The improvements in imaging technology and multi-electrode recordings now make it possible to study spatio-temporal activity in biological neural networks. I will consider a number of different models for such activity with an emphasis on slice preparations. I will show different modes of propagation of activity depending on different mechanisms and how small changes in parameters of the coupling can change the velocity of activity waves by several orders of magnitude. I derive scaling laws for the velocity of waves as a function of the rise time of synaptic potentials. I will also look at some reduced models that arise through averaging methods and how these can be used to understand partially blocked inhibition in slices.