Insulin was named before it was found. Two scientists proposed the name before anyone had isolated the hormone, and a third chose it again without knowing.
The story starts with a doctoral thesis. In 1869 Paul Langerhans wrote his on the microscopic anatomy of the pancreas. Among the gland's ordinary cells he noticed small, separate clusters. He did not suggest what they were for.
In 1893 Édouard Laguesse argued that these clusters were endocrine, releasing something into the blood. He proposed calling them the islets of Langerhans, in honor of the man who first described them. The image was apt. Under the microscope they look like little islands in the surrounding tissue.
By the 1900s, researchers believed the pancreas had an internal secretion, a substance released into the blood. In 1909 the Belgian researcher Jean de Meyer apparently became the first to suggest a name for it: insuline, from Latin insula, "island." In 1916 E. A. Schafer, in Britain, independently suggested the same name. English recorded insuline by 1914.
Neither man had the substance itself. That came in Toronto. Banting began research with Best on 17 May 1921. Leonard Thompson, a severely diabetic 14-year-old, got a first injection on 11 January 1922. A new series began on 23 January, and he responded immediately. Their colleague Macleod suggested the Latin root for "islands," without knowing of the earlier proposals. He presented a full summary of the Toronto work in Washington on 3 May 1922.
The Latin insula has a wide family in English. It gave isle, through Old French. It gave insular, "of or pertaining to an island." Through Italian isolato, "made into an island," it gave isolated. In ancient Rome an insula was also a block of buildings.
So one of medicine's best-known hormones is named for a geographic image: a scatter of islands in the pancreas.