Two of the transductions were to construct double-mutant E. coli strains carrying both the ppdD::lacZ fusion (reporter for activity of the CRP-S promoters, and thus of Sxy) and either a sxy::kan knockout or a crp::kan knockout. The goal is to test whether the high baseline expression of the reporter is due to Sxy-dependent CRP-dependent activity of the CRP-S promoter, or to baseline promoter-independent expression (e.g. from other sequences in the fusion).
Both of these transductions produced lots of colonies on the Amp+Kan selective plates. But one of the negative controls (same cells, but no P1) produced just as many, suggesting that something other than transduction was responsible for the colonies. As controls for the selection I had streaked the reporter and knockout cells onto the same plates, and onto Amp or Kan plates. These told me that the Amp+Kan plates were faulty, perhaps because the Amp was old.
I would have concluded that I should just do it again, with fresh plates, but my other transduction was a positive control for transduction (one that I knew should work), and it didn't work. This experiment tried to transduce the lacY gene from the wildtype strain W3110 into the lacY mutant C600, selecting for Lac+ by plating on minimal salts with lactose as the only sugar. The negative control was C600 with no P1. The positive control of plating the cells on minimal salts plus glucose worked well - everyone grew fine. I can't remember whether I also plated the donor cells on minimal lactose - this would have been a good control. Colonies on the lactose plates were initially very tiny, and the negative control (no phage) cells gave just as many as did the cells with phage (in fact more, probably because the phage killed some cells). So there's no evidence that this positive-control transduction worked either.
What to do next? Check growth of W3110 on the lactose plates? Repeat the lacY transduction? First I must recheck the titers of my supposedly-transducing lysates (plated last night). And go back over my notes, looking for any corners I might have cut. This is just another example of the truth of my favourite saying:
"Most scientists spend most of their time trying to figure out why their experiments won't work."