Lots of people (myself included) have an odd fascination with the superlatives of just about any topic. As it relates to highways, a common question is "What is the longest US route?" (answer) And then related questions branch off from there:
But first, a caveat: this article only takes mainline US routes into consideration. There are lots of bannered US routes (e.g. Business US xx, or Spur US xx) that are very short (some even less than a mile). There are also suffixed US routes (e.g. US xxA, or US xxN) that are very short. But none of those are mainline US routes; rather they are all auxillary US routes (i.e. their numbers are the same as the mainline route to which they connect). This article excludes bannered and suffixed US routes, and focuses only on mainline US routes.
The shortest US route, currently
In AASTHO's most recent route log (1989), two routes were listed at 42 miles each: US 266 and US 730. However, the mileages published by AASHTO were rounded to the nearest integer, and were therefore subject to a non-trivial error margin. Using a more precise method of measuring, we find that US 266 is 43.1 miles in length, while US 730 is only 41.3 miles long. So currently US 730 is the shortest US highway.
The shortest US route, future
The difference between US 730 and US 266 is less than two miles. Washington State DOT has planned a change in the alignment of US 12, which will presumably create a need to extend the US 730 designation by at least 3.5 more miles in length. Assuming that project gets completed, one unintended result will likely be that US 266 becomes the shortest mainline US route.
The shortest US route, historically
Among highway historians it is sometimes claimed that US 630 was the shortest-ever mainline US route. And that would have been true... if US 630 had been an actual route. It showed up on the Nov. 1926 map of official US routes approved by AASHO (now AASHTO), so it was definitely planned. If we measure its intended route, it would have been only 2.9 miles long. How does our modern measurement compare to the length of US 630 as listed on the early route logs? Well... turns out it never appeared on any route logs. By the time AASHO published their first US route log just a few months later (Apr. 1927), US 630 was already gone. Why? Because Idaho and Oregon objected; they wanted that short segment to be part of a longer auxiliary route that they signposted as US 30N (more details on this page). So US 630 was never signposted, and therefore it doesn't count as a "real" US route.
Well then, what was the shortest "real" US route? In AASHO's Apr. 1927 route log, US 530 was the shortest route listed, at 25 miles:
But it should be noted that there were a few others right on its heels. Here are the shortest five on that list:
The next AASHO route log was from 1929. In that log, it was US 154 (not US 530) that was indicated as the shortest. These were the shortest five, as listed...
The next route log of which I am aware was from 1935. By then US 650 was gone, and only three routes were listed at less than 30 miles:
Timeline: the shortest mainline US routes
US 530 remained the shortest US route until 1938, which is when its designation was replaced with a segment of the new US 189(ii). But a few other very short routes were also decommissioned at or near the same time, including US 320, US 366, US 110, and US 189(i). So after 1938, a series of other US routes claimed the dubious honor for varying lengths of time. Here is a full timeline:
*US 177 was commissioned in 1928. The 1929 route log listed it at 28 miles, and the 1935 log had it at 30 miles. If those figures were correct, then US 177 would have been the shortest US route from 1942 (when US 730 was extended) to 1964 (when US 177 was extended). However, the only way US 177 could have been that short is if it ended in Tonkawa. But neither route log mentions Tonkawa; both list the south terminus as Three Sands Jct. Furthermore, OKDoT maps from 1942-1964 clearly show US 177 extended south of Tonkawa, ending at Three Sands. So US 177 could not have been only 30 miles long. It actually would have been 40 miles long, and therefore would never have been the shortest US route.
What about the shortest two-digit US route?
So far this entire discussion has been about three-digit US routes. These were originally envisioned to be shorter "child" routes that branched off from longer "parent" routes. In other words, it is not all that surprising that some three-digit US routes are quite short. But since two-digit routes are presumed to be longer, which one is the shortest?
Currently, the 13th-shortest US highway is a two-digit route. US 46 measures a surprisingly short 75 miles in length, and it is entirely in one state (although there is an interesting explanation for that).
What was the shortest two-digit US route historically?
In the 1927 route log, US 48 was the shortest two-digit route listed, at 67 miles...
...and already by 1929, US 48 had been truncated on both ends, such that it was then only 50 miles long:
That was the shortest-ever two-digit US route. But US 48 was decommissioned in 1931. So what was the shortest two-digit US route after that? In 1931, US 94 [Naples to Miami] became the shortest two-digit route at 110 miles. (Incidentally, the next two shortest at the time were also east-west intra-state routes in Florida: US 92 at 163 miles [Tampa to Daytona], and US 98 at 193 miles [Pensacola to Apalachicola]). US 94 held the title for only five years, until US 46 was commissioned in 1936. At that time US 46 became the shortest two-digit, and has remained so ever since.