The Ghosts of Roebuck Bay by Ian W. Shaw – Extract

The Ghosts of Roebuck Bay

Chapter 1

The Port of Pearls
By the 1940s, Broome had become a nondescript town on a lonely coast, falling into apathy and sunbleached of its colour and rapidly losing traces of its past.
– John Bailey, The White Divers of Broome

Broome has always been open to refugees and newcomers, to fortune-seekers and adventurers. In the first few weeks of 1942, however, Broome was just a shadow of its former self. Its glory days as one of the most colourful sea-ports in the southern seas were just fading memories for the few original inhabitants still living there. The small peninsula on which the township sits was one of the first areas of the Australian continent to be visited by Europeans. The buccaneer turned explorer, William Dampier, visited twice, in 1688 and 1699. Roebuck Bay, the inner stretch of water that framed the peninsula, was named after Dampier’s ship on his 1699 voyage, while the creek at its head, Dampier Creek, was named in his honour.

Dampier had been singularly unimpressed with what he found in the northwest of the continent, but other Europeans who followed just over 200 years later found riches that Dampier could not have known existed. The waters around the peninsula were home to enormous natural pearl beds. Shortly after the discovery of this underwater treasure, the township of Broome was firstly surveyed and then named after the then-governor of Western Australia, Sir Frederick Broome. From around 1880, Broome became a magnet for fortuneseekers from around the world. Some sought the pearls, while others prized the shells the pearls grew in; mother-of-pearl was in demand in Europe and elsewhere for use as buttons and brooches, and as inlays in furniture and fittings. Between 1880 and 1920, 80 per cent of the world’s pearl shell came from Broome. The town became home to more than 8000 people, every one of them from somewhere else.

The white Europeans learned early on that while the industry itself was attractive, the same could not be said for the process of retrieving the pearl shell from the seabed. At first, local Aboriginal people were hired or simply taken to dive for the shells. As the shallower waters were worked out, the Indigenous people were replaced by those who could dive deeper: Manilamen from the Philippines, Malays and Koepangers from Timor, specifically the area around Koepang (Kupang) in West Timor. The South Asian people were in turn replaced by Japanese divers but many remained in the industry as deckhands on the pearling luggers. The Japanese divers were also better equipped with the new technology that was entering the industry in the form of air pumps and diving suits.

The changing nature of the industry straddled the years when the bottom fell out of the pearl shell market. The disruption caused by World War I was followed by the development of cheap plastic alternatives to pearl shell during the 1920s which, in turn, was followed by the Great Depression. By the late 1930s, pearling luggers were still operating out of Broome, but where they once numbered in their hundreds, there were now merely dozens of craft. The white male Europeans in their white cotton suits were gone, too, replaced by Japanese owners and agents on the luggers and in the shops.

The Japanese dominated the pearling industry, albeit through a series of dummy companies which disguised that dominance; the White Australia Policy was still the law of the land. They also held a strong position in the Broome community as a result of this dominance. They even had their own club in the town, the Japanese Club, housed in one of Broome’s largest buildings, and it quickly became one of the most popular venues in town. Broome still has Australia’s largest Japanese cemetery; it is the final resting place for around 900 Japanese people who died seeking pearls and pearl shell.

By December 1942, though, the town’s population had dropped from a peak of over 8000 to around 1600, of which just 450 were European. The majority were Asian – Chinese, Japanese, Malays and Koepangers – and a large number were Aboriginal and mixed-race residents.

With the decline of the pearling industry came the realisation that Broome was nothing more than an outpost of civilisation on the edge of a vast and empty continent. The only overland connection to Perth was via a track that was rough at best and impassable at worst, especially during the wet season. The railway tracks in the region were as small as the route they traversed was short, from the far side of Broome to the end of the long jetty. They were used by the small steam engine that pulled flatbed carriages out onto the jetty with petrol and other supplies and, on their return trip, the bags of pearl shell that the luggers brought in.

The ships and luggers that used that jetty were Broome’s lifeblood until the early 1930s when an airfi was established on the edge of town on the track to Cable Beach. The town’s fi regular air service began in 1932, connecting Broome and Perth in a trip that took two-and-a-half days. In February 1942, there was a regular passenger and mail service connecting Broome with the cities and towns to the south. The contract to deliver that service had been won by MacRobertson Miller Airlines (MMA) and the pilot assigned to it, Jimmy Woods, was one of the most popular people to visit the town.

The coming of war in September 1939 brought increasing change to Broome; slow, incremental change at first, but small changes that were cumulative. By the end of 1941, virtually all the young men and women from the town had either enlisted in the armed forces or had moved to the large cities in the south to work in the jobs vacated by others who had enlisted. In a small town like Broome, their loss was a major blow.

If the war came to Broome, the town’s defence would be in the hands of the local equivalent of a home guard, the Broome Volunteer Defence Corps (VDC). Unfortunately, the VDC was not particularly impressive, on paper at least. Formed in July 1940, it could count on just 30 or so volunteers, who had only six weapons in their armoury, all World War I–vintage Lee-Enfield rifles, with 500 rounds of ammunition for each.

Their leaders were of the same vintage, but all had impressive records from the Great War. Their commanding officer and their intelligence officer, captains Harry Macnee and Lou Goldie respectively, had both served at Gallipoli and on the Western Front, and both had been commissioned from the ranks. Both had served in the 10th Light Horse Regiment, as had the units’ drill sergeant, Beresford Bardwell. So far, their activities had been confined to regular training and occasional guard details at the wireless station and the jetty, the main points of strategic interest in Broome. They prepared as best they could but nothing could have prepared them for the fateful events ahead.


There was a distinct social order in Broome at this time and those at the top were the families of ‘Old Broome’ when Broome was the ‘Port of Pearls’, the town where fortunes were made with the opening of a single shell. By 1942, those days were long gone. The mansions were boarded up or falling down, and most of the luggers had disappeared. The most important people in town were no longer the pearlers but those who had authority rather than wealth; with the passing of the pearl trade came the passing of power from the commercial to the administrative. Those who now exercised that power were Broome’s senior public servants, the most prominent of whom were the town doctor and senior policeman.

The town doctor was the English-born Dr Alexander Thomas Hicks Jolly, 32 years of age and married to Margaret with a small daughter. Jolly had only arrived in Broome in July 1940, and was still feeling his way in local society. As well as carrying the designation of resident medical officer, Jolly was also the town’s resident magistrate. While time proved him to be a competent doctor, Jolly would never really become part of the town’s establishment the way previous doctors had. The town may have been a little too ‘frontier’ for Jolly and his family, or perhaps he was a little too urban for that particular time and place.

The senior policeman was Inspector James Duff Cowie, a former city detective who had joined the Western Australian Police Force in 1910, and served with distinction since. Cowie was not especially happy about his Broome posting. It was a long way from where the real action was in the south, and the tropical climate had already affected his health through several bouts of dengue fever. Just to confuse everyone, the senior police sergeant at Broome was also named James Cowie – no relation, but enough of a coincidence to regularly raise a laugh. Another senior bureaucrat was the Broome inspector of Aborigines, the local representative of the state Native Affairs Department. The position was occupied by Laurie O’Neill, a former police sergeant who had spent most of his police career in the northwest, at places like Fitzroy Crossing, where he had been the district’s sergeant of police. O’Neill was efficient and popular with both the Aboriginal people he was responsible for and the Europeans he had to negotiate with.

Also close to the top of social hierarchy was Beresford Bardwell. As well as his role as VDC drill sergeant, Beresford also headed up the Broome Public Works Department and was a member of the Broome Roads Board. Beresford and his brother Bernard did not really qualify as ‘Old Broome’ as they had originally come to Broome around 1910, drawn from their native Melbourne by promises of adventure and riches from the sea. They had some of the former but little of the latter before the Great War broke out in 1914. Beresford enlisted and served with some distinction before being invalided back to Australia in 1918 because of wounds suffered in France. He and Bernard continued their pearling partnership after the war, joined by other family members, and then in 1920 their fortunes improved dramatically.

Beresford was out on one of the company’s luggers, going through pearl shells brought up by one of his Japanese divers. Opening one shell, he spotted a pearl the size of his thumbnail. Concerned about what might happen if the rest of the crew saw the pearl, Beresford slipped it into his mouth, and continued opening shells until he could ease away into his cabin where he locked the pearl securely away. That pearl later sold for £4000 (about $700,000 in today’s currency), and was one of the most valuable ever discovered in the northwest.

In the mid-1920s, the Bardwells got out of the pearling trade like so many others in Broome, seeing no future for the industry. Unlike many who left pearling, they didn’t leave Broome. Beresford and his wife, Biddy (Marjory), had three children – two sons and a daughter – and were determined to stay and make a life for themselves and their children.

In the 1930s Beresford was appointed to the position of harbour master for the port of Broome. When the long jetty there caught fire in 1937, his quick thinking prevented the destruction of the entire structure. When he arrived on the scene, a fire had taken hold in the centre of the jetty. He sent someone away for dynamite, which he used to blow out sections of the jetty on either side of the fire. The jetty was damaged, but it was quickly repaired in a few days.

Finally in the hierarchy of the town, there were a number of ‘new’ Broomites, families such as the Milners. Harry and Catherine Milner had come to Broome in the 1920s. Harry was an engineer, but in 1924 left that job and went into partnership with Leonard Knight to run the open-air cinema in Broome, the Sun Theatre. They did well. Their family now numbered seven – five daughters and two sons – and, like most in the town, they participated in the town’s social life. Catherine was the Girl Guide district commissioner and, through their children, involved in most things that happened in town. Unfortunately, Harry died in 1940. Catherine bought out Leonard Knight, took over the running of the theatre and tried to carry on with her life.


In 1941, Australia’s defence planners recognised that Broome had another point of potential strategic significance – its small airfield. Because it was there, in place, and was being used on a regular basis, Broome’s little airfield was designated an Advanced Operational Base (AOB), as were similar small airfields at Wyndham and Derby in the northwest.

A wireless telegraph station was built near the airfield on the same side as the town. Unfortunately, the messages it was designed to send and receive were in code, and no trained operators were available until sometime in 1942. Until then, the military authorities in Broome had to use the civilian telegraph with its copper wires running to the nerve centres of defence in the south.

It was the same with the plans for the airfield. The proposed upgrade would only be to the extent that it could handle two medium-sized aircraft arrivals and departures a day rather than the one or two that it now handled each week. There were no plans for radar, no plans for anti-aircraft defences and there were no plans for fighter aircraft to protect the facility.


The very thing that threatened Broome’s existence – its geographic location – suddenly made it very important in the first weeks of 1942. The tides of war were lapping ever closer to northern Australia as success followed success for the Japanese.

In Malaya, the Japanese landings on the northwest of the peninsula were a preamble to a measured advance to the southeast, through successive British defence lines towards the ultimate prize of Singapore. It was an advance characterised by a total domination of the air and by the use of dozens of small boats the British forces had failed to destroy as they retreated. The boats, packed with Japanese soldiers, would sail at night to landing places behind Allied lines. The troops would wade ashore to create havoc in the rear echelon areas. Australian defence planners could do nothing about the Japanese air dominance that emerged early in the fighting as they had no aircraft capable of matching the performance of the Japanese Zero fighter, but they were able to address the issue of the potential Japanese use of abandoned boats should the Japanese ever be so bold as to invade the Australian mainland.

In January 1942, Broome pearlers were informed that all their luggers would be purchased by the Australian government. A fair price would be paid for them, and any deemed unseaworthy would be destroyed. The remainder would be sailed south, to Fremantle most likely, where they would be put to work in supporting the war effort. Unspoken but implicit was the fact that removing all luggers from the northwest would also remove the possibility of them being captured and used by the Japanese. Eventually, 44 luggers were purchased at a cost of £80,000. A further sixteen luggers, found to be unseaworthy, were purchased for almost £9000, and burnt where they lay.

The task of crewing the remaining luggers and organising their despatch to Fremantle fell to RAN Reserve Lieutenant

D.L. ‘Beau’ Davis, who flew into Broome from Melbourne on 9 February after being fully briefed at Defence headquarters. The middle-aged sailor was a good choice for the role. He had many years experience in the northwest and at one time had owned and operated a fleet of six pearling luggers out of Broome. When there was a neap tide on 17 February, Davis floated his luggers on Roebuck Bay, giving them, literally, a water test before fitting them out for the voyage south. He had no idea when that voyage would begin, however, because he was struggling to find crews for his little armada.


While Beau Davis was selecting and preparing the luggers to sail south, the war situation was steadily worsening. On 15 February, Singapore and the 90,000-plus Allied defenders surrendered to the Japanese. In the Philippines, American ground troops had been forced back onto the Bataan Peninsula and their own island fortress of Corregidor. Their army, air force and navy units had all retreated steadily to the south, first to the Netherlands East Indies (NEI), and then to Australia where they joined other American forces who had arrived there. A reinforcement route for the units that remained to face the Japanese in Java had been established along the old Qantas Empire Airways (QEA) flying boat route, via Timor and Bali into Java itself.

The NEI slowly but inexorably fell under Japanese control. Japanese forces attacked Borneo in December, then Sumatra, the Celebes (Sulawesi) and elsewhere across the archipelago. Regular air raids on Kupang in West Timor and its associated airfield at Penfui and the shooting down of a QEA flying boat, the Corio, effectively closed the air route through Timor.

The only alternative reinforcement route was a direct route between Australia and Java. The two closest points were the south coast of Java and the northwest coast of Australia. There were several possible departure points in Java – Bandung and Jogjakarta for land-based aircraft, and Tjilatjap for flying boats – but only one realistic destination in Australia.

That destination was Broome, and in a few short days, Broome went from being almost an afterthought, an alternative landing point for aircraft in times of inclement weather, to perhaps the most important port town in the western half of the continent.

A lot of work had to be carried out in a very short time. First was the airfield, which was simply too small for many of the aircraft it was expected to handle; an upgrade of its main runway was needed. The main contractor would be Bell Brothers, an earthworks company based in Perth. Co-owner Alec Bell travelled to Broome, his heavy equipment following him a few days later, and there he found a ready-made workforce awaiting him. It being the wet and the luggers being laid up, there were deckhands and support workers who now had nothing but time on their hands.

Most of the workers were Koepangers or Malays. Bell scooped up 180 of them. An iron aircraft hangar and a number of small huts were constructed at the airfield in quick time, while a lot of work went into extending and, where possible, strengthening the runway. Maintaining the runway soon proved to be a never-ending task. While it coped with the lighter domestic aircraft that had been using it for a decade, the runway surface struggled with the heavier two- and four-engined aircraft that began to arrive in the second half of February. The twin-engined DC-3 Dakotas made depressions a couple of centimetres deep every time they landed; the big American B-17s (Flying Fortresses) and B-24s (Liberators) could gouge out 20 centimetres or more during a rough landing. When the airfield got busy, which it soon did, Bell arranged for many of the 180 labourers’ meals to be brought to them as they worked as maintaining the runways had become a full-time task.

The Japanese launched a massive air raid on Darwin on 19 February causing immense damage and substantial casualties there. That air raid destroyed the morale of most residents of Darwin, civil and military alike, because it showed them in harsh detail just how isolated and exposed they actually were. While disastrous for Darwinites, it did galvanise both civil and military authorities into further action. Shortly after news of that raid reached Melbourne, the War Cabinet meeting there issued its Minute Number 1916, which said,

War Cabinet confirmed the order for compulsory evacuation of women and children from Broome and approved of the proposal for the use of civil aircraft for this purpose.

The actual evacuation was organised locally at Broome. The War Cabinet Minute applied only to European women and children and some of them had already been sent south to Perth by air. In Broome, two of the senior local officials, Inspector James Cowie and Beresford Bardwell, took immediate and decisive action. Cowie’s police and Harry Macnee’s VDC assisted the process and on 21 February more than 120 women and children departed for the south aboard the steamship Koolinda. One who refused evacuation was Biddy Bardwell, who was able to convince the relevant official – her husband – that her work at the Broome telephone exchange was vital to the town’s operations. As there was no-one to replace her, Beresford agreed she could stay until someone else was trained to take over.

Broome’s non-Europeans were not subject to the same order. The Japanese were already gone, held (under very lenient conditions) in the Broome gaol before being sent to internment camps in the south, their lovely club shut down. The Asians who remained – the Chinese, Malays and Koepangers – were by and large ignored. Most of them were not Australian citizens and many were not even Australian residents. They had shown, however, that they could cope with just about anything the northwest could throw at them and the majority had spent much of their lives in Broome. Even if it were possible to return them to their places of origin, most of those places were now either behind Japanese lines or sat directly in the Japanese line of advance. They were left to fend for themselves.

The Aboriginal and what was officially termed ‘part-Aboriginal’ population were subject to a different set of rules and regulations. Leprosy had been a scourge in the Aboriginal community and, because of this, a state law prohibited the movement of local Aborigines below the twentieth parallel of latitude. Local officials and the Native Affairs Department had factored this into their planning. Between 250 and 350 Aboriginal people, mainly women and children, were transported to the Beagle Bay Aboriginal mission, run by German monks from the Pallotine order, and located some 150 kilometres to the north of Broome.

The influx from Broome more than doubled the Beagle Bay population, and the Native Affairs Department made arrangements to assist the missionaries with building materials for the construction of new dwellings and with the promise of regular deliveries of additional rations to the mission.

Not all the Aboriginal and part-Aboriginal population were relocated to Beagle Bay. Those who had ongoing work in the town were allowed to make up their own minds about whether they would stay or go while others, sometimes described as ‘indigent natives’, were allowed to remain in their camps at Bones Well and Fishermen’s Bend, just outside the town boundaries. The Native Affairs Department indicated that it would also supply both those camps with rations on a regular basis.

All of this was undertaken according to a prepared timetable. However, it was the wet, a season when timetables could become hit-or-miss affairs. A tropical cyclone, the third of the season, had swept through the area a couple of weeks earlier, and there had been heavy rains since. When those rains came, roads were impassable. Under the new timetable, the last truck would carry just a few remaining children from Broome to Beagle Bay – orphans from the newly built orphanage – and they would be accompanied by the one other white woman who had stayed behind, Sister Catherine Hayes. Sister Hayes and her charges were to leave Broome sometime during the morning of Tuesday, 3 March.


Qantas, through QEA, had responded almost immediately to Japan’s explosive entry into the war. Singapore was the terminus of QEA operations to the west, and the company soon abandoned all its regular commercial flights to the island. In January, with the loss of Singapore looking more likely, Batavia (Jakarta), the capital of the NEI, located on the north coast of the island of Java, became the westernmost port for QEA flights, with Bali, Kupang in Timor and Darwin remaining the regular stops on that route. Disaster struck on 30 January when one of the QEA flying boats, the Corio, was shot down by Japanese fighters near Kupang on its mission from Darwin to Surabaya, killing thirteen of its eighteen passengers and crew, mostly Dutch women and children evacuees.

In the wake of that loss, the civil aviation department immediately instructed Qantas to bypass Kupang, and to organise an ‘effective connection’ from Java to the most convenient point of entry along the Western Australian coast. Qantas sent Lewis Ambrose, its senior QEA pilot, to Batavia to liaise with the relevant NEI authorities to determine which port to use.

In Batavia, NEI civil aviation authorities told Ambrose that the pace of the Japanese advance meant that both Batavia and Surabaya would soon be within range of land-based Japanese aircraft, and recommended the use of Tjilatjap (now Cilacap) on the southern side of Java. Ambrose agreed and organised for Malcolm Millar, a senior QEA representative in Singapore, plus some ground crew to be sent there. Millar was an ideal choice for the task. An experienced Qantas and QEA administrator, it had been Millar who established the Singapore base for QEA’s operations, and who had, in the years since, built up a strong network of contacts throughout the region. Within a few days, the picturesque little town, a short distance upriver from the Indian Ocean, would become one of the busiest ports in the world.

With agreement on Tjilatjap, Broome’s selection as QEA’s Australian reception point was a mere formality. While Millar and his small staff set up at Tjilatjap, Qantas despatched one of its best men, Captain Lester Brain, to Broome to oversee the operations there. Again, it would be hard to find a more qualified person for the role. Brain was widely regarded as the best all-round pilot in both Qantas and QEA, and he had pioneered many of their domestic and international routes. A man of medium height and build, with a bright, open face, Brain’s appearance sometimes belied the fierce energy which burned within. Brain coordinated the movement of ammunition and supplies to the NEI and the evacuation of personnel from there and, specifically, from Java. His task was a relatively simple one, or so it had seemed when it was presented to him. Lester Brain was to put together anything and everything necessary to anchor the Australian end of the QEA evacuation program at Broome, its Australian point of entry. He arrived in Broome on Saturday 21 February and immediately set to work. A week later, Brain would celebrate his 40th birthday; he hoped it would be a good one.

At the other end of the air route, Millar approached the Allied evacuation centre to apprise them of QEA’s progress. He was not surprised to learn that the first batch of high-value evacuees had been identified and were standing by for departure on short notice. He was surprised, though, to learn that they were all female secretaries to high-ranking Allied officers.


Brain’s orders from Qantas headquarters in Sydney and the Civil Aviation Department in Melbourne were that Brain and his staff would cooperate with the US Army wherever possible, but Brain would retain responsibility for the QEA flying boats. It was hardly an ideal situation, but it did mean that, in one area at least, Australian interests were being looked after by Australians.

Brain’s QEA operations in Broome would eventually directly involve at least fifteen people from Qantas’s air traffic, engineering and marine divisions, and at least one other manager. Many of those staff arrived on QEA flying boats from their original bases in Singapore, Batavia, Bali or Kupang. A fully furnished cottage was rented as a central point for QEA staff and operations, and outfitted to accommodate up to six QEA aircrew at a time. The remainder, if any, would stay in one of the town’s hotels. To coordinate the ground operations, accommodation and catering, Brain had one of QEA’s senior stewards, John Oram and a purser named Baron.

Brain also turned his attention to the complexities of the flying-boat operation on Roebuck Bay. Shortly after being requisitioned by the Royal Australian Navy (RAN), the coastal lugger, the Nicol Bay, and its captain, Harold Mathieson, were put to good use. The Norwegian-born Mathieson and his Fremantle-built lugger had spent most of the previous decade working up and down the North West coastline acting as a lighter for a Perth company. The tidal range in that part of Australia was extreme, and little ships like the Nicol Bay were needed to carry and deliver freight to places larger vessels were unable, or unwilling, to travel. Working primary out of Port Hedland, Mathieson and his locally-based crews had become a fixture in that part of Australia.

The Nicol Bay was instrumental in creating three flying-boat moorings at Brain’s direction. Heavy anchors and heavy chains held the mooring floats in place so that aircraft could land on the water, taxi to the floats and tie up to them. Passengers, luggage and cargo would then be ferried to the jetty by small craft.

That was the theory. The moorings had to be placed in deep water because of the tidal range. Brain soon discovered that, at low tide, the seaward end of the jetty stood almost ten metres above the sand and mudbanks. To get to and from the moorings then involved a climb up or down the stairs at the end of the jetty, a walk of several hundred metres to deeper water and then a boat ride to wherever the aircraft was moored.

Passengers and crew would have to carry their own luggage for the whole distance. Even if the loading and unloading took place at high tide, there were complications.

To Brain’s dismay, there was a distinct shortage of suitable boats in this port town. He could locate only a couple of rowboats, one small motor dinghy with an unreliable engine and the fleet of luggers waiting to be sailed to the south. After some enquiries, he was able to find and buy a motor launch and a motorised dinghy.

The Nicol Bay was ideal for refuelling the QEA flying boats, and with a stripped-down lugger in tow, could carry more than enough fuel for all flying boats it needed to service. If the tides were right, it could also ferry passengers and crew to and from the jetty. Brain’s final plan was to minimise delays to the shuttle service. Each high tide was to be used to the full and, where possible, the flying boats would be refuelled either immediately after their arrival or just before departure.

The entire operation was in place by the night of 22–23 February, and it worked as well as Brain had hoped it would. After the experience with the Corio shoot-down, radio silence was enforced between Tjilatjap and Broome, with each base simply signalling aircraft departures and arrivals. Ambrose, who was familiar with the route, flew it as often as possible. On the outward leg from Broome, the flying boats carried such things as medical supplies and aircraft spares, while inbound flights from Tjilatjap carried up to 25 passengers, mostly Allied servicemen.

Just a couple of days into the service, Brain was informed that evacuation had taken precedence over reinforcement and resupply, so the big flying boats began flying to Java carrying nothing but their aircrew. On 27 February, Ambrose flew the Coriolanus into Broome and informed Brain that Allied headquarters in the NEI believed that a Japanese invasion of Java was imminent.

Brain was well aware of the forebodings. His diary entry for 26 February reveals that he was aware of just how precarious the situation in Java actually was, but also noted that there were political imperatives to keep the evacuation route open. He concluded his diary entry with: ‘Millar agrees we will remain there and that we shall continue on a day to day basis.’ His diary entries around that time also reflected his growing pessimism. A practical man, Brain recognised that shutting down the evacuation operation too early would leave QEA, and Australia, open to criticism from the Americans who were still operating their own evacuation flights from Bandung. It was a decision he agonised over making, noting that: ‘The position of Java is apparently hopeless and it is now a case of getting as many useful people out as possible.’

In the end, the decision was taken out of his hands. The next day, Millar in Tjilatjap received a radio message saying that two flying boats, the Corinthian and the Circe, had departed from for Broome. The Corinthian arrived safely on 27 February but no trace of the Circe or the twenty passengers and crew she carried were ever found. Later that day, Millar and his staff were told that civil aviation had suspended the QEA shuttle and recalled all aircraft to the west of Broome. The Coriolanus, which had taken off earlier for Tjilatjap, returned to Roebuck Bay.

Millar and his staff were directed to report to US Army authorities in Jogjakarta for their evacuation to Australia.

Shortly after 11 p.m. on 1 March, Millar and his team boarded a B-17 Flying Fortress at Jogjakarta airport for Broome. There, they found Brain, suffering from dengue fever, waiting for them. They also found the rest of the town waiting for something to happen, but not knowing what it was they awaited.


Australia’s efforts to extricate its people from Java to Broome depended very much on the individual efforts of Lester Brain, Malcolm Millar and Lewis Ambrose and their small band of air and ground crews in Tjilatjap and Broome. Fortunately, they were supported by the people of Broome – those who had not been evacuated anyway, for Australia’s military and civilian leadership could offer little beyond words of encouragement and formal directives. In many ways, this was in direct contrast to the American approach, which identified impediments to the desired outcome and then simply bulldozed them out of the way – mostly figuratively, but sometimes literally. For instance, they paid more for anything they wanted than the locals could afford. It might not have made them many friends, but it was effective.

The US Army wanted their air force servicemen evacuated; most were members of various USAAF squadrons who had been rushed to the NEI from Australia to shore up Allied resolve and resistance in Java. A few remnants of US forces from the Philippines did fall back onto the NEI; several Catalinas from PatWing 10 (Patrol Wing 10) had escaped to Ambon and, when they were bombed out of there by the Japanese, fell back again to Surabaya and then to Tjilatjap. The US personnel would eventually be concentrated in and evacuated from either Jogjakarta in East Java or Bandung in the west of the island. They were flown out in big four-engined bombers, the B-17s and the B-24s, which could fly to Broome without refuelling.

The fi decision to evacuate the remaining 450-plus USAAF specialist flight and ground crew in Java was made by Lieutenant Colonel Eubank, commander of all the US forces there.

On 22 February, the United States Army Air Force (USAAF) Colonel Edward Perrin flew in to Broome to coordinate the American forces’ transit through Broome’s airfield. Perrin believed he was in command of this part of Australia because of an anomaly in Allied command arrangements. In an earlier attempt to coordinate Allied efforts against the Japanese, ABDA (American, British, Dutch, Australian) command had been established. Under its terms of reference, ABDA command held discussions with the governments concerned and then took nominal responsibility for a large slab of the Australian mainland – the area lying northwest of a line drawn from Onslow on the Western Australian coastline to the southeast corner of the Gulf of Carpentaria. Broome and Roebuck Bay fell clearly within that area.

After surveying the facilities, Perrin sought out Lester Brain to outline what he would take responsibility for. They were broad orders he carried, said Perrin, but they could be easily summarised: he was in charge of everything, and that included the loading and movement of all aircraft into and out of Broome. Brain said simply that he would seek some direction from senior officers in government about the best way to coordinate their various operations.

Perrin then sent out an urgent request for assistants. From Java came Lieutenant John Rouse and a captain named Schwanbeck, both B-17 pilots. Schwanbeck was appointed maintenance officer and given a crew of mechanics and technicians whose role was to ensure the aircraft were refuelled, serviced and sent on their way as quickly as possible.

Schwanbeck quickly learnt that turnaround time depended on a number of factors, some of which were well beyond his control. Fuel, for instance, had to be brought to the airfield from the dispersed fuel dumps by two old, civilian-registered trucks and then hand-pumped into the empty aircraft. The trucks’ European drivers and their Aboriginal assistants seemed to speak a language that only they understood, and were only able to work at a single, constant pace – slow. A lack of servicing also caused frequent breakdowns and tyre failures on the aircraft. Schwanbeck and his men faced trying times.

John Rouse felt the same. He flew in from Java on 24 February and assumed a role as rationing officer. Rouse took over the now vacant Broome school, which he set up as a mess area, with cooking facilities in the playground behind the schoolhouse. He seconded some American-enlisted men to help him and, for the first couple of days at least, relied on the assistance of a couple of local volunteers and at least one Catholic nun.

The facilities Rouse established seem to have worked quite well for the first few days, but proved inadequate when significant numbers of evacuees began to pass through the town. He had extra staff and kitchen equipment flown up to Broome from Perth. Once he could look beyond the next meal to be served, Rouse decided to bury food caches behind the town in case of a Japanese landing. He marked the location of the caches on a Caltex road map he bought from a local shopkeeper.

The first American evacuees arrived in Broome on 25 February, and the following day two B-17s arrived, each carrying twelve evacuees. These successful flights were followed by more, then more, and within a week every US serviceman on their priority list had been evacuated from Java, through Broome and on to either Perth or Melbourne. The first aircraft from those southern cities flew into Broome on 26 February.

The Americans also evacuated other Allied military personnel and, in the case of QEA at least, civilian non-combatants were among the earlier evacuees despatched to Broome. The circumstances of the evacuation could be challenging. In the big B-17s and B-24s, passengers had to cram as far forward as they could for take-off – if there was too much weight towards the rear of the aircraft, the pilot simply couldn’t get the tail off the ground. Battle stations were manned after take-off and the gunners remained in their turrets until the flight was halfway to Broome and presumably safe from Japanese fighters. Most flights left Java late at night and at Jogjakarta, American ground crews would create a flight path by tossing kerosene flares onto the sides of the runway from a moving car.

While its overall success cannot be discounted, the American operation was not without its own controversies, caused primarily by a lack of communication. From the beginning of its Broome operation, the USAAF believed that Allied agreements made elsewhere gave it the imprimatur to take command of all military assets in Broome. While this may have been clear to those who made the agreements, those who were on the ground, thousands of kilometres and several weeks away, were never given the same direction. The Americans assumed everyone knew they were in charge; the Australians (and Dutch, at times) assumed that the Americans were again being arrogant and overbearing. The lack of communication was in turn needlessly complicated by a lack of demarcation and a glaring absence of clear lines of authority and reporting. ABDA command was dissolved on 25 February, and when it was, the Americans simply stepped into the vacuum and assumed authority. Unfortunately, and in the words of one of the Americans present in Broome during those critical days, ‘The US Army Air Force did not particularly mix with the Navy or the Australians or the Dutch or the civilians or anyone else.’

That observation was made by Second Lieutenant John Minahan, a 27-year-old bombardier with the 7th Bomb Group who flew into Broome from Java on 27 February. Minahan was ordered to remain in Broome by Perrin to assist with the evacuation operation. A keen diarist, Minahan opened his observations of Broome with, ‘Our principal purpose was to save the combat force’. Everything else – Allies, foreign countries, civilians, cooperation – was subservient to this purpose, an approach that would come to grate with many people.

On the afternoon of Sunday, 1 March, Lieutenant Colonel Richard Legg flew into Broome to relieve Perrin as the US commanding officer. Those who knew Legg, whether as an equal in rank or as a junior, seem to have shared an opinion of him, and that was not particularly positive. A short man who seemed to try to project a larger, more aggressive version of himself to others, Legg was also prone to both hyperbole and self-aggrandizement. Generally, he was a man who generated neither respect nor confidence in other people. Legg had previously been with the 17th Pursuit Squadron in Java, and may even have spent time in transit at Broome before being sent there to take charge. Legg looked around the town, spoke to some people, and decided that he didn’t really like Broome and its inhabitants. Broome was a small town in a large continent which was part of even larger war. It was a long way from the nerve centres of that war, and was therefore somewhere Legg preferred not to be. Legg’s feelings were reciprocated almost immediately; most of the Australians who met Legg took an instant dislike to him.


Purely in terms of evacuation to and through Broome, until 1 March, the most affected by what was happening in the NEI – the Dutch colonists, civil and military – had been the least involved. For 300 years, the NEI had been administered by the Dutch and protected by Dutch arms. The longevity of the Dutch occupation may have encouraged feelings of invincibility among the Dutch colonists and military forces. If so, they were in for a rude awakening; rude, but slow. The first Japanese troops to invade the NEI did so at Miri, in Borneo, on 17 December 1942. After that, more landings forced the Dutch to fall back onto Java, the NEI heartland.

Even then, there was a belief that the NEI might survive if Java could hold out long enough. Two events shattered that illusion. The first was the naval Battle of the Java Sea, which began on 27 February and lasted, with subsidiary clashes, for two days. In that battle, the Japanese destroyed what was left of the Dutch Navy’s NEI fleet, and so ended any faint hope the Dutch may have held over preventing Japanese control of all the sea lanes the way they now controlled the air.

The second was the invasion of Java at two points on the island’s north coast on 1 March, landings that met with little more than token resistance. With the fall of the NEI now in sight, it was every man for himself. The Dutch rush to Broome had begun.

Excerpted from The Ghosts of Roebuck Bay by Ian W Shaw. Copyright © 2014 by Ian W. Shaw.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Pan Macmillan Australia solely for the personal use of visitors to this web site.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s