The way I suggest you study this book is to play through the main games once, relatively quickly, and then start playing the variation in actual games. Playing an opening in real games is of vital importance - without this kind of live practice it is impossible to get a 'feel' for the kind of game it leads to. There is time enough later for involvement with the details; after playing your games it is good to look up the line.
It's good advice, and I've used it myself and with my students for years. Looking at variations, especially sub-sub-sub variations, is unlikely to help until one has that feel Davies mentioned; but once one has it, the details are transmogrified from boring fine print into useful, appreciated information.
I'd perhaps add one more step or, rather,refinement of his approach. For those with databases, I recommend finding "heroes" for particular openings. Once you've picked them, create a mini-database of their games in that opening or variation, and scroll through all of them (or at the very least, a large chunk of them) in a single sitting and at a fairly high speed. (Repeat regularly.) That, in conjunction with (and subsequent to) replaying the games in Davies' or whomever's book, will help create a "proto-feel" going into one's own initial forays. By looking at a slew of games in a given variation rather than just a handful, you can start to see which middlegame factors are relevant and which aren't - and that's not just useful, it's essential to really understanding an opening!
To recap, here's the Davies model, with my modification:
I. Before playing a new opening/variation:
A. Quickly and fairly superficially play through the games in a text devoted to that opening.
B. Form a "heroes" database and rapidly scroll through its games.
II. Playing the new opening/variation (self-explanatory)
III. After playing the new line: now compare your play to the decrees of theory, examining relevant subvariations, etc.
Try it, it works!


