FLOAT

The FLOAT keyword can be used in several statements to designate a variable or value as a 32-bit floating point number. See Data Types and DIM for more information.