1906 San Francisco earthquake

At 5:12 a.m. on Wednesday, April 18, 1906, one of the most catastrophic natural disasters in American history struck the city of San Francisco. The earthquake hit the coast of Northern California with an estimated moment magnitude of 7.9 and a maximum Mercalli intensity of XI (Extreme). The earthquake ignited more than 50 separate fires around the city that burned for three days and destroyed nearly 500 city blocks. The earthquake and fires killed an estimated 3,000 people — with some scholarly estimates ranging as high as 6,000 or more when surrounding areas are fully accounted for — and left half of the city's approximately 410,000 residents homeless. The event remains the deadliest earthquake in the modern history of the United States and a defining moment in the history of San Francisco.

Geological background and fault rupture

The earthquake originated along the San Andreas Fault, the great tectonic boundary running the length of California. The San Andreas Fault is a continental right-lateral strike-slip transform fault that forms part of the tectonic boundary between the Pacific Plate and the North American Plate. Along this fault, the Pacific Plate moves northwestward relative to the North American Plate, and the accumulated elastic strain energy released in 1906 was enormous. The earthquake ruptured the northernmost 296 miles (477 km) of the San Andreas Fault, from northwest of San Juan Bautista to the triple junction at Cape Mendocino, confounding contemporary geologists with its large horizontal displacements and great rupture length. By comparison, the 1989 Loma Prieta earthquake had a rupture length of only 25 miles (40 km).

At 5:12 a.m. local time, a foreshock occurred with sufficient force to be felt widely throughout the San Francisco Bay Area. The great earthquake struck approximately 20 to 25 seconds later, with an epicenter in the Pacific Ocean just 2 miles (3.2 km) west of San Francisco. The amount of horizontal slip, or relative movement along the fault, varied from 2 to 32 feet (0.6 m to 9.7 m). Violent shocks punctuated the strong shaking, which lasted some 45 to 60 seconds. The earthquake was felt from southern Oregon to south of Los Angeles and as far inland as central Nevada. Seismologists estimated the average speed of the rupture along the San Andreas Fault to the north of the epicenter at approximately 8,300 miles per hour (3.7 km/sec), and approximately 6,300 miles per hour (2.8 km/sec) to the south.

For years, the epicenter was assumed to be near the town of Olema, in the Point Reyes area of Marin County, due to local earth displacement measurements. In the 1960s, seismologist Bruce Bolt of UC Berkeley proposed that the epicenter was more likely offshore of San Francisco, to the northwest of the Golden Gate, a position supported by reanalysis of the original seismographic records. The most recent analyses support an offshore location for the epicenter, placing it approximately 2 miles (3.2 km) off the San Mateo County coast near Mussel Rock, although some uncertainty remains.

One important characteristic of the shaking intensity documented in the Lawson Commission's 1908 report was the clear correlation of intensity with underlying geologic conditions. Areas situated in sediment-filled valleys sustained stronger shaking than nearby bedrock sites, and the strongest shaking occurred in areas where ground reclaimed from San Francisco Bay failed during the earthquake. This early recognition of site amplification effects — the tendency of soft sediments to intensify ground motion — was a foundational contribution to the field of earthquake engineering.

The fires and immediate destruction

San Francisco had experienced notable earthquakes in 1864, 1898, and 1900, but nothing approaching the scale of the 1906 event. The shaking was devastating in its own right: cable cars abruptly stopped, the city's brand-new City Hall crumbled to its steel frame, and the Palace Hotel's glass roof splintered and littered the courtyard below. San Francisco's brick buildings and wooden Victorian structures were especially hard hit. It's worth understanding, though, that the earthquake itself — as destructive as it was — caused only a fraction of the total damage. Estimates from historians and engineers put fire destruction at roughly 80 percent of the total property loss. The fires, not the shaking, were the true engine of destruction.

More than 50 fires broke out immediately across San Francisco, ignited by broken gas lines, fallen power lines, and overturned cooking stoves and lanterns. One of the most notorious ignition points was a blaze in the South of Market district, reportedly started by a woman cooking breakfast on a damaged flue — it became known as the "Ham and Eggs Fire" and grew into one of the most destructive of all the blazes. The disaster was compounded almost immediately by the failure of the city's water distribution system, whose underground mains had been shattered by the shaking, leaving firefighters with little ability to fight the spreading blazes. Fire Chief Dennis Sullivan, the one man in the city most prepared to lead a coordinated response, was fatally injured when chimneys from the adjacent California Hotel collapsed through the roof of his quarters at Engine Company No. 1 — he died four days later without regaining consciousness.

What followed was a conflagration of extraordinary scale. Fires swept from the business section near Montgomery Street and the South of Market district toward Russian Hill, Chinatown, North Beach, and Telegraph Hill, consuming block after block of densely built neighborhoods. In some areas, city and military authorities resorted to dynamiting buildings to create firebreaks, though these efforts sometimes spread the fires further rather than containing them. The blaze continued for four days, until its smoldering remnants were finally extinguished by rain.

The destruction of City Hall

Among the most striking images of the disaster was the ruin of San Francisco's City Hall, a grand structure completed in 1899 after 27 years of construction at a cost of $6 million — a figure that had already made it a symbol of municipal corruption. The earthquake stripped away the building's ornate stone cladding in seconds, exposing an interior steel skeleton that had been built with far less iron and steel than contracts required; corrupt contractors had substituted rubble and sand for the materials they were paid to provide. The dome survived largely intact, but the hollow shell of a building standing beneath it made the corruption visible to every San Franciscan who passed by. The wreck was demolished in the cleanup that followed. A new City Hall, designed in the Beaux-Arts style and considered one of the finest civic buildings in the United States, was completed in 1915 in time for the Panama-Pacific International Exposition.

Scale of physical destruction

The human response to the disaster was immediate and, in some instances, severe. At 7 a.m. on the morning of the earthquake, U.S. Army troops under Brigadier General Frederick Funston — who had acted without direct orders from Washington — reported to the Hall of Justice and began patrolling the stricken city. Mayor E.E. Schmitz issued a proclamation imposing a dusk-to-dawn curfew and authorizing soldiers and police to shoot to kill anyone found looting. The military's role in the disaster response was extensive, encompassing firefighting support, maintenance of order, and the organization of refugee camps, though Funston's unilateral deployment of federal troops without a presidential order remained legally controversial.

More than 500 blocks in the city centre — covering some 4 square miles (10 sq km) — were leveled. The inferno destroyed approximately 28,000 buildings, and the total property value loss was estimated at $350 million (equivalent to roughly $12 billion in 2025 dollars). It was the fires, far more than the shaking itself, that accounted for the majority of the physical destruction.

Casualties and the displaced population

The true human cost of the disaster was significantly underreported for decades. Early official counts placed deaths as low as 375 to 500, figures that were deliberately minimized in part by civic and business interests anxious to limit the perception of San Francisco as dangerous and ungovernable. Hundreds of fatalities in Chinatown went unrecorded, and the deaths of working-class residents in South of Market were similarly undercounted. The frequently cited figure of 700 deaths is now understood to underestimate the actual loss of life by a factor of three or four. Most of the fatalities occurred within San Francisco, with 189 additional deaths reported elsewhere in the Bay Area; nearby cities such as Santa Rosa and San Jose also suffered severe damage. Santa Rosa, proportionally, was among the hardest-hit communities in the region — a significant share of its downtown was destroyed and more than 100 of its residents were killed.

In 2005, the city's Board of Supervisors voted unanimously in support of a resolution — written by novelist James Dalessandro and city historian Gladys Hansen — to recognize a figure of 3,000 or more as the official death toll. Some modern scholarly estimates, particularly those that attempt to account for undocumented immigrant communities and transient populations, place the total considerably higher, with figures above 6,000 appearing in some analyses.

Between 227,000 and 300,000 people were left homeless out of a population of approximately 410,000. Half of those who evacuated fled across the bay to Oakland and Berkeley. Aid poured in from around the country and the world. The Salvation Army, responding to the disaster with every available resource it could deploy, exhausted its entire reserve funds in the relief effort — an expenditure so total that the organization resolved it would never again commit so completely to a single disaster response. Survivors slept in tents in city parks and the Presidio, stood in long lines for food, and were required to cook in the street to minimize the risk of additional fires. The distribution of relief was not equitable: Chinese residents were systematically excluded from many of the official camps and relief programs, forced instead into makeshift arrangements in the ruins of Chinatown or across the bay in Oakland, where a substantial community relocated permanently.

Relief cottages

To house the tens of thousands left without shelter, the city and federal government constructed a system of small prefabricated structures that became known as "relief cottages" or "earthquake cottages." More than 5,600 of these modest wooden homes were built in temporary refugee camps across city parks, including Golden Gate Park, the Panhandle, and several smaller sites. The cottages typically measured about 10 by 14 feet and were designed to be portable — residents could eventually purchase them for a nominal fee and have them moved to permanent lots. Thousands of San Franciscans lived in these camps for months, and some for years, as the city rebuilt around them. A small number of the original relief cottages survive to the present day, relocated to various neighborhoods and now recognized as rare physical relics of the disaster. Several have been identified in the Outer Sunset and other western neighborhoods, though preservation efforts have been inconsistent; at least one known survivor was demolished without public notice in recent years. The remaining cottages are considered significant pieces of San Francisco's social and architectural history.

Scientific legacy

The 1906 earthquake fundamentally transformed the scientific understanding of seismology and fault mechanics, and it remains one of the most consequential seismic events in the history of the discipline. Its importance derives not only from its sheer destructive power but from the extraordinary body of scientific knowledge it produced. In the immediate aftermath, the disaster prompted the founding of the Seismological Society of America in 1906, an organization established specifically to advance the scientific study of earthquakes in the wake of the disaster.<ref>{{cite web |title=1906 Earthquake |url=https://seismo.berkeley.edu/outreach/1906_quake.html |work=UC Berkeley Seismology Lab |date=2006 |access-date=2026-