Apparently, it had previously been shown that ammonium repressed nitrogen fixation in Azotobacter vinelandii, and even when fixing nitrogen, cells would immediately take up ammonium when it was given, but would not immediately start fixing nitrogen if they ran out of ammonium. They wanted to look at this lag period more closely.
What They Did
They grew A. vinelandii OP (aka CA) in Burk's nitrogen-free medium, and actually this is the paper most people later cited as the best recipe for Burk's, the standard medium for growing A. vinelandii.
So they grew the cells, sometimes with ammonium acetate or potassium nitrate as fixed nitrogen sources, sometimes with chloramphenicol to prevent protein synthesis. They also did enzyme activity assays with nitrogenase, using 15N2. And determined protein content of cells.
What They Observed
The first figure, taken from Strandberg's master's thesis, shows that when A. vinelandii is grown in a nitrogen-free atmosphere with ammonium, growth eventually levels off; if N2 is then added, cells start growing again after a short lag, 30-60 minutes. But if ammonium is added instead, there's no lag; the cells start growing again immediately. If N2 was present the whole time, the cells switch to nitrogen-fixing pretty quickly when ammonium runs out, with a small decrease in growth rate.
A better demonstration for this lag was nitrogenase activity assays: it showed right when nitrogen fixation activity started, about 1 1/4 hours after ammonium ran out. Though oxygen levels and temperature possibly weren't ideal. It could be as little as 45 minutes later.
Another interesting observation was that when ammonium ran out and cells were in an environment of 40% oxygen (with the rest 60% helium or hydrogen), they didn't start producing nitrogenase, but they did start when oxygen was only 20%. The hydrogen level didn't seem to matter.
One problem they encountered was that there were small amounts of nitrogen in their gas tanks of oxygen, helium, and hydrogen, which could've been enough to affect the results. They tried to make purer oxygen by electrolysis (splitting water), though there was still a bit of nitrogen; still, it wasn't clear whether nitrogenase production was induced by the presence of nitrogen or merely repressed by ammonium. My guess would be the latter, since cells wouldn't normally encounter N2-free environments in nature. But regulation can be complicated.
They noticed a slight rise in turbidity even after ammonium ran out, but speculated it could be due to color change that cells go through (from reddish brown to dark brown) when fixing nitrogen. The small amount of nitrogen in the gas flow was enough to get cells to produce nitrogenase, but not enough for them to use it. But cell-free extracts didn't show different absorbance for the two kinds of cells, despite the visible difference.
When they added chloramphenicol, an antibiotic that inhibits protein synthesis, obviously this inhibited nitrogenase formation. If the enzyme was already present, in vitro, the antibiotic didn't inhibit it. But it did inhibit it in cells, possibly because ammonium built up with no way to use it, repressing nitrogenase.
They tried adding 150 mg N (as ammonium) per liter to a culture of nitrogen-fixing cells, and saw that nitrogenase activity dropped off within about 3 hours. Not as fast as I would expect. They interpreted this to mean that the enzyme is not inhibited immediately, just diluted out as the cells stop producing it while continuing to multiply; but it seems to be inactivated faster than just by dilution, so there might be some inactivation or degradation going on.
