Dear Rick,
Thank you very much. You were right with the RAM. The first time I ran it with 1GB RAM Today I tried a different computer with 2 GB RAM. For one .BRIK file (99 MB) 2GB were also not enough (it stopped at #147 out of 280) For the smaller .BRIK file (47MB) the program continued until the last subbrick (#131) but then I got the error message:

Filesize limit exceeded

The outputfile was 2GB and when I tried to open it with AFNI, nothing worked any more. I had to to kill AFNI.

When I tried:

3dAllineate -base noskull+orig -1Dmatrix_apply matrixav.aff12.1D -source
epi+orig -master epi+orig -final quintic -prefix EPI_aligned
(instead of -master noskull+orig)

it worked

Now I´m not sure whether it is a good idea to write the output dataset on the same grid as the EPI dataset. What is the difference between the 2 output datasets (-master=mprage, -master=epi) and is it ok to do the alignment this way? Or is there a way how to solve the problem I had with -master=noskull+orig?

I would really appreciate your help!!
Thank you very much!

Beate