The double keyword is used to declare double precision floating-point variables. Also see the data types page.