Hi, Using this table
| Grand total | 19.55 | | |------------------+-------+----| | <2013-05-09 Thu> | 1.23 | v2 | | <2013-05-04 Sat> | 5.76 | v1 | | <2013-05-14 Tue> | 3.78 | v1 | | <2013-04-24 Wed> | 8.78 | v2 | #+TBLFM: @1$2=vsum(@I..@>);%.2f I want the formula to be relative to the first hline so, according to the manual, it should be possible to change @1$2 into @I-1$2 | Grand total | 19.55 | | |------------------+-------+----| | <2013-05-09 Thu> | 1.23 | v2 | | <2013-05-04 Sat> | 5.76 | v1 | | <2013-05-14 Tue> | 3.78 | v1 | | <2013-04-24 Wed> | 8.78 | v2 | #+TBLFM: @I-1$2=vsum(@I..@>);%.2f But this produces very strange results. Hitting C-c * with the cursor in @>$2 produces: | Grand total | 19.55 | | |------------------+-------+----| | <2013-05-09 Thu> | 1.23 | v2 | | <2013-05-04 Sat> | 5.76 | v1 | | <2013-05-14 Tue> | 3.78 | v1 | | 2939977.00 | 19.55 | v2 | #+TBLFM: @I-1$2=vsum(@I..@>);%.2f Why does it change <2013-04-24 Wed> into 2939977.0, and 8.78 into 19.55? Hitting C-u C-c * with the cursor in @>$2 messes up the complete table and results in: | 0.00 | 30.32 | | |------------+--------+----| | 0.00 | 30.32 | v2 | | 4409973.00 | 59.41 | v1 | | 0.00 | 113.06 | v1 | | 7349950.00 | 222.34 | v2 | #+TBLFM: @I-1$2=vsum(@I..@>);%.2f Am I correct to assume that this is at least unexpected, and probably erroneous behaviour? Guido -- "Consider a spherical bear, in simple harmonic motion..." -- Professor in the UCB physics department