nested ANOVA - data frame

[Liz von Moos <lizvonmoos@gmail.com>]

Hello,

I'm working with S-PLUS 8.04 in Windows XP.

I just figured out that I can use the GUI to do a nested ANOVA, so my
life is a little simpler. However, I don't know how to organize my
data or maybe, I'm doing everything right and don't realize it.

Here is what my data set looks like:

reading treatment       well
0.878   22      1
0.865   22      2
0.924   22      3
0.91            22      4
0.9             22      5
0.985   22      6
0.912   22      7
0.857   22      8
0.835   22      9
0.884   22      10
0.89            22      11
0.932   22      12
0.874   22      13
0.957   22      14
0.877   22      15
0.899   22      16
0.953   200     1
0.931   200     2
0.917   200     3
0.972   200     4
0.908   200     5
0.962   200     6
0.931   200     7
0.946   200     8
0.862   200     9
0.9             200     10
0.814   200     11
0.969   200     12
0.903   200     13
0.955   200     14
0.904   200     15
0.917   200     16
0.842   500     1
0.834   500     2
0.829   500     3
0.84            500     4
0.815   500     5
0.872   500     6
0.822   500     7
0.854   500     8
0.813   500     9
0.8             500     10
0.859   500     11
0.854   500     12
0.814   500     13
0.87            500     14
0.828   500     15
0.85            500     16

I want to look at the between-treatment variation (22, 200, 500) and
the within-treatment variation (wells 1-16 within a treatment).

This is what I get:

** Analysis of Variance Model ***

Short Output:
Call:
            aov(formula = reading ~ treatment + well %in% treatment,
data = cyto.gal.stack, na.action = na.exclude)

Terms:
                               treatment       well %in% treatment
    Residuals
 Sum of Squares     0.06076288        0.00195740                 0.05236504
Deg. of Freedom          2                     3
           42

Residual standard error: 0.03530987
Estimated effects may be unbalanced

                                     Df            Sum of Sq
Mean Sq         F Value         Pr(F)
            treatment            2            0.06076288
0.03038144     24.36779     0.0000001
  well %in% treatment      3            0.00195740        0.00065247
  0.52332       0.6686214


            Residuals          42            0.05236504         0.00124679

Is this right? I'm so lost. I expected well%in%treatment to have 15
degrees of freedom since I have 16 wells so I suspect my formula or
data set up is wrong. Also, why does it suggest that the "estimated
effects may be unbalanced"?

Please please help :(

Liz