In situ is commonly used to determine where in a tissue of interest a particular mRNA is expressed. The older, more common techniques utilize 35S, 32P, or 33P labeled oligos or riboprobes - these are the isotopic methods. They usually work quite well in my experience. There are also kits available that utilize HRP/chemiluminescense instead of radioactivity. They usually invovle labeling a nucleic acid probe with dig, hybridizing it to the tissue, and then using a HRP conjugated anti-dig antibody to visualize the hybridized probe. Some of these methods work pretty well but in my experience they do not work as well as the older methods. That being said there are many labs who generate beautiful in situs using the non-isotopic methods - C. Thisse comes to mind. The amount of radioactivity you use in the older protocols is nominal (especially compared to a nuclear accident) and does not present a risk to the experimenter. As the OP said - you just have to be more careful with the isotopes.