Late spring and late summer NSC concentrations per unit of dry mass in leaves, aboveground woody tissues, and roots in seedlings of species diff ering in light requirements, established in the understory of a cold rainforest and a Mediterranean forest in Chile. Bars represent mean values and error bars represent standard error; hatched and nonhatched areas symbolize starch and soluble sugar concentrations, respectively. The x-axis indicates abbreviated species names according to Table 1 . 

Contexts in source publication

Context 1

... both forests, patterns of root and aboveground wood NSC and starch (but not of soluble sugars) in relation to shade tolerance were aff ected by sampling dates (i.e., signifi cant interaction Tolerance x Date, Table 2 , Figs. 1, 2 ) . However, late spring NSC and starch concentrations in woody tissues were not higher in shade tolerant species than in less tolerant species. In both forests, root NSC concentrations were similar between shade tolerant and intol- erant species in the late spring, and lower in shade tolerant than shade intolerant species in the late summer ( Fig. 1 ). In the cold rainforest, the latter was caused by a decrease in the NSC concen- tration of shade tolerant species in the late summer. In the Medi- terranean forest, in contrast, this pattern was driven by an increase in NSC and starch concentra- tions of shade intolerant species in the late summer ( Table 2 , Fig. 1 ). Furthermore, shade in- tolerant species in the Mediter- ranean forest increased their aboveground wood starch con- centrations in the late summer ( Table 2 , Fig. 1 ). As an excep- tion, aboveground wood NSC concentrations in the cold rain- forest were unaff ected by sam- pling date. Within a given shade toler- ance category, there was a high interspecifi c variation in NSC, starch, and soluble sugar con- centrations, and in the sampling date eff ects ( Fig. 3 ). Likewise, NSC fractions were very vari- able within tolerance categories, especially for roots and above- ground wood ( Fig. 4 ). Except for leaf NSC fractions, the interspe- cifi c variation in NSC was not apparently related to shade tol- erance ( Figs. 3, 4 ) ...

Context 2

... both forests, patterns of root and aboveground wood NSC and starch (but not of soluble sugars) in relation to shade tolerance were aff ected by sampling dates (i.e., signifi cant interaction Tolerance x Date, Table 2 , Figs. 1, 2 ) . However, late spring NSC and starch concentrations in woody tissues were not higher in shade tolerant species than in less tolerant species. In both forests, root NSC concentrations were similar between shade tolerant and intol- erant species in the late spring, and lower in shade tolerant than shade intolerant species in the late summer ( Fig. 1 ). In the cold rainforest, the latter was caused by a decrease in the NSC concen- tration of shade tolerant species in the late summer. In the Medi- terranean forest, in contrast, this pattern was driven by an increase in NSC and starch concentra- tions of shade intolerant species in the late summer ( Table 2 , Fig. 1 ). Furthermore, shade in- tolerant species in the Mediter- ranean forest increased their aboveground wood starch con- centrations in the late summer ( Table 2 , Fig. 1 ). As an excep- tion, aboveground wood NSC concentrations in the cold rain- forest were unaff ected by sam- pling date. Within a given shade toler- ance category, there was a high interspecifi c variation in NSC, starch, and soluble sugar con- centrations, and in the sampling date eff ects ( Fig. 3 ). Likewise, NSC fractions were very vari- able within tolerance categories, especially for roots and above- ground wood ( Fig. 4 ). Except for leaf NSC fractions, the interspe- cifi c variation in NSC was not apparently related to shade tol- erance ( Figs. 3, 4 ) ...

Context 3

... e lack of a positive relation- ship between C storage in woody tissues and shade tolerance in the late spring could indicate that C allocation to other sinks (e.g., growth, defenses) is more benefi - cial for low light persistence in the examined forests. Baltzer and Th omas (2007) found that light requirements to survive were highly correlated with light re- quirements to growth in Bornean rainforest seedlings, suggesting that some level of growth is criti- cal for low light persistence. Con- sistent with this suggestion, in a temperate southern Chilean for- est, Nothofagus nitida had similar storage, but higher survival and growth rate than the relatively more shade intolerant Nothofa- gus dombeyi when seedlings of both species were grown under experimental shade ( Piper et al., 2009 ). On the other hand, invest- ment in chemical and structural defenses could be suffi cient to protect shade tolerant species from disturbances in some cases. For example, two winter deciduous species with contrasting light requirements signifi cantly diff ered in their concentrations of leaf chemical defenses, but not in C storage or growth when they were compared in low light ( Imaji and Seiwa, 2010 ). Similarly, Poorter et al. (2010) found that low light survival aft er damage was strongly related to wood density and only mod- estly related to C storage in tropical species. Evidence from Chilean Mediterranean forests suggests that shade tolerant species are much better defended than less tolerant species: understory her- bivory in Aristotelia chilensis was more than four times higher than in Persea lingue and Cryptocarya alba ( Simonetti et al., 2007 ). Overall, I found that NSC concentrations in woody tissues of these species were much more similar ( Fig. 3 ). A high investment in defenses (i.e., herbivory resistance) could make herbivory tolerance unnec- essary. If so, traits related to herbivory tolerance (e.g., C storage) will not be selected. Alternatively, high investment in both defenses and storage could be incompatible if they compete for limiting re- sources ( Leimu and Koricheva, 2006 ) (e.g., C availability in low ...

Context 4

... both forests, patterns of root and aboveground wood NSC and starch (but not of soluble sugars) in relation to shade tolerance were aff ected by sampling dates (i.e., signifi cant interaction Tolerance x Date, Table 2 , Figs. 1, 2 ) . However, late spring NSC and starch concentrations in woody tissues were not higher in shade tolerant species than in less tolerant species. In both forests, root NSC concentrations were similar between shade tolerant and intol- erant species in the late spring, and lower in shade tolerant than shade intolerant species in the late summer ( Fig. 1 ). In the cold rainforest, the latter was caused by a decrease in the NSC concen- tration of shade tolerant species in the late summer. In the Medi- terranean forest, in contrast, this pattern was driven by an increase in NSC and starch concentra- tions of shade intolerant species in the late summer ( Table 2 , Fig. 1 ). Furthermore, shade in- tolerant species in the Mediter- ranean forest increased their aboveground wood starch con- centrations in the late summer ( Table 2 , Fig. 1 ). As an excep- tion, aboveground wood NSC concentrations in the cold rain- forest were unaff ected by sam- pling date. Within a given shade toler- ance category, there was a high interspecifi c variation in NSC, starch, and soluble sugar con- centrations, and in the sampling date eff ects ( Fig. 3 ). Likewise, NSC fractions were very vari- able within tolerance categories, especially for roots and above- ground wood ( Fig. 4 ). Except for leaf NSC fractions, the interspe- cifi c variation in NSC was not apparently related to shade tol- erance ( Figs. 3, 4 ) ...

Context 5

... both forests, patterns of root and aboveground wood NSC and starch (but not of soluble sugars) in relation to shade tolerance were aff ected by sampling dates (i.e., signifi cant interaction Tolerance x Date, Table 2 , Figs. 1, 2 ) . However, late spring NSC and starch concentrations in woody tissues were not higher in shade tolerant species than in less tolerant species. In both forests, root NSC concentrations were similar between shade tolerant and intol- erant species in the late spring, and lower in shade tolerant than shade intolerant species in the late summer ( Fig. 1 ). In the cold rainforest, the latter was caused by a decrease in the NSC concen- tration of shade tolerant species in the late summer. In the Medi- terranean forest, in contrast, this pattern was driven by an increase in NSC and starch concentra- tions of shade intolerant species in the late summer ( Table 2 , Fig. 1 ). Furthermore, shade in- tolerant species in the Mediter- ranean forest increased their aboveground wood starch con- centrations in the late summer ( Table 2 , Fig. 1 ). As an excep- tion, aboveground wood NSC concentrations in the cold rain- forest were unaff ected by sam- pling date. Within a given shade toler- ance category, there was a high interspecifi c variation in NSC, starch, and soluble sugar con- centrations, and in the sampling date eff ects ( Fig. 3 ). Likewise, NSC fractions were very vari- able within tolerance categories, especially for roots and above- ground wood ( Fig. 4 ). Except for leaf NSC fractions, the interspe- cifi c variation in NSC was not apparently related to shade tol- erance ( Figs. 3, 4 ) ...

Context 6

... e lack of a positive relation- ship between C storage in woody tissues and shade tolerance in the late spring could indicate that C allocation to other sinks (e.g., growth, defenses) is more benefi - cial for low light persistence in the examined forests. Baltzer and Th omas (2007) found that light requirements to survive were highly correlated with light re- quirements to growth in Bornean rainforest seedlings, suggesting that some level of growth is criti- cal for low light persistence. Con- sistent with this suggestion, in a temperate southern Chilean for- est, Nothofagus nitida had similar storage, but higher survival and growth rate than the relatively more shade intolerant Nothofa- gus dombeyi when seedlings of both species were grown under experimental shade ( Piper et al., 2009 ). On the other hand, invest- ment in chemical and structural defenses could be suffi cient to protect shade tolerant species from disturbances in some cases. For example, two winter deciduous species with contrasting light requirements signifi cantly diff ered in their concentrations of leaf chemical defenses, but not in C storage or growth when they were compared in low light ( Imaji and Seiwa, 2010 ). Similarly, Poorter et al. (2010) found that low light survival aft er damage was strongly related to wood density and only mod- estly related to C storage in tropical species. Evidence from Chilean Mediterranean forests suggests that shade tolerant species are much better defended than less tolerant species: understory her- bivory in Aristotelia chilensis was more than four times higher than in Persea lingue and Cryptocarya alba ( Simonetti et al., 2007 ). Overall, I found that NSC concentrations in woody tissues of these species were much more similar ( Fig. 3 ). A high investment in defenses (i.e., herbivory resistance) could make herbivory tolerance unnec- essary. If so, traits related to herbivory tolerance (e.g., C storage) will not be selected. Alternatively, high investment in both defenses and storage could be incompatible if they compete for limiting re- sources ( Leimu and Koricheva, 2006 ) (e.g., C availability in low ...