Or at least rediscovered…

A 25-acre Phelps’ Lake in San Francisco’s Panhandle?

I’ve just solved a mystery described in my previous research on the south area of Divisadero street. Back when it was a winding path through the dunes, Devisadero, as it was known, connected the Mission Dolores to the Presidio. The incorrect story had settled into this version over the years: San Souci Lake, located at Divisadero north of Hayes Street, burst its banks in 1862 and flowed to 7th and Market where it destroyed Pioche’s house—an impossibility by gravity alone, since it’s a different watershed! Thus the mystery. But now I’ve found that a second lake existed along Divisadero, just to the south. I see that my conjecture was correct: the flood was toward the Mission Dolores, instead, and destroyed a different residence of Francois Pioche than his 7th Street location.

My newfound solution clears up some mysteries and debunks errors found in the files of many libraries and archives, biographies,  articles (pdf) and books (search the linked page for ‘San Souci’). Of course, my discovery creates other layers of mystery.

The topic connects with the gigantic storms of 1861-62 that I’ve been studying. I just sent a letter to Janet Sowers, who is the hydrologist in charge of the SF-PUC historic watershed map, asking her to consider including the lakes on the map.

The two Divisadero lakes may be considered “vernal” lakes, meaning formed by seasonal rains, but they may have lasted years or come back every year. To be clear, the two lakes are Phelps’ Lake and San Souci Lake. Phelps’ Lake seems to have existed only briefly—possibly only a few months in 1862, but probably also repeatedly after rains in other years. It may be the lake shown in the middle of a birds-eye view by George Goddard (detail shown).

San Souci Lake may have existed for a few years or even many years. In this post, I’ll describe full chapter and verse of evidence for Phelps’ Lake only, as I have covered some aspects of San Souci Lake previously. I think San Souci was a more enduring lake, and was mentioned more often in later documents, but it’s not to be found on any image I know—unless it’s also one of the Goddard map anomalies shown here in the detail. San Souci Lake is, however, mentioned some in the articles presented below, and guess-drawn in on my old blog post where I began describing my research on this topic.

These two lakes were apparently (based on 1850s coast surveys) separated by a linear dune about 60 feet high running along what’s now Hayes Street to the west from Alamo Square.

Evidence for Phelps’ Lake

This serious accumulation of water may have only existed after strong rains, in varying shapes depending on dune shifts and rain depths. As far as I can tell, it (or something like it) was only recorded by Americans as having existed after the big storms of 1827 (mentioned in Article 4, below) and the extreme months of deluge in 1861-62. I’ve found specific dating of its presence for about three months, after which it was reported to have drained suddenly and catastrophically on March 15, 1862 at 1:00 a.m.

I’ve found four written mentions, all quite detailed and provided below, of a long lake, sometimes linked to the Abner Phelps home, or as threatening the Francois Pioche home. It’s sometimes described as in the Mission mountains—the term frequently used for hills in the outskirts of early San Francisco. The Phelps home still stands, though it has been moved a block or so from its original location at what is now Divisadero and Oak streets. (Perhaps the move was in reaction to the formation of the lake.) The Pioche home location is still unclear, but it was near Church & Market streets of today. Pioche was a financier and owner of Market Street Railway. By January 19, 1862, a long lake one quarter mile wide had formed in the dunes. The details of its demise are better accounted than its location.

This account of a long lake may clear up the heretofore unexplained body of water of that approximate shape and location drawn on a very detailed George Goddard birds-eye view. The Goddard view was published in 1868.

I found it most useful to refer to the Coast Survey of 1857 & ’59 to see the land forms that controlled the water flow at the time. Note the long dune west from the Orphan Asylum, along what is now Page Street, and another parallel dune, as mentioned, north of that at about Hayes. The gap in the Page dune at Fillmore would have allowed the water to flow toward Pioche’s property near 14th & Market, although I don’t know the exact spot of his home, yet, so it could have been a little farther north.

Many later sources incorrectly describe the flow from the burst Phelps Lake as having been from San Souci Lake, and as having inundated Pioche’s other property at 7th and Mission. They are proven wrong by these articles.

Article 1

January 19, 1862 article in the Daily Alta California
CITY ITEMS [see 2nd item]
A New Lake — The recent heavy rains have formed a lake of considerable size in a basin high up in the Mission mountains, north-east of the Mission Dolores, and about midway between the same and the Protestant Orphan Asylum. So great was the pressure of the accumulated waters, early yesterday morning, that the residents in the vicinity procured a gang of twenty laborers and proceeded to strengthen the weak parts to prevent a crevasse and overflow. The danger threatened the elegant grounds and residence of Mr. Pioche, formerly occupied by the late Mr. Hart [located somewhere above Dolores street current and below what’s now the Lower Haight], as well as the residences of Mr. Haight, and some six or seven others. Work was kept up without intermission all day ; and although the waters had subsided since, watch was maintained last night. The above lake, we are informed, is nearly a mile long, by over a quarter of a mile wide ; but being located amidst the sand-hills, it is expected it will subside in a few days. [It didn’t subside until it suddenly broke through the reinforced sandbank March 15, 1862 at 1:00 a.m., based on the articles below.] The gullies and basins of the Mission mountains and large sandy tract west of the city, between them and the Lone Mountain, are all full of water, and an immense volume of water is pouring into the Lobos Creek, and the various tributaries of Mission Creek ; but, beyond the overflow at the Willows, little or no damage has as yet occurred.

The above article describes the location as NE of the Mission but there is no basin NE of the Mission, so I assume they mean NW. People in San Francisco were often quite vague on locations “out behind the Mission” or “near the Orphan Asylum.” A line drawn equidistant from the Asylum and the Mission crosses through the basin at the Panhandle, right beside the Phelps house. This article doesn’t mention Phelps, but a follow-up article, below, seems to identify this lake with Phelps. The exact match between the further details in the two articles makes it clear that the two article reference the same lake, and the more accurate one (implied in the article to be “inspected” by the author—maybe in a visit to the site) says the lake was half a mile west of the Asylum, putting it near the Phelps home and in the same watershed as the threatened Hart/Pioche home.

Article 2 — a tiny blurb

January 23, 1862 article in the Daily Alta California
CITY ITEMS [see last item]
Drained: the lake that formed in the Mission hills behind Mr Pioche’s residence has been successfully drained from its northwest extremity.

Article 2 seems to indicate that the lake level was lowered in a controlled way, although the use of ‘NW’ seems another directional mistake. I explain both mistakes in the above two articles as follows: People thought of Mission Bay as the “bottom” of the map, since the area was always approached from that side by SF residents. So calling “up” north, when it’s actually west, would explain why the lowest elevation edge would be called the NW extremity and why the position would be described as NE of the Mission.

As for the lake being drained: More extreme rains followed, and water must have risen again, judging by articles 3 and 4. Also, the surrounding hills gradually released rainwater and would have refilled the lake, regardless of new rain.

Article 3

March 15, 1862 article in the Sacramento Daily Union
The Reported Triumph at Manassas — Excitement and Rejoicing — Destructive Flood — Arrivals. [see 4th paragraph]
A lake about half a mile long and a quarter of a mile wide, among the hills near the Mission Dolores, broke through its bank at one o’clock this morning, and precipitated itself into the valley below, utterly crushing and destroying the splendid residence of M. Pioche, with the fine garden, stable and carriage houses, and carrying away one hundred feet of the Market Street Railroad [which ran on Valencia Street]. The grounds and gardens of Woodward are damaged to the amount of four thousand dollars, being buried in nearly five feet of sand and mud. Pioche’s damage is twenty thousand dollars. Great damage was done to the gardeners, whose early crops were nearly ready for market, and which are now covered with two or three feet of water. The total damage by the flood is estimated at fifty thousand dollars. The persons in Pioche’s house narrowly escaped with their lives. There are fears that another lake in the vicinity [likely San Souci Lake] will break through, and workmen are embanking it.

The above is a short enhancement to the below article, which contains a lot of detail

Article 4

March 15, 1862 article in the Daily Alta California
CITY ITEMS [see 6th item]
Terrible Flood — Destruction of Property.
Early yesterday morning news was brought to town that an immense amount of property had been destroyed, and more seriously injured, by a flood in the neighborhood of the Mission Dolores. The information had not been exaggerated, and to-day the scene of the disaster corroborates the statement.

ORIGIN OF INUNDATION.
In the coast range of hills, to the northwest of the Mission, are, at this season, some ten ponds of greater or lesser dimensions. (Incredible!) One of these, situated in a valley one half mile west of the Protestant Orphan Asylum, has been known as Phelp’s (sic) Lake, an ex-Assembly man of that name occupying a residence at its head. This body of water, up to yesterday, embraced an area of twenty-five acres, and was about fifteen feet in depth. For a long time past fears have been entertained of this superincumbent mass of water bursting through and deluging the valuable real and personal property lying below. For the purpose of avoiding so serious a calamity, some six weeks ago a dam was constructed, and, adjoining thereto, a ditch cut to lead off gradually the superfluous water of the lake. [See Article 2 of January 23, above.] This dam has been pretty closely inspected and guarded, and, Thursday evening, there appeared to be no immediate danger of its giving way. The lake, at this point, was nearly fifteen feet in depth.

THE FLOOD.
About midnight F. L. A. Pioche, who occupied the elegant gothic cottage (not a cottage by today’s use of the word) a quarter of a mile below the foot of the lake (only 1/4? Was his house near the dune gap at what became Fillmore and Haight?), was about retiring, when, hearing the sound of rushing waters, went out to discover the cause. He at once saw an unusual quantity of water on his grounds, and hastened back, aroused the sleeping inmates (old term for occupants), who had barely time to escape before the torrent swept under the foundations of the house, which almost instantaneously settled and crushed to atoms. The invading stream had divided above the house, one branch pouring down the road in front and the other in the rear of the grounds. The various outhouses [outbuildings]—stable, carriage house, etc.—were first overwhelmed and completely wrecked. The beautiful yard immediately before the dwelling, on which Mr. Pioche had expended some ten or twelve thousand dollars, was cut up by the circling eddies into trenches, and to render the work of demolition complete, the banks caved, carrying with them much valuable shrubbery. The antics which the waters played were indeed curious. They did not sweep off the building, but undermined it in such a manner as to sink, and crush it like an egg shell. Of course, the destruction of furniture, and other contents of the dwelling, was heavy. In addition to costly furniture ruined or seriously damaged, a large number of superb paintings, elegant ware, cabinets of minerals, shells, vases, mirrors, frames and innumerable articles of vertu, rare and costly, were ruined. No value in figures can be put upon these latter— they cannot be replaced with money. Mr. Pioche seems to regret their loss more than all other effects destroyed. The house had lately been repaired, repainted, and greatly improved, and the grounds constantly and carefully cultivated. Incontestible proofs of the resistless force of the stream are seen in the bulky articles which were swept down the roaring current. A handsome piano forte was borne below nearly to the Railway [at Valencia Street], and two beautiful vehicles carried out of the carriage house, and buried beneath the water and sand. A number of casks and barrels, some filled with choice liquors, were swept quite down to the flat, and one carried as far as Judge Cowles’ residence, on McLaren [now named what?], near Mission street. The costly silverware supposed at first to be lost, was subsequently recovered. The total losses sustained by Mrs. Hart, the owner [actually former owner’s widow, I think, and apparently still living on the land] of the residence, and Mr. Pioche, in furniture, pictures, improvements on grounds, etc., cannot fall short of $30,000.

FURTHER DAMAGES BELOW.
The stream, after leaving the above scene of devastation, took a circuitous route for another quarter of a mile, when it encountered the kitchen, and out-houses of the public house [saloon] called L’Ermitage. [Pioche’s home near the Mission was often later referred to as the Hermitage, perhaps related to this saloon, which may have been his, too. The 1864 Lang directory lists “l’ermitage Saloon” at SW corner Dolores and Market, but I suspect it was not right on the corner.] The soil here, as above, is very sandy, and vast pieces of the banks crumbled and fell into the stream. These deposits were hurried down to the many patches of cultivated ground of the gardeners, causing the ruin of their crops of vegetables, just ready for the market. The tract immediately lying on the railway [at Valencia Street between 14th & 15th] was covered with water on the previous day [March 14] to the depth of three or four feet [Other reports, in the Daily Alta of March 13, describe the serious flooding in Hayes Valley and areas along the railway that existed before this inundation], but this has now disappeared, and a sterile sheet of sand been substituted in its stead.

BREAKS IN THE RAILWAY.
The tremendous current rushing right against the railroad embankment at right eagles, speedily forced a passage through it, leaving a chasm of ninety feet wide, but the rails withstood the pressure and were not carried off. The Superintendent was promptly advised of the accident, and at an early hour had a strong force at work repairing damages. By noon to-day, the trains will be running as usual.

OTHER EFFECTS OF THE FLOOD.
Just east of the railway the stream washed unceremoniously into the magnificent grounds of Mr. R.B. Woodward, tearing up fences, uprooting shrubbery and covering the earth with heavy deposits of sand and slime to the depth of three feet. The gardens [locations unknown] of Mr. Judson, of the Chemical Works, of Dr. Ashe, and others contiguous to the railway, have been greatly damaged. Between Phelps’ Lake and the Sans Souci House is another pond of five acres [called San Souci Lake, generally]. For a number of weeks past this has been full, and the water has encroached into the house itself [shown at the north corner of the small triangular basin, on the Coast Surveys], where it stands some three feet in depth. There has been danger that this, too, would break through its sandy barrier and precipitate itself into the basin below. Thursday night, when the flood came, many supposed that the swollen stream derived its supply from this source. This, however, was not true, but at 12 o’clock, yesterday, an outlet was made, and the superabundance of water gradually drained off. At sunset, last evening, this lake [San Souci] had fallen about one foot. No damage is apprehended of more destruction of property, as pretty much all the harm which could be done was done previously. Besides, a gang of men are at the breach checking any great efflux at this point. The depth of water in this pond is fourteen feet. What was Phelps’ Lake last evening presented a bed of black, earthy deposits, with a small creek coursing along the southerly bank of sand. [Wow! Cole Valley and the hills were still saturated, oozing that rainwater.] It appears that this tract has not been deemed arable land in the dry season hitherto. Mr. Phelps, however, believes that it is now improved so materially by these deposits, as to be tillable this season. We learn that in 1827, which was a season similar to the present, one of the numerous lakes in this vicinity broke through its confines and flooded the country below, causing great damage to such lands as were then under cultivation. And furthermore, that these lower grounds were, at that distant period, buried under masses of sand to the depth of several feet. Various opinions an entertained as to the immediate breaking through of the water at this last scene of destruction. Some aver that the bank was cut by some cowardly miscreants, whilst others assert that the gradual yielding of the sand, the waters of the lake easily percolated through, and so started the rush which only ended with the drainage of the pond itself.

The Goddard sepia map can be seen and explored in full resoultion and zoom on David Rumsey’s site. It is most interesting that the ocean road cuts right through the lake in Goddard’s print, and that a second lake, not looking much like the precise location of San Souci, but perhaps San Souci Lake expanded by the storms, is beside it. The ocean road that cuts through the lake is the road that was used to go west from Hayes Valley on McAllister street, veering onto Fulton street near where it passes the San Souci Roadhouse at the Devisadero Road (archaic spelling). Although the scaling is off, the two long lakes shown seem to be in the approximate location of the Panhandle, between Lone Mountain and the San Miguel Hills. But the key locating feature is the ocean road.

Pioche purchased the property that was destroyed from the widowed Mrs. Hart in 1857 and and sold at least part of it again in October 1862 to the Pacific Homestead Union (a developer, I suspect), to be subdivided—as “unions” are wont to do?! The Daily Alta carried an ad for selling or leasing the premises “to homestead unions and others” and called it “lately the residence of Pioche”, “near the Mission Dolores”, “Beyond the Willows” property. The Oct. 25th issue says it’s Pacific Homestead Union Property now, to be subdivided into lots of 50 x 114 ft. at $140 for each lot.

Pioche himself was in Europe and/or New York for much of that year, starting April 21 through at least September. Thanks to the brand new transcontinental telegraph, he was able to keep in touch from NY. Pioche was, incidentally, one of the few major figures involved at a high level in the early development of San Francisco who was living openly, by some reports, as a gay man, with his partner Robinson.

The last bit of this story, that I’ve seen so far, is reported a few months later when crews sent by various “entertainment houses” to fix the destroyed paths and roads near Pioche’s got into fights. One guy (Dennis Meagher) killed another (Francis N. Jay) with a shovel on May 16th, and the trial was covered a few times (August 6 & Sept 1, 1862 Daily Alta).

There you have it. A new lake documented in all ways except what we really wish for: a photo! What do you think? Do you have any specifics about San Souci to share? If you read this far, you’re a “serious researcher” and I’d love to know your comments, below.

Filed under: Articles