APPLICATION OF PHYTOSOCIOLOGICAL TECHNIQUES 327 



although basically of these classes, are actually evergreen, perennating for 

 several seasons without dying back to ground level or below. Not having 

 year-around experience with the vegetation of the stations studied, we have 

 had to use our best judgment in cases where local persons were unable to 

 advise us. 



There is no "normal spectrum" for leaf-size classes and, as pointed out 

 earlier, very few data of any kind, so we merely present a summary table 

 for the four rain-forest stations reported on here (table 26). With the stations 



Table 26. Comparison of Raunkiaerian leaf-size classes in Brazilian rain forest 



Alto do Horto 



Class Mucambo Caiobd Palmital Botanico 



Aphyllous 



Leptophyll » 2.3 



Nanophyll 3.2 



Microphyll 15.1 



Mesophyll 68.3 



Macrophyll 11.0 



Megaphyll 



arranged in the same order as in table 25, we find that the three smaller leaf- 

 size classes (and the aphyllous class) increase with departure from the cen- 

 tral Amazonian station and that the three larger leaf-size classes diminish 

 steadily in the same direction. There is no question of the validity of these 

 shifts in leaf-size percentages. 



Rain-forest phanerophytes are strongly mesophyll except at Horto Bot- 

 anico, where the vegetation is strongly microphyll. There is some tendency 

 for smaller leaf-size classes to have a higher percentage in taller strata than 

 in lower ones. This shows up at Mucambo and Alto do Palmital. 



SUMMARY 



On a basis of sample plots, Brazilian rain forest has been studied in its 

 equatorial form at Belem, Para, in coastal mountain form at Caioba, Parana, 

 as plateau, temperate forest at Foz de Iguagu, Parana, and as gallery forest 

 in its southernmost extension at Pelotas, Rio Grande do Sul. Several concepts 

 and methods of phytosociology that are commonly applied in studies of 

 temperate vegetation have been employed in this study of tropical rain forest. 

 These studies, although not extensive, show that the complex rain-forest 

 vegetation can be analyzed by these methods and that often they bring out 

 features that are not readily appreciated in non-quantitative studies. 



