I think this part is the key; if the orchids can have a wide sufficiency range, then it becomes what Ray is saying. From the perspective of deficiency, it is the system where one limiting factor determines the growth rate (within the limitation of the plant physiology). I know that you all know this, but here is the quote from the growing media book (p. 147):
"... As far as nutrients are concerned, each nutrient is a separate necessity. One cannot substitute for another. Plant growth is limited by that nutrient in lowest supply... Increasing the supply of that nutrient will increase growth, as in Fig 14.3, until it is again limited by the nutrient in second lowest supply, ...."
So as long as there are sufficient elements, then photosynthesis (C supply) becomes the limitation. So the actual ratio doesn't matter as long as there is enough of each (Ray's view). But too much of some elements can cause the toxicity (via direct physiological problem or indirect effect of interference among minerals). If the sufficiency range is relatively narrow as Mike thinks, then the ratio starts to become more important.
I've been thinking more closer to Ray's perspective; epiphytes (slow growers) are quite flexible in sufficiency range, so as long as there is enough of each, they should be ok (but we all know that high EC can cause the problems with water uptake and root dehydration). But it's just my gut feeling.
Anyway, thanks for the gypsum topic, Mike. I started to read the chapter of the grow media book, and I agree with Mike that it is something interesting to experiment with. I'll wait to see what you will find out! 
Also Mike's recommendation of the grow media book was great! Here is the link to the book. The amazon price in the US has come down recently (but it is starting to go up again), and it covers a wide range of topics. It is quite a bit easier to read, and more practical than Marrchner's Mineral Nutrition of Higher Plants (which goes into scientific literature at a deeper level).
Lance's link to N/S ratio of Phal bellina is pretty interesting. Thanks for finding this, Lance! The abstract seems to be an overstatement (1:15 is the best). Statistically, it seems that they should have said that as long as S is lower than 1:15, the plants grew well. The grow media book mentions that S requirement vary widely among among taxonomic groups (p.197), so it is a good info!
