No that wasn't my point (that the native code is less trustworthy than the Java runtime.) My point was just that any security measures that restrict applets do not restrict anything that an applet causes to happen via a native method. For example one security restriction is that applets aren't allowed to read files. If an applet calls a native method then that native method can read any files it wants. I'm talking about the model, not about the quality of implementation. I'm not saying it's a bad or untrustworthy thing to do (call native methods), I just thought it was worthwhile to point out that once you call a DLL from an applet, you have effectively chosen to disable the application level SecurityManager. It's your call as to whether this is a problem or not. Marianne