I have a strong feeling that someone out there who hates me submitted my email address to the king of spammers. At any single day, I would receive an average of 30 unsolicited messages from various spammers (offering me stocks, viagra, penis enlargers and vibrating cock rings, although I actually inquired about he last one -- kidding!). It's gotten annoying, to say the least. The bad part is that Apple Mail.app's built-in junk filter doesn't catch them, thanks in part to the fact that spammers have been using image attachments to deliver their messages. I tried creating my own rule, but couldn't come up with one that could effectively segregate the illegitimate messages from my Inbox. A couple of months ago, I came across this entry in Hawk Wings, which suggested a rule to catch image spam. It's fairly simple. As the page states "The only tricky thing here is selecting the “Edit Header List…” from the list of conditions and then entering “Content-Type” in the next window. “Content-Type” will now appear in the list of conditions. You will need to select it and enter as its content “multipart/related”." That's the one condition I didn't figure out, and it makes this rule really work. Here's how my rule looks: For the first few weeks though, there were many instances of false-positives, so I tweaked it a bit. I added "set color" to distinguish Junk that was filtered by Mail automatically, from the ones moved to the junk folder by this rule. Also, I added the "Any attachment name ends with .gif" (not a lot of people send .gifs anyway). If and when you decide to try this out, you'll need to regularly visit the Junk folder to check for any legit email that may have been filtered. A simple drag and drop of legit messages back into the Inbox will train Mail to let messages from that person through the next time. It's like the reverse of junk mail training. For the past month, I noticed that I rarely get false positives. IOW, it's been working well. Now I visit the junk folder just once a week. Try it out. I hope it helps you (that is, if you have an image spam problem).