The Oak Island money pit

Oak Island lies in Mahone Bay about 60 kilometres south-west of Halifax. The money pit is near the east end, on the lots that the Oak Island Treasure Company and its successors held; Smith's Cove is the beach about 150 metres east of it⁠[6]⁠[4]. Bedrock is Mississippian Windsor Group limestone and gypsum under the south-east half of the island and Cambro-Ordovician Halifax Formation slate under the north-west, with glacial till above. The only public geological survey document is a preliminary Geological Survey of Canada interpretation of a 1996 multibeam sonar survey, which notes circular seabed depressions that may be gas pockmarks or sinkholes over the Windsor rocks and warns that sonar alone cannot identify man-made features⁠[8]. Consulting reports by Warnock Hersey and Golder Associates (late 1960s and early 1970s) and a 1995 Woods Hole Oceanographic Institution study exist but were commissioned privately and have never been published⁠[11].

After two centuries of digging, the original pit no longer exists as a feature. The area has been excavated to 134 feet with a crane (Dunfield, 1965 to 1966), bored to 235 feet (Triton Alliance's Borehole 10-X, 1971), and criss-crossed by dozens of shafts and tunnels dug by searchers, so that anything found there today must first be shown not to be a searcher's own work⁠[4]⁠[10].

Everything attributed to the 1795 to 1804 digs is known only from accounts written sixty years later. McCully (1862) and "A Member" (1864) describe flagstones a few feet down, oak platforms about every ten feet, layers of charcoal, putty and coconut fibre, and at about 90 feet (95 in one account) a flat stone cut with characters⁠[1]⁠[3]. The stone is the only one of these finds that anyone later claimed to have seen. It was said to be built into a chimney in the 1860s; an 1864 correspondent said the marks looked scraped rather than cut; the 1893 prospectus said it had been taken to Halifax and read as "Ten feet below are two million pounds buried"; a 1911 visitor described a hard stone with no visible symbols; and a 1935 statement placed it as a beating stone in a Halifax bookbindery that closed in 1919. It has not been seen since⁠[6]⁠[4]⁠[5]. No drawing made from the stone itself survives; the familiar cipher is a later reconstruction.

The Truro Company's auger of 1849 was said to have passed through wood, "loose metal" and more wood at about 98 feet and to have brought up three links of a gold chain, a claim first printed thirteen years later⁠[1]. An 1897 drilling was said to have recovered a scrap of parchment with the letters "vi" or "wi" in ink; the 1931 Chappell shaft produced an axe and a miner's pick; Triton's 1971 borehole camera produced footage described as chests and a severed hand that no one else has been able to make out⁠[4]. None of these objects was recorded archaeologically, and most come from fill that earlier searchers had already disturbed⁠[10].

At Smith's Cove, searchers in 1850 found what they described as box drains of flat stones converging on a point, covered with coconut fibre and eelgrass, and read them as the intake of an engineered flood system. A 2010 study by Dennis King argued they are the remains of an early salt works; others see nineteenth-century wharf or searcher work⁠[4]. Coconut fibre from the cove and the pit exists in several hands; a 1970 analysis reported three of four samples as coconut, and radiocarbon dates have been quoted for it, but the chain of custody from any original context is not documented⁠[9].

The Lagina era (2006 to the present) has produced a lead cross from Smith's Cove (2017), human bone fragments from boreholes, a wooden "slipway" and "wharf" structures at Smith's Cove, and many small finds of iron, pottery and coin, all announced on television⁠[7]. The one statement found outside the programme is from the University of New Brunswick, whose laser-ablation lab measured the lead cross's isotopes and reported a European lead source; the lab did not date the object⁠[12]. No Lagina-era find has been published in a peer-reviewed or museum report, and the finds remain with the licensees under the Oak Island Treasure Act, which requires sworn reports to the Minister but no archaeological publication⁠[13].

The earliest document is a traveller's 1857 note of debris at an abandoned dig; nothing from 1795 to 1856 mentions a treasure pit⁠[14]⁠[5]. The 1861 dig by the Oak Island Association was promoted and attacked in the Halifax press that year, which is also when the flood-tunnel claim first appears in print, and when a boiler explosion killed the first searcher⁠[15]⁠[16]⁠[4]. McCully, who had organised the 1849 company, wrote the first full narrative in 1862; a second appeared in 1864⁠[1]⁠[3]. DesBrisay's county history of 1870 and 1895 passed the story into print as local history, and twentieth-century writers built on it⁠[17]⁠[2].

The later record is better. Companies dug in 1893 to 1900 (Maynard Kaiser died in 1897), 1909 to 1911 (Bowdoin, who left unconvinced), 1931 (Chappell), 1935 to 1939 (Hedden), 1959 to 1965 (Restall), 1965 to 1966 (Dunfield) and 1967 onward (Triton Alliance)⁠[4]. On 17 August 1965 Robert Restall, his son Robert, Karl Graeser and Cyril Hiltz died in a shaft on the island, apparently from gases; the family's own account is in Lee Lamb's book⁠[18]. Triton's work in the 1970s produced the consulting reports and Borehole 10-X; Triton and Frederick Nolan litigated over lots through the 1980s⁠[4]⁠[11]. Rick and Marty Lagina bought into Oak Island Tours in 2006, and the province replaced the Treasure Trove Act with the Oak Island Treasure Act in 2010 to regulate their search⁠[13]⁠[4].

What is not known is anything about 1795. No deed, diary, letter, court record or newspaper from before 1857 has been found that mentions the pit, the men, or a treasure; the names McGinnis, Smith and Vaughan are attested on the island as settlers, which is all the early record shows⁠[5].

The treasure reading holds that someone, before 1795, dug a shaft at least 90 feet deep, built platforms at intervals, and protected the deposit with tunnels that flood from the sea. Candidates have included Captain Kidd, Spanish or French colonial authorities, British Army paymasters during the American war, and the Knights Templar⁠[4]⁠[10]. The reading predicts a datable worked structure below the searchers' own disturbance and an engineered channel from the shore. Neither has been shown. Two centuries of digging have reached 235 feet without a documented chamber, and the Woods Hole dye study of 1995, as reported by those who have read it, attributed the flooding to natural groundwater movement⁠[11]⁠[4].

The natural reading holds that the original depression was a sinkhole over a dissolved cavity in the Windsor gypsum, that the "platforms" were fallen timber and debris in the collapse, and that the flooding is simply the island's water table in porous rock. It has the geology on its side: the bedrock, the 1878 ox sinkhole, other sinkholes on the island and mainland, and the consulting reports as read by geologist Steven Aitken⁠[8]⁠[9]⁠[11]. It does not by itself explain the coconut fibre, which is why the fibre's provenance matters.

The hoax or embellishment reading, argued by Joltes, holds that the story was concocted or inflated in the 1840s to 1860s by men selling shares, and that each detail (the tackle block, the regular platforms, the stone's message) first appears when a company needed money⁠[5]. Nickell adds that the tale follows the stock pattern of nineteenth-century treasure legends and reads like a Masonic allegory of the secret vault⁠[10]. This reading predicts that no pre-1857 document will ever surface and that the earliest accounts will contradict the later ones, which they do on several points, including the depth of the stone and the dryness of the pit below the platforms in 1861⁠[16]⁠[5].

These readings are not exclusive. A natural sinkhole could have been dug into by 1790s settlers and then mythologised by 1840s promoters, which is close to what the sceptical writers propose.

Three things would move the question. First, a document from before 1857 that mentions the pit: a deed, a letter, a court or militia record, or a newspaper item. Lunenburg County records for 1795 to 1850 have never been systematically searched for this purpose⁠[5]. Second, publication of the physical data that already exists: the Warnock Hersey, Golder and Woods Hole reports, the Triton core logs, and the borehole and sonar data gathered since 2014, so that geologists outside the search can test the sinkhole reading against the actual stratigraphy⁠[11]⁠[8]. Third, for any object claimed to be old, a recorded find context, a conservation record and an independent dating published outside the television programme, which the Treasure Act does not require⁠[13]⁠[12].

A find that would settle it in favour of a deposit is a sealed, dated, worked chamber below the searchers' disturbance. A finding that would settle it against is a complete set of cores showing undisturbed natural collapse breccia and till from the surface to bedrock everywhere the pit is said to have been.