The residuals table for nonlinear regression only has X values for the first data set, so Prism cannot plot the residuals from the other data sets.

This happens with Prism 3 and 4 when you have entirely separate X values for each data set. For example, data set A might be in rows 1-10, and data set B in rows 11-20. This way of entering data lets you have different X values for each data set. But when you ask for residuals from nonlinear regression, Prism puts all the residuals (Y values) into the results table, but only puts in X values for the first data set. The X values for dataset B and later are missing (empty).

Here is an easy work around: Insert a new row at the top of the table (select the first row, then choose Insert...Row). If row 1 is entirely blank, Prism won't get trapped by the bug and the residuals table will have X values for all the datasets.