Well, I would like to make a few more comments in this post, relating to the matters recently discussed. I wrote almost all of this post before seeing Chris’s latest post, but I thought I would go ahead and post it anyway, because it sort of fleshes things out a little.
I formerly owned a hand-press, and set a certain amount of type by hand, arduously, and printed a few miniature magic-related booklets -- but honestly I'm not sure that helped much in discussing these issues.
I am not trying to support or oppose any of the recent comments -- just stating a few things. One of the problems is that printing methods have evolved over time, and have always been subject to variation, and it is difficult to make statements that are true in all places at all times.
Back in that era (the turn of the century, circa 1900), a printing forme with set type, for printing without electrotype or other plates, could be extremely heavy. An article by Vernon Possnett in The British Printer for January-February 1901 alludes to formes that would be difficult for two men to lift, and makes suggestions as to the preferred method for removing the form from the table.
Based on estimates discussed below, I could see that such a forme (with say 16 pages of type) for The Expert at the Card Table might easily weigh around 200 pounds, especially when one includes the chase and probably other hardware.
And if then there would be maybe a dozen of these, with a total weight of perhaps well over a ton. My impression is that those are not the kind of thing that you would want to move any noticeable distance.
But under usual printing terminology, it would not be best to refer to those formes as plates.
So, that’s if The Expert at the Card Table were printed directly from set type, which (or so it seems) nobody thinks it was. Instead, the consensus appears to be that it was printed from plates, presumably stereotype plates or electrotype plates.
The main reason I think the book was printed from plates is that some of the damage to the type that is evident in early printings appears to me to be consistent with damage to molds created during the platemaking process. John Bodine sent me images of the backs of the title pages on several copies of the first edition, and they all exhibited the same damage.
A book that is often cited regarding nineteenth-century printing methods is The Harper Establishment, by Jacob Abbott (1855), a reprint of which is on the Hathi Trust site. Some factors:
Abbott intimates that a plate weighed roughly one-fifth the weight of the set type, and that plates might be about 3/16 of an inch thick. That thickness is about one-fifth the height of type, so that estimate makes sense.
According to what Abbott says, the norm for Harper was one plate per page.
Set type would be about 0.9186 inches in height.
Lead weighs about 708 pounds per cubic foot. Type metal (an alloy of lead and other metals) would be somewhat less.
After spending quite a bit of time working with numbers, my estimate of the weight of a plate plus corresponding wood block in the case of The Expert at the Card Table is something like 2.2 pounds, actually probably somewhat less. That is in very close agreement with the estimate that Brad gives above. This assumes one page per plate. There are lots of variables, though.
Still, even if you figure a total of 400 pounds, and 205 or so plates -- that’s pretty heavy and would take up a moderate amount of space, unless, for example, as in cases mentioned by Abbott, the plates were removed from the blocks for storage.
But -- given the McKinney bankruptcy, and S.W. Erdnase’s possible disappearance soon after the book was printed, and the fact that it was years before Drake printed his own first printing, it seems simpler to assume that the plates, if any existed, were separated from the blocks and melted down, or disposed of in some way. This is somewhat related to the weight of the plates, but it doesn't matter much whether we are talking a ton, or 400 pounds.
On the other hand, it still seems quite possible to me that the first edition was not printed from plates, in which case the possibility that McKinney would just leave to type standing seems remote, since the absence of plates could well have been predicated on the assumption that future printings would not be needed.
Assuming one plate per page, with a type-height of 0.9186 inches, the blocks would make a conceptual stack nearly 16 feet tall--or eight stacks two feet tall.
Eight stacks, two feet tall, weighing a total of 400 pounds . . . definitely not impossible to move, but, I would think, extremely inconvenient.
I just quickly looked at a 1946 Powner copy of the book. It is very hard to tell, but it looks as though there are six gatherings, which would mean an average of a little more than 34 pages per gathering. That borders on impossible (for reasons that are a little beyond the scope of this post), so let’s assume 32 pages per signature. That would mean 16 pages per forme. Based on Abbott, the usual practice would be to lock the 16 (or however many) blocks into place in the forme (or perhaps more precisely, into the chase, creating the forme). Then they could be released therefrom after printing.
Again, I’m not trying to argue with anyone -- just stating a few of my perceptions.