Insulin secretion was studied using adult rat intact islets, dissociated isolated islet cells, and isolated islet cells that had been allowed to reaggregate. The three preparations were maintained for 2 h in tisue culture medium at 37 C in order to allow for equilibration after the isolation procedures and to permit restoration of contacts between the reaggregated cells. The isolated cells appeared well preserved at the ultrastructural level. Evidence for intimate contacts between the reaggregated cells was demonstrated by the reappearance of gap junctions between adjacent cells. Basal insulin release (2.8 mM glucose) during 30 min was elevated in the isolated cell suspension but was restored to the level found in intact islets when isolated cells from the same population were reaggregated. The elevated basal release did not appear to be due to cell damage since there was a commensurate increase in the rate of insulin biosynthesis relative to intact islets. Although there was a marked stimulation of insulin release from the intact islets at 16.7 mM glucose, the response was smaller in the reaggregated cells and in the isolated cell suspension. All three preparations responded to elevated cAMP levels evoked by glucagon in the presence of 16.7 mM glucose. Similar results were obtained when insulin release was studied in various preparations derived from newborn rat pancreatic endocrine cells. Thus, basal insulin release was elevated in isolated cells, whereas cultures with cell contacts displayed lower basal insulin release. Taken together, the restoration of lower basal insulin release and the parallel appearance of contacts between the reaggregated cells suggest that these two phenomena are interrelated. Thus, cell contacts may be important in regulating basal insulin release. Whether such regulation is a consequence merely of cell association or of direct communication between cells remains to be established.