"In releases prior to JDBC 2.0 every database session requires a new connection and login even if the previous connection and login used the same table and user account. If you are using a JDBC release prior to 2.0 and want to improve performance, you can cache JDBC connections instead."
The JDBC API changed in big ways between version 1 and 2 so they require different caching strategies.
One word of warning, the book you are reading was written over 10 years ago. The JDBC API alone has gone through 2 more major revisions. The Java API has had 4 major revisions.