Robert

What question are you asking? Are you making comparisons between groups? I assume you are simply comparing the groups correct? If that is the case you can't use all of your measurements as if they are independent measurements because within an individual they are pseudo-replications i.e. each measurement is correlated with each other measurement. What you have to do is average the values and then make your comparison between groups. So you will have ONE VCAM intensity measurement per mouse (the average of the 10 sections per individual). With a Student's t-test 10 observations is still a reasonable n size. If you were interested in say the precision of your measurements or something like that I suppose you could look at coefficient of variation of the data ASSUMING your variable is continuous and not a count or percentage or something like that. You might have other reasons for looking at the variability in the VCAM intensity if the sections were taken along some specific points of whatever organ you were taking the samples from and you want to make say a sort of average "profile" of that organ. If that is the case then you need to detail how that was done so your statistical tests can reflect the experimental design. Two things: let me know if this makes sense AND DON'T take my word for it. I've told you what I think. Get some advice from someone in your stats department or a biostatistician. I've done work like this and have some experience but always get a second opinion.

You are comparing two groups of mice. The two group means and variances are to be used. The mean and variance of the group A mice and the mean and variance of group B mice will be calculated and used in the computation.

If aortic sinus sections are to be compared for both groups it means comparing 100 group A values and 100 group B values. The mean and variance of the group A 100 values and those of group B 100 values will be calculated and used in the t or z test.