"colorbar.plot" <-
function (x,y,strip, strip.width = 0.1, strip.length=4*strip.width, 
zrange=NULL, adj.x=.5,adj.y=.5,col=tim.colors(256),horizontal=TRUE,...) 
{

# coerce to be one column matrix if it is a vector
if( !is.matrix( strip)) { strip<- matrix( c( strip), ncol=1)}

m<- nrow( strip)
n<- ncol(strip)

# find common range across strips if not specified
if( is.null(zrange)) { 
   zrange<- matrix( range(c( strip),na.rm=TRUE), nrow=n, ncol=2, byrow=TRUE) }

# see help( par) for background on graphical settings 
    ucord <- par()$usr
    pin <- par()$pin

if( horizontal){
    dy<- strip.width*(ucord[4] - ucord[3])
    dx<- strip.length* pin[2]*(ucord[2] - ucord[1])/(pin[1])}
 else{
    dx<- strip.width*(ucord[2] - ucord[1])
    dy<- strip.length* pin[1]*(ucord[4] - ucord[3])/(pin[2])}


#
# dx and dy should have the correct ratio given different different scales 
# and also different aspects to the plot window 
#

n<- ncol( strip)
m<- nrow( strip)

# create grids in x and y for strip(s) based on the users
# coordinates of the plot and th positioning argument (adj)

if( horizontal) { 

    xs<- seq(0, dx,, m+1) +x - adj.x*dx
    ys<- seq(0, dy,,n+1) +y - adj.y*dy }
 else{
    xs<- seq(0, dx,, n+1) +x - adj.x*dx
    ys<- seq(0, dy,,m+1) +y - adj.y*dy }


#
# plot image row by row to allow for different zlim's 
# see image.add for a fields function that just plots the whole image at 
# once. 

    for( k in 1:n){
if( horizontal){    
  image( xs, c(ys[k],ys[k+1]), cbind(strip[,k]), 
                    zlim= zrange[k,],add=TRUE, col=col,...)}
else{
  image( c(xs[k],xs[k+1]), ys, rbind(strip[,k]), 
                    zlim= zrange[k,],add=TRUE, col=col,...)}

    }

}