Drawing Richmond's 1925 streetcar lines
Share
About this post: it was written with AI. Claude, a model made by Anthropic, drafted it from isoscape’s code and notes and the sources listed at the end, and I reviewed it before it was published. The maps themselves I draw with isoscape, a tool I built with Claude; my artist statement says more about how I work with AI. — Chris
By May 1888, Frank Sprague's electric cars were running on the whole of Richmond's Union Passenger Railway. It was the first large electric street railway that really worked. Over the next few decades the city spread out along the lines into places like the Fan, Barton Heights, Highland Park and Woodland Heights, and some of them were advertised by how fast the car got you to Seventh and Broad. The last car ran in November 1949.

None of that system is left to measure, so this map is built from records. Here is what I used and where it's approximate. Each of the fourteen stops has its own sources on the streetcars page in the explorer.
The routes
The starting point is a 1943 map by Harland Bartholomew & Associates, the St. Louis planning firm working for Richmond's City Planning Commission. It's called Early Transit Routes, 1888 & 1925, and it shows the car lines in both years along with the city limits at each date. The Library of Virginia has it online in its Mapping Inequality exhibition.
It's a small planning map, though, so I checked each line against histories of that line and wrote it out as a list of street corners, like Floyd Avenue at Robinson Street, then Grove Avenue at Three Chopt Road. The program finds those corners in OpenStreetMap and follows the streets between them, sticking to the streets the corners name. Where a line crossed a viaduct that's since been torn down, it's drawn straight across. Where the sources don't say exactly where a line ended, the end is approximate, and the page notes that.
The buildings
The buildings come from two places. Most are buildings still standing that the City of Richmond's assessor dates to 1925 or earlier. The assessor gives one year per parcel, so that year only goes to a few buildings on a lot, not to every building on a campus. The rest are buildings that are gone now, which I traced from Sanborn fire-insurance maps of downtown.
Small streets only show up where there was a house within 75 meters in 1925, so a subdivision that had been laid out but not built on shows its main roads and not much else. One known gap: Henrico County doesn't publish construction years the same way, so the suburbs past the 1925 city line look emptier on the map than they really were.

The fair grounds, and a correction
The Sanborn Map Company started in 1867 and made detailed maps of American cities for fire insurers, with every building outlined and colored by material: pink for brick, yellow for frame, blue for stone or concrete. The Library of Congress has the largest collection and has been putting the public-domain sheets online for free. For Richmond, I line each sheet up on today's streets using known corners, separate the colored buildings from the paper and lettering, and check later editions to see when each building disappeared.
That matters most at the State Fair Grounds, which is also where I got the first version wrong. The fair was on the grounds along the Boulevard from 1906 until 1941, where The Diamond is now. In the first printing I drew the grounds as a rectangle lined up with north, with a small oval inside. It didn't match the streets or the sources.

Sheet 468 of the 1925 Sanborn atlas shows it correctly. The grounds are a triangle between the Boulevard and Hermitage Road, the track is a full mile (my original note said half a mile), and the fair's buildings line the edges. For version 2 I fixed the sheet in place at Hermitage Road and Sherwood Avenue, took the track's south turn straight off the sheet, drew the north turn so the track comes to a mile, and traced thirty-six of the buildings. For the years the fair was there, today's streets and buildings on the site are removed. The 1934 USGS survey shows the same layout, with a long oval track inside the triangle and the exhibition buildings at the south end.


I keep a dated list of fixes like this one, with each piece's version number, on the Changes page.
Scale
I made two changes so the lines can be seen, and both are noted on the print. On the big map, the lines and cars are three times their real size. A streetcar at true scale would be a speck on a map of a city 13 kilometers across. On the single-stop prints they're one and a half times. The cars are also about twice as wide as a real 1920s car so their light roofs stand out against the red line.
The fourteen stops and their sources · Explore the 1925 map · The set in the shop
Sources
- Bartholomew & Associates, Early Transit Routes, 1888 and 1925 (Library of Virginia)
- Richmond City Planning Commission plates, 1942–46 (Library of Virginia research guide)
- A Streetcar Undesired (VCU Magazine, Winter 2025)
- Gems of Symmetry and Convenience: Sprague in Richmond (American Heritage)
- Sanborn Fire Insurance Map, Richmond, 1925, vol. 4 (Library of Congress)
- About the Sanborn Maps collection (Library of Congress)
- Mapping the U.S. block by block: the Sanborn maps go online (Library of Congress blog)
- Sanborn maps: history and use, founded 1867 (University of Florida Libraries)
- USGS Richmond quadrangle, 1:24,000, 1934 (PDF)
- Every stop, with its own sources (isoscape explorer)
- How it's made: Time, and streetcar lines