just a couple more comments... a little late.
Here's some food for thought.
Think of the plastics industry history (going back into mid 1800s) and the volumes and varieties of composition used in accessory items: thermoset, bakelite, casein, lucite, celluloid, and on and on. A few degrees of this and a little more of that, and you basically have a new plastic type.
The reason I bring this point up is that, personally, I hate to see items hot pin tested for the sake of trying to identify a plastic. It damages the item. Period. I know there are different schools of thought about it, but as a jewelry seller I totally disagree with defacing and item to fine-pick a composition. I cannot count the number of times I've seen a botched not pin test that put a permanent ugly little scar on something.
There are clues and tests to easily tell bakelite. That is very important BECAUSE of the collector market.
Some of the "lesser identifiable" plastics are easier to date from era, style, findings, surface characteristics etc. If you have a quality 1930s necklace (like your wonderful beads that started this thread) - what is most important in determining value: item? --age? style? jewelry appeal? composition? condition?
Not that we shouldn't persue identity questions, but please take it into perspective that there are always a sea of attributes that make the item special and valuable. In the world of collectible plastic jewelry, composition is not always the key.
Like I said... just 2cents worth.