(original review, 1993)
As a maths and physics graduate, I observe that most compilers of the best books of all-time lists are, self-evidently, not from my side of the cultural divide. They should at the very least, it seems to me, be required to read C. P. Snow's "The Two Cultures" and "The Scientific Revolution", not to mention Arthur Koestler's "The Sleepwalkers" (a masterly history of how mankind has viewed, and developed ideas about, the universe) - and then put them on the lists, as they deserve to be there IMO. At least they do if you buy into the idea of a top 100 in the first place, which, in my view, would be like trying to rank great composers/mathematicians/physicists - inherently rather ridiculous. It might have been better though to have just two lists from either side of the cultural divide. It would not be difficult to come up with 100 tomes from the sciences after all.
If you're into stuff like this, you can read the full review.