How do I display my germination rates? Is it count data? Not sure! I have 4 treatments, each with 5 Petri dishes (so total 20), and each Petri dish has 25 seeds in it. I am looking to discuss how many seeds have germinated. Should I get the mean germination of each treatment and then discuss it as a percentage of 25 seeds? So if in the root treatment, 4,4,6,6,0 germinate, so my mean is an average of 5 seeds out of 25, how do I best display/discuss this?

> ARpetri<-read.table(file.choose(),header=TRUE)
> attach(ARpetri)
> names(ARpetri)
[1] "Distilled" "Marigold"  "Mixed"     "Root"     
> ARpetri
  Distilled Marigold Mixed Root
1         2        1     1    5
2         2        1     0    3
3         3        0     0    3
4         3        1     0    4
5         0        1     1    1


the counts on each row do not belong together, just the columns and the first column is excel generated and not my own. I hope that helps clarify. Many thanks for all the advice and I will definitely check out the R mailing list as suggested. Any recommendations for where I can look up how to create the plot pictured above? I think it would be useful to re-create with my other counts and "eyeball" the data. Thanks all!
 A: Your outcome is discrete insofar as the response can only assume values 0/25 (1/25) 25/25 or 0(0.04)1 or 0(4)100%, but in principle your rate is a probability. Had you used 250, 2500, ... seeds your response would have been more and more finely measured and it's fruitful (pun intended) to think that you're modelling a quantity on a continuous scale. 
Tukey called numbers like 4/25 counted fractions, not that anyone really needs a special name. 
GLM software that I know of will handle this properly as a model with binomial link and total 25. Display is not a problem as you have 20 counts which can be shown directly on a single graph with both count and probability or percent scales. 
If you post the data, example analyses are likely to follow. 
EDIT Some small questions remain about the data, but on the assumption that the row identifiers just arbitrarily label replicates, and have no other meaning, then this kind of plot shows all the information in the data. The horizontal lines show means. 
Note that my software sorts Distilled Marigold Mixed Root alphabetically but it could be that there is a more meaningful order. I am not a plant scientist and in any case I guess that it needs the OP to explain here. 

