Maybe it's hard to imagine a book on maths as something you would read in your spare time. I did, however, and it was one of the most fascinating and intellectually stimulating books I have ever read.
The weird thing about mathematics is that it is incredibly simple in its origins. You just start with one number and if you keep adding ones you get the essential series of 1, 2, 3, 4, 5 etc. (And of course you can further complicate it by making numbers negative, or fractions like 0,25 and 1/4, but those are just modifications that don't change a lot to the basic system)
But when you start looking at the relations between the numbers, unexpected patterns emerge. For instance, Euclid defined something like a 'perfect number'. It is a number that is equal to the sum of its divisors. 6 can be divided by 1, 2 and 3. Oddly enough, 6 is also the sum of 1+2+3, thus making it a perfect number. The next one is 28 (1+2+4+7+14) and then they quickly become rare, with 496 and 8128 as the next perfect numbers in line. Euclid also discovered a rather simple formula involving primes to find the next perfect numbers.
That proves a strange relation between primes and the arbitrary concept of a 'perfect number'. Primes themselves are even a lot weirder, but I can't go into that here. Other interesting parallels can be found between the Golden Ratio and the Fibonacci sequence: 0, 1, 1, 2, 3, 5, 8, 13, 21 and so on.
I guess there are many accessible introductions to mathematics around, some good, some bad. Of Devlin's book I can say that it is an excellent one, very readable and though it sometimes demands a little effort to understand a problem, you are always rewarded by a fascinating new insight.