Variance in Laboratory Experiments

Given the expected frequency of a certain outcome of a replicate in an experiment, we can estimate the expected variance around that mean (either by deriving it or performing simulations). I have heard that laboratory experiments tend to have greater variances than expected due to conditions not included in the model (ie, we can't control for every variable in an experiment) when determining the expected variance. I am looking for a citation that addresses the issue of variance in laboratory experiments. Specifically, I am interested in an article that deals with higher than expected variance in experimental replicates. Ideally, the article would deal with biological experiments, and it would be even better if it dealt with breeding experiments. Any suggestions would be greatly appreciated and can be left in the comments or emailed to me.