St. Francis Dam — the Foundation Mulholland Trusted, and the 1928 Flood That Ended Him

Summary

Two and a half minutes before midnight on 12 March 1928, the St. Francis Dam — a curved concrete gravity dam standing 185 feet above the streambed of San Francisquito Canyon, some 40 miles northwest of Los Angeles — broke apart and emptied its reservoir of roughly 38,000 acre-feet down the Santa Clara River valley to the Pacific. The flood, a wall of water that began near 140 feet high at the dam, killed at least 431 people across a 54-mile path and remains the second-deadliest dam failure in American history. The concrete did not fail under load. The rock beneath and beside it did: a slide-prone mica schist on the east abutment, riding above an undetected ancient landslide, and a conglomerate on the west that softened to mud when wet.

The dam was the work of William Mulholland, the self-taught engineer who had built the Los Angeles Aqueduct and run the city's water department for a generation. He was the sole authority over its design — no independent geologist examined the site, no formal foundation investigation was made, and no licensing body reviewed his work, because municipal dams in California were then exempt from state oversight. Mulholland trusted the ground because he trusted his own eye for ground, and the ground he chose betrayed him in the two specific ways that geology can betray a dam: by sliding, and by dissolving.

The reservoir filled slowly over two years, reaching capacity for the first time on 5 March 1928. New cracks and leaks appeared, and on the morning of 12 March Mulholland and his assistant inspected a muddy leak at the west abutment and judged it normal. That night the dam was gone, almost the entire structure swept away, leaving only a central monolith — Block 35, soon nicknamed "the Tombstone" — standing alone in the canyon.

At least eight inquiries followed, the most consequential a coroner's inquest that exonerated Mulholland of crime while condemning the system that had let one man decide everything. The disaster destroyed his career and ended forever the era of the unreviewed municipal dam. Within sixteen months California had stripped the municipal exemption, given the state authority over every non-federal dam above 25 feet, and made the lone, unchecked engineer a thing of the past. St. Francis is the canonical American demonstration that a dam is no stronger than the rock it is founded on — and that no engineer, however eminent, should be the only person who looks at it.

---

Timeline

The Build: A Gravity Dam Raised Twice, Founded on Two Bad Rocks

The St. Francis Dam was conceived as insurance. Los Angeles drew its water through a single aqueduct vulnerable to drought and to the dynamite of aggrieved Owens Valley farmers, and the city wanted a reservoir holding roughly a year's supply close to the city. Mulholland chose San Francisquito Canyon and a curved concrete gravity dam — a wall that resists the reservoir chiefly by its own dead weight bearing on the canyon floor, its modest curvature giving only secondary arch action. A gravity dam is conceptually simple and forgiving, but it has one absolute requirement: the foundation must be strong, watertight, and able to carry the wall's weight plus the uplift of water trying to float it off its base.

That requirement was not met, and it was not met in two independent ways. The canyon's two walls were different rocks, both unsuitable. The west (right) abutment was the reddish Sespe-formation conglomerate, which looked competent when dry but slaked when wet — its clay cement dissolved on contact with water until samples placed in a glass disintegrated into sand and mud within days. The east (left) abutment was a foliated mica schist of the Pelona formation, fissile and weak along its layering, and — unknown to anyone in 1928 — it formed the toe of a vast prehistoric landslide. The contact between the two formations ran diagonally through the foundation along the inactive San Francisquito fault, so the wall straddled a seam between a rock that would dissolve and a rock that would slide.

Mulholland did none of the things that might have revealed this. No independent engineering geologist examined the site, and no deep borings characterized the schist or the buried slide. And during construction the design was raised twice, by ten feet each time, to a final height of 185 feet — without widening the base. A gravity dam's stability depends on the ratio of its base width to its height; raising the wall while keeping the base fixed steepened the resultant load, increased the water pressure and the uplift at the heel, and narrowed the margin against sliding and overturning. The single most important structures on the site — the abutment rocks and the dam's own footing — were the least understood and the least conservatively designed.

The Failure Sequence: Slide on the Left, Slake on the Right, Float in the Middle

For two years the reservoir was an incomplete test. As it rose, water permeated the foundation. On the east abutment it seeped along the foliation of the mica schist; on the west it dissolved the cement of the conglomerate; beneath the dam it built uplift pressure under the heel that the design had not adequately relieved — only a fraction of the foundation was protected by relief wells, and there was no proper cutoff trench or grout curtain to intercept the flow. The cracks and muddy leaks of 1926–1928 were the foundation announcing itself, but they were read as ordinary concrete behavior.

On 5 March 1928 the reservoir reached capacity for the first time, and the abutments carried full thrust and full uplift together for the first time. The mechanism that modern analysis — chiefly the reinvestigation by geologist J. David Rogers — identifies as the trigger is the reactivation of the ancient landslide beneath the east abutment. Saturated by the full reservoir, the buried slide mass began to move, and the east end of the dam moved with it. As that abutment gave way, the failure unzipped the structure. Water tore into the foundation, eroding the slaking conglomerate on the west and undercutting the wall, while uplift reduced the dam's effective weight and let the reservoir lift and shove the monoliths downstream.

The dam disintegrated in seconds just before midnight on 12 March 1928. It did not fail as concrete fails — by crushing or cracking under stress — but as a foundation fails: the rock supporting it slid, dissolved, and let go, and the wall went with it. Roughly 38,000 acre-feet, about 12.4 million cubic metres, burst into the canyon. The initial wave stood near 140 feet at the dam and roared 54 miles down the Santa Clara valley, sweeping away the powerhouse, the construction camps, and the floodplain towns before reaching the Pacific near Ventura around three in the morning. At least 431 people died; many bodies were carried to sea and never found, and remains were recovered as late as the 1950s. Only Block 35 remained upright in the empty canyon — proof that the concrete itself had been strong enough, and the ground had not.

Contributing Factors

Aftermath

The St. Francis Dam killed at least 431 people — the official 1928 count was 385, but bodies recovered for decades pushed the modern figure to 431 or more, with some estimates higher still — making it the second-deadliest dam failure and one of the worst civil-engineering disasters in American history. Los Angeles paid millions in claims, the wreckage was dynamited and removed (the lone Block 35 "Tombstone" was toppled in 1929 as a hazard to sightseers), and the site is now a national memorial. The lasting legacy was law. On 14 August 1929 California granted the Department of Public Works — the body that became the Division of Safety of Dams — authority to review and approve every non-federal dam in the state above 25 feet high or 50 acre-feet, eliminating the municipal exemption that had let Mulholland act unreviewed. The same year the state established formal civil-engineering registration. The model spread, and the federal and state dam-safety regimes that govern American dams today trace their lineage to this failure. Within the profession the dam became the permanent byword for two propositions inseparably joined: that a dam is a foundation problem before it is a concrete problem, and that no engineer should be the only person who decides whether the ground will hold.

---

Lessons

References

• [St. Francis Dam]( — Wikipedia
• [St. Francis Dam disaster]( — Encyclopaedia Britannica
• St. Francis Dam (California, 1928) — Dam Failures and Lessons Learned
• [The St. Francis Dam Collapse: Los Angeles' Forgotten Disaster]( — HISTORY
• [St. Francis Dam: Worst Civil Engineering Failure of the 20th Century]( — Geoengineer.org