I am performing this simple experiment: I have one variety of grass and 8 different fungi (say #1 to #8). I am going to put 10-20 grass plants in each one of 18 containers, and then I will put each one of the fungi in 2 containers and let 2 with no fungus ("negative control"). After some time I will count how many plants have been killed by each one of the fungi in each one of two replicates for that fungus. I want to compare the results for each one of the fungi with respect to the negative control. I think I can do that using the Fisher exact test, but I don't know if I have to combine the counts of the two replicates into a single count (like using a single container with all the plants together) or if I can keep those count separated. Also, I would like to know if I have to run the test doing pairwise comparisons (controls vs #n) or if I should run a single test considering all the results at the same time. May I do that also with the Fisher exact test? Do I have to make a multiple comparison adjustment?