Why aren’t two-by-fours two inches by four inches?

Dear Cecil: Why is a two-by-four not actually two inches by four inches? I went to our local hardware store (no Home Depots in our little burg) and asked for a piece of lumber that was, as I had measured it, three-quarters of an inch by three and a half inches. The girl looked at me funny and said, “You mean a one-by-four?” I said, a little embarrassed, “Um, yeah, I guess that’s what I need.” Then I got to looking around, and it dawned on me that lumber sizes have nothing to do with their actual dimensions. With all their professing to “measure twice and cut once,” why don’t carpenters seem to care about the actual size of the stuff with which they work? Hunter, Craig, Colorado

Cecil replies:

Oh, they care. In fact, they’ve developed a special vocabulary to deal with the situation. Those in the building trades know one-by-four, two-by-four, and so on are “nominal” dimensions — that is, in name only. If you want an honest-to-Jesus one-by-four-inch board rather than the usual anorexic stick, the magic words are “true size.” Better yet, tell the clerk you want “five-quarter” stock. Five-quarter boards, commonly used for exterior trim, are actually four-quarters of an inch thick — that is, one inch true size.

That’s absurd, you say. Few carpenters would argue. However, they have reality to contend with. Also greed. You’ll appreciate these two factors show considerable overlap.

First reality. Years ago, cutting logs into lumber involved a lot of guesswork. The chief variable was the moisture content of the wood — green lumber shrinks as it dries. How much depends on how wet it was to start with. Typically a two-inch green board loses an eighth-inch of thickness once seasoned, but the actual difference may be more or less.

When sawmill operators adjust the “set-off” on their equipment — that is, the amount the log is advanced after each pass through the blade — they must allow for the kerf (or width) of the blade plus shrinkage. In the 19th century, they lacked an accurate way to gauge moisture content. So they made the set-off a little wider than the nominal size, knowing the true size of the seasoned lumber would probably be a little less. The difference between nominal and true size was known as “scant” allowance.

The process was far from exact. I know this from examining the boards in the museum of antique lumber known as my house, which was built in the early 1890s. The nominal thickness of rafters and such clearly was two inches, and I’d say on average true size was slightly less than that. But there’s quite a bit of variation, from one and three-quarters inches to two and an eighth.

Carpenters in the 1890s dealt with this as best they could, judging from my house. They used thicker lumber as headers — that is, the horizontal boards above windows or on top of a line of studs, which carried a lot of weight.

As time went on, builders began demanding lumber of uniform dimension, so sawmill operators began planing boards after cutting them. Assuming you waited till the boards dried out and adjusted your planer accordingly, you’d wind up with a product of reliable size. However, it was also thinner. Now the true size of lumber wasn’t slightly less than nominal, but a lot less.

How much less? That’s where the greed comes in.

Whether you’re milling lumber or making Hershey bars, smaller is cheaper. The driving concern at the turn of the century wasn’t so much the wood itself but the cost of freight. The virgin forests close to civilization had been cut down, and lumber had to be shipped from increasingly distant locations. In the early 1900s you might pay $10 per thousand board feet at the mill and $20 to ship it. It occurred to lumber tycoons that if they did all their finishing out in the woods, thereby reducing the product’s bulk, they’d save a ton of money on freight. So that’s what they did.

But a problem soon emerged: once nominal and true sizes parted ways, everything was up in the air. Southerners argued that southern yellow pine was stronger than northern white pine, and therefore could be cut thinner. Thus while two-by stock was one and three-quarters inches thick in most of the country, southern yellow pine manufacturers made theirs an inch and five-eighths. As lumber became a national commodity, builders complained about getting wood of different sizes. After World War I a push for standardization began.

Years of wrangling ensued, as lumbering regions jockeyed for competitive advantage and debated arcane issues, at one point arguing over a thirty-second of an inch. It wasn’t until 1963 that modern sizes were agreed on, but the standard has endured ever since. Today nominal one-by-fours are three-quarters by three and a half inches, while nominal two-by-fours are one and a half by three and a half — confusing for novice carpenters, and a bother for those rehabbing century-old houses, but otherwise a triumph of rationality over nature and the buck.

Cecil Adams

Send questions to Cecil via cecil@straightdope.com.