An accurate evaluation of antiepileptic medication is never easy. Gowers was fully aware of this in 1881: "There is no point in therapeutics... more open to fallacy, or on which more generalizations have been published which subsequent observation has proved to be inaccurate" (in proof of which was his own half-hearted endorsement of anticonvulsant properties in digitalis and zinc).1 In the continuing search for better drugs for seizures the troublesome question can still be raised: Has any satisfactory system for drug appraisal yet been devised?
Preliminary information regarding anticonvulsant activity of a new compound is obtained at first hand from laboratory experiments. These can show that the drug is not obviously damaging to vital organs and functions of the test animal and that it has a sufficiently prolonged period of action. A rough assessment is made by observing its effects upon the threshold or intensity of seizures induced