Comparing means problem


I have run an experiment where I am comparing the temperature required to cause a febrile seizure in several different animal models of epilepsy. The problem that I am having is that the study design limited me to a fever temp of 41.5 - 42.5 - which it turned out was not sufficient to produce seizure in my control animals (well only one).

This leaves me with a problems of statistical comparison because I need to see if there is a significant difference between the controls (which don't have a seizure temperature) and the 3 epilepsy genotypes.

Any help would be much apreciated!