FIGURE 26. Experimental planting established with transplants shown in 

 Figure 25, May 1981. 



Data Analysis 



Data were analyzed using the Statistical Analyses System (SAS) 

 programs for analysis of variance and least significant difference 

 (LSD) (Barr et al., 1976). All statistically significant differences 

 were determined at the .05 level. Variability was generally high and 

 not all data could be analyzed statistically. We feel that in field 

 experiments of the type conducted on disturbed marsh sites that overall 

 observations, photographs, and at times fragmentary data have to be 

 interpreted and used as best they can even when statistical 

 significance cannot be documented. Consequently, because of these 

 conditions and the fact that data of this type are not readily 

 available, we have included data in this report that we consider 

 important even though the variability is high. 



381 



