Functional characteristics of corticolous lichens in the understory of a tropical lowland rain forest


Author for correspondence: Michael Lakatos Tel & Fax: (+49) 631 205 2496 Email:


  • • In tropical lowland forests, corticolous crustose green algal lichens are abundant and highly diverse. This may be related to adaptation to prevailing microenvironmental conditions including, for example, high precipitation and low light intensities.
  • • In the understory of a tropical lowland rain forest in French Guiana, we studied the morphology of crustose green algal lichens and measured gas exchange and chlorophyll a fluorescence.
  • • We found that (i) periods of thallus suprasaturation with water were reduced by the presence of water-repelling surface structures of mycobiont hyphae at the thallus surface and the medulla; (ii) photosynthesis was adapted to the low light intensities present in the understory; and (iii) photosynthesis was rapidly activated in fluctuating light.
  • • The combination of these three mechanisms enables corticolous lichens to implement specific morphological and physiological strategies, which may favour growth in the limiting understory habitat of tropical lowland rain forests.


The diversity of lichenized fungi in tropical rain forests is estimated to be 300 species per km2 (Aptroot & Sipman, 1997) or even per hectare (Komposch & Hafellner, 2000). For example, Aptroot (1997) reported 173 lichen species on a single tree in a tropical montane rain forest of Papua New Guinea. Although lichenized fungi are not considered a distinctive feature of lowland tropical rain forest, they are very common in this ecosystem. They frequently cover the bark of trees, lianas and saplings (corticolous lichens) as well as leaf surfaces (foliicolous lichens). In tropical forests, lichens show characteristic distribution patterns related to their growth forms. In montane and premontane rain forests, a high percentage of foliose (leaf-like) and fruticose (shrub-like) lichens can be found. In lowland rain forests, these growth forms are rather infrequent while crustose lichens occur abundantly (Cornelissen & Ter Steege, 1989; Sipman & Harris, 1989; Montfoort & Ek, 1990; Sipman & Tan, 1990; Lakatos et al., 2004). To date, no convincing explanation has been suggested for the observed dominance of crustose lichens in tropical lowland rain forest, particularly in the understory.

Several studies have conclusively shown that ecophysiological investigations have considerable potential to explain vegetation patterns and the diversity of lichens. While physiological mechanisms in lichens in extreme environments such as hot arid (cf. Lange et al., 1997) and cold polar habitats (cf. Kappen et al., 1990) have been intensively studied, there is a lack of knowledge in this respect for the humid tropics. The virtual absence of foliose and fruticose green algal lichens in lowland tropical forests has puzzled ecophysiologists and lichenologists over the last 20 years. Regarding a similar distribution pattern among bryophytes, Richards (1984) hypothesized that, during the tropical nights, high temperatures lead to a nocturnal respiratory CO2 loss, which cannot be compensated by daily photosynthesis because of limitations associated with their poikilohydric nature. As bryophytes and lichens share many physiological characteristics (cf. Green & Lange, 1995), this hypothesis was later extended to macrolichens (Lange et al., 1994; Zotz & Winter, 1994; Zotz et al., 1998, 2003; Zotz, 1999; Zotz & Schleicher, 2003). Studies on cyanobacterial macrolichens in a Panamanian lower montane rain forest (Lange et al., 2000, 2004) revealed that nocturnal respiration losses are often higher than carbon gains during the day, despite relatively low temperatures at this altitude. Hence, the virtual absence of macrolichens in lowland forests might be explained by negative carbon balances as a result of higher nighttime temperatures (Lange et al., 1994; Zotz & Winter, 1994; Zotz et al., 2003). Transplantation experiments from a montane to a lowland forest did not result in short-time acclimation but in immediate death (G. Zotz, pers. comm.). However, new results have demonstrated the ability of temperate and polar lichens to acclimate their respiration rates to temperature changes and that their temperature optima are often related to the prevailing microclimatic conditions of their habitat (Friedmann & Sun, 2005; Lange & Green, 2005; Sun & Friedmann, 2005).

Suprasaturation with water is an additional phenomenon that limits photosynthesis. Progressive water uptake in lichens leads to increasing CO2 diffusion resistances within the thallus, which is explained by a progressive closure of interhyphal spaces caused by expanding hyphal cell walls and the blockage of pores (Cowan et al., 1992). When the water absorption capacity of the lichen thallus is exhausted, continuous water films eventually appear on the thallus surface. Both processes considerably lower CO2 gas diffusion and thus reduce carbon profit, which can reduce assimilation to less than 20% of the maximal net photosynthesis (Lange et al., 2001). In the understory of tropical forest, this may cause problems for corticolous lichens because of frequent stem water flow ranging from 0.4% (Cavelier et al., 1997) to 8% of the net precipitation (Jordan & Heuveldrop, 1981). Reduced photosynthetic profit as a result of desiccation of the lichen thallus, as shown in tropical montane forest (Lange et al., 2000, 2004) and in the canopy of lowland forest (Zotz & Winter, 1994), is thought to play a minor role in the more humid understory.

Another limiting factor for carbon gain in the dense tropical forest understory is light. Light conditions in forest understory habitats are variable and mainly composed by diffuse light with light intensities below 25 µmol m−2 s−1 (Pearcy, 1987; Chazdon, 1988; Oberbauer et al., 1989; Lakatos, 2002). The low light levels are frequently interrupted by sunflecks increasing photosynthetic photon flux density (PFD) by over two orders of magnitude. Sunflecks may contribute 10–85% of total daily PFD in evergreen tropical forest understory habitats (cf. Pearcy, 1990). Approximately one-third of the daily carbon gain of higher plants in the understory can be attributed to sunfleck periods as a result of photosynthetic induction and low light compensation points (Valladares et al., 1997). To date, it has not been investigated whether lichens growing in the understory possess similar photosynthetic features and are able to utilize sunflecks similarly to higher plants.

Therefore, precipitation and low PFD are the two colimiting climatic factors of major physiological importance for the growth and survival of crustose lichens in tropical lowland habitats. These two climatic factors are particularly unfavourable for lichens in the understory of tropical lowland rain forests (Lakatos, 2002).

Here, we present a detailed ecophysiological study on strategies implemented by corticolous crustose green algal lichens growing in a tropical lowland forest. These lichens represent a large and diverse but fairly difficult group, in an evolutionary, ecological and physiological perspective, which has been rather under-studied. In order to elucidate potential mechanisms, we investigated their morphology as well as their physiological response to ecological parameters in the understory of a neotropical lowland rain forest in French Guiana. We evaluated the three hypotheses that corticolous lichens (i) possess morphological features that prevent thallus suprasaturation and, hence, reduce photosynthetic limitation as a result of water saturation events, (ii) are adapted to very low light intensities, and (iii) can improve their photosynthetic induction by exploitation of sunflecks.

Materials and Methods

Experimental site and study organisms

Measurements were performed in a tropical evergreen lowland rain forest at the Les Nouragues field station (4°05′N, 52°40′W, 160 m above sea level) in Les Nouragues National Park, French Guiana, during October and November 1999. Mean annual rainfall is 2920 mm, with an indistinct dry season around September to October and a slight drop in the amount of rainfall around February and March (Grimaldi & Riéra, 2001). For a detailed description of the study site and the abiotic conditions, see van der Meer (1995). Experimental measurements and collection of lichen samples were conducted in the understory 100 m west of the station at the base of two trees (Eriotheca sp. and Virola michelii) at a height of 70–200 cm.

To characterize general structural categories of corticolous lichen thalli, several sampling sites were randomly chosen in the national park. On the basis of this screening, we selected the four corticolous green algal lichen species, Thelotrema alboolivaceum (Nyl.) Hale (Thelotremataceae), Cryptothecia rubrocincta (Ehrenb.) Thor (Arthoniaceae), Phyllopsora cf. corallina (Eschw.) Müll. Arg. (Bacidiaceae) and Coenogonium linkii Ehrenb. (Coenogoniaceae), each representing a different aspect of the crustose growth form. The corticolous lichen species are strictly endoperidermal; they penetrate into the dead cork but not into the underlying living tissues of the bark. All four species are commonly found on several tree species and frequently share the same trunk. For physiological measurements, west-facing lichens were chosen. T. alboolivaceum, C. rubrocincta and C. linkii form an association with Trentepohlia sp. as a photobiont, while P. corallina is associated with Pseudochlorella sp. (Brako, 1991).

Morphology and anatomy

Thallus morphology and anatomy and the response of the thallus to water were documented macroscopically and microscopically as well as by electron microscopy.

To determine the hydrophobicity of the thalli, the contact angle of a water drop (φ) was measured by the sessile drop method with a drop shape analysis system (G10/DSA10; Krüss, Hamburg, Germany). The optical contact angle is used to estimate the wetting properties of a localized region on a solid surface. The final height and width of a water drop are determined only by the surface energy of the liquid (the surface tension between solid and liquid, γSL) and the surface energy of the solid (solid–vapour, γSV). The viscosity of the liquid and gravity (liquid–vapour, γLV) affect the spreading of the drop. The angle between the baseline of the drop and the tangent at the drop boundary is measured and calculated using Young's equation: γSV − γSL =γLV cos φ. If the surface energy of the solid exceeds that of the liquid, the drop spreads out to a film with zero angle. A contact angle of 180° denotes complete unwettability (superhydrophobicity), i.e. when the droplet theoretically contacts the surface at only one point.

The anatomy of desiccated and fully hydrated thalli was studied using low-temperature scanning electron microscopy (LTSEM) after cryofixing for 60 s in liquid nitrogen slush-cooled below −196°C. The frozen specimens were transferred to a cold stage (CTS 1500C; Oxford Instruments, Eynsham, UK) and fractured with a pre-cooled knife-edge manipulator under vacuum. Fractured specimens were either immediately sputter-coated in an argon atmosphere with c. 10–20 nm gold (20 mA, 0.1 bar, 1.5–2.0 min) or directly transferred under vacuum to the scanning electron microscope (SEM) (S-3200 N; Hitachi, Pleasanton, CA, USA) cooled specimen stage. Uncoated and coated specimens were observed at accelerating voltages of 5 and 15 kV, respectively, and micrographs were recorded digitally.

Specific thallus parameters

Lichen thalli were scanned and the projected surface area was calculated using a graphics program (Adobe Photoshop; Adobe Systems, Munich, Germany). This procedure allows higher resolution and accuracy than a leaf area meter, especially for filamentous thalli and remains of bark. The projected thallus area ranged from 4 to 17 cm2[mean = 9.1 cm2; standard deviation (SD) = 3.6 cm2; n = 37]. The exact thallus dry weight could only be analysed for C. linkii (mean = 53.1 g m−2; SD = 2.4 g m−2; n = 6) because of the tightly adnate contact of the other crustose thalli with the bark. Fresh weight was determined in the field using a portable electronic digital balance (SAC 64; Scaltec, Heiligenstadt, Germany; accuracy 10 mg) at the beginning and at the end of the experiment. Data were linearly interpolated to the respective times of measurement. Lichen thallus water content (WC) was determined from the difference between fresh and dry weights (R 160 P; Sartorius, Göttingen, Germany; accuracy 1 mg) divided by thallus area and hence expressed as millimetre ‘precipitation-equivalent’ because of the difficulties of separating the lichens from their substrates. This WC relates to an equivalent amount of precipitation taken up by the lichen per square metre. As obtained from desiccation curves, mean optimal WC of the lichen thalli (n = 3–6 per species) ranged between 0.2 and 0.4 mm H2O (Fig. 1), which, for example, corresponded to a WC of 300–600% dry weight (DW) for C. linkii.

Figure 1.

Net photosynthetic CO2 uptake rate (NP) related to thallus water content (WC) at saturating photosynthetic photon flux density (PFD) (150 µmol m−2 s−1) and in darkness, with mean maximal and minimal values for three to six samples per species selected from sequential measurements during drying periods as the lichens passed through a range of water contents (n = 6–15). (a) Thelotrema alboolivaceum, (b) Cryptothecia rubrocincta, (c) Phyllopsora corallina and (d) Coenogonium linkii.

The chlorophyll a+b (Chl) contents of the samples were determined by heating the chopped material twice for 45 min in dimethyl sulphoxide (DMSO) at 60°C in the presence of MgCO3 to avoid breakdown of chlorophyll (Ronen & Galun, 1984; Barnes et al., 1992). After extraction and centrifugation (12 000 g for 5 min at 20°C) the supernatant was analysed spectrophotometrically (Lambda 2; Perkin Elmer, Wellesly, MA, USA). The mean Chl contents ranged from 118 mg Chl m−2 for P. corallina to 148 mg Chl m−2 for C. rubrocincta, with intermediate values of 122 mg Chl m−2 for C. linkii and 140 mg Chl m−2 for T. alboolivaceum, and the chlorophyll a to chlorophyll b ratio was 2.3–3.2 (SD < 25%; n = 5–15).

The stable isotope ratio of carbon (δ13C) was analysed in the lichen organic matter according to Máguas & Brugnoli (1996). For determination of δ13C, samples were dried at 60°C and homogenized in a ball mill (MM 200; Retsch, Haan, Germany) and aliquots (∼0.2 mg) were weighed into tin capsules for analysis (micro balance S3D; Sartorius). Stable isotope analysis was performed with a stable isotope ratio mass spectrometer (IRMS) (SIRA II; VG ISOGAS, Manchester, UK) working in continuous-flow mode. For this, an element analyser (EA 1110; CE Instruments, Milano, Italy) was coupled to the IRMS. A urea carbon standard was calibrated to the Vienna-Pee Dee Belemnite (V-PDB) standard using International Atomic Energy Agency (IAEA)-CH-6 and IAEA-CH-7 reference materials (International Atomic Energy Agency, Vienna, Austria) and was run with each sample analysis. The average precision for the batch runs was better than 0.13.

Experimental methods

To assess photosynthetic efficiency, CO2 gas exchange was measured with a portable CO2/H2O porometer (HCM-1000; Walz, Effeltrich, Germany). Chlorophyll a fluorescence was measured with a miniaturized pulse-amplitude modulated photosynthetic yield analyser (Mini-PAM; Walz). To enable chlorophyll a fluorescence and CO2 gas exchange of corticolous lichens to be obtained simultaneously and with a fast system response, a newly designed cuvette with a volume of c. 40 cm3 was used (Lakatos, 2002). The response of the gas exchange system to a step change in CO2 injection into the cuvette was 21 s delayed with the system response completed within 3–4 s. This settling time was aligned to the fluorescence and gas exchange measurements. Only freshly collected material was used for physiological measurements, as it is not possible to obtain reliable data from these sensitive tropical green algal lichens after storage or desiccation below a water content of approximately 30%. The photosynthetically inactive bark with an attached lichen sample was carefully removed from the phorophyt. The samples were then cleaned and any aerophile algae, liverworts and other living organic materials were removed using a stereomicroscope. While kept in darkness, the samples were soaked in rain water for 30 min. Subsequently, continuous measurements were performed. To ensure that samples were free of any other assimilating material, gas exchange and chlorophyll a fluorescence measurements were performed on randomly chosen and comparably prepared bark samples after removal of lichen thalli, on dry lichen thalli, and on the wetted underside of the bark.

CO2 gas exchange

CO2 gas exchange was measured in continuous open flow mode of 600 mL min−2 at 80% relative humidity in order to obtain slow thallus desiccation and to prevent condensation inside the cuvette (based on Lange et al., 1984). As cooling of the cuvette caused interference with gas exchange signals at low assimilation rates of the lichen samples, a constant temperature could not be maintained under the tropical field conditions. Over the course of sequential measurements during desiccation, the temperatures of single thalli increased maximally up to 3°C. To some extent, this lowered the net photosynthesis (NP) values achieved at high PFD because of increased respiration. The light source was a scattering halogen lamp (35 W; Osram, Berlin, Germany) coated with a diffuser disc and a combination of neutral density filters (No. 209, 210, 211 and 299; LEE Filters, Hampshire, UK). Measurements were performed with ambient air drawn in via a perforated tube from a location adjacent to the bark surface (c. 0.2 m2) along the trunk of a phorophyt. Ambient CO2 concentrations ranged from 380 to 360 ppm with a shift of 9 (± 6) ppm during measurements. NP was related to total chlorophyll content as well as to thallus surface area.

Chlorophyll a fluorescence

Because chlorophyll a fluorescence is less affected by temperature (Green et al., 1998) or WC (Leisner et al., 1997) than gas exchange measurements, we used chlorophyll a fluorescence to quantify photosynthetic efficiency. The effective quantum yield of photosystem II (PS II) (ΔF/inline image) during steady-state photosynthesis of the light-adapted lichen was calculated as (inline image − F)/inline image, where F is the fluorescence of the light-adapted sample and inline image is the maximum light-adapted fluorescence when a saturating light pulse is superimposed on the prevailing environmental light levels (Genty et al., 1989; Schreiber et al., 1995). ΔF/inline image characterizes the efficiency of PS II in transporting electrons into the electron chain in light. The intensity (2000 µmol m−2 s−1) and duration (600 ms) of the light pulse were optimized to fully saturate fluorescence emission without photoinhibiting the sample. The apparent rate of photosynthetic electron transport through PS II [the electron transport rate (ETR)] was obtained as 0.5 × (ΔF/inline image) × PFD. The factor 0.5 accounts for distribution of the excitation to both PS II and PS I.

The light reflectance factor (absorption coefficient) is not known numerically for these lichens. Thus, no correction was made for the reflectance factor in order to enable comparison of the present results with data reported in the literature. After dark adaptation, Fv/Fm[(Fm − F0)/Fm] is well verified as an index of the potential quantum yield of PS II, where Fm is the maximal fluorescence and F0 is the minimal fluorescence of the dark-adapted lichens. Nonphotochemical fluorescence quenching (NPQ) was quantified as (Fm − inline image)/inline image (Krause & Weis, 1991; Bilger et al., 1995).

Light–response curves

Light–response curves were measured with hydrated dark-adapted samples. For samples enclosed in the dark cuvette, respiration was measured before a saturating flash was given to determine the maximal fluorescence yield after dark adaptation (Fv/Fm). At regular intervals of 3 min – sufficient time to attain steady state – PFD was increased in seven steps from 2 to 350 µmol m−2 s−1. Simultaneously, CO2 gas exchange and chlorophyll fluorescence parameters were recorded. The samples were removed from the cuvette and weighed for water content. Measurements were checked and fitted (P < 0.001) using the Marquardt algorithm (SigmaPlot 7.0; SPSS, San Rafael, CA, USA) by a nonrectangular hyperbola (Leverenz & Jarvis, 1979):

θP2 − (φI + Pmax)P + φI Pmax = 0(Eqn 1)
θ ETR2 − (φI + ETRmax)ETR + φI ETRmax = 0(Eqn 2)

(P, the measured rate of photosynthesis; Pmax, maximal net photosynthesis; ETR, the apparent rate of electron transport; I, the incident irradiance; θ, the convexity value or bending factor; φ, the apparent quantum efficiency of CO2 reduction or electron transport.)

At light limitation, the initial slope of the light–response curve describes the apparent quantum efficiency of CO2 reduction or electron transport (φ), whereas at light saturation and carboxylation limitation the maximum rates of electron transport (ETRmax) and CO2 fixation (Pmax) are reached. Saturating PFD values (PFDsat) are defined for 95% of Pmax and ETRmax. The transition zone between these two phases of photosynthetic limitation is characterized by the convexity value or bending factor (θ). The convexity is a parameter that estimates the degree of light use efficiency in the intermediate light range (Ögren, 1993; Leverenz, 1994). High efficiency and high θ values are attained when the colimitation of electron transport and carboxylation is low. This occurs when electron transport capacity is high relative to Rubisco capacity (Ögren & Evans, 1993). All calculated regressions were significant (P < 0.05) and passed the normality and constant variance tests. The data were analysed by the Shapiro–Wilk test and one-way analysis of variance (ANOVA) and checked assuming equal variances by Fisher's least significant differences (LSD) post hoc test.

Experiments under fluctuating light

The natural light intensities intercepted by lichens in the understory of this forest are below 10 µmol m−2 s−1 for at least 95–99% of the daytime period (Fig. 2). Artificial sunflecks (lightflecks) were simulated as rapid changes between dark and light intervals of durations 30 and 180 s (short and long lightflecks). During the interruptions by dark periods no background illumination was given to avoid additional conditioning effects (Kirschbaum & Pearcy, 1988). The PFD at light periods was set to 15 µmol m−2 s−1 in order to simulate low light intensity sunflecks and to 150 µmol m−2 s−1 to simulate high light intensity sunflecks. Preliminary measurements showed that at light intensities of 15 µmol m−2 s−1 photosynthetic compensation was reached, while 150 µmol m−2 s−1 saturated photosynthesis (see Figs 7 and 8 below). Short lightflecks were applied for 20 min with low light intensities or for 10 min with high light intensities. After the period of short lightflecks, three long lightflecks with the preceding light intensity were applied for a total of 18 min. Before measurements, lichen thalli were dark-adapted and hydrated within the optimal WC. Gas exchange and chlorophyll a fluorescence were measured continuously during the dark and light periods. We measured three samples per species. To characterize the efficiencies of photosynthetic response to lightflecks, we chose the following parameters: mean ΔF/inline image, ETR, NPQ, maximal gross photosynthesis (GPmax), which steady state was reached during long sunflecks, and the percentage of GPmax after short sunflecks of 30 s (%GPmax). The slope of ΔF/inline image during sunfleck exposure indicates the velocity and extent of adaptation and recovery.

Figure 2.

Microclimatic measurements in close vicinity to the studied lichens on four representative days (25–28 October) at the study site in the tropical understory of the National Park Les Nouragues. The left-hand axes have a logarithmic scale: maximal photosynthetic photon flux density (PFDmax; µmol m−2 s−1) in a west-facing vertical (solid line) and horizontal (dotted line) sensor position. The right-hand axes have a linear scale: mean ambient air temperature (°C) at 1.5 m above the ground.

Figure 7.

Experimental sunfleck simulation at low light intensities [photosynthetic photon flux density (PFD)] of 15 µmol m−2 s−1 for the representative lichens (a) Cryptothecia rubrocincta and (b) Coenogonium linkii at an optimal water content between 0.2 and 0.4 mm precipitation equivalent (n = 3). Simultaneous measurements of CO2 gas exchange (left panel) and chlorophyll a fluorescence (right panel) were carried out. Light–dark intervals started with lightflecks of 30 s illumination and, after 20 min, switched to lightflecks of 3 min duration for 18 min. Left panel: maximal CO2 uptake within 30 s, expressed as estimated per cent of maximal gross photosynthesis (%GPmax) relative to the mean GPmax that was reached at steady-state photosynthesis of given PFD within 3 min. Right panel: effective quantum yield (ΔF/inline image) within 30 s. ΔF/inline image characterizes the efficiency of photosystem II in transporting electrons into the electron chain in the light (lower values) and in recovering in the dark (higher values). In the dark, the slope of ΔF/inline image displays the velocity and extent of recovery, with higher values indicating better recovery. An overview of all cardinal points of photosynthesis obtained during artificial sunfleck treatments is given in Table 2.

Figure 8.

Experimental sunfleck simulation at high light intensities [photosynthetic photon flux density (PFD)] of 150 µmol m−2 s−1 for the representative lichens (a) Cryptothecia rubrocincta and (b) Coenogonium linkii at an optimal water content of 0.2–0.4 mm precipitation equivalent (n = 3). Simultaneous measurements of gas exchange (left panel) and chlorophyll a fluorescence (right panel) were carried out. Light–dark intervals started with lightflecks of 30 s illumination and after 10 min switched to lightflecks of 3 min duration within 18 min. For an explanation of the panels, see Fig. 7.


Morphological and anatomical adaptations to thallus water supersaturation

According to our observations, the growth forms of the majority of corticolous green algal lichens in the study site could be assigned to four basic and simplified morphological types (Fig. 3, types A, B, C and D). The most common growth form (type A) was the crustose type with a thallus closely attached to the bark, a smooth cortex layer, and a more or less distinct thallus margin without prothallus (Figs 3a, 4a–d). Type A lichens were often fertile and included species of the genera Dimerella, Lecidea, Thallotrema, Thelotrema and Porina. Also frequent were crustose type B lichens with a well-defined prothallus, exclusively formed by loosely interwoven mycobiont hyphae. In type B lichens the prothallus was visible along the margin and was often located underneath the loosely attached thallus (Figs 3b, 4e–g). A characteristic feature was the absence of a distinct cortex, which resulted in a powdery-like habitus. This type was represented by the lichen genera Chiodecton, Cryptothecia and Dichosporidium. The rare type C lichens, such as those in the genera Biatora, Crocynia, Phyllopsora and Psoroma, exhibited a larger byssoid prothallus. Here, the lichen thallus was reduced to small squamules, rarely larger than 5 mm, which were located on top of the prothallus (Figs 3c, 5a–d). The filamentous type D lichens were only represented by the genus Coenogonium. The filaments of its green algal photobiont Trentepohlia were loosely enveloped by fungal hyphae (Figs 3d, 5f,g,k). In this study one lichen species as representative of each morphological type was selected: type A, T. alboolivaceum; type B, C. rubrocincta; type C, P. corallina; type D, C. linkii.

Figure 3.

Classification of growth forms of common corticolous green algal lichens in the tropical understorey into four basic morphological types and their basic responses to water. (a) Type A lichens have a closely attached thallus; (b) type B lichens have a loosely attached thallus with a marginal and underlying prothallus; (c) type C lichens have a squamulous crustose thallus, which is reduced to exposed scales, and (d) type D lichens are filamentous lichens of the genus Coenogonium. (a–c) Types A, B and C repel water drops on the surface; (b) type B additionally collects water with the prothallus and drains it through little channels; (c) type C absorbs water with the prothallus, and (d) type D takes up water through capillary forces which then evaporates as a result of the large thallus surface area (see text for further explanation).

Figure 4.

Diagrams (a, e), microphotographs and micrographs obtained by low-temperature scanning electron microscopy (LTSEM) of Thelotrema alboolivaceum (a–d) and Cryptothecia rubrocincta (e–i). (b) Cross-section showing the distinctive cortex, algal layer and measured contact angle (ϕ) indicating hydrophobicity. (c) Top view of the surface showing apothecium and (d) LTSEM enlarged image of the surface. (f) Cross-section showing the algal layer and ϕ, without a cortex. (g) Top view of the surface showing a reddish prothallus and greenish thallus. (h) Cross-section (LTSEM) showing water-filled channels (Ch) and the medulla (M) extending to the surface. (i) LTSEM enlarged medulla hyphae covered by microscale bulbos-like crystals.

Figure 5.

Diagrams (a, f), microphotographs and micrographs obtained by low-temperature scanning electron microscopy (LTSEM) of Phyllopsora corallina (a–e) and Coenogonium linkii (f–k). (b) Cross-section of thallus squamules showing the cortex, algal layer and ϕ. (c) Top view and LTSEM enlarged image (d) of the surface showing the black prothallus and green squamules. (e) The prothallus with absorbed water. (g) Microphotograph of the filamentous symbiont Trentepohlia. (h) Top view of the structure showing one apothecium. (i) Overview of the thallus. (j) The thallus with surplus water coloured using ink; the tips are not covered with water even at the highest suprasaturation. (k) The algal filament enveloped by three to four hyphae enlarged by LTSEM.

The morphological–anatomical structures of the thalli of these lichens exhibited different features for avoidance of suprasaturation by water. Types A, B and C had different hydrophobic layers (Fig. 3a–c) indicated by contact angles (ϕ), and neither closed water films on their surface nor free water inside the fully hydrated thalli could be observed in various LTSEM sections (Figs 4, 5a–d).

Type A. T. alboolivaceum effectively repelled water from the thallus surface. Its smooth cortex is composed of a dense, flattened pseudo-parenchymatic epinecral layer (15–20 µm thick) creating a hydrophobic surface with contact angles (ϕ) of 130–135° (Fig. 4a–d).

Type B. C. rubrocincta also repelled water from the surface. In contrast to T. alboolivaceum, the superhydrophobic behaviour (ϕ: 175° ± 2, n = 3) of the surface was not caused by a cortex, but by an extreme enlargement of erected and tightly meshed hyphae from the medulla (Fig. 4f–h). The hyphal surface is covered with many microscale bulbous-like crystals (Fig. 4i). In addition, this type is characterized by collection of water with its hydrophilic, nonlichenized prothallus (ϕ: 0°) and draining of this water through little channels passing underneath the thallus (Fig. 4h). It was frequently observed that, after rain, water exuded continuously from experimentally induced injuries to the surface of the lichenized thallus part. The draining channels were also confirmed by several LTSEM observations showing a complex anatomy of the lichenized thallus. The channels were observed to be lined with water-repellent filaments and filled with liquid water (Fig. 4h; Ch). Hard evidence of this specific water movement has yet to be obtained in additional experiments. No free water could be found in the interhyphal spaces within the thallus (Fig. 4h, medulla, and Fig. 4i).

Type C. The hydrophilic prothallus of P. corallina had a high water absorption capacity and absorbed water like a sponge (Fig. 5a–e). This and the hydrophobic cortex (ϕ: 100° ± 4, n = 3) of the oval-shaped lichenized squamules prevented the elevated thallus scales from being covered with a water film.

Type D. As a consequence of its horizontal exposition (Figs 3, 5i) and of the evaporating surface of its filamentous thallus structure (Fig. 5h–k), a closed water film did not occur in C. linkii. The filament of the photobiont was envelope by three to five hyphae (Fig. 5f–k). When supplied with liquid water, the hyphae swelled up from approximately 2 to 4 µm in diameter without covering the photobiont completely, thus allowing gas diffusion (Fig. 5a,k). Even at the highest suprasaturation, the tips were not covered by water films, as demonstrated under experimental conditions with a surplus of water coloured using ink (Fig. 5j).

The absence of regularly occurring CO2 limitation by suprasaturation was also confirmed by the isotopic composition of the organic lichen material. Mean and standard deviation of organic δ13C values were −33.2 ± 0.4, −33.4 ± 0.6, −33.2 ± 0.9 and −31.8 ± 0.9 (n = 4–7) for T. alboolivaceum, C. rubrocincta, P. corallina and C. linkii, respectively.

Photosynthetic capacity

The light–response curves of photosynthesis showed that the studied lichens have low light compensation (LCP) and saturation (Psat) points (Fig. 6). CO2 gas exchange measurements showed LCP between light intensities of 7 µmol m−2 s−1 (T. alboolivaceum) and 23 µmol m−2 s−1 (C. rubrocincta and C. linkii) and Psat ranging from 39 µmol m−2 s−1 (T. alboolivaceum) to 193 µmol m−2 s−1 (C. linkii) when samples were within their optimal range of WC but sometimes exceeded optimal temperature conditions (Table 1). Although the higher values of θ (> 0.85) and φ (> 0.03) for T. alboolivaceum and P. corallina indicated a higher light use efficiency in the intermediate light range (Table 1), they were only significant for T. alboolivaceum (ANOVA: F(θ)3,10 = 14.08; F(φ)3,10 = 15.18; P < 0.05). Maximum rates of average CO2 exchange ranged from 0.07 to 0.59 µmol CO2 m−2 s−1, which corresponds to 0.8–3.8 nmol CO2 (mg Chl)−1 s−1 with significant differences compared with C. linkii (ANOVA: F3,10 = 34.7; P < 0.001; Table 1).

Figure 6.

Light–response curves of the corticolous lichens (a) Thelotrema alboolivaceum, (b) Cryptothecia rubrocincta, (c) Phyllopsora corallina and (d) Coenogonium linkii (n = 3–5) corresponding to Table 1. Photosynthesis is illustrated as a function of light [photosynthetic photon flux density (PFD), µmol photon m−2 s−1] at an optimal water content of 0.2–0.4 mm precipitation equivalent. In the panels of the left, the gas exchange of each species is indicated as net photosynthesis (NP, µmol CO2 m−2 s−1). In the panels on the right, the apparent electron transport rate (ETR, µmol electrons m−2 s−1) is shown. All panels indicate the saturation point and data are fitted using Eqns 1 and 2 with r2 significant at P < 0.05.

Table 1.  Cardinal points of mean (± standard deviation) photosynthetic capacity [maximal net photosynthesis (Pmax), maximal apparent rate of electron transport (ETRmax) and saturating photosynthetic photon flux density (PFDsat)], convexity (θ), light compensation point (LCP) and apparent quantum efficiency (φ)
 Thelotrema alboolivaceumCryptothecia rubrocinctaPhyllopsora corallinaCoenogonium linkiiUnits
  1. Values were determined using light response curves (Fig. 6) and calculated using Eqns 1 and 2 (see the Materials and Methods section) and are the means of three replicas for each sample (n) per indicated species (the uppercase letter represents the growth form type) within temperature ranges indicated (T).

  2. Different letters within a row indicate a significant (P < 0.05) difference (least significant differences post hoc test).

Pmax0.12 ± 0.02a0.17 ± 0.02a0.07 ± 0.01a0.59 ± 0.04bµmol CO2 m−2 s−1
Pmax2.79 ± 0.87ac1.43 ± 0.36ab0.84 ± 0.12b3.84 ± 0.28cnmol CO2 mg Chl−1 s−1
LCP7 ± 423 ± 714 ± 323 ± 4µmol photons m−2 s−1
PFDsat39 ± 11a175 ± 54b76 ± 16a193 ± 35bµmol photons m−2 s−1
θ0.91 ± 0.09a0.70 ± 0.04b0.87 ± 0.13b0.72 ± 0.15bDimensionless
φ0.055 ± 0.01a0.009 ± 0.002b0.031 ± 0.005b0.022 ± 0.002bµmol CO2 µmol photon−1
ETRmax10.2 ± 1.27a7.55 ± 0.55a7.52 ± 0.59a14.8 ± 1.36bµmol electrons m−2 s−1
PFDsat85 ± 24a72 ± 20a52 ± 28a341 ± 49bµmol photons m−2 s−1

The simultaneous chlorophyll a fluorescence measurements also revealed a significantly higher photosynthetic capacity for C. linkii (ANOVA: F3,10 = 4.13; P < 0.05) with a mean ETRmax of 15 µmol electrons m−2 s−1 (Fig. 6d) in comparison with values of 10, 7.6 and 7.5 µmol electrons m−2 s−1 for T. alboolivaceum, C. rubrocincta and P. corallina, respectively (Fig. 6a–c; Table 1). With the exception of P. corallina, single thalli achieved ETR of more than 16 µmol electrons m−2 s−1. Light saturation points (PFDsat) of ETR slightly differed from those obtained by gas exchange measurements: for T. alboolivaceum, C. rubrocincta and P. corallina light saturation was attained at a PFD of 85, 72 and 52 µmol m−2 s−1, respectively, whereas C. linkii required significantly higher light intensities of 341 µmol m−2 s−1 to achieve saturation (ANOVA: F3,10 = 8.83; P < 0.05; Fig. 6 and Table 1). In terms of most of the cardinal points of photosynthetic capacity (Table 1), these four lichens may be classified into two groups: (i) species favouring lower light conditions (T. alboolivaceum, C. rubrocincta and P. corallina) and (ii) species favouring higher light conditions (C. linkii).

Photosynthetic reaction in fluctuating light

The response to artificial short-duration (30 s) and long-duration (3 min) lightflecks with both low (15 µmol m−2 s−1) and high (150 µmol m−2 s−1) PFD revealed diverse physiological reactions in the efficiency of CO2 uptake, ETR and recovery of PS II. In Figs 7 and 8, lightfleck characteristics are shown for C. linkii and C. rubrocincta, representing the low and high light groups, respectively, and an overview of all achieved cardinal points is presented in Table 2. Light induction during short lightflecks was characterized by a stepwise increase of ΔF/inline image and GP. After five short lightflecks, the studied lichens reached at least 50% of GPmax and 90% of mean ΔF/inline image. For lightflecks of low light intensity (15 µmol m−2 s−1), C. linkii reached only about 55% of GPmax within 30 s whereas T. alboolivaceum, C. rubrocincta and P. corallina achieved approximately 70% of GPmax (Table 2). The time to achieve GPmax ranged from 39 s (P. corallina) to 54 s (C. linkii). At high PFD (150 µmol m−2 s−1), C. rubrocincta revealed the best response, achieving approx. 85% of GPmax, whereas T. alboolivaceum, P. corallina and C. linkii reached about 60% of GPmax within 30 s (Table 2). C. rubrocincta attained GPmax most rapidly, with a time of 32 s, whereas T. alboolivaceum, P. corallina and C. linkii required 52–54 s. Generally, C. rubrocincta converted light energy to CO2 fixation most rapidly, and all the studied lichens attained GPmax within 1 min, independent of the light intensity. The absolute gross CO2 uptake rates under high light conditions were 0.3 µmol CO2 m−2 s−1 for C. linkii and approximately 0.2 µmol CO2 m−2 s−1 for T. alboolivaceum, C. rubrocincta and P. corallina.

Table 2.  Cardinal points of mean (± standard deviation) photosynthesis obtained during artificial sunfleck treatment with photosynthetic photon flux density (PFD) of 15 and 150 µmol m−2 s−1 (Figs 7, 8)
PFD (µmol m−2 s−1)VariableThelotrema alboolivaceum Type ACryptothecia rubrocincta Type BPhyllopsora corallina Type CCoenogonium linkii Type DUnits
  1. Values shown are the percentage of maximal gross photosynthesis that was achieved within 30 s (%GPmax) after full induction, effective quantum yield (ΔF/inline image), electron transport rate (ETR), and nonphotochemical quenching (NPQ) based on three samples per indicated species (‘Type A’ etc. represent growth form types). The evaluation of the recovery of photosystem II is indicated by + for faster and – for slower response. Different letters within rows indicate a significant (P < 0.05) difference (least significant differences post hoc test).

15 %GPmax68 ± 467 ± 470 ± 554 ± 5%
ΔF/inline image0.40 ± 0.020.44 ± 0.060.49 ± 0.020.46 ± 0.03Dimensionless
ETR3.0 ± 0.13.4 ± 0.33.6 ± 0.23.4 ± 0.5µmol electrons m−2 s−1
Recovery of PS II++ 
150%GPmax54 ± 783 ± 559 ± 556 ± 4%
ΔF/inline image0.09 ± 0.00a0.04 ± 0.01a0.09 ± 0.01a0.24 ± 0.07bDimensionless
ETR6.5 ± 1.1a2.7 ± 0.9a6.7 ± 0.7a17.7 ± 1.5bµmol electrons m −2 s−1
NPQ0.18 ± 0.04a1.17 ± 0.38b0.26 ± 0.05a0.40 ± 0.12aDimensionless
Recovery of PS II+ 

At the same time, PS II efficiency was characterized by the chlorophyll a fluorescence parameters ‘maximal apparent ETR’ and ΔF/inline image. For lightflecks of low light intensity all four lichen species revealed similar mean ΔF/inline image and ETR values (Table 2). This confirms the similarities in electron transport at low light conditions found for light–response curves (Table 1). At saturating light (Fig. 8), C. linkii showed the highest ETR (18 µmol electrons m−2 s−1) corresponding to a ΔF/inline image of 0.24. With a ΔF/inline image of 0.09, i.e. an ETR of 6.5 µmol electrons m−2 s−1, P. corallina and T. alboolivaceum attained medium values, in contrast to a ΔF/inline image of 0.04, corresponding to an ETR value of 2.7 µmol electrons m−2 s−1, for C. rubrocincta (Table 2). Conversion of high light energy into electron transfer was best achieved in C. linkii, whereas C. rubrocincta seemed to be photoinhibited at 150 µmol m−2 s−1 with a NPQ value of 1.17 (Table 2; F3,8 = 61.5; P = 0.000). The efficiency of PS II in C. rubrocincta and particularly in P. corallina showed a fast recovery as indicated by a stepwise increase of ΔF/inline image towards initial values at low PFD, while that of the other species improved more slowly (Table 2; Figs 7, 8).


Morphology and anatomy reduce water saturation events

Crustose lichens growing on tree bark maintain a high diversity under the particular conditions of a tropical forest understory. Based on morphology and anatomy, we identified four morphological types A to D (Fig. 3) representing the prevalent thallus growth forms at the base and lower trunk areas in an evergreen neotropical lowland rain forest. The criteria for this new and more detailed classification of crustose types are based not only on the traditional growth form classifications of previous studies (Cornelissen & Ter Steege, 1989; Montfoort & Ek, 1990; Wolf, 1995) but also on functional microstructure analysis such as hydrophobicity, gas exchange and LTSEM measurements. Here, we were able to demonstrate that, in the studied representatives, mycobiont hyphae on the thallus surface together with certain features of the medulla and the prothallus have water-repelling properties that reduce periods of suprasaturation with water.

Hydrophobicity  The measurement of the contact angle of a sessile water drop (ϕ) as an indicator of wetting properties revealed high hydrophobicity of certain thallus structures of the studied species, with the exception of C. linkii. Surfaces covered by natural compounds such as cellulose or lipids possess ϕ of 55–104° (Reichenbach-Klinke, 1999). The contact angles of higher plants with smooth leaf surfaces are between 10° and 110° and increase to the maximum of 160° when epicuticular crystals and papillose cell structures are present (Barthlott & Neinhuis, 1997). The ϕ of mycelial mats of filamentous fungi ranged between hydrophilic (ϕ < 30°) for two deuteromycetes and hydrophobic (c. 75–115°) reaction for the ascomycete Cladiosporium sp. and three studied basidiomycetes (Smits et al., 2003). The studied crustose lichens showed a high hydrophobic response with ϕ from 100° to 135°, and C. rubrocincta showed an unusually high hydrophobic response of 175°. This superhydrophobicity may be caused not only by physical enlargement of the surface and a crystalline coating, but probably also by the polymerized fungal protein hydrophobin (Honegger, 1997; Wessels, 2000; Wösten, 2001). To our knowledge, the contact angles of fungal surfaces of lichen thalli have not yet been investigated, but several studies have shown hydrophobic reactions of fungal hyphae caused by hydrophobin (Scherrer et al., 2002; Scherrer & Honegger, 2003). In lichens, hydrophobins commonly occur on hyphae of the medulla layer (Scherrer et al., 2000). Hyphae of the medulla layer also form the surface of C. rubrocincta (Fig. 4h). The hydrophilic reaction of the prothallus of P. corallina (Fig. 5e) is caused by a differently structured surface of the fungal hyphae which is mainly papillose and may also be associated with water or nutrient absorption (Brako, 1991). In contrast to these findings, most foliose and fruticose lichens exhibit hydrophilic thallus surfaces (cortex) that absorb water and nutrients, while the water-repellent hydrophobe layer is restricted to inside the medulla and the photobiont layer (Scherrer et al., 2002), preventing the interior of the thallus from becoming waterlogged (Honegger, 1997).

Functional strategies  At high water content, the NP of the studied lichens showed no significant reduction as a result of CO2 limitation, except for C. linkii, whose maximal photosynthesis was reduced by approximately one-third (Fig. 1). Hence, the morphological and anatomical properties of the mycobiont apparently prevented continuous water films from forming on the surface and also prevented free water from penetrating inside the thallus, effectively thus preventing photosynthetic limitation as a result of increased CO2 diffusion resistance. Vertical exposition of the thallus also facilitates a rapid run-off of water. In addition, corticolous lichens exhibit further anatomical features to improve ventilation of the photobiont. These features may simply result in the repelling of water through hydrophobic surface properties (types A to C) or may additionally provide specific water drainage (type B). The photosynthetically inactive prothallus seems either to direct water through the thallus (type B) or to store water (type C). Regarding water storage, type D lichens (C. linkii) adopt a physiological conflict to hold as much water as possible, which simultaneously may create CO2 limitation, to balance the high evaporation rates as a result of the enlarged filamentous thallus. Through adhesive forces, it utilizes stem water flow to build up a water reservoir between the lower thallus side and the tree trunk. Although CO2 diffusion resistance is increased in this thallus region, the impact on the photobiont is small because the remaining larger part of the filamentous thallus is infrequently enveloped with hyphae and is for the most part exposed to the air. This arrangement leads to high evaporation rates and rapid desiccation of the thallus (Lakatos, 2002). The experimental results clearly showed that artificial addition of water did not lead to a complete covering of the photobiont Trentepohlia (Fig. 4k) and the resulting thin water film did not extend to the thallus tips (Fig. 4j). This thallus structure is probably the morphological–anatomical reason why photosynthetic limitation by suprasaturation hardly occurred in Coenogonium interplexum in the study of Thomas et al. (1996) and occurred only to a small extent in the present study. Moreover, C. linkii can accumulate throughfall water as a result of its horizontal exposition (Fig. 5i). Despite high evaporation, prolonged photosynthetic activity may result from this additional exploitation of water resources. In addition, this horizontal exposition is linked to a higher light gain (see ‘Functional diversity in an ecological context’).

As an integral indicator, the organic δ13C composition of the studied lichens of between −33 and −31 confirmed that they were not regularly CO2-limited. Free-living terrestrial Trentepohlia algae show isotopic composition of −30 (M. Lakatos et al., unpublished). As an increased CO2 diffusion resistance would lead to a δ13C enrichment of the thallus (Máguas & Brugnoli, 1996), the observed depleted δ13C values do not indicate a substantial CO2 limitation by water (Lange & Ziegler, 1986; Lange et al., 1988) and is in accordance with findings for other tropical Trentepohlia lichens (M. Lakatos et al., unpublished). Lichens with frequent suprasaturation events such as the supralitoral Lichina pygmaea13C: c. −10) or the homogenous Collema crispum13C: c. −16) have highly enriched δ13C values in comparison with the mean of −22 for other lichens with CO2-concentrating mechanisms (M. Lakatos et al., unpublished).

Consequently, morphological–anatomical developments to enhance water exploitation seem to play an important role in optimizing the required balance between desiccation, which causes inactivation, and suprasaturation, which limits photosynthesis. Particularly during the high precipitation periods that frequently occur in tropical rain forests, these functional properties prevent corticolous lichens from being highly photosynthesis-limited. This is in contrast to many other lichens in which such a net photosynthetic depression occurs (Lange et al., 2001). Our findings regarding the anatomical diversity of corticolous lichens suggest that future studies should also carefully consider structural properties in order to be able to differentiate more precisely within one classified group and to improve our understanding of the ecophysiological functioning of these lichens. However, the main findings of this study emphasize the causality between structure and function in the lichens predominating in tropical understories.

Exploitation of low light intensities

Light–response curves and the resulting cardinal points of photosynthesis allow quantitative assessment of light acclimatization status by determination of rates and efficiencies of photosynthetic processes (Ögren & Sundin, 1996). Referring to the cardinal points, the studied green algal lichens showed low light compensation and saturation points, thus revealing that photosynthesis was well adapted to low light intensities. The comparison of CO2 gas exchange and chlorophyll a fluorescence measurements revealed that the two processes were nonlinearly related. Differences between chlorophyll a fluorescence and CO2- or oxygen-based measurements of photosynthesis are expected and observed (Green et al., 1998; Franklin & Badger, 2001; Figueroa et al., 2003) Discrepancies are caused by processes such as the respiration of the mycobiont, photorespiration, electron transfer to oxygen, and electron cycling around PS I.

Light compensation point  In the present study, absolute values of gas exchange obtained by artificial light–response curves have to be considered with caution, as optimal conditions of acclimated growth temperature could not be attained throughout the measurements. The infrequent measuring points at 31–34°C exceeded the maximum temperatures to which corticolous lichens are naturally exposed (Fig. 2), and the present study represents the worst case scenario for these lichens. Gas exchange rates of C. linkii under controlled temperatures indicated optimum CO2 uptake rates at 28°C of 1.1 µmol CO2 m−2 s−1 and 8.2 nmol CO2 (mg Chl)−1 s−1 as well as lower LCP and Psat (Lakatos, 2002). Hence, the gas exchange measurements presented here may underestimate the natural photosynthetic capacity. Interestingly, even under these unfavourable conditions with high respiration rates, the corticolous lichens showed a very low LCP, which was also indicated by high quantum efficiencies (φ) (Table 1). Acclimation of macrolichens to a low light regime has been reported previously. In the understory of a tropical rain forest in Panama, the foliose Leptogium azureum revealed an LCP of 7 µmol m−2 s−1 (Zotz & Winter, 1994). The highly shade-adapted foliose Pseudocyphellaria dissimilis from a temperate evergreen forest in New Zealand showed an LCP of 1 µmol m−2 s−1 (Green et al., 1991). Studies on macrolichens in a lower montane rain forest demonstrated LCPs of 12–40 µmol m−2 s−1 (Lange et al., 2000, 2004). The lichens in these studies were associated with cyanobacteria (Nostoc spp.) as photobionts. For green algal lichens with a CCM or with additional cyanobionts, LCPs of 35 µmol m−2 s−1 in Panama (Zotz et al., 1998), 4 to 29 µmol m−2 s−1 in New Zealand (Green et al., 1997) and 5 to 10 µmol m−2 s−1 in Sweden (Sundberg et al., 1997) have been observed. Foliose and fruticose trebouxioid lichens (with CCM) from a deciduous woodland attained an LCP of 7–18 µmol photons m−2 s−1 (Smith & Griffiths, 1998), and similar values were recorded for the endolithic crustose Verrucaria baldensis (Tretiach & Geletti, 1997). The present study demonstrates that microlichens are also photosynthetically highly adapted to the low light regimes existing at the deeply shaded forest trunks of a tropical evergreen forest, irrespective of the presence (P. corallina) or absence of a CCM.

Exploitation of fluctuating light

The phenomenon of photosynthetic induction of higher plants by short sunflecks has been the subject of several studies over the last two decades (e.g. Chazdon & Fetcher, 1984; Chazdon & Pearcy, 1986, 1991). Studies on ferns (e.g. Gildner & Larson, 1992) and aquatic algae (Ibelings et al., 1986; Mallin & Paerl, 1992; Wagner et al., 2006) examined the effects of fluctuating light. The lichens investigated here showed a different use of sunflecks depending on their photosynthetic adaptations and anatomical properties.

The four corticolous lichens studied had LCPs ranging from 7 to 23 µmol m−2 s−1. For 90–99% of the daytime period the predominating light intensity does not exceed PFD values of 15 µmol m−2 s−1 (Fig. 2; Lakatos, 2002). With such prevailing light intensities the lichens are only able to compensate part of their respiratory losses. Accordingly, during 1–10% of the day, the PFD is caused by sunflecks that are intercepted at this site with values of 15–1200 µmol m−2 s−1, and which may potentially be used to reach a positive carbon gain. Thus, it even appears to be an ecological need for the lichens to exploit these short periods of incident light in order to achieve positive carbon balances. Indeed, the studied lichens reached GPmax within 32–54 s (Table 2). As most sunflecks have durations of 30 to 60 s (Chazdon, 1988) these lichens may gain 50–100% of their gross photosynthesis during sunfleck events. Absolute CO2 uptake rates after lightfleck treatment were sometimes higher than those recorded for steady-state light–response curves. However, this observation and the varying respiration rates during the dark periods of the experiments, which may be a result of decreasing temperatures, need further investigation. The increased values of %GPmax and ΔF/inline image after lightfleck treatment, which persisted even until illumination after 3 min of darkness, indicate an activation of the photosystem of the photobionts. This persisting induction improves photosynthetic gain under fluctuating light conditions. Thus, the combination of adaptation to low light intensities, a fast photosynthetic response within the duration of a sunfleck and persisting induction are crucial for corticolous lichens to optimize their light use efficiency in a light-limited understory habitat.

The present study shows that, in light-limiting environments, the photosynthesis of corticolous lichens shows similarities with that of tropical understory plants. Higher plants in similar environments exhibit lower LCPs than the studied lichens. However, in both types of organism, the sunfleck-induced activation of the photosynthetic apparatus leads to a specifically improved light use efficiency. Compared with vascular plants, the response of lichens appears to be faster, a complete activation of Rubisco taking place within 3 min, independent of light quantity. In vascular plants, in contrast, the CO2 uptake rate is limited for about 1 min as a result of ribulose-1.5-bisphosphate regeneration (Sassenrath-Cole & Pearcy, 1992). Hence, a rise in Rubisco activity is usually completed within 5–10 min and stomatal opening may continue to increase for 30–60 min (Kirschbaum & Pearcy, 1988). Post-illumination CO2 fixation (Chazdon & Pearcy, 1986), which is inversely related to the duration of sunflecks (Pearcy, 1990), could not be detected in lichens. Considerable work on sunfleck reactions remains to be done in order to obtain a better understanding of the photosynthetic behaviour of these poikilohydric plants.

Functional diversity in an ecological context

Our measurements represent the first data on the gas exchange of crustose tropical lichens. These results provide several lines of evidence that tropical corticolous lichens are well adapted to the unfavourable microclimatic conditions in the understory of tropical lowland rain forests. On the one hand, the four lichens studied share common ecological features. They possess functional strategies to avoid photosynthetic limitation by water suprasaturation and they have the ability to exploit low light intensities with equal efficiency. In an ecological context, as 90–99% of the daytime incident PFD is usually below 15 µmol m−2 s−1, this unusual similarity in apparent quantum efficiency and the low LCP found in all the lichens indicate optimal exploitation of the predominating energy source. On the other hand, the four studied lichens exhibited differences in their responses to varying light qualities (low or high light-adapted) and in morphological strategies of water release and absorption (four types). Combining the different efficiencies of physiological and morphological features in an ecological context suggests that C. linkii is preferably adapted to high PFD and long sunflecks (approx. 40–60 s). Its more horizontal exposition allows C. linkii to have higher interception of both light and throughfall water. C. linkii also showed differences from the other lichens regarding persistence of photosynthetic activity as, on the one hand, liquid water absorption took place very rapidly but, on the other hand, desiccation occurred quickly because of its open thallus construction. Thus, the beneficial effects of exposure to radiation are counteracted by limited activity. The other three lichens are mainly adapted to low PFD. T. alboolivaceum revealed the lowest LCP and Psat as well as the longest activity, as complete desiccation could not be attained in the field. C. rubrocincta performed photosynthesis with shorter lightflecks and at low PFD, while higher PFD led to slight photoinhibition. P. corallina was found to have a low compensation and saturation point, and hence it is also adapted to low light intensities. It usually desiccated more rapidly than C. rubrocincta and therefore had shorter periods of photosynthetic activity. However, compared with C. linkii, both of the latter species had longer potential activity as a result of slower desiccation.

In conclusion, the major contributions of the present study to this field of research are twofold. Firstly, it provides evidence for the differential use of sunflecks by lichens. Secondly, it contributes to our understanding of how tropical crustose lichens cope with the unfavourable conditions in the understory of tropical lowland forests, via a combination of structural and physiological adaptations.


We are grateful to Dr Pierre Charles Dominique for the opportunity to stay at the well-equipped Les Nouragues field station (UPS 656, CNRS) in French Guiana. The authors gratefully acknowledge helpful comments and remarks on the manuscript by Dr Rainer Wirth, Dr Gavin West, and anonymous referees. We thank Stephanie Dojani, Antje Siegel and Wilfried Harter for assistance in improving a new cuvette system for the CO2 gas exchange system. We also thank Professor Wolfgang Krumbein and Renate Kort for the LTSEM analysis, Cornelia Müller for the drawings and Melanie Hoffmann for assistance with contact angle measurement. The study was supported by the EC in the framework of the ‘European Tropical Forest Large Scale Facility Program’ (DG/FJ/042/98) and the German exchange program DAAD.