Introduction
Phenotypic plasticity, defined as the ability to produce different phenotypes in response to different environmental conditions (Bradshaw, 1965; Massimo Pigliucci, 2005). The expression of a phenotype under a certain environment is the result of the integration of local responses of many modules or traits and their interactions (Pigliucci & Preston, 2004). Consequently, the evolution of a given trait and its plasticity may be subject to restrictions due to its genetic correlations with other traits (Agrawal & Stinchcombe, 2009; Gianoli & Palacio-López, 2009). Not all species or traits are plastic, probably because the extent of phenotypic plasticity is limited by both intrinsic and extrinsic factors (DeWitt et al. , 1998; Givnish, 2002; Valladares et al ., 2007).
A process that may limit phenotypic plasticity is developmental canalization, which is generally assessed by coefficient of variation (CV) in traits (Woods et al. , 1999). Developmental canalization, or robustness, is the property of an organism that buffers development against environmental and genetic perturbations to produce a consistent phenotype (Waddington, 1957), including environmental canalization and genetic canalization (Wagner et al. , 1997). Environmental canalization indicates the insensitivity of traits to environmental perturbations in variable environments (Debat & David, 2001; S.C. Stearns et al. , 1995; Wagner et al. , 1997), like a process in opposition to phenotypic plasticity (Stearns et al. , 1995; Wilkins, 1997). But both phenotypic plasticity and environmental canalization may reflect the ability of an organism to adjust phenotypic expression appropriately in dealing with environmental changes, at individual and population levels respectively (Reed et al. , 2010; Schlichting & Pigliucci, 1998). Genetic canalization, refers to the buffering of the effects of genetic variation (Wagner et al. , 1997). The developmental canalization here may include both genetic and environmental canalization. It is widely appreciated that stress conditions increase phenotypic variability in traits (Woods et al. , 1999). If both plasticity and instability of traits increase with environmental stress, it is reasonable to infer some common mechanisms in charge of the two kinds of variabilities (Meiklejohn & Hartl, 2002). Wagner et al.(1997) hypothesized that selection for environmental canalization may facilitate the evolution of genetic canalization, given the ample evidence for genetic canalization from empirical studies (Dun & Fraser, 1958; Dworkin, 2005; Polaczyk et al. , 1998; but see Hermisson & Wagner, 2004). However, relevant studies did not find changes of environmental canalization affect the level of genetic canalization (Debat et al. , 2009; Ian Dworkin, 2005; Meiklejohn & Hartl, 2002). In the unpredictable environments of Mediterranean ecosystem, environmental stress favors both plasticity and high degree of genetic variations across species in oak seedlings (Valladares et al. , 2002), implying negative correlations between plasticity and canalization. Studies on the relationships between trait plasticity and canalization are rarely seen, even less on how environments affect their relationship, especially for plant species.
Phenotypic integration may also play a role as an internal constraint to phenotypic plasticity (Gianoli, 2001, 2003; Pigliucci et al. , 1995; Schlichting, 1986, 1989; Valladares et al. , 2007). Phenotypic integration refers to the pattern and magnitude of character correlations (M. Pigliucci & Preston, 2004), which results from genetic, developmental and/or functional connections among traits (Pigliucci & Preston, 2004; Schlichting & Pigliucci, 1998), and is often expressed in terms of the number of significant phenotypic correlations between traits (Pigliucci, 2002; Pigliucci & Marlow, 2001; Schlichting, 1989). Evidence has shown that the degree of plasticity in response to shading or drought in a given trait decreased with the increase of the number of its correlations with other traits in two local species from Chile, suggesting the restriction of phenotypic integration to the degree of plasticity (Gianoli & Palacio-López, 2009), and the greater the constraints of genetic correlations for a trait, the lower its ability to respond to environments. On the other hand, however, the strength of phenotypic integration may also increase with environmental stresses (García-Verdugo et al. , 2009; Gianoli, 2004; Schlichting, 1989; Waitt & Levin, 1993), which also induce plastic responses in traits, implying the positive correlations between phenotypic integration and plasticity in traits. Since most empirical studies have only studied patterns of phenotypic integration in a single environment (Pigliucci & Preston, 2004; but see Liu et al. , 2007; Pigliucci et al. , 1995), we still know surprisingly little about how the environment influences levels of phenotypic integration (Mallitt et al. , 2010), even less on the relationship between phenotypic integration and plasticity and its variation with environmental conditions.
Population density, as one of the major natural stresses that result in size variations in plants, its variation can result in the heterogeneity of multiple environmental factors, inducing complex responses in traits. Both variation among individuals and phenotypic plasticity can impart integration among morphological traits (Klingenberg, 2014). Thereby such responses to density may also correlate with developmental canalization and phenotypic integration. Correlations can vary with different abiotic conditions and growth stages, due to their influences on trait plasticity (Wang et al. , 2017), integration and canalization (Damiánet al. , 2018; Goswami et al ., 2015). We need more detailed studies on integration, canalization and plasticity to generalize about the relationships among them (Kavanagh, 2020), especially at different stages of plant growth or under different abiotic conditions. As traits may differ in the magnitude of their plastic responses to different resources, resulting in a “hierarchy” of responses (White, 1979), the importance of considering a wider range of plant traits is underscored (Ryser & Eek, 2000). Here we conducted a field experiment on an annual herbaceous species ofAbutilon theophrasti , to analyze phenotypic plasticity, canalization and integration for a number of allocation and morphological traits and their relationships in response to density for plants under two contrasting soil conditions at two growth stages. We aimed to test the following hypotheses: 1) trait plasticity, canalization and integration increase with higher densities; 2) there are positive correlations among the three processes, which intensified with greater densities; 3) soil conditions and growth stage can influence responses of their correlations to density.