In the 1930s, when streetcars were alive and well — over a decade before any sort of BART planning occurred, and three decades before the start of BART construction — San Francisco planned a network of subway tunnels extending across the City that would serve as a rapid transit spine that streetcars could funnel in and out of on their trips to and from downtown. It was not the first time such a system was proposed, nor would it be the last, but it was a milestone of sorts in San Francisco’s transit history. Three older streetcar tunnels — the Stockton Tunnel (completed 1914), the Twin Peaks Tunnel (completed 1918), and the Sunset Tunnel (completed 1928) — were built not for the express purpose of providing rapid transit, but rather, to break through San Francisco’s famous hilly barriers. A fourth tunnel, the Fort Mason Tunnel, was also completed in 1914 in connection with the Panama-Pacific International Exposition the following year. Streetcars led the way in the development of new neighborhoods — or in the urbanization and densification of existing neighborhoods, by making them more accessible — much as cable cars had done a few decades earlier, though the streetcars often burrowed under hills in addition to climbing over them. Meanwhile, streetcars converged from all over the City onto Market Street — but on the surface. The surface of Market was equipped with two tracks in each direction, and it’s not hard to imagine what the chaos of Market Street in the early 20th century must have been like: hordes of pedestrians, and lines of streetcars. A Chronicle editorial from the 1930s once described Market Street as an “elongated four-track car barn.” Clearly, something needed to be done to relieve streetcar congestion. But that was not the only good reason to build a downtown subway.
The plans and studies for a subway coalesced into a formal proposal in 1936, around the same time as both the Bay Bridge and the Golden Gate Bridge were completed. These two bridges, by connecting the City to the nearby mainland, extinguished the need for ferry service, and they had a profound effect on the way people moved about the region. Even in the 1930s, when over two-thirds of San Franciscans were transit-dependent, there was a worry that transit riders would turn to the automobile and abandon the slow, congested streetcars, thereby only adding to the congestion problem. The traffic that the new bridges could dump everyday onto city streets made this worry that much more acute. The realization that an investment in rapid transit had to be made to keep streetcar service competitive led to this subway proposal. In that sense, their concerns back then were not all that different from our concerns today.
The proposed subway system was rejected. In 1937, voters quite definitively turned down the $49.3 million rapid transit bond measure. In fact, the measure failed to earn even a simple majority of the votes (just 42% voted in favor of the bond), probably because of concern and uncertainty in approving such a large bond issue during economically tenuous times. (The $5 million sewer bond on that ballot also failed. In contrast, the much smaller $2.9 million Airport expansion bond, and the $1.6 million hospital improvement bond — on the same ballot as the rapid transit bonds — both passed.) Like the Bay Bridge and the Golden Gate Bridge, the subway also would have had a profound, though quite different, effect on travel patterns — by reducing transit travel times, encouraging San Franciscans to stick with their streetcars, and offsetting the impending love affair with the automobile that was to truly set in just a decade later. But rapid transit was not yet to be.
If the proposed subway were built, here is a sketch of what it would have looked like:
The map only includes proposed construction. Then-existing surface streetcar lines that would have connected to the subway are omitted. Also omitted is the Twin Peaks Tunnel, which predates this proposal.
One thing that is immediately striking about this map is how of much of it we ended up building eventually — sort of:
- We do have a Market Street subway, which connects directly into the older Twin Peaks Tunnel — San Francisco’s first subway, which had already been in use for almost two decades by the time of this proposal. The two tunnels combined form one long tunnel that extends along most of the diagonal of San Francisco’s square of seven by seven miles. In the 1936 proposal, the subway only extended as far west as Church Street, without connecting the two tunnels. Streetcars running in the Market Street Tunnel would have resurfaced at Church, and then used surface tracks to access either the Sunset Tunnel or the Twin Peaks Tunnel on the way to their outer terminals. The planned subway connection from Market to Howard via Fremont Street (which would have served the Transbay Terminal, completed in 1939, a few years after this proposal) was never built. An underground pedestrian walkway will be built in the future to connect the planned Transbay Transit Center to Market Street.
- A subway for the Mission District was proposed with the same alignment now used by the BART tunnel — but under this proposal, streetcars would have emerged from the tunnel near 27th and Dolores Streets, and then followed the Southern Pacific right-of-way through the Bernal Cut as a surface line, from Dolores Street to Ocean Avenue. This proposed surface extension was basically realized a half century after it was proposed, when the J-Church was extended from its former 30th Street terminal to the Balboa Park BART station and to the yard adjacent to the station. The two alignments are quite close, but not identical.
- Even in the 1930s, San Francisco dreamed of a north-south subway line to tunnel underneath the City’s densest quarter north of Market, in which the destinations of so many downtown-bound riders were (and still are) to be found. This proposal places a subway under Montgomery Street, filling in a missing subway connection to the north side of the Financial District. The tunnel would have run less than one mile from Market Street to Columbus Avenue, where streetcars would have resurfaced. More recently, the dream of a north-south subway has reappeared a few blocks to the west, as the planned Central Subway, whose Stockton Street alignment misses the Financial District.
But what major piece of this would-be subway system still remains unaccounted for? That would be Geary. The fact that a Geary subway was proposed as early as the 1930s testifies to Geary’s long-standing importance as one of San Francisco’s primary transit corridors. But its tale is one of unrealized promises. It has periodically entered the limelight, poised to finally receive the attention it deserves, only to be ultimately deprived of the favor. This pre-BART rapid transit proposal is interesting in its own right, and there is more that could be said about it. But one reason I brought it up now was to segue into a long-overdue discussion about Geary and its bus rapid transit corridor, currently being planned. (The other reason for writing this now is that a friend has been clamoring at me to post about transit history, so hopefully now he’ll get off my back.) The story of transit on Geary, for all intents and purposes, began in the year 1880 — and that is where we’ll take up the story, as well. Check back soon for further posts on Geary.
Excellent post, tell your friend thanks for getting you to write this!
Were you able to find any maps? Where did you find this information? This is fantastic stuff! Is the bond measure available online or do we have to go to the library?
Man, that foresight to reject transit bond measures was really smart. We could have never afforded 50 million today!
Apologies in advance for this, but: where’d you dig this one up, Eric?
I’m assuming there were planned stations that aren’t shown here. Would Geary cars have really run express all the way from Fillmore to Market? No stop at Van Ness?
It’s fascinating to imagine how this might’ve changed things. The system itself would no doubt have been modernized and expanded over the years; along Market, it would probably look a lot like it does today (albeit with much older stations), and I imagine there would be stops at Geary & Van Ness, 16th & Mission, and maybe a few other places. If there were a tunnel under Montgomery, you can bet there would still be streetcars on Columbus. And of course there’d still be streetcars on Geary.
Construction of such a system likely wouldn’t have prevented removal of the Key System or construction of BART, but it’s easy to imagine that BART would look very different–BART planners might have been able to fulfill their dream of running express service direct from Civic Center to Glen Park, bypassing the Mission.
Anyway–great post, Eric.
I don’t think i’ve ever posted a comment, but I follow your blog regularly. I lived in the Inner Richmond for a year and a half, and everyday when I sat on the 38 Geary, I wondered how it could be that no one had built a subway out to the Richmond yet. I look forward to reading more about what you have to say regarding our most transit-neglected corridor. Cheers!
jfds: OK, will do ;-)
Brunoboris: Most, but not all, ballot measures from the last century are available online, on the library’s website. The bond measure description of this project doesn’t go into all the details, but gives an outline of the scope of the project. Here is a link to a 9 MB PDF that includes the bond measure.
Overhead Wire: haha yeah, nowadays, we’d probably blow $50 million just on “studies.”
Steve: I probably should’ve been clearer about that on the map. The marked intersections were just meant to mark the boundaries of the tunnel, not to be a complete system map. (That’s why most of the surface extension south of Dolores was left out.) Oh, and transfer points (Market/VN and Market/Geary). But yes, there would definitely be stations that were not marked on here. Sorry for the confusion — that was what I meant by “sketch.” But yeah, it is fascinating to imagine how things would be different had they gone through with this. Not only would BART look different, but we wouldn’t be talking about Geary BRT or the Central Subway.
As for where this came from: I remember reading a book years ago (I can’t remember which now for the life of me) that mentioned this proposal in passing, and I was intrigued enough at the time to go hunting for more. Turned out, the library has hidden away several of O’Shaughnessy’s old plans and reports, and I’ve been tempted to return periodically throughout the years, including recently. Just fascinating stuff to sift through.
Earlier versions of the plan looked slightly different: rather than having one Geary tunnel, there were a couple tunnels that merged into the Market Street tunnel, like the N and J do now at Duboce. Those tunnels were placed at O’Farrell (for Geary streetcars) and McAllister. Another possible future expansion was an additional downtown subway under Mission Street for when the Market Street tunnel had too much traffic.
Daniel: Thanks for reading, and for introducing yourself. I hope you enjoy the next couple of Geary posts.
Great post! I’m really looking forward to more information on past proposals for Geary rail. Its hard to even find information on the original Bart plans that included that corridor.
I wonder what the city would have looked like if they hadn’t built the GG at Bay bridges.
duncan: Funny you mention BART. Another project that’s been on my to-do list is a series of posts on BART history, especially prior to the start of construction. Not sure when it’ll happen, though; probably not before November, the way the election looks to be shaping up.
BART’s Geary plans were in flux, and were eventually dropped after Marin County was out of the picture. There were different proposals, but the BART line could’ve gone out as far as Presidio, 11th Ave, or 25th Ave. I will, say, though, that this upcoming Geary history post has gotten really long, so the BART description is fairly abbreviated and doesn’t really go into it much.
Thanks for the link, Eric. I’ve spent the past half-hour Googling for more with remarkably little success. I’d read before that a Geary subway had been voted on in the ’30s, but it’s sort of amazing that in this age of information overload there could still be so little out there about something of such historical importance. I mean, this could’ve been a real turning point in the city’s history. Anyway, you’ve done a real service by bringing it to everyone’s attention.
Of course, in looking for more on the ’37 plan I came across a bunch of stuff about the streetcar abandonments that followed in short order … it’s incredible to think that within a few years of this vote, we were ripping out tracks with, well, abandon.
This has got me to thinking about other turning points. What if BART hadn’t been built? As much as we like to criticize BART, Seattle, Detroit, Denver and Los Angeles all said no to similar systems around the same time, and you can see where they are now. I’d like to think that if BART hadn’t passed we might’ve eventually built a system that was a better fit–but more likely, we’d be arguing over critical (and by now, enormously expensive) lines and not relative luxuries like San Jose BART.
I did find a couple of interesting tidbits. One is here:
[PUC General Manager E.G. Cahill] also was defeated, by popular vote in 1937, in plans to construct a three line rapid transit system. This defeat was not helped by labor’s belated and half hearted support of the measure. To improve the dilapidated condition of most of the streetcar lines necessitated an enormous outlay of capital. Bus and trolley conversion was cheaper.
There’s a footnote attached to that. It reads:
The bond issue was endorsed by The Labor Council by a vote of 182 to 112 on October 1, 1937. The affirmative action followed four months of debate, with the MSR’s union leading the argument against based on possible loss of employment.
I’ll leave that for others to comment. But I will note that $50 million in 1937 translates to around three-quarters of a billion today, or a little more than half the cost of the Central Subway.
Great tidbits, there, Steve, and thanks for linking that article. It fleshes out why this bond was defeated, something I didn’t find that much on. When I was browsing through some old newspaper articles from the time leading up to the 1937 election, they confirm your excerpt about Market Street Railway, and their opposition certainly makes sense.
I don’t know if you’ve come across it, but I found this report quite helpful when doing some research into BART history a few years ago. It was done in 1980, though, so I wouldn’t be surprised if it’s somewhat out of date, or if there are additional sources available now that would change some of its conclusions. It’s kind of remarkable that there hasn’t been more written about the history of rail transit in the Bay Area (and BART in particular).
I posted – or thought I posted – a comment here late last night, but it doesn’t seem to have gone through. Anyway, thanks for this post and the upcoming Geary and BART posts.
As far as the precise locations of the stations are concerned, the 1931 plan, at least is deliberately vague:
“In order to secure ultimately the benefits of rapid transit, it is necessary to space stations about one-quarter of a mile apart, otherwise little advantage in time saving can be obtained over frequent stop operation on the surface. At the time that subway construction is undertaken, careful consideration will have to be given to the location of these stations. Any announcement of proposed subway station locations at this time would be premature and might result in unwarranted real estate speculation based on possible pedestrian traffic to and from these stations.” (p. 49)
Meanwhile, in the 1948 transportation plan, the stations on the Post Street subway would have been: Leavenworth/Hyde, Mason, Stockton/Grant, Montgomery/Market, 2nd/Mission
The vagueness of the map is a direct result from both that paragraph Eric quoted and a desire to keep it relatively uncluttered — although, quite frankly, for the discussion here, the stations are less interesting than just the tunnel itself. The tunnel, and the possibility of a true transfer station at Market & Geary. That said, those Post Street stations are awfully frequent.
Personally, I find the 1936 Geary tunnel (the Geary/North Beach alignment, rather than the Geary/Transbay Terminal alignment) to be the most intriguing of the bunch.
Here in Southern California we had the “Kelker-DeLeuw” study back around 1925. It recommended all sorts of rapid transit construction, and (as I recall) a Union Station for the steam railroad passenger trains. We did finally get a Union Station in 1939, but the rapid transit proposals gathered dust. Unlike San Francisco, we lost all our local rail service (and no, it was not Judge Doom and Cloverleaf Industries). There was a “last chance” effort to upgrade rail service around 1948, when there was still a lot of Pacific Electric and LARy track in service, but when it came time to pay for all the improvements, all the various government entities were “looking out the window”. Private capital wasn’t interested–PE had been losing money for years, and the Huntington estate sold LA Ry (the narrow-gauge “Yellow Cars”) to a subsidiary of National City Lines (there will be a short pause for booing and hissing by passionate trolley fans).
At least in MuniLand, the City kept five of its streetcar lines running, and the Geary Street trolleys lasted until 1956 (several Los Angeles lines went bus during the mid-50’s–it was not a happy time for electric railway enthusiasts.)
(background info: I lived along a PE line, and even as a ten-year-old, I could tell that the tracks were worn out and needed a complete rebuilding)