secondary stability revisited

The other day, I finally figured out what people mean when they talk of secondary stability.
A little bit of background: Go the the internet, that repository of all things worth knowing and you will find an absolute lack of definition for the concept of secondary stability. The truth is that everybody kind of knows what it means but nobody can define it.
Let's back up some more. There is another term used frequently by kayakers and that is primary stability. This term actually has a definition. Primary stability is the slope of a curve at 0 degrees of heel that plots resistance to heeling vs. degrees of heel. In other words, primary stability of a boatis how fast its resistance to heeling increases as you heel it over. The resistance to heeling that the boat offers is also referred to as righting moment and is measured in foot pounds. For conventionally shaped boats like a kayak, righting moment increases with degree of heel, maxes out at some angle and then decreases down to zero at some greater angle. That angle where the righting moment becomes 0 is where the boat capsizes unless corrective measures are applied.
Now back to secondary stability. When kayakers use the term it is usually used as follows. The kayaker has just tried out a new kayak, paddled it, made turns, heeled it over and braced and is now on back on dry land an offers his report on the stability of the kayak. "Primary stability is poor but secondary stability is good." What this statement really says is that although the boat has low primary stability, resistance to heeling increases to some tangible level once the boat is heeled over far enough.
If I had to specify secondary stability in terms that could actually be measured, I would probably do it as a pair of numbers, (maximum righting moment, degree of heel at maximum righting moment.) And there you have it, probably the first non-evasive definition on the internet of secondary stability.