R help - Jan 2009 - high values of my response variable get chopped off of lattice dotplots

I am trying the following lattice graphics code:
> library(lattice)
> attach(junk)
> summary(junk)
   mean.barr        variable      value
 Min.   : 3.50   item15 :15   Min.   :0.0000
 1st Qu.: 9.50   item14 :14   1st Qu.:0.0000
 Median :11.00   item16 :14   Median :0.0000
 Mean   :11.64   item1  : 0   Mean   :0.6744
 3rd Qu.:16.00   item2  : 0   3rd Qu.:1.0000
 Max.   :20.00   item3  : 0   Max.   :3.0000
 NA's   :18.00   (Other): 0

Here are a few cases in junk

    rad1 rad2 mean.barr variable value
553   11   NA        NA   item14     0
556   16   16      16.0   item14     2
562   NA   NA        NA   item14     2
565   NA   NA        NA   item14     0
568   20   20      20.0   item14     0
571   17   NA        NA   item14     0
574   13   11      12.0   item14     3
580   15   NA        NA   item14     0
583   18   15      16.5   item14     0
586    8   11       9.5   item14     1
589   12    5       8.5   item14     0
592    8   14      11.0   item14     1
. . . . cut off for brevity . . .
> dotplot(mean.barr~value | variable, data=junk)
The highest value of mean.barr appearing anywhere on my dotplot is 8.
But there are higher values in the dataframe, at all 3 levels of
variable.  Any idea how I can find out what happened to the values > 8?

Thanks.

--Chris

-- 
Christopher W. Ryan, MD
SUNY Upstate Medical University Clinical Campus at Binghamton
40 Arch Street, Johnson City, NY  13790
cryanatbinghamtondotedu
PGP public keys available at http://home.stny.rr.com/ryancw/

"If you want to build a ship, don't drum up the men to gather wood,
divide the work and give orders. Instead, teach them to yearn for the
vast and endless sea."  [Antoine de St. Exupery]