R help - Jul 2007 - Basic missing values question...

I am trying to run the power.t.test function to calculate sample size for a 
data frame of variances (i.e, 228 rows x 4 cols).  Many of the fields are 
missing values, which crashes the function.  I have tried na. options in the 
function, but the function doesn't recognize them.  I have also tried an 
if-else loop to skip over fields with missing values.  How can I deal with 
these missing values?

Here's the code:

var.df<-read.table("Transect_variance.csv",sep=",",header=TRUE,
fill=TRUE)
n.df<-data.frame(cbind(var.df[,1:2],rep(-1,228),rep(-1,228),rep(-1,228),rep(-1,228)))
	names(n.df)<-c("Strata","Transect","n35","n45","n36","n46")

#var.df[1:2,]
#	Strata 	Transect      	 ID		Var35    	Var45 		Var36    	Var46
#	Herb    	3-1 		Herb 3-1 	2.024264      	NA 		7.182       	NA
#	Herb    	4-1 		Herb 4-1 	2.232552 	2.26 		2.360 		2.393387

for i in 1:6 {
	for j in 1:228 {
			if ( var.df[i,j]=="") {n.df[i,j]==""  } else {
			power.t.test(delta = 2.5, sd = var.df[i,j], sig.level = 0.1, power = 0.9, 
type = "two.sample", alternative = "two.sided")
			}

	}
}