Ade,
Interesting what you say about the Severn. Living in Bewdley I've got the same advantages on the Middle river as you have in relation to the Wensum and fish short sessions when I think conditions are best. I have found that the period after it has been dark for a few hours is actually the most productive. My equivalent figures are 374 barbel in 337 hours or a barbel every 54 minutes to be precise. On the other hand on the very rare occasions when I can commit myself to longer sessions they tend to be further afield on the Lower Severn, but not necessarily in the best conditions, and as a result the catch rate drops to 8 from 26 hours or one every 3 hours 15 minutes.
For what its worth my mates who fish the Trent all report the same night feeding pattern as you identified.
My guess is that this issue of the day/night feeding ratio is related to the barbel population density.
Let me give an example. The Teme in summer is very prolific. The barbel will feed all day long on a much more regular basis than on the Middle Severn. I think the catch rates are so high because many barbel move in from the Lower Severn to spawn and stay all summer because of the better oxygenated water and the cover. Come winter the Teme tumbles down from Wales at a greater velocity than the Severn and it becomes a moody and difficult river. Catch rates plummet as the shoals drop downstream and spread out into the Severn down to Tewksbury if not beyond.
Though the Trent is a prolific venue it is, unlike the Severn and Teme, a natural barbel habitat in which the species has co-existed with others for millions of years and the barbel population is more 'in balance' with the rest of nature.
Dusk and night feeding are natural patterns in barbel, day time feeding less so. To my mind day time barbel feeding is angler induced behaviour which it is easier to stimulate when you have more fish per square metre.
Tom