A standard question on freshman math tests when I was in college was "a real tunnel." A straight-line tunnel from Washington DC to Boston (430 miles) would of course be sloped down for the first half of the distance and then rise for the rest of the way to the end. A "train" starting at one end would thus accelerate going downhill for the first half of the distance and then coast uphill, losing speed to come to a stop at the other end. (Friction and wind resistance are generally ignored for freshman spot-tests. Sophomores get a tougher test.)

The usual questions included calculating how deep the tunnel would be at the center, maximum speed at the half-way point, and how long it would take the train to make the trip. For extra credit sometimes one could calculate how much shorter the tunnel was than a "great circle" route on the surface.

Recollection is that it would be about a half hour trip (a simple pendulum period calculation). Speed at the center was "very fast" but I don't recall just how many zeros were in the answer, and I'm too lazy to work it out again.

That would be closer to a real New York to New England tunnel, although New York to Boston is only about half as long (210 miles?).

John