Nothing. A Java or C++ float follows the IEEE single-precision floating points standard. A Java or C++ double follows the IEEE double-precision floating-point standard.
(However, our textbook also consider smaller-bit-length floating-point numbers since they are easier to work with.)