Earhart Project Research Bulletin #64
September 4, 2012

“Better Than Average Luck”

Recent research by members of TIGHAR’s Earhart Search Forum has uncovered documentation of an extraordinary event that occurred in the Southwest Pacific just nine months before Earhart and Noonan disappeared.

“That we survived at all is, of course, an amazing piece of good fortune. With only average luck the machine would have just vanished…”

The crew of an aircraft similar in size and performance to Earhart’s Electra, on the second try of an attempted record-setting flight, got lost on an over-water leg due to radio direction finding problems.

Unsure of his position and running low on fuel, the pilot spotted a coral reef that looked dry and smooth enough to land on. He elected to make a wheels-down landing and the aircraft rolled to a stop undamaged. The crew then used their radio to call for help but could not establish communication with the would-be rescuers. When the aircraft failed to arrive at its intended destination, government authorities launched a sea and air search that proved fruitless. Luckily, there was a fishing boat in the area. The aircraft was abandoned but the crew was rescued and lived to tell the tale, complete with photographs.

Over the years, with no knowledge of this incident, TIGHAR pieced together from archival, photographic, analytical, and anecdotal evidence, a sequence of events describing what Earhart and Noonan may have done after the last in-flight radio message heard by Itasca. The uncanny parallels between the documented events of October 7, 1936 and TIGHAR’s hypothetical reconstruction of events of July 2, 1937 do not prove that TIGHAR’s scenario is accurate but they do demonstrate that it is plausible.

Croydon Crew

On October 7, 1936 British pilot Harold H. “Timber” Wood, engineer L. Davies, radio operator C. Gilroy, and the aircraft’s designer, Frederick F. Crocombe, took off from Darwin, Australia for Koepang on the island of Timor. They had recently completed a flight from London to Melbourne but failed to set a hoped-for record due to a long hiatus in Karachi to repair damage from a hard landing. The return flight was intended to redeem their reputation by beating the existing Australia-England record of 5 days, 15 hours.

The aircraft was the General Aircraft Ltd ST-18 Croydon, the one-off prototype of a design aimed at stemming the rising tide of American dominance of the commercial aircraft market.

The crew of the ST-18 Croydon. At far left, F. F. Crocombe; third from left, “Timber” Wood. We do not know which of the other two men is which. Photo courtesy National Library of Australia.

General Aircraft Ltd ST-18 Croydon   Lockheed Aircraft Model 10-A Electra
Configuration Twin-engine, low wing monoplane with retractable landing gear   Twin-engine, low wing monoplane with retractable landing gear
First flight November 16, 1935   February 23, 1934
Crew 3 (pilot, co-pilot, radio operator)   2 (pilot, co-pilot)
Capacity 10 passengers   10 passengers
Length 43 ft 3 in   38 ft 7 in
Wingspan 59 ft 6 in   55 ft
Height 13 ft 15 in   10 ft 1 in
Wing area 454 ft2   458.3 ft2
Empty weight 8000 lb   6454 lb
Gross weight 11,350 lb   10,500 lb
Powerplants 2 × Pratt & Whitney R-985 Wasp Junior SB 9-cylinder radial piston engine, 450 hp each (option for P&W Wasp R-1340 S3H1, 550 hp each)   2 × Pratt & Whitney R-985 Wasp Junior SB 9-cylinder radial piston engine, 450 hp each (Model 10E: P&W Wasp R-1340 S3H1, 550 hp each)
Maximum speed 203 mph   202 mph
Range 900 miles   713 miles
Service ceiling 19,500 ft   19,400 ft

The most striking and esthetically unfortunate feature of the Croydon was the sacrifice of a cantilever wing for the sake of passenger convenience. Lockheed Model 10 and Boeing Model 247 passengers had to step over the main wing structural member to get to the forward-most seats. Croydon passengers enjoyed a clear aisle at the expense of drag-inducing external wing struts.

Croydon in flight

Wood and his three crew members departed Darwin in pre-dawn darkness for the anticipated two and a half hour flight across the Timor Sea to Koepang. They soon noticed a discrepancy between their compass course and back-bearings provided by the Darwin direction finding station that indicated they were straying north of course. Suspecting compass trouble, they altered their course southward. When land did not appear on schedule they tried for another back-bearing but Darwin was out of range.

In an article titled “A Reef In Time” published in the December 10, 1936 issue of Flight magazine, the airplane’s designer Frederick Crocombe described what happened next.

We descended to 3,000 ft., the visibility being hindered by the fact that there were still some clouds about, and it was inclined to be misty with the rays of the sun shining directly into them.

Earhart, too, descended to stay below the clouds and was probably hindered by sun glare.

At 0800 hrs. we sighted a reef, circular in form, and apparently just below the surface of the sea. We took this as an indication that land might not be too far away and commenced to search for it, using the reef as a base. We discovered that there were three reefs spread over a fair distance in a northerly direction. In conversation with Wood as to his course, I formed the impression that the northerly course was the most likely to produce results, and we concentrated our search mainly in this direction. Gilroy was hard at it on the radio trying to contact somebody so as to find the location of the reefs, but with no result. Koepang seemed to be off duty, and we were out of touch with Darwin. The fuel supply was rapidly decreasing. We had started out with 235 galls., which should give us 5.22 hrs. duration at a consumption of 45 gall./hr., which was a fair figure in view of our previous consumption observations.

Earhart had was attempting a much longer flight with far more fuel, but a fuel crisis is a fuel crisis and an airplane low on fuel must find someplace to land.

At the northernmost reef we observed a fishing boat and, coming low, we threw out a can containing a message asking for the direction of Koepang. There was no response. We were by now really worried, and it was 0900 hrs. We had just about one hour’s fuel left, and at no time during the search from the reefs had we seen any sign of land. We did not know where Koepang was except that instinct indicated a northerly direction. This was hardly sufficient justification since, if land was more than 180 miles away, we should have to come down into the sea. On the other hand, the reef was below and a landing looked possible. Also, there was a boat which might be able to help us.

Replace “Koepang” with “Howland” and, except for the fishing boat, these words could have been written by Earhart.

reefWood and company landed on Seringapatam Reef  in the Timor Sea. Google Earth shows it to be a circle of coral that is not quite an atoll, with no real land and no real lagoon but, like Nikumaroro,  the reef surface does dry sufficiently at low tide to permit an aircraft to land.

By 0915 hrs. we had definitely decided to land and made a trial approach after extending the chassis, just touching the wheels to test the surface; this was found to be firm, but covered with a thin coating of water. Wood then circled for the last time, Gilroy and Davies being sent aft, leaving the two of us forward. I put the wing flaps half down, and Wood flattened out to make a fine three-point landing, pulling up quickly on the rough surface, and managing to avoid scattered groups of boulders.

The photos show that Wood made a successful landing on a reef that was considerably rougher than Gardner’s.

ST18 on reef

Courtesy Flightglobal. Used by permission.

Niku Reef

The reef looked higher a little further on and, since we appeared to have landed undamaged, we started to taxi towards it, but it was rough going for the tail wheel. There was a crack aft, the tail dropped, and the elevator control went soggy, so that was that. Wood had at first sight treasured the remote idea of getting fuel sent to the reef, and having a shot at taking off again. After looking at the surface close to, we surmised that this would probably have been fatal.

TIGHAR has postulated that, after landing, Earhart taxied to a higher part of reef and we’ve wondered whether she had hopes of contacting Itasca and getting them to bring fuel so that she could take off and fly to Howland and continue the world flight.

We scrambled out to look at the damage; the tail wheel had fallen through a thin crust of coral, fracturing the yoke casting, and the machine was about ten yards from the seaward west side of the reef. Here the reef fell away very steeply, and heavy breakers were pounding.

There are no “thin crusts of coral” on Gardner’s reef so the kind tail-wheel failure the Croydon experienced seems unlikely, but the aircraft’s distance from the reef edge (ten yards) and description of the west side of the reef are identical to what we have envisioned Earhart and Noonan experiencing.

Our immediate task was to try to establish W/T communication with Koepang, so we ran out a trailing aerial, using the two oars of our collapsible boat to insulate it from the reef. Gilroy could hear both Koepang and Sourabaya W/T stations, but was unable to make contact with either of them, despite the fact that the radiation in the aerial was just over 2 1/2 amps. In an endeavour to improve the transmission and recharge the aircraft battery, which was running down, we started up the port engine, which was equipped with a generator.

Of course, Earhart and Noonan did not have a trailing wire antenna but there is evidence that they were able to hear some transmissions from Itasca and the powerful commercial stations in Hawaii, presumably over the Bendix MN-5 loop antenna. There is also compelling evidence that, like the crew of the Croydon, Earhart and Noonan tried to call for help and ran their generator-equipped engine to recharge the batteries.

At 1115 hours we decided that we must be out of range of the W/T stations in the vicinity, and closed down, our position still unknown. The tide had by now begun to rise. It was obvious that our best policy was to try to get on board the fishing boat in the lagoon. Our subsequent adventures were not so much of aeronautical as of geographical importance.

Sadly, there was no fishing boat standing by to rescue Earhart and Noonan but there was an island to which they could retreat.

That we survived at all is, of course, an amazing piece of good fortune. With only average luck the machine would have just vanished, and a verdict of structural failure, or of engine trouble, would have been our epitaph. Even after a successful landing on the reef we would not have survived long if there had been no means of getting away from it: we might have lasted four days on our emergency rations, with the high tide level of the water over the reef rising every day. We actually landed on the reef at low water, during a period of neap tides.

The same was true of tidal conditions at Gardner Island nine months later.

At high tide on that day the reef was covered to a depth of 3 ft. A week later this would have increased to 14 ft., which meant that the aircraft would then be covered, and the heavy ocean swell would soon break it up. By the time we were able to get information through to the outside world it was therefore too late, and the reef too inaccessible for any salvage work to be done.

High tides at Gardner never approach 14 feet but heavy ocean swells are common.

Surf on Niku

With the machine relatively undamaged, due to Wood’s magnificent work, this was tragic, and it was one of the hardest things I ever had to do to leave the aircraft stranded, through no fault of its own, standing up like a monument in the clear atmosphere until the sea eventually claimed it.

The ultimate fate of the ST-18 Croydon is somewhat in doubt. In a 1995 interview La Ode Ndoke, an elderly local man who had been on another boat in the area that day remembered:

“We went to have a look at the plane afterwards and measured the wingspan — it was 8 depa [fathoms] long. The frame of the plane is still there to this day.”

A fathom is six feet so Ndoke measured the wingspan as forty-eight feet. The Croydon’s wingspan was over fifty-nine feet, but some of the wing may have been washed away. 

Tarawa debrisAirplane structures and engines can survive on coral reefs for many decades, as evidenced by debris on the reef at Tarawa dating from 1943 and photographed by TIGHAR fifty-eight years later in 2001. But these objects are far from the reef edge and protected from the effects of heavy swells. That anything of the ST-18 Croydon would survive on Seringapatam Reef after fifty-nine years is difficult to believe.

tarawa propPerhaps the most intriguing question about this incident is whether Earhart and/or Noonan was aware of it.  The article in Flight was published in December 1936 at a time when Earhart was actively planning her world flight. If she knew about the Croydon’s successful emergency landing on a dry coral reef did it influence her actions when presented with a similar situation less than a year later?

Electra on reef

Archived Research Bulletins Earhart Project Home Page

About TIGHAR Join TIGHAR TIGHAR Projects TIGHAR Publications Contract Services


Copyright 2021 by TIGHAR, a non-profit foundation. No portion of the TIGHAR Website may be reproduced by xerographic, photographic, digital or any other means for any purpose. No portion of the TIGHAR Website may be stored in a retrieval system, copied, transmitted or transferred in any form or by any means, whether electronic, mechanical, digital, photographic, magnetic or otherwise, for any purpose without the express, written permission of TIGHAR. All rights reserved.

Contact us at: info@tighar.org  •   Phone: 610.467.1937   •   JOIN NOW