For parenteral products administered in large volumes, such as saline infusions, evenminute volumes of endotoxin are unacceptable. Historically, large volume parenteral manufacturers have been foremost in developing tests for bacterial endotoxin assays due to the critical need for very sensitive detection of endotoxin in their products. However, many of today’s problems in developing suitable endotoxin tests are not issues of sensitivity, as dosesa are often small, but rather revolve around the recovery of control standard endotoxin spike, the difficulty of which is exacerbated by the chemical nature of the drug materialsa. Smallvolume parenteral drugs often contain high drug concentrations which interfere both with the physiology of rabbits in the rabbit pyrogen assay and with spike recoveries in the Limulus amebocyte lysate (LAL) assay (1). Some common types of problem compounds encountered in developing endotoxin assays for smallvolume parenterals include water-insoluble drugs, drugs containing activity that mimics that of endotoxin, drugs containing endotoxin (that it may be desirable to remove prior to method validation), bulk drugs with variable potencies, multiple drugs in a given container, and potent, highly interfering drugs such as chemotherapy drugs. Now that the science of LAL testing has been firmly established, the challenges that remain often reside in difficult, product specific applications. Perhaps the last great challenge encountered in each parenteral analytical laboratory is the development of, not just an LAL test, but also a rugged, reproducible, and perhaps automatable test that will stand the test of time in routine use.