The hypergeometric distribution models sampling without replacement from a finite population. It gives the probability of getting x successes in n draws from a population of size N that contains a number of successes. The mean is equal to n(a/N) and the variance is equal to n(a/N)(1-a/N)(N-n)/(N-1). When the sample size n is small compared to the population N, the binomial distribution is a good approximation to the hypergeometric.