EVIDENCE is accumulating that in some mammalian species there may be a direct effect of short chain fatty acids on the secretion of insulin from the pancreas. Thus it has been shown that octanoate causes a marked release of insulin from slices of rat pancreas1. Furthermore, butyrate was found to evoke a rapid and sustained elevation of serum insulin when infused into adult sheep2. These results contrast with the lack of effect on insulin secretion when longer chain fatty acids, such as palmitate, are incubated with pancreas slices3.