In case you're interested, I could tell you a little bit about the theory behind the feat.
At the time I wrote it, I was interested in spell point systems. I was especially fond of a system where a cantrip was worth 1 point, a first level spell was worth 2, a second level spell worth 3, and so on. Using this weighting, I found that the following point assignments gave a very good agreement with the wizard's spells per day charts:
1st level: 2
2nd level: 2+2=4
3rd level: 4+3=7
4th level: 7+4=11
and so on. Basically, you start with 1 spell point (as an apprentice, presumably), and every time you went up a level, you got one spell point per level. So when you got to first level, you got 1 point. When you gained 2nd level, you got 2 points, and so on.
A first level spellcaster could cast 1 first level spell. At second level he could cast 2. A third level spellcaster could cast 2 first level spells and 1 second level spell, and so on. It worked pretty well, I thought.
In such a system, a third level spell (4 points) and a second level spell (3 points) together were the equivalent of a sixth level spell (7 points). So I wrote a feat saying that two spells could fit in one slot, as long as the combined level was less than the size of the slot.
Similarly, a sixth level spell (7 points) is worth less than two third level spells (4 points a piece), so I added a line to the feat that allowed spells to be divided among spell slots as well as combined.
In other words, the spell exchange feat is an attempt to smuggle a spell point system into the spells/day chart.
I now feel that this much flexibility is more than what should be granted by a feat, and so I do not encourage the use of Spell Exchange in a standard D&D campaign.
