Matching Klein's Documents, Waldman Exhibits 4 vs 3

Inspired by David Josephs' paper "The Evidence IS the Conspiracy: The Klein’s Rifle, Seaport Pistol and Backyard Photos", Part 1: The Klein’s Rifle, where he points out the weird grouping of serial numbers in Klein's copy of their receiving document, listing the infamous C 2766 Carcano rifle, I though it would be interesting to match up all the serial numbers in the Waldman 4 exhibit to the serial numbers on the 10 slips supposed to originate from 10 cartons in the particular shipment containing C 2766. The exhibits in question are:

Waldman Exhibit 3, first page, www.history-matters.com/archive/jfk/wc/wcvols/wh21/html/WH_Vol21_0359b.htm

Waldman Exhibit 4, first page, www.history-matters.com/archive/jfk/wc/wcvols/wh21/html/WH_Vol21_0362a.htm

A thread discussing the Klein's documents and testimonies can be found here and this post shows how the serial numbers within a slip (~carton) jumps all over the place. Based on an FBI report quoting Mitchell J. Scibor, general operation manager of Klein's, the rifles were handled and tagged one at a time as they were picked from the carton. This would lead one to assume the serial numbers would appear in groups of 10 according to the slips.

To map out the grouping across all serial numbers I basically set up two database tables, W4 and W3, containing data from Waldman Exch. 4 and Waldman Exch. 3. Joining the tables on serial number using W4 as the driving table gave the sequence of cartons picked, which brings the authenticity of Waldman 4, or Waldman 3(!) in to question. The raw data can be found in these links to back pages (stay open for viewing):

After correcting typos and other mishaps I was still left with a dozen or so W4 serial numbers that didn't get a match. As can be seen from my slip notes a good number of serial numbers do not agree! Some of the W3 numbers are almost illegible and I accepted the W4 version but some are actually different. In some cases likely a misread stamping on the rifle, like 'B' and 'R' being confused, or '7' and '9'. What we do know is that either of them is wrong!

If all this stuff is real, at least 5% of the serial numbers are wrong, so much for gun control...LOL

As for the grouping, it can be seen that carton '2' seems to have been opened first but then the process descends into chaos with something like five boxes being picked randomly although towards the end the '4' had "survived" and is finally being emptied. 

Still a mystery to me what's going on since, as the argument would usually go, if you want to fake it why not do it properly and go by the slips?