An Archaeological Reconnaissance of McKean Island, Phoenix Group, Kiribati
Thomas F. King
The International Group for Historic Aircraft Recovery (TIGHAR)
In September and October of 1989, The International Group for Historic Aircraft Recovery (TIGHAR) undertook its initial fieldwork in the Phoenix Islands, as part of its Amelia Earhart Search Project. The bulk of TIGHAR’s effort was devoted to Nikumaroro, formerly known as Gardner Island, but a single day was spent on nearby McKean Island. This brief report summarizes TIGHAR’s work at McKean.
Fieldwork was directed by Richard Gillespie, with the author serving as Project Archaeologist in charge of operations on land. The land team included John Clauss, Veryl Findlayson, Russ Matthews, Mary DeWitt, William Decker, Julie Williams, and Jessica Krakow. The dive team was supervised by Joseph Latvis and comprised Tommy Love, Mike Bowman, and Dutch Kluge. Transportation and logistical support were provided by M/V Pacific Nomad, captained by Mr. Victor Jhone, and Mr. Kotuna Kaitara of the Kiribati Customs Department represented the government of Kiribati.
McKean Island lies at Lat. 3o 36' S, Long. 174o 07' W. It is a raised coral island – in essence, a once-living coral reef that has been elevated above sea level by tectonic forces. It is roughly circular, about 3/4 mile across, surrounded by a fringing reef. Its maximum elevation above sea level is about five meters. J.D. Hague, who devoted most of a paragraph to its description in 1862, suggests that it at one time had a lagoon but that uplift had raised it to a point at which it had become a soggy depression in which bird guano had accumulated for thousands of years.
McKean has been an officially declared bird sanctuary since 1938, and today is protected as such under Kiribati law. The island is virtually devoid of vegetation; thin colonies of pigweed (Sesuvium sp.) and other low, shrubby plants grow in low-lying areas where a thin soil has formed, and around the guano "lagoon." The island is home to tens of thousands of sea birds, including terns (Gygis sp.), Frigate Birds (Frigata sp.), and gulls (Larus sp.). Approaching McKean by ship, one can hear its birds before the island itself comes into view, and if one is downwind one can easily smell them.
The north and east sides of the island exhibit pronounced storm surge ridges, with a linear depression behind them. Along the north shore this depression is two to three meters deep and about ten meters wide, making it, after the guano "lagoon," the island's most prominent geographic feature.
"Discovered" by American whalers by the late 1820s, McKean was given its name (after the crewmember who first sighted it) by Commander Charles Wilkes of USS Vincennes in 1840. At the time, bird guano was an important source of phosphate for explosives, fertilizers, and other uses. As a result, the U.S. Congress in 1856 enacted the Guano Act, "bonding" a large number of central Pacific islands including McKean for exploitation by U.S. commercial interests. In 1859 the island was claimed for the U.S. by Capt. Thomas Long of the schooner E.L. Frost, and in the same year C.A. Williams & Co. (later the Phoenix Guano Co., a subsidiary of the American Guano Company) filed claim to it under the Guano Act. Mining operations got underway in the summer of 1859. Twenty-nine Hawaiian laborers worked the island under the direction of a Mr. A.M. Goddard worked the guano deposits. The first shipload left the island aboard the schooner Modern Times in mid-August 1859.
By the early 1860s, when J.D. Hague visited the island to study its guano deposits, mining was well advanced in the large, irregular, sometimes water-filled depression at the center of the island. Although it lacked a channel to the sea, the guano depression's water level was influenced by the tides, standing at about 60 cm. at high tide. At low tide the guano deposits were merely soggy; presumably this was when mining was carried out. The deposits mined comprised a soft, amorphous layer of calcium phosphate and calcium sulfate — the product of mixing bird dung and evaporated sea water. Hague does not specify the depth of this ooze on McKean, but notes that a similar deposit on Jarvis Island was about two feet (ca. 60-65 cm) deep. About a foot (ca. 30 cm) of coral mud lay atop the guano deposit.
Presumably, mining methods on McKean were similar to those described as being carried out around 1900 on Banaba or Ocean Island (which was a British-run operation, however, and a good deal larger than the McKean mine). The procedure on Ocean Island was:
"first to remove the vegetation and a little surface rubbish, then loosen up about three inches of the surface of the deposit, and keep stirring it until dry. The phosphate was then collected, and by means of barrows loaded into trucks which were run along by small steam locomotives to the storage bins"
Equipment used on Ocean Island at the turn of the century included spalling hammers, picks, drills, wheelbarrows and dynamite, as well as the steam locomotives and "trucks" (ore carts) referred to above. Once dry, the phosphate was bagged and lightered out to the ship that would carry it away. On Ocean Island in later days a facility was constructed for loading directly into ships' holds, but there is no evidence of such a facility on McKean.
Getting and keeping the excavated phosphate dry was a major challenge on Ocean and other guano islands, and presumably was so on McKean. Charles Crosby, a shipping firm executive who visited the Ocean Island mine, reported that:
"…the practice is for the men to be sent to the field at 6:00 a.m. to turn over the surface of the field with a view to it being dried by the sun, and about noon or a little later if the ground was wet when they started, it is run into the bins up to 5 o'clock, but if there had been much rain the night before it is impossible to get any stuff properly dried in the course of one day, so they wait for the next. What frequently happens is that after the man have been working for a few hours a shower comes on and their labour is all lost"
In the late 1860s, the guano deposits ran out, and the last workers were removed from the island in late 1870. McKean has been uninhabited since that time, but for occasional visitors. In 1936 H.M.S. Leith visited the island to explicitly annex it to the British Empire, and in 1937 McKean with the rest of the Phoenix Islands was incorporated within the Gilbert and Ellice Islands Colony. Captain O. Bevin of Leith left a notice board and typed message (in a sealed tin can) proclaiming the island the property of His Britannic Majesty King Edward VIII. Harry Evans Maude, founder of the Phoenix Islands Settlement Scheme, visited the island in 1937 and left his own message plus a flagstaff flying the Union Jack (see Structure 6, below). In 1940, Maude could find no record of other visits by British ships, or of visits by government officials other than himself and his colleague Eric Bevington.
In the 1960s and early '70s, the Smithsonian Institution sponsored a number of ornithological expeditions to the island, typically involving three to five people and lasting two days. These studies resulted in a series of unpublished journals held by the Smithsonian's Museum of Natural History, Ornithology Department, which TIGHAR was able to examined in preparation for the 1989 expedition through the courtesy of Dr. Roger Clapp.
TIGHAR subjected McKean Island to one day of archaeological survey on October 3, 1989. The total lack of higher vegetation made it seem unlikely that people or an airplane stranded there could have escaped the view of the search pilots from U.S.S. Colorado who flew over the island in July of 1937. However, the possibility that the aircraft could have been sunk in the guano pit or covered with birds, and its crew dead or immobile and similarly covered convinced us that at least a cursory inspection was necessary. As it turned out, a cursory inspection was all we were able to do, and all we concluded was needed.
Maps and airphotos indicated that there were only two locations on McKean where a pilot might attempt to land an aircraft — along what appeared to be a beach lining the northeast shore, and in the guano lagoon. Accordingly, the dive team's priority was to inspect the reef face along its northeast quadrant, while the land team was to inspect the northeastern part of the island and the "lagoon." If these priority targets could be dealt with in good time, additional areas could be searched.
There is no landing as such at McKean, and no channel across the reef flat as there is at Nikumaroro. Landing was made on an uninviting strip of the lee shore, slightly more accessible than other areas, marked by a pylon just inshore from the beach. Landing was made at high tide, when it was possible to float the aluminum skiff from Pacific Nomad, our expedition vessel, across the reef flat to the beach. The surf was running quite high, and the landing was not without risk, but all land team members reached shore without serious damage. En route back to the ship the boat was shadowed by a large tiger shark, which luckily had not expressed interest in the dive team.
Although plagued by mechanical difficulties and fairly high seas, the dive team covered the reef face from the northwest corner of the reef flat down to the southeast corner, in two extended dives. Divers were deployed from approximately 10 meters to 40-45 meters depth; shallower waters were avoided because of the large numbers of feeding sharks there. Some 30-40 white tip sharks circled the divers for the entire day. Despite the distractions presented by this accompaniment, the divers felt they had given adequate attention to the half-circumference they were able to inspect. Results were entirely negative; the only artifacts noted were pieces of fishing longline.
The land team began by sweeping across the island to the northeast shore, spread out at roughly 20-30 meter intervals and inspecting the ground, while trying to avoid the very numerous nests, eggs, and angry birds. Reaching the northeast shore, we gave special attention to the linear depression, which we reasoned might have provided welcome shelter from the wind for anyone marooned on the island. The beach that we had expected to find along the northeast shore did not exist; in its place the island was bordered by a steep, coral-armored storm surge ridge. We inspected this area and then concentrated on the linear depression, walking 15 meter transects with full metal detection, marking turning points with day-glow orange spray paint on the rocky surface. We then proceeded southwest around the top of the guano lagoon, then south to the southern tip and back to the landing, concentrating on visual inspection with spot-checks by metal detector. Two- and three-person teams split off and inspected the southeast shore and the fringes of the "lagoon."
Our central task was looking for evidence that might be associated with Earhart and Noonan, of course, but within available time an effort was made to record the island's cultural features generally. Objects noted on the surface were inspected, and quick measured sketch-maps were drawn of all ruins. Ruins were also located in relative space by compass, with approximate ranges established by pacing. No artifacts were collected.
The intrepid Mr. Gillespie essayed an inspection of the guano pit, venturing out onto its caked surface some three to four meters before falling through. At this point he found the semi-liquid guano to be about a meter deep or more; he was pulled out before he fully plumbed the depths. Continuing around the periphery of the depression he and two colleagues extended metal detectors out over the sludge as far as they safely could. Somewhat less than one-half of the "lagoon's" surface area was inspected in this manner, all along its edges. The detectors consistently produced fairly strong readings, suggesting the presence of a good deal of metal in the pit. We speculated that this might represent steel rails, ore carts, and similar mining equipment, but had neither the wherewithal nor the inclination to find out.
No aircraft parts of any kind were observed, nor was anything else found that seemed likely to have been left by stranded aviators. A few scattered aluminum tubes, the largest about a meter long, are most easily accounted for as the leavings of recent shipwreck survivors and/or ornithologists. The remains of several more or less recent wrecked fishing boats lay on the reef and shores. Occasional tin cans and bottles were noted, together with a refillable propane can, which were doubtless left by their temporarily stranded crews. Fragments of rail, ore cart wheels, and similar pieces of mining equipment are scattered over the island as a result of the 19th century mining.
The linear depression apparently had been used as a campsite at least once; several tarp grommets were found there, together with a scotch whiskey bottle top and several indeterminate lumps of badly rusted ferrous metal. In a 1942 publication on islands in the area Edwin H. Bryan implies that this depression, with its stone wall (Structure 1; see Figure 3 below) might be some sort of prehistoric cultural feature, but we saw nothing to suggest that this was the case. The depression appears to be a natural artifact of the adjacent storm surge ridge, and the wall is similar to those found elsewhere on the island as remnants of the 19th century guano operation.
The ruins of ten dry-laid coral stone structures were recorded (Figures M-1, M-2). These ruins have been noted by virtually every previous 20th century visitor to the island, and are certainly the remains of the guano mining operation. Typically, these structures are of dry-laid stacked coral slabs, interspersed with standing slabs and boulders. Similar structures have been described on Manra (Sydney Island), where they apparently were built by workers in John Arundel's guano operations around the turn of the century.Structure 1 (Figure 3) is a low, "J" shaped wall about 80 cm. high and 50 cm. wide, in the linear depression along the north shore. A 6-meter long wall segment crosses the depression near its southwest end. A short segment to the north and a long (ca. 20 meters) segment to the south extend along the sides of the gully to the northeast.
Structure 2 (Figure 5) appears to be fragmentary. It lies in a field of coral rubble that may represent collapsed walls, and consists of a "J"-shaped wall segment about 10 meters on its long axis, three on its short. About 100 cm. high and 35 cm. thick, it is associated with a large slab of cemented coral.
Structure 3 (Figure 6) is a rather impressive circular wall about 120 cm. high, 35 to 50 cm. wide, and about 22 meters in diameter. The wall is for the most part made up of stacked coral slabs, occasionally interspersed with large coral boulders or standing slabs. Three piles of coral are semi-symmetrically organized inside the circle, near its center. Metal detecting revealed an undefined lump of ferrous metal just outside the southeast arc of the circle, and a "hit" that could not be located on the inside of the circle nearby. Structure 3 may represent a guano drying facility.
Structure 4 (Figures 7 and 8) is a complex curvilinear structure made up of twelve rooms or cells, some without evident entrances, others interconnected, with a pathway leading from an outside opening from northeast of the largest rooms, toward Structure 5. Wall construction is similar to that of Structure 3. The walls are 1 to 1.5 meters high, 35-50 cm. thick. Several building episodes appear to be represented; some walls contact but are not integrated with crossing walls. Low cement steps ascend one wall from a sort of narrow courtyard whose outside edge is defined by several low coral rubble piles. The whole structure is some 50 meters long and up to 20 wide. Metal detector sweeps yielded
a number of "hits," most of which could not be found, but large amounts of thin ferrous metal were noted in small concentrations on the surface, together with a large brass(?) food serving tray and some ferrous spikes with ring ends. Apparent post-guano period material consisted of a 10 cm. long aluminum tube in the southeastern room, a 1 meter long aluminum pipe in the northwestern room and a propane bottle with Japanese characters on it in the large west-central room. Structure 4 looks like it might have been a residential facility for the Hawaiian guano workers.
Structure 5 (Figure 9), connected to Structure 4 by a fairly straight pathway through the coral rubble, is a straightforward rectilinear structure with an open end facing southwest toward the landing and a smaller opening in the northeast wall. About 20 meters long and 12.5 meters. wide, Structure 5 has walls over 2.5 meters high. The northeast end of the ruin appears to have been floored with a pavement of coral cobbles. Ferrous spikes, straps, a claw hammer head, a chain link, and a brass spike and strap were found in metal detector sweeps, along with hits that could not be identified. Structure 5 appears to have been a warehouse or storage facility of some kind associated with the landing, readily accessible from Structure 4.
Structure 6 (not sketched, but seen in a picture in Maude's book, "The Colonization of the Phoenix Islands") is a simple circular stone cairn on the landing beach, 2.5 meters in diameter. Nearby is a concrete base, apparently for a flagstaff.
Structure 7 (Figure 10) is an odd, semi-rectangular structure with a half-oval apse or bay, southeast of Structure 5. It is about 10 meters wide and 20 meters long, oriented northeast-southwest, with walls similar to those of Structure 4.
Structure 8 (Figure 11), a short distance northweast of Structure 7, is a small rectangular structure about three meters on a side, including massive corral boulders in its walls.
Structure 9 (Figure 12) is another rectilinear structure, 21 meters long and 7 meters wide, oriented northwest-southeast, with the usual stone walls. Its southeast end is curved, and there is a short cross-wall that segments the southeast end from the rest of the structure. The north wall, if it ever existed, is mostly gone, but for a single tier of stones forming an indeterminate pattern on the rubble ground. A ferrous wheel lay on one of the walls, and a heavy braided steel cable was found inside the southeast end room.
Structure 10 (Figure 13) is a simple square ruin, 6 meters on a side that incorporates a massive coral boulder in its northeast corner. Its walls are the usual construction, about 1 meters high, and there is an 80 cm. wide doorway through its southwest wall.
About midway between Structures 9 and 10 (Figure 13) is a substantial pile of ferrous spikes and rods, many of the latter bent and twisted, together with at least one ferrous metal wheel about 20 cm. in diameter, probably from an ore cart.
One more structure was noted but recorded only cursorily by the "lagoon" survey party, on the edge of the guano depression toward its north end. This structure is rectangular, divided into two rooms, and opens toward the south into what appears to be a road that runs south and then curves west toward Structure 3. What appeared to be (and presumably were) narrow-gauge railbeds were also noted running down from the vicinity of the structures into the guano depression, but these were not mapped.
The lack of a beach along the northeast shore in 1989 does not mean there was not one there in 1937, but if there was, we found no evidence that anyone tried to land an airplane on it. Nor was there any evidence of an aircraft, or aviators, elsewhere on the island. As TIGHAR speculated before going into the field, it would be possible for an airplane to have "landed" in the ooze-filled pit (which from a distance resembles a lovely green field, or a shallow lagoon) and been obscured from view to the U.S.S. Colorado’s search pilots both by the ooze itself and by the birds that would inevitably alight on it. If this were the case, of course, it could account for the metal detector readings indicating metal in the pit. The alternative explanation for the readings — discarded mining equipment — accounts for the readings at least as well, however. Maude reports that when he and Bevington inspected the island on October 26, 1937 — less than four months after the Lockheed's disappearance — the "lagoon" was "almost pure salt water." Neither he nor Bevington report seeing any evidence of the Electra, Earhart, or Noonan. Considering these near-contemporary observations, the fact that we found nothing in the reports of the Smithsonian expeditions to suggest the presence of aircraft wreckage, and the negative results of our survey but for the ambiguous metal detector readings in the guano pit, it appears very unlikely that McKean Island was where the World Flight ended.
Though McKean does not appear to merit further work for purposes of the Earhart search, its ruins (and perhaps whatever lies in the guano pit besides the obvious) could yield useful data on the technology and sociology of 19th century guano mining, should questions about such mining arise that are amenable to archeological study. Study of the ruins might also provide some insight into the character of expatriate Hawaiian miner communities during the period. Certainly the ruins are worthy of preservation, particularly since such preservation requires no action other than discouraging the extremely rare visitors to the island from disturbing them.
Bigelow, Henry W., Jr.
- 1949 Some Notes on Sydney Island. ms, Department of Anthropology, Bernice P. Bishop Museum, Honolulu
Bryan, E.H., Jr.
- 1941 American Polynesia; Coral Islands of the Central Pacific. Tong Publishing Co., Honolulu
- 1935 Ocean Island and Nauru: Their Story. Angus & Robertson Ltd., Sydney, 1935
- 1862 On Phosphatic Guano Islands of the Pacific Ocean. American Journal of Science (2nd series) 34(101):224-243
King, Thomas F., Randall Jacobson, Karen Burns and Kenton Spading
- 2001 (2nd Ed. 2004) Amelia Earhart’s Shoes, AltaMira Press, Walnut Creek CA
- 1949 "Hull Island," ms, Bernice P. Bishop Museum, Honolulu
- 1938 Report on the Colonization of the Phoenix Islands by the Surplus Population of the Gilbert and Ellice Islands. F.W. Smith publishers for the High Commission for the Western Pacific, Suva, Fiji
- 1940 Report (Classif. Secret) on The Phoenix and Line Islands with Special Reference to the Question of British Sovereignty. F.W. Smith publishers for the High Commission for the Western Pacific, Suva, Fiji
Williams, M., and B. Macdonald
- 1985 The Phosphateers: A History of the British Phosphate Commissioners and the Christmas Island Phosphate Commission. Melbourne University Publishing, Ltd., Melbourne