The arrival of any piece of unsolicited and unwanted email, commonly referred to as spam, is a source of annoyance to many email users. It results in real costs to individuals and organisations. Spam also contributes to a reluctance to use email by some individuals. Currently most spam prevention techniques rely on methods that examine the whole email message at the mail server. This thesis details research that aims to deny spam entry into the internal network, stopping it ever reaching the mail server. A system is described that can identify current potential spammer IP addresses in real-time and then inform all network gateways to block emails from those addresses. Various tests of the system's timeliness and efficiency are then illustrated, leading to a final conclusion of the system's viability and overall usefulness. This is followed by a discussion of various areas in which future research could be carried out.