I've had a couple of customers ask about this topic. What follows is by no means a complete discussion on the subject. It's a brief outline of what tagging is about.
Before tagging was used, cancelling the stamps on letters was a very manually intensive process. A postal clerk had to face the letter the right way so that it was fed through the cancelling machine and the stamps were properly cancelled. As mail volume increased, this required a huge amount of labor for the USPOD (as it was known back then). To reduce the labor costs, the solution was to automate the process with a machine.
Around 1954, the USPOD began experimenting with fluorescent compounds. These compounds were mostly phosphor based. These compounds "glow" when exposed to ultraviolet light (UV rays).
By applying fluorescent materials to stamps and developing special machines to detect the glow of the fluorescent material, a machine could "see" where the stamps were at and face the letter so that it was fed into the cancelling machine in the right side and direction.
In 1963, Scott #C64a, the 8¢ Airmail issue, was produced with tagging and was the first stamp the USPOD used for experimenting with the new canceling machines. In 1963, the 5¢ City Mail Delivery issue (Scott #1238) was the first commemorative stamp produced with tagging applied to all stamps produced. Since then, hundreds of US stamps have appeared with tagging.
An ultra-violet (UV) light is essential to determining the type of taggant. You can buy a UV light that has either a short-wave filter or a long-wave filter. I recommend buying a lamp that has both filters in a single unit. The short-wave detects the type of tagging and long-wave detects paper types used for printing a stamp.
Tagging is a clear compound. Think of it as an "invisible" ink to the naked eye. Over the years, the USPS has experimented with several different ways of applying the tagging compound. First, the stamp is printed and then the tagging is printed on top of the stamp design. Second, the tagging compound is mixed in with the paper when the paper is manufactured. And third, the tagging compound is mixed in with the stamp printers' ink.
Overall tagging comes in several different forms. There is continuous tagging where the tagging compound was applied from edge to edge of the sheet of stamps. There is block tagging where there are untagged gaps between the tagged areas. Why block tagging? Well, the tagging compound is abrasive to the tiny perforating pins, causing them to wear out faster. The USPS switched to block tagging so that most of the face of the stamp was covered with tagging. But the area between the stamps where the perforations would fall was left untagged so that the perforation pins didn't wear out so quickly.
Suffice to say, there are several different ways that tagging was applied to stamps as the USPS experimented with different compounds, applications, etc to find a method that worked best. I'm not going to cover all of the types in detail. That's an exercise left to the reader!
Some stamp issues come in both untagged and tagged versions. In other words, the stamp was originally printed in an untagged form. But in later printings when tagging was available, the later printings are tagged. These issues are not errors, just normal varieties of the same stamp issue. Scott #C64 is one such stamp. Scott #1209 and #1213 are two other regular issue stamps with both tagged and untagged copies.
Because tagging is invisible to the naked eye, it's very difficult to inspect stamps for production problems. Some stamps are known with tagging omitted. Untagged errors are usually when the tagging compound reservoir runs empty and the printing press operator doesn't notice it. These are errors because that stamp was intended to be only released as a tagged issue. Scott #1238a is one such tagging omitted error. Tagging errors are listed in the Scott catalog. Some tagging omitted errors are common and some are very rare. There are also a wide variety of tagging freaks and oddities. For example, if the reservoir was running low on tagging compound, the tagging on the stamp may be spotty or uneven. Or for block tagging, the sheet of stamps may have become shifted so that the untagged areas cross the face of the stamp design instead of being between the stamp designs. Tagging varieties is a fertile field for research and there are discoveries just waiting to be made for someone with the time and patience to search through a lot of stamps looking for varieties.
Tagging glows reddish or bluish-green when exposed to short wave UV light. Reddish tagging was only used for air mail stamps from 1963 to 1978. Beginning with the 1978 31¢ Wright Brothers (Scott C91-C92), bluish-green tagging was used for all subsequent air mail stamps.
If you want to learn more about tagging, I strongly recommend the "Handbook on United States Luminescent Stamps" by Alfred "Tag" Boerger and John Stark. Originally published in 1971, this book is still the authority on the subject. This title should be readily available at any of the philatelic literature dealers for about $12 per copy. If you're interested in tagged stamps, this book is worth its weight in gold.