Growth and developmental rates determine the size and age of metamorphosis,
respectively. Both can be affected by the interaction between intrinsic hormonal levels and external environmental factors such as food availability, predation, population density, competition, desiccation, salinity, chemical cue and temperature (Newman &
Dunham, 1994; Altwegg & Reyer, 2003; Rose, 2005; Relyea, 2007; Wells, 2007;Wu &
Kam, 2009; Cabrera-Guzman et al., 2013). Temperature is the most powerful cues to larval anurans due to its direct influence on the metabolic rates that drive development and growth (Smith-Gill & Berven, 1979; Atkinson, 1996). Because developmental rate is accelerated more than growth, high temperature often causes larvae to undergo metamorphosis earlier (Harkey & Semlitsch, 1988; Beck & Congdon, 2000; Alvarez &
Nicieza, 2002b; Castañeda et al., 2006; Walsh, Downie & Monaghan, 2008a). Besides, a long-term field research indicated that a difference of 5°C in the average temperature is associated with about three weeks late in reaching metamorphosis (Loman, 2002). In addition, temperature fluctuation regimes may also have influences on metamorphic traits of larvae. Compared with stable temperature treatment, although larvae from the fluctuating treatment displayed earlier metamorphosis with smaller size, they had better
locomotor performance such as faster swimming and greater jumping ability (Niehaus, Wilson & Franklin, 2006;Měráková & Gvoždík, 2009).
Food availability is another key factor on larval growth and development in amphibian. Generally, tadpoles fed on sufficient or high quality food (i.e. protein-rich food) accelerate their growth rates, thus metamorphose at larger sizes than those fed on restricted or low quality food. Additionally, food assimilation is deeply influenced by temperature, thereby affect tadpoles' metamorphic time and size. In fact, several empirical studies showed that the interaction of food and temperature had
comprehensive effects on metamorphic physiology, morphology and performance, and varied from populations to species (Beck & Congdon, 2000; Alvarez & Nicieza, 2002b;
Merila et al., 2004; Arendt & Hoang, 2005; Benavides et al., 2005; Arendt, 2006;
Castañeda et al., 2006; Castano et al., 2010; Gomez-Mestre et al., 2010; Dahl et al.,
2012; Liess et al., 2013).
Previous studies had proved that larvae raised at warm temperature with higher quantity or quality food not only had larger metamorphic weight, but also had the advantages of locomotor performances such as superior sprint speed and endurance within population (Beck & Congdon, 2000; Alvarez & Nicieza, 2002a; Arendt & Hoang, 2005; Castano et al., 2010). Castañeda et al. (2006) revealed that digestive plasticity was highly sensitive to temperature, but not diet quality. Under warm condition, tadpoles had smaller body mass caused by the decreased intestinal content, gut length and enzyme activity. Moreover, when compared between two distinct lineages
(Agalychnis callidryas and Xenopus laevis), Gomez-Mestre et al. (2010) found that low food level and low temperature both led larvae to extend larval periods. Although the influences of food and temperature had similar trends between two species, the two factors resulted in different morphology changes. Restricted food made metamorphs shorter-legged with a higher degree of ossification, whereas lower temperature made
metamorphs longer-legged with a lower degree of ossification, respectively. However, these allometric differences did not affect froglrets’ locomotor performance. Merilä et al.
(2004) indicated that interaction of temperature and food made allometric differences of larvae morphology. Under the stress environment (lower temperature with lower food),
R. temporaria tadpoles decreased body length more than tail length which resulted in
increased mean relative tail length. Besides, the results proposed that the degree relative tail length changed in response to temperature could reflect the latitudinal thermal optima among populations.
Geographic variations
Several studies have observed that geographic variations in environmental temperature are associated with the divergence of larval growth and development within species, especially for widely distributed species. For example, tadpoles at high elevation experience cold temperatures and grow slower, with longer larval periods, and metamorphose at larger sizes than lowland ones (Berven, Gill & Smith-Gill, 1979;
Berven & Gill, 1983; Lai et al., 2002). The temperature-dependent patterns have also been detected along latitudinal clines (Blouin & Brown, 2000). In a series studies on R.
temporaria (the most widespread anuran in Europe), researchers demonstrated the
plasticity of metamorphic traits in response to different temperature regimes by using the populations across southern to northern Sweden (Stahlberg, Olsson & Uller, 2001;
Laugen et al., 2003; Merila et al., 2004; Liess et al., 2013). Because of the larger temperature variation and severe time constrain in northern Sweden, tadpoles from northern population had higher development rate and growth rate than extreme southern
ones under high temperature (Stahlberg, Olsson & Uller, 2001). In addition, although the developmental rates in the field did not show latitudinal order, the developmental rates increased strongly and linearly with increasing latitude in the laboratory
experiments (Laugen et al., 2003).
Larval amphibians can also be affect by the spatial variation in food availability, and the interaction between temperature and food (Merila et al., 2004; Benavides et al., 2005; Liess et al., 2013). Liess et al. (2013) showed that development rate of R.
temporaria larvae was only affected by temperature, while growth rate was affected by
the interaction of temperature and food. Northern (Arctic) population had higher developmental rate than southern (Boreal) one under cold condition regardless of food quality. Whereas southern tadpoles fed on high quality food under cold condition had faster growth rate and larger metamorphic weight than northern ones. However, a study on Bufo spinulosus showed a different pattern (Benavides et al., 2005). Assimilation efficiency of B. spinulosus tadpoles were affected by the interaction of temperature and diet quality, not by their geographic origin. Under low quantity food, larvae raised at higher temperature had significant greater assimilation efficiency than lower ones, while larvae had no differences between temperatures with high quantity food.
Temporal variations
Although considerable researches have been devoted to spatial divergences among populations, little attention has been paid to temporal divergences in amphibian life history. Some researches investigated how anuran species over-wintered as larvae (Lai et al., 2002; Chuang, 2006; Hsu, Kam & Fellers, 2012). They found that cold winter
temperatures retarded development and slowed growth. As spring came, growth rate increased more than development rate did. Thus, over-winter larvae metamorphosed at larger sizes than those that completed metamorphosis before winter. In fact, many environmental factors vary seasonally. Temporal variations in environmental conditions, such as temperature and food availability should be a challenge faced by all prolonged amphibian breeders. However, few studies have explicitly investigated the temporal pattern of tadpoles' phenotypic plasticity within population.
Amphibians that are called prolonged breeders because their breeding seasons span several months. However, it is usually not clear if individuals can breed multiple times.
Due to amphibian energy limits or climate condition such as rainfall, there may be temporal ''breeding cohorts'' which sequentially join in the breeding activities (Beebee, 1979). In a study on German natterjack toad, Bufo calamita, researchers found that a sexually mature male only joined one of three distinct breeding cohorts (from April to
August) due to its energetic limits of vocalization (Sinsch, 1988). Interestingly, by using radio telemetry and mark-recapture, they also found that the metamorphs and toadlets would come back to the same pond and join the same breeding cohort at their first breeding. And only reproductive males showed a lifelong fidelity to the site, whereas females did not prefer certain breeding sites. Moreover, the genetic distances using allozyme analyses were greater between temporal populations (cohorts) than between spatial ones, implying temporal reproductive isolation (Sinsch, 1992; Sinsch, 1997). In addition, the first spring cohort had a longevity (potential reproductive lifespan) one year more than the summer cohort. Therefore, the early breeding cohort of B. calamita potentially had greater fitness (Leskovar et al., 2006). Even though there are detail demographic studies of B. calamita, the previous researches have not addressed whether different cohorts have different life history strategies in response to temporal
environment variations.