Here is the first order. I see nothing special. I placed the ones with the most detail on the bottom row. The 1921 on the top row is not a D. I also notice the best obverses have less desirable reverses.
Yes, a 4% weight loss isn't surprising. And a 1921-P is nothing to sneeze at, even if it's a details coin due to reverse scratches. Definitely worth a positive feedback, IMHO. Don't forget to check any 1942-D dimes for the 1942/1-D overdate.
4% loss seems excessive, since .715/.7234 is only about 1.2%. 90% silver US coins contain .7234 ounces of actual silver weight per $1 face value. They are treated as .715 ounces of actual silver weight per $1 face value due to expected wear. This makes me wonder why your Mercuries have more than triple the expected weight loss.
This was my formula 2.5g(mint weight) x 50 = 125 - 120.4(actual weight)= 4.6 g. loss... 4.6g x .0321g(troy oz conversion)=0.14766(total loss in troy)x.0723(content in troy)=0.010676(silver loss in troy)-3.617(expected troy oz per 50 mercs.)=3.606 Troy oz. of silver
I've seen worse. Heck, I've got worse in an album or two. It's one of my limits as a collector: I'll never pay up for a nice problem-free example if I can get the same date with problems near melt. I'm not proud about it, but I'm also not in a rush to change.
1929D, ODV-010...the date extends further past the bust than 1928D (ODV-009), and it's a lazy 9. Refer to Mr. Wiles' ODV listings.