The absorption and disposition of rectally administered secobarbital was studied in ten epileptic children, ages 2-13 yrs. Five subjects received secobarbital rectally in solution, and the other five received secobarbital suppositories. Concentration of secobarbital in serum was serially determined during 48 hrs after a single rectal dose of about 5 mg/kg. The rate of absorption of secobarbital, as measured by the time to reach peak serum concentration, was much more rapid from the solution than the suppository (0.92 +/- 0.47 hr vs 4.60 +/ 2.30 hr). The peak serum concentration of secobarbital in the solution group was consistently higher than in the suppository group (2.26 +/- 0.37 micrograms/ml vs 1.35 +/- 0.24 microgram/ml). None of the individual peak serum concentrations exceeded 3 micrograms/ml, which is well below the previously reported minimum toxic concentration of secobarbital (ie, 6 microgram/ml). The elimination half-life of secobarbital varied over a wide range, from 2.7 to 13.5 hr, and is, on the average, shorter than estimates previously reported for adult volunteers or poly-drug abusers. Also, the mean elimination half-life did not differ between the solution and the suppository groups. The extent of rectal absorption of secobarbital, as assessed by the area under the serum concentration time curve, was not significantly different between the solution and the suppository treatments. If rectal secobarbital is considered for treatment of prolonged seizure, a rectal solution may offer a more rapid and consistent onset of action than with the suppository preparation.