous leaves and are the only extant terrestrial species of Tracheophyta lacking stomata. One of these terrestrial species is I. [Stylites] andicola, which has roots extending [is greater than] 2 m in depth and a below-round:above-ground biomass ratio [is greater than] 15. These plants obtain most of their carbon from the sediment by diffusion through hollow roots and are CAM. These patterns have been verified experimentally (Keeley et al., 1984, 1994) and with isotopes; depletion in [sup.14]C, relative to contemporary atmospheric levels, supports the conclusion of sediment-based nutrition, and high deuterium verifies the importance of CAM (Sternberg et al., 1985). Retention of CAM (Table V) in these neotropical terrestrial Isoetes would be favored by the accumulation of lacunal [CO.sub.2] at night and by the highly cutinized astomatous leaves, which provide diffusive resistance to [CO.sub.2] leakage during daytime deacidification.
The fact that these terrestrial species have retained the conservative lacunal leaf architecture suggests an aquatic ancestry for these species. These terrestrials are all high polyploids (2n = 44-132: J. Hickey, pers. comm.), and a polyphyletic origin for this syndrome is supported both by flavonoid patterns between terrestrial and nearby lacustrine species and by the presence of this terrestrial syndrome in widely disjunct Isoetes in South America and Papua New Guinea (Keeley et al., 1994).
B. PATTERNS OF RADIATION IN CRASSULA
All terrestrial perennial species of Crassula have the CAM pathway, although stomatal behavior and gas exchange patterns are plastic (Pilon-Smits et al., 1995), and nearly all are restricted to Southern Africa (Tolken, 1977). Annual species, on the other hand, occur throughout the world and include both aquatic and terrestrial plants. Aquatic annuals from four continents, occurring in both seasonal pools and lakes, have been tested: All are CAM (Table I) and all are closely related in the subgenus Disporocarpa (Tolken, 1977, 1981; Bywater & Wickens, 1984). Two terrestrial annual species in Disporocarpa lack CAM and CAM can not be induced (Table V), and these are perhaps the only members of the family completely lacking the CAM pathway (cf. Pilon-Smits et al., 1995). Arguments similar to those proposed above for the loss of CAM in Temperate Zone terrestrial Isoetes would apply to these terrestrial Crassula, which occupy similar seasonal environments.
The present distribution of Crassula suggests a South African origin for the group and long-distance dispersal of the annual species or their progenitors, likely accounts for their global distribution. Such dispersal is most probable for aquatic species, which are distributed in habitats more likely to be frequented by migrating birds, and the seeds (dispersed into the mud) have a high probability of sticking to long-distance dispersers (Raven, 1963). Thus, the terrestrial annuals are probably secondarily terrestrial and secondarily Cs. Since the rest of the Crassulaceae family is both terrestrial and CAM, it would perhaps be prudent to suggest that CAM was present in terrestrial ancestors giving rise to aquatic CAM species. However, species in Disporocarpa are apparently basal to the genus (Tolken, 1977), which makes it at least plausible that terrestrial CAM plants in Crassula may be derived from aquatic CAM species.
XIII. Conclusions and Areas for Future Research
CAM is a [CO.sub.2]-concentrating mechanism. The immediate or proximal selective advantage is the provision of an endogenous [CO.sub.2] source for photosynthesis. This has arisen in two environments with different selective forces. On land the ultimate selective factor has been to enhance water use efficiency, and in aquatic habitats the ultimate selective factor has been to diminish the threat of carbon starvation--the "desiccation vs. starvation" dilemma of Luttge (1987). As a concentrating mechanism, a primary function of CAM is to enhance the [CO.sub.2](Pi) sufficiently to overcome photorespiratory effects. This requires daytime decarboxylation of overnight malate stores in a system with sufficient diffusional resistances to allow accumulation of [CO.sub.2] and prevent leakage. In terrestrial plants this requires increasing stomatal resistance, whereas in aquatic plants this is largely effected by the [10.sub.4] greater diffusional resistance of the water. An additional factor may be the relatively thick cuticle characteristic of most Isoetes, although little is known about their permeability characteristics. A valuable contribution would be comparative studies of resistances contributing to [CO.sub.2] disequilibria in aquatic plants.
In both terrestrial and aquatic CAM plants, dark [CO.sub.2] fixation may result in net carbon uptake plus the conservation of carbon by re fixation of respiratory [CO.sub.2]. In aquatic plants, CAM's contribution to the total carbon budget is variable. Exemplary studies of the contribution of CAM to the carbon budget, such as those by Boston and Adams, Madsen, and Robe and Griffiths for lacustrine species, are needed in a greater range of habitats. Quantitative estimates of the CAM contribution to the carbon budget are likely to provide more insights than attempts to typologically categorize variation with terms such as "idling," "cycling," AAM, SCAM, TAAM, and so forth.
Although we have a reasonably good understanding of the selective factors favoring CAM in seasonal pools and oligotrophic lakes, there are other habitats (Section VII.C) where the role of CAM is not apparent. These species need to be examined in greater detail.
Future research should focus on species with predictable diel acid fluctuations, but with characteristics that do not fit recognized criteria for CAM. Of particular interest is the seasonal pool species Downingia bella (Campanulaceae), which may reflect an innovative CAM mechanism. Other roles for dark [CO.sub.2] fixation should be examined. Dark [CO.sub.2] fixation may be important as a source of carbon skeletons for both carbon and nitrogen assimilation, particularly in nutrient-poor habitats.
Of practical concern is the manner in which lake acidification and eutrophication alter carbon budgets (e.g., Robe & Griffiths, 1994). Also, in many parts of the globe aquatic CAM species are threatened: I. andicola of Peru, for instance, is clearly threatened by habitat loss (Leon & Young, 1996), and two of the three primitive Isoetes, morphologically similar to the extinct Isoetites, are apparently extinct (Hickey, 1986). At the other extreme, the aquatic CAM Crassula helmsii is an aggressive alien (Dawson & Warman, 1987), in need of further studies such as those of Newman and Raven (1995) in a greater range of habitats.
Isoetes, being the oldest lineage of CAM plants, potentially holds further interesting discoveries with respect to photosynthetic patterns. The most primitive species in the group are distinct in their lack of the typical terete "isoetid" leaf. These species are restricted to isolated sites in South America and have seldom been collected. They are apparently basal to the group, sharing the laminate leaf characteristic with the extinct and possibly ancestral Isoetites (Hickey, 1986). The hypothesized amphibious origin for CAM suggests the possibility that these primitive species may lack CAM. Further study of the photosynthetic metabolism and habitat characteristics of these would be a stimulating contribution to the story of aquatic CAM photosynthesis. Here, and in other aspects of aquatic CAM photosynthesis, a multitude of possibilities are presented with new molecular genetic techniques, now being applied to terrestrial CAM plants (Cushman & Bohnert, 1997).
Thanks to John Raven, C. Barry Osmond, Renee Gonzalez, Darren Sandquist, and Teresa Swida for direct contributions to this work; to Renee Gonzalez for the Spanish resumen; and to the many other authors contributing to the literature cited here. This work was supported by grants from the National Science Foundation, the American Philosophical Society, the National Geographic Society, and the John Simon Guggenheim Foundation.
XV. Literature Cited
Akagawa, H., T. Ikawa & K. Nisizawa. 1972a. Initial pathway of dark 14[CO.sub.2]-fixation in brown algae. Bot. Mar. 15: 119-125.
--, -- & --. 1972b. [sup.14][CO.sub.2]-fixation in marine algae with special reference to the darkfixation in brown algae. Bot. Mar. 15: 126-132.
Allen, E. D. & D. H. N. Spence. 1981. The differential ability of aquatic plants to utilize the inorganic carbon supply in fresh waters. New Phytol. 87: 269-283.
Allsopp, A. 1951. The sugars and non-volatile organic acids of some archegoniates: A survey using paper chromatography. J. Exp. Bot. 2: 121-124.
Ash, S. R. & K. B. Pigg. 1991. A new Jurassic Isoetites (Isoetales) from the Wallowa terrane in Hells Canyon, Oregon and Idaho. Amer. J. Bot. 78: 1636-1642.
Aulio, K. 1985. Differential expression of diel acid metabolism in two life forms of Littorella uniflora (L.) Aschers. New Phytol. 100: 533-536.
--. 1986a. CAM-like carbon pathway in submerged aquatic plants. Biol. Pl. 28: 234-236.
--. 1986b. CAM-like photosynthesis in Littorella uniflora (L.) Aschers.: The role of humidity. Ann. Bot. 58: 273-275.
Axelsson, L., S. Carlberg & H. Ryberg. 1989. Adaptations by macroalgae to low carbon availability. I. A buffer system in Ascophyllum nodosum, associated with photosynthesis. Pl. Cell Environm. 12: 765-770.
Baskin, J. M. & C. C. Baskin. 1979. The role to temperature in the vegetative life cycle of Isoetes butleri. Amer. Fern J. 69: 103-110.
Beer, S. & R. G. Wetzel. 1981. Photosynthetic carbon metabolism in the submerged aquatic angiosperm Scirpus subterminalis. Pl. Sci. Lett. 21: 199-207.
--, K. Sand-Jensen, T. V. Madsen & S. L. Nielsen. 1991. The carboxylase activity of Rubisco and the photosynthetic performance in aquatic plants. Oecologia 87: 429-434.
Black, C. C., J.-Q. Chen, R. L. Doong, M. N. Angeiov & S. J. S. Sung. 1995. Alternative carbohydrate reserves used in the daily cycle of crassulacean acid metabolism. Pages 31-45 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
Bloom, A. J. & J. H. Troughton. 1979. High productivity and photosynthetic flexibility in a CAM plant. Oecologia 38: 35-43.
Bold, H. C., C. J. Alexopoulos & T. Delevoryas. 1980. Morphology of plants and fungi. Harper & Row, New York.
Borland, A. M. & H. Griffiths. 1995. Variations in the phases of crassulacean acid metabolism and regulation of carboxylation patterns determined by carbon-isotope-discrimination techniques. Pages 230-249 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
Boston, H. L. 1986. A discussion of the adaptations for carbon acquisition in relation to the growth strategy of aquatic isoetids. Aquatic Bot. 26: 259-270.
-- & M. S. Adams. 1983. Evidence of crassulacean acid metabolism in two North American isoetids. Aquatic Bot. 15: 381-386.
-- & --. 1985. Seasonal diurnal acid rhythms in two aquatic crassulacean acid metabolism plants. Oecologia 65: 573-579.
-- & --. 1986. The contribution of crassulacean acid metabolism to the annual productivity of two aquatic vascular plants. Oecologia 68:615-622.
-- & --. 1987. Productivity, growth, and photosynthesis of two small "isoetid" plants Littorella uniflora and Isoetes macrospora. J. Ecol. 75: 333-350.
--, -- & T. P. Pienkowski. 1987. Utilization of sediment [CO.sub.2] by selected North American isoetids. Ann. Bot. 60: 485-494.
--, -- & J. D. Madsen. 1989. Photosynthetic strategies and productivity in aquatic systems. Aquatic Bot. 34: 27-57.
Bowes, G. & M. E. Salvucci. 1989. Plasticity in the photosynthetic carbon metabolism of submersed aquatic macrophytes. Aquatic Bot. 34: 233-266.
Broadmeadow, M. S. J. & H. Griffiths. 1993. Carbon isotope discrimination and the coupling of [CO.sub.2] fluxes with forest canopies. Pages 109-129 in J. R. Ehleringer et al. (eds.), Stable isotopes and plant carbon--water relations. Academic Press, San Diego.
Browse, J. A., J. M. A. Brown & F. I. Dromgoole. 1980. Malate synthesis and metabolism during photosynthesis in Egeria densa Planch. Aquatic Bot. 8: 295-305.
Bywater, M. & T. E. Wickens. 1984. New World species of the genus Crassula. Kew Bull. 39: 699-728.
Chellappan, K. P., S. Seeni & A. Gnanam. 1980. Photosynthetic studies with mesophyll protoplasts from Notonia grandiflora, a crassulacean acid metabolism plant. Physiol. Pl. 48: 403-410.
Christopher, J. T. & J. A. M. Holtum. 1996. Patterns of carbon partitioning in leaves of Crassulacean acid metabolism species during deacidification. Pl. Physiol. 112: 393-399.
Church, M. R., R. R. H. Cohen, C. L. Gallegos & M. G. Kelly. 1983. Evidence for carbon uptake and storage in the dark with subsequent photosynthetic fixation by cultures of mixed Chlorophyceae. Arch. Hydrobiol. 98: 509-522.
Cockburn, W. 1981. The evolutionary relationship between stomatal mechanism, crassulacean acid metabolism, and [C.sub.4] photosynthesis. New Phytol. 24: 3-24.
Collins, C. D., R. B. Sheldon & C. W. Boylen. 1987. Littoral zone macrophyte community structure: Distribution and association of species along physical gradients in Lake George, New York, U.S.A. Aquatic Bot. 29: 177-194.
Coudret, A., F. Ferron, P. Jolivet, G. Tremblin, E. Bergeron & M. C. Sourie. 1992. Fate of metabolites issued from [sup.14]C dark fixation into brown seaweeds: Connection with their respective biology. Photosynthetica 26: 235-245.
Cushman, J. C. & H. J. Bohnert. 1997. Molecular genetics of crassulacean acid metabolism. Pl. Physiol. 113: 667-676.
Dawson, F. H. & E. A. Warman. 1987. Crassula helmsii (T. Kirk) Cockayne: Is it an aggressive alien aquatic plant in Britain? Biol. Conserv. 42: 247-272.
DiMichele, W. A. & R. M. Bateman. 1996. The rhizomorphic lycopsids: A case-study in paleobotanical classification. Syst. Bot. 21: 535-552.
Dittrich, P., W. H. Campbell & J. C. C. Black. 1973. Phosphoenolpymvate carboxykinase in plants exhibiting crassulacean acid metabolism. Pl. Physiol. 52: 357-361.
Eggleton, P. & R. I. Vane-Wright (eds.). 1993. Phylogenetics and ecology. Academic Press, San Diego.
Ehleringer, J. R. & R. K. Monson. 1993. Evolutionary and ecological aspects of photosynthetic pathway variation. Annual Rev. Ecol. Syst. 24: 411-439.
Farmer, A. M. 1987. Photosynthetic adaptation in Lobelia dortmanna and other isoetids. Photosynthetica 21: 51-55.
-- & D. H. N. Spence. 1985. Studies of diurnal acid fluctuations in British isoetid-type submerged aquatic macrophytes. Ann. Bot. 56: 347-350.
-- & -- .1986. The growth strategies and distribution of isoetids in Scottish freshwater lochs. Aquatic Bot. 26: 247-258.
--, S. C. Maberly & G. Bowes. 1986. Activities of carboxylation enzymes in freshwater macrophytes. J. Exp. Bot. 37: 1568-1573.
Fischer, A. & M. Kluge. 1985. Light period of crassulacean acid metabolism: Control of carbon transfer from malic acid to carbohydrates by [CO.sub.2] concentration. Physiol. Pl. 63: 327-331.
Gacia, E. & E. Ballestros. 1993. Diel acid fluctuations in pyrenean Isoetes species: The effects of seasonality and emersion. Arch. Hydrobiol. 128:187-196.
-- & --. 1994. Production of Isoetes lacustris in a Pyrenean lake: Seasonality and ecological factors involved in the growing period. Aquatic Bot. 48: 77-89.
-- & J. Penuelas. 1991. Carbon assimilation of Isoetes lacustris L. from Pyrenean lakes. Photosynthetica 35: 97-104.
--, E. Ballestros, L. Camarero, O. Delgado, A. Palau, J. L. Riera & J. Catalan. 1994. Macrophytes from lakes in the eastern Pyrenees: Community composition and ordination in relation to environmental factors. Freshwater Biol. 32:73-81.
Gibson, A. C. & P. S. Nobel. 1986. The cactus primer. Harvard University Press, Cambridge.
Grauvogel-Stamm, L. 1993. Pleuromeia sternbergii (Munster) Cords from the lower Triassic of Germany-Further observations and comparative morphology of its rooting organ. Rev. Palaeobot. Palyn. 77: 185-212.
Griffiths, H. 1988. Crassulacean acid metabolism: A re-appraisal of physiological plasticity in form and function. Adv. Bot. Res. 15: 43-92.
--. 1989. Carbon dioxide concentrating mechanisms and the evolution of CAM in vascular epiphytes. Pages 42-86 in U. Luttge (ed.), Vascular plants as epiphytes: Evolution and ecophysiology. Springer, Berlin.
Groenhof, A. C., N. Smirnoff & J. Bryant. 1988. Enzymic activities associated with the ability of aerial and submerged forms of Littorella uniflora (L.) Aschers. to perform CAM. J. Exp. Bot. 39: 353-361.
Helder, R. J. & M. V. Harmelen. 1982. Carbon assimilation pattern in the submerged leaves of the aquatic angiosperm: Vallisneria spiralis L. Acta Bot. Need. 31:281-295.
Hickey, R. J. 1986. The early evolutionary and morphological diversity of Isoetes, with descriptions of two new neotropical species. Syst. Bot. 11:309-321.
--. 1990. Studies of neotropical Isoetes L. I. Euphyllum, a new subgenus. Ann. Missouri Bot. Gar & 77: 239-245.
--, W. C. Taylor & N. T. Luebke. 1989. The species concept in Pteridophyta with special reference to Isoetes. Amer. Fern J. 79: 78-89.
Holaday, A. S. & G. Bowes. 1980. [C.sub.4] acid metabolism and dark [CO.sub.2] fixation in a submersed aquatic macrophyte (Hydrilla verticillata). Pl. Physiol. 65:331-335.
Holbrook, G. P., S. Beer, W. E. Spencer, J. B. Reiskind, J. S. Davis & G. Bowes. 1988. Photosynthesis in marine macroalgae: Evidence for carbon limitation. Canad. J. Bot. 66: 577-582.
Holtum, J. A. M. & K. Winter. 1982. Activity of enzymes of carbon metabolism during the induction of crassulacean acid metabolism in Mesembryanthemum crystallinum. Planta 155: 8-16.
Hostrup, O. & G. Wiegleb. 1991a. The influence of different [CO.sub.2] concentrations in the light and the dark on diurnal malate rhythm and phosphoenolpyruvate carboxylase activities in leaves of Littorella uniflora (L.) Aschers. Aquatic Bot. 40: 91-100.
-- & --. 1991b. Anatomy of leaves of submerged and emergent forms of Littorella uniflora (L.) Ascherson. Aquatic Bot. 39:195-209.
Jackson, S. T. & D. F. Charles. 1988. Aquatic macrophytes in Adirondack (New York) lakes: Patterns of species composition in relation to environment. Canad. J. Bot. 66: 1449-1460.
Johnston, A. M. & J. A. Raven. 1986. Dark carbon fixation studies on the intertidal macroalga Ascophyllum nodosum (Phaeophyta). J. Phycol. 22: 78-83.
Joshi, G., T. Dolan, R. Gee & P. Saltman. 1962. Sodium chloride effect on dark fixation of [CO.sub.2] by marine and terrestrial plants. Pl. Physiol. 37: 446-- 449.
Kalt, W., C. B. Osmond & J. N. Siedow. 1990. Malate metabolism in the dark after [sup.13][CO.sub.2] fixation in the crassulacean plant Kalanchoe tubiflora. PI. Physiol. 94: 826-832.
Karrfalt, E. 1977. Substrate penetration by the corm of Isoetes. Amer. Fern J. 67: 1-4.
--. 1980. A further comparison of Isoetes roots and stigmarian appendages. Canad. J. Bot. 58: 2318-2322.
--. 1984. The origin and early development of the root-producing meristem of Isoetes andicola L. D. Gomez. Bot. Gaz. 145: 372-377.
Keeley, J. E. 1981. Isoetes howellii: A submerged aquatic CAM plant? Amer. J. Bot. 68: 420-224.
--. 1982. Distribution of diurnal acid metabolism in the genus Isoetes. Amer. J. Bot. 69: 254-257.
--. 1983a. Crassulacean acid metabolism in the seasonally submerged aquatic Isoetes howellii. Oecologia 58: 57-62.
--. 1983b. Lack of diurnal acid metabolism in two terrestrial Isoetes species. Photosynthetica 17: 93-94.
--. 1987. The adaptive radiation of photosynthetic modes in the genus Isoetes (Isoetaceae). Pages 113-128 in R. M. M. Crawford (ed.), Plant life in aquatic and amphibious habitats. Blackwell Scientific, London.
--. 1988. Photosynthesis in quillworts, or, Why are some aquatic plants similar to cacti? Plants Today (July-August): 127-132.
--. 1989. Stable carbon isotopes in vernal pool aquatics of differing photosynthetic pathways. Pages 76-81 in P. W. Rundel et al. (eds.), Stable isotopes in ecological research. Springer, New York.
--. 1990. Photosynthesis in vernal pool macrophytes: Relation of structure and function. Pages 61-85 in D. H. Ikeda & R. A. Schlising (eds.), Vernal pool plants: Their habitat and biology. Studies from the Herbarium, no. 8. California State University, Chico.
--. 1991. Interactive role of stresses on structure and function of aquatic plants. Pages 329-343 in H. A. Mooney et al. (eds.), Response of plants to multiple stresses. Academic Press, New York.
--. 1996. Aquatic CAM photosynthesis. Pages 281-295 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
--. 1998a. Diel acid fluctuations in aquatic grasses. Photosynthetica (in press).
--. 1998b. Photosynthetic pathway diversity in a seasonal pool community. Funct. Ecol. (in press).
-- & G. Bowes. 1982. Gas exchange characteristics of the submerged aquatic crassulacean acid metabolism plant Isoetes howellii. Pl. Physiol. 70: 1455-1458.
-- & G. Busch. 1984. Carbon assimilation characteristics of the aquatic CAM plant Isoetes howellii. Pl. Physiol. 76: 525-530.
-- & B. A. Morton. 1982. Distribution of diurnal acid metabolism in submerged aquatic plants outside the genus Isoetes. Photosynthetica 16: 546-553.
-- & D. R. Sandquist. 1991. Diurnal photosynthesis cycle in CAM and non-CAM seasonal-pool aquatic macrophytes. Ecology 72: 716-727.
-- & --. 1992. Carbon: Freshwater plants. Pl. Cell Environm. 15: 1021-1035.
-- & P. H. Zedler. 1998. Characterization and global distribution of vernal pools. Pages 1-14 in C. W. Witham et al. (eds.), Ecology, conservation and management of vernal pool ecosystems-Proceedings from a 1996 conference. California Native Plant Society, Sacramento.
--, R. P. Matthews, B. Babcock, P. Castillo, B. Fish, E. Jerauld, B. Johnson, L. Landre, H. Lure, C. Miller, A. Parker & G. van Steenwyk. 1981. Dark [CO.sub.2]-fixation and diurnal malic acid fluctuations in the submerged aquatic Isoetes storkii. Oecologia 48: 332-333.
--, C. M. Walker & IL P. Mathews. 1983a. Crassulacean acid metabolism in Isoetes bolanderi in high-elevation oligotrophic lakes. Oecologia 58: 63-69.
--, R. P. Mathews & C. M. Walker. 1983b. Diurnal acid metabolism in Isoetes howellii from a temporary pool and a permanent lake. Amer. J. Bot. 70: 854-857.
--, C. B. Osmond & J. A. Raven. 1984. Stylites, a vascular land plant without stomata absorbs [CO.sub.2] via its roots. Nature 310: 694-695.
--, D. A. DeMason, R. Gonzalez & K. IL Markham. 1994. Sediment-based carbon nutrition in tropical alpine Isoetes. Pages 167-194 in P. W. Rundel et al. (eds.), Tropical alpine environments. Plant form and function. Cambridge University Press, Cambridge.
Kelly, G. J., J. A. M. Holtum & E. Latzko. 1989. Photosynthesis, carbon metabolism: New regulators of [CO.sub.2] fixation, the new importance of pyrophosphate, and the old problem of oxygen involvement revisited. Prog. Bot. 50: 87-95.
Kerby, N. W. & L. V. Evans. 1983. Phosphoenolpymvate carboxykinase activity in Ascophyllum nodosum (Phaeophyceae). J. Phycol. 19: 1-3.
Kerstiens, G. 1996. Cuticular water permeability and its physiological significance. J. Exp. Bot. 47: 1813-1832.
Kirk, J. T. O. 1983. Light and photosynthesis in aquatic ecosystems. Cambridge University Press, Cambridge.
Kluge, M. & J. Brulfert. 1995. Crassulacean acid metabolism in the genus Kalanchoe: Ecological, physiological, and biochemical aspects. Pages 324-335 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
-- & I. P. Ting. 1978. Crassulacean acid metabolism. Analysis of an ecological adaptation. Springer, New York.
Kovach, W. L. & D. J. Batten. 1993. Diversity changes in lycopsid and aquatic fern megaspores through geologic time. Paleobiology 19: 28-42.
Kremer, B. P. 1979. Light independent carbon fixation by marine macroalgae. J. Phycol. 15: 244-247.
Kurimo, H. & U. Kurimo. 1981. Distributional relations and homogeneous areas in aquatic macrophyte vegetation: A case study. Ann. Bot. Ferm. 18: 293-312.
Lee, L. & F. Belknap. 1970. Hard water as a limiting factor in the distribution Of Isoetes echinospora. Amer. Fern J. 60: 134-136.
Leon, B. & K. R. Young. 1996. Aquatic plants of Peru: Diversity, distribution, and conservation. Biodiv. Conserv. 5: 1169-1190.
Loczy, S., R. Carignan & D. Platnas. 1983. The role of roots in carbon uptake by the submersed macrophytes Myriophyllum spicatum, Vallisneria americana, and Heteranthera dubia. Hydrobiologia 98: 3-7.
Luttge, U. 1987. Carbon dioxide and water demand: Crassulacean acid metabolism (CAM), a versatile ecological adaptation exemplifying the need for integration in ecophysiological work. New Phytol. 106: 593-629.
--. 1988. Day-night changes of citric-acid levels in crassulacean acid metabolism: Phenomenon and ecophysiological significance. Pl. Cell Environm. 11:445-451.
--. 1995. Clusia: Plasticity and diversity in a genus of [C.sub.3]/CAM intermediate tropical trees. Pages 296-311 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
Maberly, S. C. & D. H. N. Spence. 1983. Photosynthetic inorganic carbon use by freshwater plants. J. Ecol. 71: 705-724.
-- & --. 1989. Photosynthesis and photorespiration in freshwater organisms: Amphibious plants. Aquatic Bot. 34: 267-286.
Madsen, T. V. 1985. A community of submerged aquatic CAM plants in Lake Kalgaard, Denmark. Aquatic Bot. 23: 97-108.
--. 1987a. Sources of inorganic carbon acquired through CAM in Littorella uniflora CL.) Aschers. J. Exp. Bot. 38: 367-377.
--. 1987b. Interactions between internal and external [CO.sub.2] pools in the photosynthesis of the aquatic CAM plants Littorella uniflora (L.) Aschers and Isoetes lacustris L. New Phytol. 106: 35-50.
--. 1987c. The effect of different growth conditions on dark and light carbon assimilation in Littorella uniflora. Physiol. Pl. 70:183-188.
--, IC Sand-Jensen & S. Beer. 1993. Comparison of photosynthetic performance and carboxylation capacity in a range of aquatic macrophytes of different growth forms. Aquatic Bot. 44:373-384.
Martin, C. E. 1995. Putative causes and consequences of recycling [CO.sub.2] via crassulacean acid metabolism. Pages 192-203 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
Medina, E., M. Popp, E. Olivares, H.-P. Janett & U. Luttge. 1993. Daily fluctuations of titratable acidity, content of organic acids (malate and citrate) and soluble sugars of varieties and wild relatives of Ananas comosus L. growing under natural tropical conditions. Pl. Cell Environm. 16:55-63.
Middelboe, A. L. & S. Markager. 1997. Depth limits and minimum light requirements of freshwater macrophytes. Freshwater Biol. 37: 553-568.
Monson, IL K. 1989. On the evolutionary pathways resulting in [C.sub.4] photosynthesis and crassulacean acid metabolism (CAM). Adv. Ecol. Res. 19: 57-110.
Moyle, J. B. 1945. Some chemical factors influencing the distribution of aquatic plants in Minnesota. Amer. Midl. Naturalist 34: 402-420.
Musselman, L. J. & D. A. Knepper. 1994. Quillworts of Virginia. Amer. Fern J. 84: 48-68.
Newman, J. IL & J. A. Raven. 1995. Photosynthetic carbon assimilation by Crassula helmsii. Oecologia 101:494-499.
Nielsen, S. L., E. Gacia & K. Sand-Jensen. 1991. Land plants of amphibious Littorella uniflora (L.) Aschers. maintain utilization of [CO.sub.2] from the sediment. Oecologia 88: 258-262.
Nobel, P. S. 1995. High productivity of certain agronomic CAM species. Pages 255-265 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
--, A. A. Israel & N. Wang. 1996. Growth, [CO.sub.2] uptake, and responses of the carboxylating enzymes to inorganic carbon in two highly productive CAM species at current and doubled [CO.sub.2] concentrations. Pl. Cell Environm. 19: 689-699.
Olivares, E., K. Faist, M. Kluge & U. Luttge. 1993. [sup.14]C pulse-chase labeling in Clusia minor L. J. Exp. Bot. 44: 1527-1533.
Osmond, C. B. 1978. Crassulacean acid metabolism: A curiosity in context. Ann. Rev. Pl. Physiol. 29: 379-414.
--. 1984. CAM: Regulated photosynthetic metabolism for all seasons. Pages 557-563 in C. Sybesma (ed.), Advances in photosynthesis research. W. Junk, The Hague.
Parker, D. 1943. Comparison of aquatic and terrestrial plants of Isoetes engelmannii in the Mountain Lake, Virginia, area. Amer. Midl. Naturalist 30: 452-455.
Pedersen, O. & K. Sand-Jensen. 1992. Adaptations of submerged Lobelia dortmanna to aerial life form: Morphology, carbon sources, and oxygen dynamics. Oikos 65: 89-96.
--, -- & N. P. Revsbeeh. 1995. Diel pulses of [O.sub.2] and [CO.sub.2] in sandy lake sediments inhabited by Lobelia dortmanna. Ecology 76: 1536-1545.
Pfeiffer, N. E. 1922. Monograph of the Isoetaceae. Ann. Missouri Bot. Gard. 9: 79-217.
Pietsch, W. 1991. On the phytosociology and ecology of Isoetes asiatica (Makino) Makino in oligotrophic water bodies of South Sakhalin. Vegetatio 97:99-115.
Pigg, K. B. 1992. Evolution of Isoetalean lycopsids. Ann. Missouri Bot. Gard. 79: 589-612.
Pilon-Smits, E. A. H., H. Hart & J. V. Brederode. 1995. Evolutionary aspects of crassulacean acid metabolism in the Crassulaceae. Pages 349-359 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
Rattray, M. R., D. R. Webb & J. M. A. Brown. 1992. Light effects on crassulacean acid metabolism in the submerged aquatic plant Isoetes kirkii A. Braun. New Zealand J. Mar. Freshwater Res. 26: 465-470.
Raven, J. A. 1977. The evolution of vascular land plants in relation to supracellular transport processes. Adv. Bot. Res. 5: 153-219.
--. 1984. Energetics and transport in aquatic plants. Alan R. Liss, New York.
--. 1986. Biochemical disposal of excess H* in growing plants. New Phytol. 104:175-206.
--. 1995. Photosynthesis in aquatic plants. Pages 299-318 in E.-D. Schulze & M. M. Caldwell (eds.), Ecophysiology of photosynthesis. Springer, Berlin.
-- & A. M. Johnston. 1991. Photosynthetic inorganic carbon assimilation by Prasiola stipitata (Prasiolales, Chlorophyta) under emersed and submersed conditions: Relationship to the taxonomy of Prasiola. Brit. Phycol. J. 26: 247-257.
-- & C. B. Osmond. 1992. Inorganic C acquisition processes and their ecological significance in inter- and subtidal macroalgae of North Carolina. Funct. Ecol. 6: 41-47.
-- & G. Samuelsson. 1988. Ecophysiology of Fucus vesiculosus L. close to its northern limit in the Gulf of Bothnia. Bot. Mar. 31: 399-410.
-- & R. A. Spicer. 1995. The evolution of crassulacean acid metabolism. Pages 360-385 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
--, L. L. Handley, J. J. MacFarlane, S. McInroy, L. McKenzie, J. H. Richards & G. Samuelsson. 1988. The role of [CO.sub.2] uptake by roots and CAM in acquisition of inorganic C by plants of the isoetid life-form: A review, with new data on Eriocaulon decangulare L. New Phytol. 108: 125-148.
--, J. Beardall & S. Roberts. 1989. The ecophysiology of inorganic carbon assimilation by Durvillaea potatorum (Durvillaeales, Phaeophyta). Phycologia 28: 429-437.
--, A. M. Johnston, L. L. Handley & S. G. McInroy. 1990. Transport and assimilation of inorganic carbon by Lichina pygmaea under emersed and submersed conditions. New Phytol. 114: 407-417.
Raven, P. H. 1963. Amphitropical relationships in the floras of North and South America. Quart. Rev. Biol. 38: 151-177.
Retallack, G. J. 1997. Earliest Triassic origin of Isoites and quillwort evolutionary radiation. J. Paleontol. 71: 500-521.
Richardson, K., H. Griffiths, M. L. Reed, J. A. Raven & N. M. Griffiths. 1984. Inorganic carbon assimilation in the isoetids, Isoetes lacustris L. and Lobelia dortmanna L. Oecologia 61:115-121.
Robe, W. E. & H. Griffiths. 1988. [C.sub.3] and CAM photosynthetic characteristics of the submerged aquatic macrophyte Littorella uniflora: Regulation of leaf internal [CO.sub.2] supply in response to variation in rooting substrate inorganic carbon concentration. J. Exp. Bot. 39:1397-1410.
-- & --. 1990. Photosynthesis of Littorella uniflora grown under two PAR regimes: [C.sub.3] and CAM gas exchange and the regulation of internal [CO.sub.2] and [O.sub.2] concentrations. Oecologia 85:128-136.
-- & --. 1992. Seasonal variation in the ecophysiology of Littorella uniflora (L.) Ascherson in acidic and eutrophic habitats. New Phytol. 120: 289-304.
-- & --. 1994. The impact of [NO.sub.3]-loading on the freshwater macrophytes Littorella uniflora. N utilization strategy in a slow-growing species from oligotrophic habitats. Oecologia 100: 368-378.
Rorslett, B. & P. Brettum. 1989. The genus Isogtes in Scandinavia: An ecological review and perspectives. Aquatic Bot. 35: 223-261.
Ryberg, H., L. Axelsson, S. Carlberg, C. Larsson & J. Uusitalo. 1990. [CO.sub.2] storage and [CO.sub.2] concentrating in brown seaweeds. I. Occurrence and ultrastructure. Pages 517-520 in M. Baltscheffsky (ed.), Current research in photosynthesis. Kluwer, The Hague.
Sand-Jensen, K. 1978. Metabolic adaptation and vertical zonation of Littorella uniflora (L.) Aschers. and Isoetes lacustris L. Aquatic Bot. 4: 1-10.
--. 1987. Environmental control of bicarbonate use among freshwater and marine macrophytes. Pages 99-112 in R. M. M. Crawford (ed.), Plant life in aquatic and amphibious habitats. Blackwell Scientific, London.
--. 1989. Environmental variables and their effect on photosynthesis of aquatic plant communities. Aquatic Bot. 34: 5-25.
-- & T. V. Madsen. 1991. Minimum light requirements of submerged freshwater macrophytes in laboratory growth experiments. J. Ecol. 79: 749-764.
-- & M. Sondergaard. 1979. Distribution and quantitative development of aquatic macrophytes in relation to sediment characteristics in oligotrophic Lake Kalgaard, Denmark. Freshwater Biol. 9:1-11.
--, C. Prahl & H. Stokholm. 1982. Oxygen release from roots of submerged aquatic macrophytes. Oikos 38: 349-354.
Sandquist, D. R. & J. E. Keeley. 1990. Carbon uptake characteristics in two high-elevation populations oft he aquatic CAM plant Isogtes bolanderi (Isoetaceae). Amer. J. Bot. 77: 682-688.
Schmitt, J. M., B. Fiblthaler, A. Sheriff, B. Lenz, M. Babler & G. Meyer. 1995. Environmental control of CAM induction in Mesembryanthemum crystallinum--A role for cytokinin, abscisic acid and jasmonate? Pages 159-175 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
Sculthorpe, C.D. 1967. The biology of aquatic vascular plants. Edward Arnold, London.
Seddon, B. 1965. Occurrence of Isoetes echinospora in eutrophic lakes in Wales. Ecology 46: 747-748.
--. 1972. Aquatic macrophytes as limnological indicators. Freshwater Biol. 2: 107-130.
Sharma, B. D. & R. Harsh. 1995. Diurnal acid metabolism in the submerged aquatic plant Isoetes tuberculata. Amer. Fern J. 85: 58-60.
Skog, J. E. & C. R. Hill. 1992. The Mesozoic herbaceous lycopsids. Ann. Missouri Bot. Gard. 79:648-675.
Smith, F. A. & N. A. Walker. 1980. Photosynthesis by aquatic plants: Effects of unstirred layers in relation to assimilation of [O.sub.2] and [HCO.sub.3] and to carbon isotopic discrimination. New Phytol. 86: 245-259.
Smith, J. A. C. & K. Winter. 1995. Taxonomic distribution of crassulacean acid metabolism. Pages 427-436 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
--, J. Ingrain, M. S. Tsiantis, B. J. Barkla, D. M. Bartholomew, M. Bettey, O. Pantoja & A. J. Pennington. 1995. Transport across the vacuolar membrane in CAM plants. Pages 53-71 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin. Smith, S. V. 1985. Physical, chemical, and biological characteristics of [CO.sub.2] gas flux across the air-water interface. PI. Cell Environm. 8: 387-398.
Smith, W. G., H. L. Boston & M. S. Adams. 1985. A preliminary study of the source and fate of carbon acquired via CAM in Littorella uniflora var. americana (Fern.) Gl. J. Freshwater Ecol. 3: 203-209.
Smits, A. J. M., P. Laan, R. H. Thier & G. van der Velde. 1990. Root aerenchyma, oxygen leakage patterns and alcoholic fermentation ability of the roots of some nymphaeid and isoetid macrophytes in relation to the sediment type of their habitat. Aquatic Bot. 38:3-17.
Sendergaard, M. 1981. Loss of inorganic and organic carbon by [sup.14]C-labeled aquatic plants. Aquatic Bot. 10: 33-43.
-- & K. Sand-Jensen. 1979a. Carbon uptake by leaves and roots of Littorella uniflora (L.) Aschers. Aquatic Bot. 6: 1-12.
-- & --. 1979b. Physico-chemical environment, phytoplankton biomass and production in oligotrophic, sol, water Lake Kalgaard, Denmark. Hydrohiologia 63: 241-253.
Spicer, R. A. 1993. Palaeoecology, past climate systems, and [C.sub.3]/[C.sub.4] photosynthesis. Chemosphere 27: 947-978.
Stebbins, G. L. 1974. Flowering plants. Evolution above the species level. Belknap Press, Cambridge.
Steinberg, S. L. 1996. Mass and energy exchange between the atmosphere and leaf influence gas pressurization in aquatic plants. New Phytol. 134: 587-599.
Sternberg, L., M. Deniro & J. E. Keeley. 1984. Hydrogen, oxygen, and carbon isotope ratios of cellulose from submerged aquatic crassulacean acid metabolism and non-crassulacean acid metabolism plants. Pl. Physiol. 76: 68-70.
--, --, D. McJunkin, R. Berger & J. E. Keeley. 1985. Carbon, oxygen, and hydrogen isotope abundances in Stylites reflect its unique physiology. Oecologia 67: 598-600.
Stewart, W. N. 1983. Paleobotany and the evolution of plants. Cambridge University Press, Cambridge.
Surif, M. B. & J. A. Raven. 1983. The occurrence of diel changes in titratable acidity of plant cell contents: Indications of CAM-like metabolism in plants native to Scotland and comparisons with plants from elsewhere. Trans. Bot. Soc. Edinburgh 45: 235-244.
Szmeja, J. 1994. Dynamics of the abundance and spatial organisation of isoetid populations in an oligotrophic lake. Aquatic Bot. 49: 19-32.
Taylor, T. N. 1981. Paleobotany. An introduction to fossil plant biology. McGraw-Hill, New York.
Taylor, W. C. & R. J. Hiekey. 1992. Habitat, evolution, and speciation in Isoetes. Ann. Missouri Bot. Gard. 79: 613-622.
Tessenow, U. & Y. Baynes. 1978. Experimental effects of Isoetes lacustris L. on the distribution of [E.sub.H], pH, Fe, and Mn in lake sediment. Verb. Intl. Verein. Limnol. 20: 2358-2362.
Toivonen, H. & T. Lappalainen. 1980. Ecology and production of aquatic macrophytes in the oligotrophic, mesohumic lake Suomunjarvi, eastern Finland. Ann. Bot. Fenn. 17: 69-85.
Tolken, H. R. 1977. A revision of the genus Crassula in southern Africa. Bolus Herbarium, Rondebosch, South Africa.
--. 1981. The species of Crassula L. in Australia. J. Adelaide Bot. Gard. 3: 57- 90.
Tryon, R. M. & A. F. Tryon. 1982. Ferns and allied plants. Springer, New York.
Turpin, D. H., G. C. Vanlerberghe, A. M. Amory & R. D. Guy. 1991. The inorganic carbon requirements for nitrogen assimilation. Canad. J. Bot. 69:1139-1145.
Voge, M. 1997. Number of leaves per rosette and fertility characters of the quillwort (Isoetes lacustris L.) in 50 lakes of Europe: A field study. Arch. Hydrobiol. 139: 415-431.
Wang Zi-Qiang. 1996. Recovery of vegetation from the terminal Permian mass extinction in North China. Rev. Palaeobot. Palyn. 91: 121-142.
Webb, D. R., M. R. Tattray & J. M. A. Brown. 1988. A preliminary survey for crassulacean acid metabolism (CAM) in submerged aquatic macrophytes in New Zealand. New Zealand J. Mar. Freshwater Res. 22: 231-235.
Wetzel, R. G. 1975. Limnology. W. B. Saunders, Philadelphia.
Willenbrink, J., B. P. Kremer & K. Schmitz. 1979. Photosynthetic and light-independent carbon fixation in Macrocystis, Nereocystis, and some selected Pacific Laminariales. Canad. J. Bot. 57: 890-897.
Winter, K. 1985. Crassulacean acid metabolism. Pages 329-387 in J. Barber & N. R. Baker (eds.), Photosynthetic mechanisms and the environment. Elsevier, London.
-- & J. A. C. Smith. 1995a. An introduction to crassulacean acid metabolism. Biochemical principles and ecological diversity. Pages 1-13 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
-- & --. 1995b. Crassulacean acid metabolism: Current status and perspectives. Pages 389-426 in K. Winter & J. A. C. Smith (eds.), Crassulacean acid metabolism. Biochemistry, ecophysiology, and evolution. Springer, Berlin.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||crassulacean acid metabolism|
|Author:||Keeley, Jon E.|
|Publication:||The Botanical Review|
|Date:||Apr 1, 1998|
|Previous Article:||Cecropia schreberiana in the Luquillo Mountains of Puerto Rico.|
|Next Article:||Are biotic factors significant in influencing the distribution of halophytes in saline habitats?|