Problem with fisher.test

[asanquer@virbac.fr]

Hello,

I have a problem with the use of fisher.test:

Hereafter the crossatbulation of my data:

Call:
crosstabs(formula =  ~ Value + Group, data = mydata)
133 cases in table
+----------+
|N         |
|N/RowTotal|
|N/ColTotal|
|N/Total   |
+----------+
Value  |Group
       |A      |B      |C      |RowTotl|
-------+-------+-------+-------+-------+
0      |25     |29     |29     |83     |
       |0.3    |0.35   |0.35   |0.62   |
       |0.54   |0.67   |0.66   |       |
       |0.19   |0.22   |0.22   |       |
-------+-------+-------+-------+-------+
1      |11     | 8     | 6     |25     |
       |0.44   |0.32   |0.24   |0.19   |
       |0.24   |0.19   |0.14   |       |
       |0.083  |0.06   |0.045  |       |
-------+-------+-------+-------+-------+
2      | 6     | 4     | 5     |15     |
       |0.4    |0.27   |0.33   |0.11   |
       |0.13   |0.093  |0.11   |       |
       |0.045  |0.03   |0.038  |       |
-------+-------+-------+-------+-------+
3      | 4     | 2     | 4     |10     |
       |0.4    |0.2    |0.4    |0.075  |
       |0.087  |0.047  |0.091  |       |
       |0.03   |0.015  |0.03   |       |
-------+-------+-------+-------+-------+
ColTotl|46     |43     |44     |133    |
       |0.35   |0.32   |0.33   |       |
-------+-------+-------+-------+-------+
Test for independence of all factors
   Chi^2 = 2.98057 d.f.= 6 (p=0.8112815)
   Yates' correction not used
   Some expected values are less than 5, don't trust stated p-value

When I want to perform a Fisher test, I obtain the following error message:

> fisher.test(mydata$Value, mydata$Group)
Problem in fisher.test(mydata$Value, ..: stack for npoin is full in column
   3 .
 Make node.stack.dim bigger
Use traceback() to see the call stack

Does someaone could explain me what is the problem, and how can I solve it?

Thanks a lot

Annaële Sanquer