Description of the illustration number_datatypes.eps { NUMBER [ (precision [, scale ]) ] | FLOAT [ (precision) ] | BINARY_FLOAT | BINARY_DOUBLE }