Thanks, folks for those observations.
Plus ça change...
I had hoped that having these trip meters would mean there was no need to do anything other than pause one and restart the other twice a year when the tyres were changed. But obviously, I am looking for rather more in my trip meters than trip meters provide.
I'm not much given to manual recording or spreadsheet keeping, but may make an exception for this. When first I had my previous Jazz, I monitored my fuel consumption in the traditional way for a short while, there being no car display to mislead. And then when I first put winter tyres on it, I did the same. So far I could see, any fuel consumption difference was negligible. That impression remained with me once I had stopped measuring it directly. The need to fill the tank always arrived more or less when I had come to expect it before the tyres were changed.
So I will see what is revealed by the rather less impressionistic approach Peteo48 suggests. Twice a year on the tyre change the readings on B will be recorded and put in a spreadsheet and Trip Meter B will then be reset. Trip Meter A, which was reset when I bought the car and has not been reset since, will provide the background average for comparison with any individual seasonal result.
I will bear in mind Stupot's, John's and TG's points, but probably not do much about them. In fact, I can't now do an actual against car-computer fuel consumption comparison before the beginning of the next tyre season. And to give the car computer a fair trial, I suppose I should spend a whole tyre season filling the tank and totting up a string of fuel used figures. I don't see myself doing that.
In fact, I'm probably more interested in recording the total mileage travelled by each set of tyres than in recording their fuel consumption characteristics. I've had a little bet with myself that the Michelin Alpins that have just gone on the car will probably wear more slowly than the OEM Dunlop SP Sports that I have just taken off.
We will see.