Prevent labels from clipping in hclust/dendrogram plot

Question

I'm producing dendrograms to compare song similarity of a bird species between states. However, I can't figure out how to prevent the state names from clipping when I produce the plot (click for example). Any ideas?

Code:

var.towhee <- read.csv(file="states.csv", header=TRUE, fill=TRUE)
rownames(var.towhee) <-var.towhee$State # Set row names to state name
var.towhee <- var.towhee[,-1] # Remove state column 

library(vegan)
library(permute)
library(lattice)

norm <- decostand(var.towhee, method="normalize") # Normalize data
dis <- vegdist(norm, method="euclidian") # Calculate distances
UPGMA <- hclust(dis, method="average") # Cluster using UPGMA method
UPGMA <- as.dendrogram(UPGMA) # Convert hclust objects into dendrogram objects

plot(UPGMA, horiz=TRUE, xlab="Song Distance")

For reference this is how my data is formatted:

            Variable 1    Variable 2    Variable 3
State 1     123           123           123
State 2     123           123           123
State 3     123           123           123

Note that I've set the row names to be states rather than numbers. This is where the plot is grabbing the labels from.

Answer 1

The problem can be solved setting mar parameters.
Here is an example of a dendrogram with clipped labels:

hc <- hclust(dist(USArrests), "ave")
hc <- as.dendrogram(hc)
par(mar=c(3,4,1,1))
plot(hc, horiz=TRUE)

and here the figure with the complete labels:

par(mar=c(3,4,1,6))
plot(hc, horiz=TRUE)