ONE OF MY hstory masters always used to qualift the dates he gave with the proviso, 'Give or take a hundred years'. I'm afraid we took this for ignorance, but I think it was his way of explaining that there was more to the subiect than the recital of dates. Later at Oxford mv d examiners asked me to give the dates of George II17s prime ministers. I thought then, and stdl do, that ths was a pretty fatuous way to evaluate three years of study under historians such as Richard Cobb and Christopher Hd. In other words, I am not a fan of hstory by numbers.

Nor, at times, is Christopher Lee. As he says, 'Only dvnasties end on time and to date. Periods and influences rarely harmonise with set times.' Quite so. So why 1603? Well, it's the year Elizabeth I died, and 'a new monarchv is never a dull moment in British historv'. I'm