The purpose of this investigation was to develop a method to quantitate insulin absorption, and to compare insulin absorption from various noninjection sites of administration. Log dose/effect curves were established for i.m. insulin in adult male rats. The effects measured were the maximum change in plasma glucose concentration and the cumulative percentage of change in plasma glucose concentrations from 0 to 4 hr. Both log dose/effect curves gave similar results when calculating the efficacy of other routes, relative to i.m. Nasal, buccal, sublingual and rectal absorption sites were isolated by ligation procedures or with physical barriers. Rectal insulin was more efficacious than nasal, buccal and sublingual insulin, when administered without an absorption-promoting adjuvant. However, the efficacy relative to i.m. insulin was low for each route, probably due to a combination of slow membrane permeation and metabolism at the absorption site. Administration in a solution containing 5% sodium glycocholate, an absorption-promoting adjuvant, increased insulin efficacy by each route. The rank order was nasal greater than rectal greater than buccal greater than sublingual, with nasal and rectal insulin being roughly half as efficacious as i.m. insulin. Orally administered insulin, at doses 5 times higher than administered by other routes, and with Na glycocholate, produced no hypoglycemic response.