As you learned in Part I, "TCP/IP Basics," the Network Access layer is necessary to relate the logical IP address, which is configured through the protocol software with the actual permanent physical address of the network adapter. The physical address is burned into the card at the factory. Data frames sent across the LAN must use this physical address to identify the source and destination adapters, but the lengthy physical address (48 bits in the case of ethernet) is so unfriendly that it is impractical for people to use. Also, encoding the physical address at higher protocol levels compromises the flexible modular architecture of TCP/IP, which requires that the upper layers remain independent of physical details. TCP/IP uses the Address Resolution Protocol (ARP) and Reverse Address Resolution Protocol (RARP) to relate IP addresses to the physical addresses of the network adapters on the local network. ARP and RARP provide a link between the logical IP addresses seen by the user and the (effectively invisible) hardware addresses used on the LAN.
You'll learn about ARP and RARP in Hour 4, "The Internet Layer."