same subset of trials in different blocks

My study includes six blocks, three of which will include judgments of a subset of stimuli (from a larger stimulus list). I have been able to have the first block select the subset at random without any problem.

Is there a way to have this specific/identical subset of stimuli used in the other blocks that record judgments? As I understand it, the way that the subset feature works doesn’t permit this sort of thing. Is there a way around this limitation?

Here would be the design:

Block 1 - straightforward
Block 2 - randomly select 50/434 possible trials, from stimulus list
Block 3 - straightforward
Block 4 - SAME 50 trials from the stimulus list, but with a different set of judgments
Block 5 - straightforward
Block 6 - SAME 50 trials from the stimulus list, again with a different set of judgments

Unfortunately this is not a capability of SuperLab.