A message like this will usualy contain an html portion with an image in it that is a single pixel in size, that is white-on-white. It doesn't show up when you look at it, but it sends a request to the sender's specified website to get the pixel, thus showing them which email accounts are active.

What would happen if an ISP's mail server were to collect the URLs in emails and then retrieve the URL. Wouldn't this create a lot of false positives for the spammer thus screwing with their business model? And what if some scumsucking bottom feeder were to create a whole bunch of fake email addresses pointing at the servers of ISPs who do this. And then this evil waste-of-space no-buts were to sell that list to spammers for a good price based on the fact that these were all good active email addresses? Criminals victimizing other criminals... oh what a surprise! And what if ISPs would get together and share the IP addresses of these 1-pixel web servers via some type of real-time feed so that they could engage in cooperative blocking of validation traffic? Would they be helping the spamming criminals or would they be helping the scamming criminals or would they be good guys?