By Emilia Wellesley · Published May 6, 2026 · Updated May 6, 2026
The Hollow Earth Theory is a clearly falsified hypothesis that nevertheless built one of the most durable storylines in modern Western thought. Modern seismology and satellite gravimetry have closed the geophysical question: the Earth is a stratified solid with a thin crust, a silicate mantle that transmits shear waves, a liquid iron-nickel outer core, and a solid inner core. The Preliminary Reference Earth Model published by Adam M. Dziewonski (1936-2016) and Don L. Anderson (1933-2014) in 1981 fixed those layers in numerical form using roughly a million seismic travel-time observations and a thousand normal-mode periods [1]. The GRACE and GRACE Follow-On satellite missions, flying since 2002, have since mapped the gravity field with enough precision to detect millimeter-scale separation changes between paired spacecraft and to confirm the Earth’s mass distribution at the level of bulk flows in the outer core [2]. There are no concentric inner shells. There are no polar holes. The interior is not lit by an internal sun.
And yet the idea persisted for three centuries, picked up by an Astronomer Royal, an Army captain, a religious mystic, a self-published metallurgist, and an entire shelf of nineteenth-century novelists. It is worth treating the Hollow Earth not as an open question, but as the history of an idea: a useful fossil for understanding how a hypothesis can be reasonable in one century, eccentric in the next, fringe in the one after, and finally available only as literary furniture. The story belongs in the broader catalog of historical and archaeological mysteries precisely because it shows the seam where pre-modern science gave way to the modern earth sciences, and what survived in the gap.
What the Hollow Earth Theory Actually Claims
In its strongest historical forms, the Hollow Earth Theory claims that the planet is not a solid sphere of rock and metal but an enclosure: a thick outer crust wrapped around one or more habitable inner shells, sometimes lit by a small interior sun, sometimes accessible through enormous openings at the poles. This is the answer in fifty words. The geological evidence against it is overwhelming and was already decisive by the early twentieth century. The historical interest is the slow process by which this answer became unavoidable.
The claim has appeared in at least four distinct varieties. First, the concentric-spheres model proposed by Edmond Halley (1656-1742) in 1692, in which an outer shell encloses progressively smaller inner spheres separated by atmospheres. Second, the polar-opening model promoted by John Cleves Symmes Jr. (1779-1829) beginning in 1818, in which the planet is hollow and accessible through wide openings at both poles. Third, the inverted or concave-Earth model put forward by Cyrus Reed Teed (1839-1908) in his 1869 conversion vision, in which humanity lives on the inside surface of a closed sphere with the sun and stars at its center. Fourth, the modern conspiratorial-fringe revival, which welds Symmes’s polar holes to UFO lore, lost-civilization narratives, and the Verne-and-Burroughs literary tradition without bothering with the geology that disproved the model a century ago.
Each variety failed for a different reason, and the reasons matter.
Halley’s 1692 Concentric Spheres: The Idea Begins as Real Science
The first respectable hollow-Earth proposal came from a sitting fellow of the Royal Society. In late November 1691 Edmond Halley read an early version of a paper to the Society; in January 1692 the longer version was ordered for printing. It appeared in the Philosophical Transactions as “An Account of the Cause of the Change of the Variation of the Magnetical Needle; With an Hypothesis of the Structure of the Internal Parts of the Earth,” volume 16, pages 470-478 [3]. Halley would later become Astronomer Royal in 1720; this was not a fringe figure speaking from the margins.
Halley’s puzzle was real. Compass readings drifted over time and varied by location in ways that no fixed-magnet model could explain. He proposed an interior of three concentric shells, each with its own magnetic poles, rotating at slightly different speeds. The slow drift between the shells, in his model, would account for the secular variation of the magnetic needle. He suggested that the inner atmospheres might be luminous and possibly inhabited, and that gas escaping from the interior could explain the aurora borealis. The diagram he printed showed a planet structured like a smaller Saturn: a sphere wrapped in concentric circles.
Read in 1692, this was reasonable. Seismology did not exist. The behavior of P-waves and S-waves through different phases of matter would not be understood for two centuries. Halley was reaching for a mechanism, and the mechanism he reached for was geometric. The hypothesis was wrong, but it was the right kind of wrong: an explicit, testable model published in a peer venue. That is the shape of legitimate science failing forward.
Symmes, Teed, Gardner: The Slow Fall into Pseudoscience
By the early nineteenth century, the geological record had begun to push back. But the geology was not yet decisive, and the polar regions were still unmapped. Into that gap stepped John Cleves Symmes Jr., an Army veteran of the War of 1812 who had read his Halley.
On April 10, 1818, writing from St. Louis, Symmes mailed five hundred copies of his now-famous Circular No. 1 to learned societies, foreign governments, and members of Congress. The text was direct: “I declare the earth is hollow, and habitable within; containing a number of solid concentric spheres, one within the other, and that it is open at the poles 12 or 16 degrees; I pledge my life in support of this truth, and am ready to explore the hollow, if the world will support and aid me in the undertaking” [4]. Symmes attached a doctor’s certificate of sanity. He twice petitioned Congress for an exploring expedition, in 1822 and 1823, and the second petition won twenty-five votes before being tabled. His follower James McBride published Symmes’ Theory of Concentric Spheres in 1826, the year before Symmes’s death. The polar holes themselves came to be called Symmes Holes, an eponym that has outlived the geology by two centuries.
By 1869, when the upstate New York physician Cyrus Reed Teed reported a divine vision and adopted the name Koresh, the position had drifted into religious cosmology. Teed’s Cellular Cosmogony taught that the universe is a closed sphere with humanity living on the inside surface, the sun a battery-operated mechanism at the center, and the stars only refractions of its light. The Koreshan Unity, his utopian community, settled in Estero, Florida in the 1890s and peaked at roughly 250 members between 1903 and 1908; the site is now a Florida state park [5]. The Koreshan Geodetic Survey of 1897, conducted on the beach at Naples, attempted to prove the concave Earth using a “rectilineator” device along the Gulf shoreline. The device measured what it was built to measure.
By 1913, the Aurora, Illinois metallurgist Marshall Blutcher Gardner self-published A Journey to the Earth’s Interior, or, Have the Poles Really Been Discovered? An expanded edition followed in 1920. Gardner discarded Symmes’s nested shells in favor of a single hollow with a small interior sun, and assembled a mass of explorer testimony to argue the polar regions were warmer than expected because of warm air venting from inside. By 1913 this was no longer a respectable position. Robert Peary had reportedly reached the North Pole in 1909; Roald Amundsen had reached the South Pole in 1911. There was no opening. The book was already pseudoscience when it was printed.
What Modern Geophysics Settles
The technical answer arrived from a discipline that did not exist when Halley wrote. Seismology emerged in the late nineteenth and early twentieth centuries, and within decades it had mapped the Earth’s interior in cross-section. The crucial observation is simple. P-waves, the compressional waves that travel through any material, propagate through the entire planet. S-waves, the shear waves that require a solid medium, are blocked at depth and create a measurable shadow zone on the side of the planet opposite a large earthquake. The shadow zone is where it should be if the outer core is liquid and the mantle above it is solid. There is no shadow zone consistent with a hollow.
The 1981 Preliminary Reference Earth Model formalized that picture. Dziewonski and Anderson inverted around a thousand normal-mode periods, five hundred summary travel-time observations, and 1.75 million phase observations from twelve years of International Seismological Centre data [1]. The model gives radial profiles of density, elastic moduli, and attenuation from the surface to the center. Earth, in PREM, is a layered solid with an iron-nickel outer core that is liquid and an inner core that is solid. Bulk density rises from about 2.6 grams per cubic centimeter at the surface to about 13 grams per cubic centimeter near the center. There is no room in the budget for an empty interior; the planet’s measured mass and moment of inertia would be impossible if the interior were hollow.
Satellite gravimetry has since added an independent line of evidence. The Gravity Recovery and Climate Experiment (GRACE), a joint mission of NASA and the German Aerospace Center launched in March 2002, and its successor GRACE Follow-On launched in May 2018, fly paired spacecraft about 220 kilometers apart in polar orbit and measure changes in their separation to roughly ten micrometers [2]. Those measurements resolve the geoid and the time-varying gravity field with a sensitivity that registers groundwater flux and ice-mass loss. The gravity field is consistent with the PREM density profile and inconsistent with any large-scale void. There is no missing mass to hide a hollow.
A third constraint comes from heat flow. The Earth radiates roughly 47 terawatts of internal heat at its surface, sustained by radioactive decay in the mantle and by the slow cooling of the core. The thermal gradient through the crust runs about 25 degrees Celsius per kilometer near the surface and rises sharply with depth. A habitable inner shell would have to overcome temperatures in the thousands of degrees and pressures of millions of atmospheres. The thermodynamics permit no such region.
The Idea That Refused to Stay Disproved
The geology closed the question. The literature did not. From the moment Symmes published his Circular, fiction discovered the hollow as a place to set adventures. Edgar Allan Poe sent his narrator toward the southern polar opening in The Narrative of Arthur Gordon Pym of Nantucket (1838). Jules Verne sent his savants down a volcanic conduit in Voyage au centre de la Terre (1864) and reached print before any rigorous seismic objection had been formulated. Edgar Rice Burroughs ran an entire decades-long Pellucidar cycle from 1914 onward, beginning with At the Earth’s Core. H. P. Lovecraft borrowed the cosmic geometry without endorsing the geophysics. L. Frank Baum sent Dorothy underground in Dorothy and the Wizard in Oz (1908). The hollow was useful in fiction precisely because it had been respectable enough, recently enough, that a reader of 1860 or 1900 could entertain it for the duration of a chapter.
The cultural-historical literature on this arc is now substantial. Peter Fitting’s Subterranean Worlds: A Critical Anthology, published by Wesleyan University Press in 2004, gathers extracts from fifteen subterranean fictions written between 1665 and 1914 and traces the genealogy from Athanasius Kircher and Ludvig Holberg through Verne and Burroughs [6]. David Standish’s Hollow Earth: The Long and Curious History of Imagining Strange Lands, Fantastical Creatures, Advanced Civilizations, and Marvelous Machines Below the Earth’s Surface (Da Capo Press, 2006) covers the same arc with more attention to American eccentrics, including Symmes, Teed, and the Gardner-era publishing scene. Standish’s verdict on the hypothesis is briskly accurate: an idea that was wrong and changed nothing, except that it changed quite a lot of fiction [7].
The fringe revival of the late twentieth century cannot claim the same intellectual lineage. Postwar Hollow Earth conspiracism welds Symmes’s polar holes to flying-saucer narratives, Nazi-survival mythologies, and various lost-civilization tropes, none of which engage with PREM, GRACE, or the heat-flow constraint. The hypothesis at this stage is no longer in dialogue with the data.
Why the History Is Worth Holding Onto
A clearly falsified theory, well documented, is more useful than a half-correct one. The Hollow Earth provides a clean case study in how a scientific hypothesis ages: born in the seventeenth century with the credentials of a Royal Society fellow, popularized in the nineteenth by an Army captain with a doctor’s note and a doctrine, absorbed into religious cosmology by a Florida prophet, recycled through a self-published 1913 monograph, retired into fiction, and reanimated as conspiracy in the late twentieth century. Each phase tells you what was known about the Earth’s interior at that moment, and what was not.
Halley was wrong because he had no seismology. Symmes was wrong because he had no firsthand polar reconnaissance and chose to ignore the geology that already existed. Teed was wrong on different grounds, having moved the question out of physics and into revealed religion. Gardner was wrong on the same grounds as Symmes, but a century too late to be respectable. The fringe revival is wrong because it elects not to look. The history is the lesson.
The dust on this artifact is mostly the print-shop residue of nineteenth-century pamphleteers. Brushed away, what remains is a clean record of how the modern earth sciences came to know the Earth’s interior, and what kinds of imaginings the slow accumulation of seismic, gravimetric, and thermal evidence eventually displaced. The interior is full of iron and silicates and pressure. The hollow lives only in the libraries.
