Vzd

month

Figure 3. (Continued) (B) Onset and cessation of activity in a cathemeral primate, Eulemur mongoz, Anjamena, Madagascar (16°03'S; 45°55'E). Trendlines indicate the negative association between onset of activity and sunset in the afternoon and cessation of activity and sunrise in the morning.

month

Figure 3. (Continued) (B) Onset and cessation of activity in a cathemeral primate, Eulemur mongoz, Anjamena, Madagascar (16°03'S; 45°55'E). Trendlines indicate the negative association between onset of activity and sunset in the afternoon and cessation of activity and sunrise in the morning.

set and sunrise, respectively, effectively shift the main activity phase into the night when daylength is short and the day when daylength is long (Curtis et al., 1999). Recent field data on E. f. collaris also report dusk acting as the primary zeitgeber (Donati and Borgognini-Tarli, 2006).

Cathemerality is an activity cycle resulting from masking of the genetically predetermined nocturnal activity rhythm. Substantial variability observed within the cathemeral activity cycle (Figure 2) in combination with the persistence of modes B and C across different habitat types and latitudes (Mode B: 12°S-24°S) indicates multiple factors modulating the endogenous rhythm.

Light Intensity

Light intensity varies greatly between day, night, and twilight periods, but also depends on cloud cover, lunar phase, and vegetation cover (Halle, 2000a). In experiments carried out on E. f. albifrons, activity changed through variation of the dark-phase light intensity (Erkert, 1989; Erkert and Cramer, 2006). The animals were nocturnal when subjected to full moon light intensities (10-1 lux), new moon light intensities (10-3 lux) inhibited much nocturnal activity and resulted in cathemeral behavior, and only when it was impossible for the animals to detect any light during the dark phase (10-7 lux; physiological darkness) were they fully diurnal.

Masking effects due to low levels of luminosity have been observed in the field in many lemurids in all habitat types and higher levels of nocturnal activity are observed around full moon, with lower levels around new moon (Colquhoun,

1998; Donati et al., 1999, 2001; Olivieri, 2002; Kappeler and Erkert, 2003; Donati and Borgognini-Tarli, 2006). Diurnal activity levels often decrease after full moon nights and increase following new moon nights (Olivieri, 2002; Donati and Borgognini-Tarli, 2006). Detailed analyses show that nocturnal activity is highest when the moon is above the horizon (waxing moon: first half of the night; full moon: all night, activity peaks during brighter middle of the night; waning moon: second half of the night) (Kappeler and Erkert, 2003; Donati and Borgognini-Tarli, 2006). The inhibitory effect of low nocturnal illumination is most dramatically reported on by Donati et al. (1999), documenting complete cessation of E. f. rufus activity during a lunar eclipse.

Nocturnal activity in E. mongoz is not affected by lunar phase, but variation in luminosity due to variable canopy cover may contribute toward an increase in diurnality when light levels are low. On Anjouan, in the Comoros, luminance was low in highland rainforests, contributing to diurnality in E. mongoz, while they were nocturnal at the same time of year in the brighter, seasonal environments of the lowlands (Tattersall, 1976). In Madagascar in seasonally dry forest, 10-fold less light penetrated the canopy during the wet season when E. mongoz was mainly diurnal than during the dry season (Curtis et al., 1999). E. m. macaco was more active during new moon nights during the dry season (when canopy cover was sparse) than during the wet season (Colquhoun, 1998). Kappeler and Erkert (2003) suggest better nocturnal light availability in higher forest strata may explain frequent observations of cathemeral lemurs feeding in peripheral regions of the canopy at night (Overdorff, 1988; Andrews and Birkinshaw, 1998; Curtis et al., 1999; Donati et al., 1999; Rasmussen, 2005).

Low nocturnal illumination levels are probably the most important masking factor which inhibits activity. Nocturnal activity runs parallel to the moonlit nighttime hours in most cathemeral lemurs, as well as cathemeral populations of Aotus azarai (Fernandez-Duque, 2003; Fernandez-Duque and Erkert, 2006)—identical to the situation documented in many nocturnal primates (Bearder et al., 2006). These lemurs are inherently dark-active (Erkert, 1989; Erkert and Cramer, 2006) and common effects of moonlight on cathemeral and nocturnal primates support this further. The effect of lunar light levels on nocturnal activity appears to be an ancient primate trait retained in many cathemeral lemurs and cannot help to further our understanding of cathemerality. A more fruitful avenue of research might be detailed investigations of the effects of light intensities due to variable canopy cover on the cathemeral activity cycle. Given the variability in the effects of illumination on activity cycles in cathemeral lemurs, other masking factors must also contribute to the production of cathemerality.

Temperature, Relative Humidity and Rainfall

Assessing the effect of climatic variables on cathemerality is problematic as they all have a seasonal component and are related to daylength as well as intercorrelated. Daylength plays a role in cathemerality (Figures 1, 2, and 3b) and in seasonal environments we would expect high rainfall, high relative humidity, and higher temperatures associated with the austral summer also to be linked to increased diurnal activity and low values for these variables during the austral winter to be linked to increased nocturnal activity. Chronobiological experiments corroborate this for temperature: According to the "circadian rule," we would expect inherently nocturnal species to be "cold-active," i.e., to increase activity at lower ambient temperatures and decrease activity when temperatures are high (Aschoff, 1979). Erkert and Cramer (2006) demonstrated this for E. f. albifrons, recording an increase in activity at ambient temperatures of 20°C and a decrease at 30°C.

There is a trend in Eulemur spp. toward diurnality with higher temperatures and nocturnality with lower temperatures in both seasonal and less seasonal habitats (Overdorff and Rasmussen, 1995; Colquhoun, 1998; Curtis et al., 1999; Donati et al., 1999; Rasmussen, 1999; Kappeler and Erkert, 2003; Donati and Borgognini-Tarli, 2006). In other cases nocturnality is associated with high temperatures (Mutschler, 1998) in lake-side reed beds (H. g. alaotrensis), diurnality with low temperatures in highland rainforest (E. mongoz) (Tattersall, 1976) or no effect is observed in rainforest (E. rubriventer) and seasonal habitats (E. f. mayot-tensis) (Overdorff and Rasmussen, 1995; Tarnaud, 2006). In the Neotropics, A. azarai increases diurnality when temperatures are low (Fernandez-Duque and Erkert, 2006).

Rainfall as a predictor of diurnal/nocturnal activity was found to be negligible in the two studies that have assessed its effects on cathemerality (Overdorff and Rasmussen, 1995; Kappeler and Erkert, 2003). Donati and Borgognini-Tarli (2006) found rainfall and humidity to be negatively associated with nocturnal activity, but link this to reduced luminosity at night during rainfall, when cloud cover is higher.

Climatic variables, in particular temperature, appear to play some role in masking the nocturnal activity rhythm in cathemeral lemurs, but no consistent pattern is discernible. More detailed data are needed to investigate the individual effects of climatic variables on cathemerality.

ADAPTIVE VALUE Thermoregulation

Tattersall (1976) first proposed a possible link between cathemerality and ambient temperature and Morland (1993) suggested that lemurs rely primarily on behavioral, rather than strictly physiological mechanisms for thermoregulation. Cathemerality could be such a behavioral mechanism, highly advantageous in "extreme" environments and permitting the animals to shift their activity and reduce thermoregulatory costs by remaining within their taxon-specific ther-moneutral zone (TNZ: range of ambient temperatures at which least energy is expended in maintenance of body temperature) (Curtis et al., 1999; Curtis and Rasmussen, 2002, 2006).

E. fulvus has a low basal metabolic rate (BMR), but high body temperature and a TNZ of 22°C to 30°C (Daniels, 1984; Erkert and Cramer, 2006). A low BMR indicates a high capacity for temperature regulation, but high body temperature rules out any capacity to lower body temperature during periods of inactivity in order to conserve energy by decreasing the temperature gradient between the environment and the body (Daniels, 1984; Müller, 1985). H. g. griseus has been reported to have a slightly lower and variable body temperature, which would result in a broader TNZ (Bourliere et al., 1956). No other information is available on lemurid BMR or body temperatures. If we extrapolate to other lemurids, then nocturnal activity in Eulemur spp. minimizes cold stress and the energetic costs of maintaining a high body temperature when ambient temperatures are below TNZ (Curtis et al., 1999). Thermoregulatory costs for HHapalemur spp. are lower during cold periods as they have some capacity for passive adaptation to low ambient temperatures due to their lower body temperature. Ambient temperature during hot periods is likely to create heat stress, requiring inactivity during the daytime and a shift of activity into the nocturnal phase (Mutschler, 1998).

Proposed thermoregulatory advantages to cathemerality are avoidance of heat stress during hot days or cold stress during cold nights by increasing either nocturnal or diurnal activity, but clear shifts in activity rhythms are also observed in the absence of strong seasonality in temperature. There are numerous discrepancies in the interpretation of the potential thermoregulatory advantages of cathe-merality, which will only be resolved when we have more data on BMR and body temperatures in these species.

Food Availability, Diet, and Digestibility Temporal Availability of Food Resources

Tattersall and Sussman (1975) tentatively linked nocturnality in E. mongoz to the temporal availability of nectar of the kapok flowers, Ceiba pentandra, which only open at night. Andrews and Birkinshaw (1998) found some food items to be more important either during nighttime or during daytime diets in E. m. macaco, but other studies on cathemeral lemurs in a variety of habitats have found few or no associations between temporal availability of food and nocturnality/diurnality (Overdorff and Rasmussen, 1995; Colquhoun, 1998; Curtis et al., 1999; Rasmussen, 1999; Tarnaud, 2006). Kappeler and Erkert (2003) suggested that a shift to diurnal activity might constitute an ecological advantage in facilitating visual detection of ripe fruit during the day, but then refuted this as unlikely since lemurs are dichromats. However, fruit consumed by lemurs is colored green, brown, tan, purplish, red (Dew and Wright, 1998). Some of these colors require only dichromatic ability for detection, so Kappeler and Erkert's suggestion might be worth further investigation.

Dietary Quality and Digestibility

Enqvist and Richards (1991) proposed a hypothesis based on the seasonal dietary shift observed in many lemurs to include more leaves in diets during periods of fruit scarcity. They suggest cathemerality is a behavioral strategy to cope with increased fiber intake employed by these small-bodied lemurs with simple digestive systems: Energy and nutrient intake is maximized by optimal spacing of food harvesting through extension of activity across the 24-hr period.

Most field data do not support their hypothesis, as either no increase in nocturnal activity is observed during the dry season (Andrews and Birkinshaw, 1998; Colquhoun, 1998) or the amount of nocturnal activity does not correlate with fibrous foods or fiber content in the diet (Overdorff and Rasmussen, 1995; Mutschler, 1998; Curtis et al., 1999; Donati et al., 1999; Rasmussen, 1999; Curtis, 2004). One study supports the hypothesis (Tarnaud, 2006), where female E. f. mayottensis increased mature leaf and fiber consumption during the daytime in the dry season when overall activity was extended into the nighttime. Overdorff and Rasmussen (1995) compared gut passage rate in three cathemeral frugivore-folivores (E. mongoz, E. fulvus, E. rubriventer) with that of a specialized folivore (H. griseus). Results indicate a reduced capacity for coping with fibrous foods in the former three species and as all four species exhibit cathemeral activity cycles, the link between cathemerality and the consumption of fibrous foods is not supported. Evidence from studies on molar morphology and digestibility of fibrous material discussed by Overdorff and Rasmussen (1995) indicates that increased fiber intake would not pose any particular problem for nonspecialist lemurids.

Predation

Cathemerality has been proposed as a mechanism to avoid predators and to minimize the risk of predation (Curtis and Rasmussen, 2002; Rasmussen, 2005; Colquhoun, 2006). Raptors, viverrids, boids, and crocodylids have been documented as predators on lemurids (Goodman et al., 1993). The greatest threat, however, is presumed to be posed by the largest living Malagasy carnivore (6.75 kg), the cathemeral fossa (Cryptoproctaferox), which exhibits varying degrees of arboreality, depending on habitat (Hawkins, 2003; Colquhoun, 2006).

Cathemeral lemurs often feed and travel higher up in the canopy at night than during the day and this has been interpreted as a strategy for predation risk minimization, as feeding in exposed parts of the canopy is safest at night when raptors are inactive (Overdorff, 1988; Andrews and Birkinshaw, 1998; Curtis et al., 1999; Donati et al., 1999; Rasmussen, 2005). Feeding and traveling high in the canopy at night may also help avoid threats from below, mainly posed by the fossa. Cryptoprocta is highly adapted for arboreal locomotion, but is less adept at moving about in the highest strata of dry forests (Hawkins, 2003) and would be restricted in access to small, peripheral branches of the canopy in all forest types due to its body size. Data on E. mongoz demonstrate how the capacity to shift between the diurnal and nocturnal phases of the day might aid in predator avoidance when infants are beginning to move about independently and are most vulnerable to predation by raptors (Figure 1). Other studies have reported no connection between cathemerality and predation: For example, Tarnaud (2004) observes that there are few predators on Mayotte and yet E. f. mayottensis is still cathemeral.

Cathemeral species cannot completely eliminate predation risk by shifting activity into either the nocturnal or diurnal phases. Slight adjustments in activity times may, however, be effective in combination with other antipredator behaviors and when the behavior and ecology of predators and other prey species are considered (Rasmussen, 2005). Effective group size is increased in some sympatric pairs of Eulemur species through polyspecific associations (Harrington, 1978; Freed, 1996) and increases protection from predation (van Schaik and van Hooff, 1983). Rasmussen (2005) proposed that small group size and cryptic habits in E. mongoz may reduce diurnal predation risk from raptors and Cryptoprocta ferox during the wet season. This strategy would offer less protection from raptors during the dry season when canopy density is lower so shifting activity to the nighttime could be beneficial. Cryptoprocta ferox poses a year-round threat that may be reduced by the different patterns of cathemeral activity exhibited by sympatric species or by the same species at different times of the year, making it difficult for this predator to optimize foraging times. Increased availability of more conspicuous and/or abundant prey species may also relax predation pressure by Cryptoprocta ferox on lemurs at certain times of the year (Rasmussen, 2005).

Interspecific Competition

Cathemerality may be one of many factors reducing competition between sympatric species and contributing toward niche separation (Rasmussen, 1999; Curtis and Rasmussen, 2002, 2006). The temporal dimension of niche separation has been neglected and merits further attention (Halle, 2000a). Competition, like predation pressure, is difficult to assess and relationships between resource competition and activity rhythms in primates are equivocal based on studies carried out to date.

The only detailed investigation of niche separation in lemurs found that microhabitat structure and food chemistry separated seven species of sympatric lemurs in rainforest habitat (Ganzhorn, 1989). The two cathemeral species in the area were E. fulvus and H. griseus, which exhibit little overlap in diet. In contrast, giant bamboo comprises 72-95% of the diets of three sympatric HHapalemur spp. in rainforest (Tan, 1999) and the temporal dimension may be an important factor in the coexistence of these lemurids. HHapalemur simus has been described as nocturnal or cathemeral, Hapalemur griseus griseus as diurnal or cathemeral, and Hapalemur aureus as cathemeral (Wright, 1986; Meier et al., 1987; Wright et al., 1987; Santini-Palka, 1994; Ratsirarson and Ranaivonasy, 2002; Mutschler, personal communication), but the activity cycles of these species have yet to be investigated in detail in the wild. For E. mongoz in seasonal forests, the most important potential primate competitor was E. f. rufus as it not only shared food resources, but also exhibited a similar activity pattern. Therefore, I predicted that for cathemerality to have any function in niche separation, different types of cathemerality would have to exist (Curtis, 1997). This was confirmed by Rasmussen (1999) (Figure 2), who found high levels of spatial and dietary convergence, but distinct patterns of cath-emerality that allowed E. mongoz and E. f. fulvus to shift peak feeding times and minimize competition. In contrast, no differences were discerned in cathemeral activity patterns in eastern rainforests in E. rubriventer and E. f. rufus, where the two species exhibited little dietary divergence, apart from during periods of food scarcity (Overdorff, 1993). In Sambirano rainforests, Freed (1996) found remarkably similar diets in E. coronatus and E. f. sanfordi, which also exhibited the same type of cathemerality. The significance of competition in shaping cathemeral activity rhythms in Eulemur species is even less clear for those populations that do not co-occur with a congener (Donati et al., 1999; Kappeler and Erkert, 2003; Donati and Borgognini-Tarli, 2006; Tarnaud, 2006).

Tattersall and Sussman (1998) note the overall tendency for pairs of Eulemur species to co-occur in northern Madagascar. They suggest that the variation in the activity cycle observed in these morphologically and ecologically similar species may have been an important factor in maintaining sympatry in a number of different habitats. However, evidence from the field is inconclusive and the only indication of a potential link between cathemerality and interspecific competition stems from seasonal forest habitat in Madagascar, where modes A and B have been observed (Figure 2).

CATHEMERALITY IN CONTEXT—DAY-NIGHT ACTIVITY IN OTHER MAMMALS

Day-night activity is widespread in mammals (16 of 24 orders) and common in the artiodactyls, perissodactyls, carnivores, rodents, and monotremes, but rare in primates (2 of 14 families). Day-night active mammals inhabit environments ranging from aquatic to terrestrial, arctic to tropical, forest to desert and are exposed to enormous variability in environmental pressures (Curtis and Rasmussen, 2006). Halle and Stensteth (2000) state this flexibility may (1) permit avoidance of unfavorable environmental conditions; (2) minimize competition; (3) maximize reproductive success; (4) increase predator efficiency; and (5) reduce predation risk. I will elaborate only on those points that permit comparisons between cathemeral primates and other mammals.

Environmental Conditions Luminosity

High nocturnal luminance suppresses nocturnal activity in most small nonprimate mammals, but either enhances or has no effect on nocturnal activity in primates (Bearder et al., 2006; Curtis and Rasmussen, 2006). Large-bodied cath-emeral mountain tapirs (Tapiridae) exhibited high levels of nocturnal activity only during full moon nights in primary rainforest, but showed no differences in less dense secondary forest (Lizcano and Cavelier, 2000). This masking effect of low light intensities in dense canopy forest is similar to that proposed for many cath-emeral primates.

Possible advantages of activity during periods of higher illumination need to be counterbalanced by the potential for increased predation risk, but many nocturnal primates are themselves visually oriented predators and increased luminosity may aid them in hunting (Bearder et al., 2006). Cathemeral lemurs are prey species rather than predators, but foraging may also be facilitated by higher light levels ( Kappeler and Erkert, 2003).

Temperature and Thermoregulation

Daily and annual changes in temperature lead to shifts from one temporal niche to another in many groups of mammals (Curtis and Rasmussen, 2006). Large-bodied herbivores inhabiting arid, hot environments reduce heat stress by being active at night (Grenot, 1992). Sloths (Bradypodidae) are nocturnal when temperatures are high and diurnal when temperatures are low, counteracting thermoregulatory constraints imposed by ineffective body temperature control (Chiarello, 1998). Likewise, echidna (Tachyglossidae) is nocturnal when it is hot and cathemeral when it is cold due to thermoregulatory constraints (Abenspergtraun and Deboer, 1992). Many arctic mammals reduce thermoregulatory costs by shifting to diurnality during cold winter months (Zielinski, 2000). These examples cover a wide range of body sizes and BMRs (Müller, 1985; Martin, 1990), but reveal a trend toward noc-turnality in cathemeral mammals when temperatures are high. When temperatures are low, a variety of strategies are exhibited, ranging from diurnality to mixed day-night activity.

These strategies are mirrored to some extent in cathemeral primates: A. azarai conserves energy through increased diurnal activity during the cold winter (Fernandez-Duque and Erkert, 2006), resembling sloths, echidna, and arctic mammals. Like sloths and terrestrial herbivores, H.g. alaotrensis may reduce heat stress by increasing nocturnality during periods of high temperatures (Mutschler, 1998). The idea that Eulemur may reduce cold stress through increased nocturnal activity is not supported (Curtis and Rasmussen, 2002). However, BMR is determined not only by body mass, but may vary according to ecological demands

(Müller, 1985). Strepsirhine BMRs are 30-60% below the mammalian mass-specific standard (Müller, 1985) and this group might exhibit thermoregulatory strategies not present in other mammals. Comparative research on mammals assessing links between body size, physiological variables (BMR, body temperature, TNZ), ecology, and activity rhythms is needed to resolve this, for which further physiological data are required.

Predation Risk

Analysis of the pattern of predation risk throughout the 24-hr day requires consideration of the pooled activity patterns of the entire predator community (Halle, 2000a). Cathemeral mammalian predators often vary diurnal and nocturnal activity levels to maximize access to diurnal, nocturnal, and cathemeral prey, and prey species also exploit temporal strategies to avoid predation (van Schaik and Griffiths, 1996; Zielinski, 2000; Hawkins, 2003). Studies on nonprimate mammals indicate that cathemerality may be effective in minimizing predation risk, can be dependent on habitat structure, but sometimes occurs only in the absence of predators, or, due to their presence.

Mustelids exhibit a tendency toward cathemerality with increasing body size in temperate regions: Small species vulnerable to predation by diurnal raptors are almost exclusively nocturnal and larger species increase diurnality to avoid predation by nocturnal foxes (Canidae) (Zielinski, 2000). Subtropical ursids exhibit intraspecific differences and smaller females and subadults avoid large nocturnal predators (felids) through shifting most activity into the diurnal phase (Joshi et al., 1999). Microtine rodents show an 18-month periodicity in diurnality, which results in predators having no predictable seasonal pattern to which their activity can be adapted. Furthermore, as these rodents are heavily predated on by diurnal raptors a tendency toward increased nocturnality was observed, the more open the habitat became (Halle, 2000b). Diurnal activity in the cathemeral tree hyrax (Procaviidae) in montane tropical forests and cathemeral fruit bats (Pteropodidae) on Pacific islands is probably only possible due to the absence of large avian predators (Milner and Harris, 1999; Brooke, 2001). Nocturnality in cathemeral sloths (Bradypodidae) in some areas may be a response to the presence of large diurnal avian predators (Chiarello, 1998).

Suggested links between cathemerality and predation in primates mirror the functional interpretations of temporal shifts in other mammals. In seasonal habitats, cathemeral lemurs can minimize predation by raptors by moving into the nocturnal phase when canopy cover is sparse, as well as by avoiding exposed areas of the canopy during the day (Overdorff, 1988; Andrews and Birkinshaw, 1998; Curtis et al., 1999; Donati et al., 1999; Rasmussen, 2005). Temporary shifts to nocturnal activity, as exhibited by E. mongoz when infants are vulnerable (Figure 1) (Curtis et al., 1999), demonstrate the link between body size and activity observed in some other mammals. Different cathemeral activity patterns in sympatric lemurs may also serve as a "predator confusion strategy," making it difficult for the cathemeral Cryptoprocta ferox to optimize its foraging times. Finally, in the Neotropics Wright (1989) suggested that Aotus azarai might shift activity to the daytime to avoid predation by the great horned owl (Bubo virgini-anus), but long-term data collected by Fernandez-Duque (2003) provide no support for a link between cathemerality and predation.

Interspecific Competition

The temporal dimension plays a role in niche separation in a number of mammalian species. Studies on carnivores (Mustelidae) in temperate regions and rodents (Muridae and Heteromyidae) in desert habitat emphasize this, and sympatric rodents compete for the most attractive time window, with the dominant species monopolizing preferred portions of the 24-hr period (Halle and Stensteth, 2000; Ziv and Smallwood, 2000; Jones et al., 2001; Marcelli et al., 2003). In temperate regions, sympatric microtine rodents (Muridae) avoid interference competition by fine-tuning activity to different times of the day and night (Halle, 2000b). Finally, Jacomo et al. (2004) found that three cathemeral canids in a seasonal tropical environment exhibited significantly different activity patterns, contributing toward niche separation.

Variation in the cathemeral activity pattern in pairs of Eulemur species in seasonal forests in Madagascar may also reduce interference competition (Curtis, 1997; Rasmussen, 1999). Curtis and Rasmussen (2006) proposed that the dominant E. fulvus might occupy the more attractive time windows, with the sympatric subordinate species (e.g., E. mongoz) adjusting activity to less favorable times, as observed in several sympatric rodents. In eastern rainforests, three sympatric species of Hapalemur that exhibit high dietary overlap may avoid competition through activity during different temporal phases (Tan, 1999). In the Neotropics, Wright (1989) suggested that the absence of competition for resources from diurnal monkeys (e.g., Callicebus) might result in cathemeral activity rhythms in Aotus.

How To Bolster Your Immune System

How To Bolster Your Immune System

All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.

Get My Free Audio Book


Post a comment