Okay I'm going to post the 3 key memos describing the microfilm with the serial numbers.
We should all be able to get to the consensus truth, which is what Flyjack has tried to describe.
First memo, I call "microfilm 11_25_71" pages 1 and 2.
says first microfilm has $250,000. $230,000 of that are twenties
page 2 has 15 start/stop serials, which are 15 bundles of twenties that weren't given to Cooper, out of the $230,000
Each packet represented by the start/stop serials, had 100 bills.
(the microfilm having the bills in the order they were in each packet, so start/stop is sufficient to identify all bills in a packet)
Second memo, I call "new microfilm 11_26_71" pages 1 and 2
new film with 6 lots of bills, $50,000 per lot, $300,000 total
The money not given to Cooper is at the end of lot 5 or 6.
They specify the starting serials for lot 5 and lot 6, for that money.
They say you can take the serials starting where they say in lot 5 and lot 6, and get all the serials to be removed from the $230,000 specified in the first microfilm
This seems redundant to me, as the start/stop serials in the first memo should be sufficient to remove $30k from the original microfilm.
Third memo, I call "bill serial numbers 12_1_71" pages 1 and 2
seems to describe the final process used. The first microfilm which as $230,000 worth of twenties described, is used, and $30,000, identified by the start/stop serials of 15 packets of twenties, are identified again in this memo. on pages 1 and 2.
It's worth looking at these start/stop serials. You can see that the start/stop serials are relatively random with respect to each pair and each other. This makes me believe the full packets were random
2nd and 3rd memos in two subsequent posts