Two remarks on amalgams of locally finite groups