Scott 2170 Odd Printing Inking Process Flaw

In a recent auction box lot purchase, there was a glassine envelope with ~80 Scott 2170 (Paul Dudley White MD, 3c denomination). Most stamps were singles, but there was one strip of 5 stamps. I wish the previous owner didn't separate these from the sheet, but there's no going back...

This was from an online auction lot, in which most of the stamps were photographed. These were part of a loose stack of glassines and stamps and weren't seen before receiving the lot. In fact, these stamps weren't the same "flavor" as the rest of the lot, and weren't expected. There was also a "bundle" of Scott 220 (2c Small Banknote) including some with the "cap" variety.

My question about the Scott 2170 stamps shown is whether this is an underinking printing process oddity. At first glance, the glassine envelope of stamps appeared to be faded stamps. Upon closer inspection with not just the lighter ink but also the vertical streaking, I thought this may have been from the printing process and not an effect of later storage and handling of the stamps.

The strip of 5 was more elucidating, showing striking variations among adjacent stamps. At least I was fortunate to find this strip, instead of all individual stamps!

Generally described as ink starvation. Sometimes it could be caused by water/condensation mixed in with the ink, or solvent used to clean the plates contaminating the ink. Seems relatively common on this particular issue, there are a couple like this for sale on eBay right now and here is a strip from my collection.

Thank you ZebraMan and Parcelpostguy for your input on this. I like the phrase "ink starvation" more than the term "underinking"! So, as an EFO, I would think the market value perhaps to be in the $1 to $5 range as an oddity, of course depending upon the demand and scarcity, this not being particularly scarce.

Ink starvation usually means exxactly that, the ink is disappearing because the ink tray has emptied. That results in a web area with a color disappearing somewhere and subseqent areas void of ink. Usually that is noticed, either by a supervisor or by electronics, or both. This web segment usually is destroyed but with this issue, a good portion of the web seems to have made it into the public.

Here we have a partial area of disappearing ink. However, the ink reappears "on the other side", so it cannot be ink starvation.
This might be a local effect of solvent smearing on the plate, or some kind of cylinder pressure problem.

This issue is known with many other problems like ink smears and misperforations in all directions so the printer must have had its problems.

DK - disagree

"Here we have a partial area of disappearing ink. However, the ink reappears "on the other side", so it cannot be ink starvation."
Depends on the direction of web travel. If it is moving in the direction of top to bottom (or bottom to top) and not left to right (or right to left), then it is not reappearing.
"This might be a local effect of solvent smearing on the plate"

Orr - yes, not partcularly scarce on this issue (I have it too).
If that were the case, the ink would be snaered rather than weak.

Pretty sure the web ran vertically (A press?). So this "almost missing ink" variety occured with ink on the left and right apparently present for quite some length of web.
Jay Stotts currently has a series of articles running in the US Specialist so he likely knows all the answers and we might get them in a future article.