http://www.cknow.com/vtutor/vthistory.htm
Narrative histories by Dr. Alan Solomon and Robert M. Slade are available. Below is an expanded summary.
1981 - The First Virus In The Wild
As described in Robert Slade's history, the first virus in the wild actually predated the experimental work that defined current-day viruses. It was spread on Apple II floppy disks (which contained the operating system) and reputed to have spread from Texas A&M. [Side note: Thanks to a pointer from anti-virus pioneer Fridrik Skulason we know the virus was named Elk Cloner and displayed a little rhyme on the screen:
It will get on all your disks
It will infiltrate your chips
Yes it's Cloner!
It will stick to you like glue
It will modify ram too
Send in the Cloner!
For more info on Elk Cloner see http://www.skrenta.com/cloner/
1983 - The First Documented Experimental Virus
Fred Cohen's seminal paper Computer Viruses - Theory and Experiments from 1984 defines a computer virus and describes the experiments he and others performed to prove that the concept of a computer virus was viable. From the paper...
On November 3, 1983, the first virus was conceived of as an experiment to be presented at a weekly seminar on computer security. The concept was first introduced in this seminar by the author, and the name 'virus' was thought of by Len Adleman. After 8 hours of expert work on a heavily loaded VAX 11/750 system running Unix, the first virus was completed and ready for demonstration. Within a week, permission was obtained to perform experiments, and 5 experiments were performed. On November 10, the virus was demonstrated to the security seminar.
1986 - Brain, PC-Write Trojan, & Virdem
The common story is that two brothers from Pakistan analyzed the boot sector of a floppy disk and developed a method of infecting it with a virus dubbed "Brain" (the origin is generally accepted but not absolutely). Because it spread widely on the popular MS-DOS PC system this is typically called the first computer virus; even though it was predated by Cohen's experiments and the Apple II virus. That same year the first PC-based Trojan was released in the form of the popular shareware program PC-Write. Some reports say Virdem was also found this year; it is often called the first file virus.
1987 - File Infectors, Lehigh, & Christmas Worm
The first file viruses started to appear. Most concentrated on COM files; COMMAND.COM in particular. The first of these to infect COMMAND.COM is typically reported to be the Lehigh virus. At this time other work was done to create the first EXE infector: Suriv-02 (Suriv = Virus backward). (This virus evolved into the Jerusalem virus.) A fast-spreading (500,000 replications per hour) worm hit IBM mainframes during this year: the IBM Christmas Worm.
1988 - MacMag, Scores, & Internet Worm
MacMag, a Hypercard stack virus on the Macintosh is generally considered the first Macintosh virus and the Scores virus was the source of the first major Macintosh outbreak. The Internet Worm (Robert Morris' creation) causes the first Internet crisis and shut down many computers. CERT is created to respond to such attacks.
1989 - AIDS Trojan
This Trojan is famous for holding data hostage. The Trojan was sent out under the guise of an AIDS information program. When run it encrypted the user's hard drive and demanded payment for the decryption key.
1990 - VX BBS & Little Black Book (AT&T Attack)
The first virus exchange (VX) BBS went online in Bulgaria. Here virus authors could trade code and exchange ideas. Also, in 1990, Mark Ludwig's book on virus writing (The Little Black Book of Computer Viruses) was published. While there is no proof, hackers are suspected of taking down the AT&T long-distance switching system.
1991 - Tequila
Tequila was the first polymorphic virus; it came out of Switzerland and changed itself in an attempt to avoid detection.
1992 - Michelangelo, DAME, & VCL
Michelangelo was the first media darling. A wordwide alert went out with claims of massive damage predicted. Actually, little happened. The same year the Dark Avenger Mutation Engine (DAME) became the first toolkit that could be used to turn any virus into a polymorphic virus. Also that year the Virus Creation Laboratory (VCL) became the first actual virus creation kit. It had pull-down menus and selectable payloads.
1995 - Year of the Hacker
Hackers attacked Griffith Air Force Base, the Korean Atomic Research Institute, NASA, Goddard Space Flight Center, and the Jet Propulsion Laboratory. GE, IBM, Pipeline and other companies were all hit by the "Internet Liberation Front" on Thanksgiving.
1995 - Concept
The first macro virus to attack Word, Concept, is developed.
1996 - Boza, Laroux, & Staog
Boza is the first virus designed specifically for Windows 95 files. Laroux is the first Excel macro virus. And, Staog is the first Linux virus (written by the same group that wrote Boza).
1998 - Strange Brew & Back Orifice
Strange Brew is the first Java virus. Back Orifice is the first Trojan designed to be a remote administration tool that allows others to take over a remote computer via the Internet. Access macro viruses start to appear.
1999 - Melissa, Corner, Tristate, & Bubbleboy
Melissa is the first combination Word macro virus and worm to use the Outlook and Outlook Express address book to send itself to others via E-mail. It arrived in March. Corner is the first virus to infect MS Project files. Tristate is the first multi-program macro virus; it infects Word, Excel, and PowerPoint files. Bubbleboy is the first worm that would activate when a user simply opened and E-mail message in Microsoft Outlook (or previewed the message in Outlook Express). No attachment necessary. Bubbleboy was the proof of concept; Kak spread widely using this technique.
2000 - DDoS, Love Letter, Timofonica, Liberty (Palm), Streams, & Pirus
The first major distributed denial of service attacks shut down major sites such as Yahoo!, Amazon.com, and others. In May the Love Letter worm became the fastest-spreading worm (to that time); shutting down E-mail systems around the world. June 2000 saw the first attack against a telephone system. The Visual Basic Script worm Timofonica tries to send messages to Internet-enabled phones in the Spanish telephone network (later in 2000 another Trojan attacked the Japanese emergency phone system). August 2000 saw the first Trojan developed for the Palm PDA. Called Liberty and developed by Aaron Ardiri the co-developer of the Palm Game Boy emulator Liberty, the Trojan was developed as an uninstall program and was distributed to a few people to help foil those who would steal the actual software. When it was accidentally released to the wider public Ardiri helped contain its spread. Streams became the first proof of concept NTFS Alternate Data Stream (ADS) virus in early September. As a proof of concept, Streams has not circulated in the wild (as of this writing) but as in all such cases a circulating virus based on the model is expected. Pirus is another proof of concept for malware written in the PHP scripting language. It attempts to add itself to HTML or PHP files. Pirus was discovered 9 Nov 2000.
2001 - Gnuman, Winux Windows/Linux Virus, LogoLogic-A Worm, AplS/Simpsons Worm, PeachyPDF-A, Nimda
Gnuman (Mandragore) showed up the end of February. This worm cloaked itself from the Gnutella file-sharing system (the first to specifically attack a peer-to-peer communications system) and pretended to be an MP3 file to download. In March a proof of concept virus designed to infect both Windows and Linux (and cross between them) was released. Winux (or Lindose depending on who you talk to) is buggy and reported to have come from the Czech Republic. On 9 April a proof of concept Logo Worm was released for the SuperLogo language by Logotron. LogoLogic-A spreads via MIRC chat and E-mail. May saw the first AppleScript worm. It uses Outlook Express or Entourage on the Macintosh to spread via E-mail to address book entries. Early August, the PeachyPDF-A worm became the first to spread using Adobe's PDF software. Only the full version, not the free PDF reader, was capable of spreading the worm so it did not go far. September, the Nimda worm demonstrated significant flexibility in its ability to spread and used several firsts. While not new in concept, a couple of worms created a fair amount of havoc during the year: Sircam (July), CodeRed (July & August), and BadTrans (November & December).
2002 - LFM-926, Donut, Sharp-A, SQLSpider, Benjamin, Perrun, Scalper
Early in January LFM-926 showed up as the first virus to infect Shockwave Flash (.SWF) files. It was named for the message it displays while it's infecting: "Loading.Flash.Movie...". It drops a Debug script that produces a .COM file which infects other .SWF files. Also in early January Donut showed up as the first worm directed at .NET services. In March, the first native .NET worm written in C#, Sharp-A was announced. Sharp-A was also unique in that it was one of the few malware programs reportedly written by a woman. Late May the Javascript worm SQLSpider was released. It was unique in that it attacked installations running Microsoft SQL Server (and programs that use SQL Server technology). Also in late May the Benjamin appeared. Benjamin is unique in that it uses the KaZaa peer-to-peer network to spread. Mid-June the press went wild over the proof-of-concept Perrun virus because a portion of the virus attached itself to JPEG image files. Despite the hype, JPEG files are still safe as you must have a stripper program running on your system in order to strip the virus file off the image file. On 28 June the Scalper worm was discovered attacking FreeBSD/Apache Web servers. The worm is designed to set up a flood net (stable of zombies which could be used to overwhelm one or more systems).
Your choice from here; read some detailed histories of the early years or move on to Virus Protection techniques.