HomeWelcomeUpdatesCompanyAircraftPeopleProjectsEquipmentOperationsPhoto AlbumGuest BookSearchAdastrianaQuestions


by Wal Bowles

VH-AGX landing at Madang on 27th June 1962 with the starboard engine shut down.
Photo: Wal Bowles Collection


The Madang fire tenders await the arrival of VH-AGX on 27th June 1962.
Photo: Wal Bowles Collection



"Congratulations on handling of Lae landing

Lou Pares

General Manager

Adastra Aerial Surveys"

At the time this telegram arrived in Lae, Papua New Guinea, I felt I could have done without it, even though I appreciated the thought behind it.  Telegrams are of course confidential, but I didn't want information to drift surreptitiously into areas of officialdom which might prompt more paperwork.  The flight from Madang to Lae and the landing were still vivid in my mind; Jack's "fright" was shared by all three on board, a flight which could easily have ended in disaster.

The telegram referred to a landing in Lockheed Hudson aircraft VH-AGX on 31 July 1962.  I was the pilot-in-command, and Jack McDonald and Brian Smith were innocent passengers.  Jack was the Chief Engineer of Adastra Aerial Surveys and Brian was the engineer assigned to service the aircraft as part of the field crew.  Lou Pares sent the telegram a day or two after the landing, soon after Jack returned to Adastra Head Office, Sydney.  Jack never flew in a Hudson aircraft again!

To write with accuracy of an event which occurred some 30 years ago is a little daunting.  However the impact of the moment was so profound that I still recall the circumstances with a mixture of feelings, not the least being wonderment that we survived.  Jack still raises the subject occasionally in the convivial atmosphere which exists whenever ex-Adastra crew members get together.

About ten years ago at a civil aviation function a person who looked familiar to me, but whose name I could not remember, introduced himself and said: "The last time I saw you was at Lae in 1962 when you landed a Hudson aircraft in atrocious weather conditions.  It should have been written up as an epic flight.  I was an air traffic controller in the tower at the time." It must have made an impression on this experienced controller to remember it as he did all those years ago; it certainly made an impression on me!

When several people bear witness to the same event, invariably there are discrepancies, or at least shifts of degrees of emphasis of some aspects.  Should others who were involved read this account, their recollections may vary somewhat from my own.

But it is not so much the detail of the event which has remained in my memory, even though this was a life-threatening situation with all the attendant internal stirring of the glandular orchestra; rather it was a sudden unexpected change in myself which is difficult to describe - perception? consciousness?  I do not know, but it seemed to arrive of its own accord apparently from my internal reactions.

To this day I remain astonished by this unforgettable experience and I still have difficulty in explaining it, even to myself.

Most pilots experience a "close go" now and then.  And "automatic cool" is an expression sometimes used by pilots to describe what they can hear on radio communication channels in the voices of their fellow aviators in emergency circumstances or in circumstances requiring a more than ordinary degree of concentration to avoid a possible disaster.  This was something different.  But I have little doubt that what I experienced is something which is common to all people who accept a life threatening situation from which there seems little chance of escape.

In 1962 I was the unit manager of an aerial survey project based in Lae.  I had flown Lockheed Hudson aircraft VH-AGX from Sydney to Lae in January of that year.  The Hudson was originally a twin-engined light bomber of a type used during World War Two.  For its time, it was an aircraft of high performance.

Our task was the methodical photographic coverage of much of Papua New Guinea on an opportunity basis, as often as weather conditions would allow.  Completely cloudless conditions were a requirement for the quality of photographs demanded, but completely cloudless conditions rarely exist in the tropics.  A degree of cloud was tolerated for survey flying in PNG, the same area sometimes being photographed more than once with the expectation that surface features obscured by cloud on one survey attempt might be discernible on another.

In PNG the warm, moist, tropical air and high, steeply rugged mountain ranges set the stage for an interplay of factors which result in weather conditions ranging from mist and fog in river valleys to deep, dense cloud masses, large areas of persistent torrential rain and violent tropical thunderstorms.  Changes in weather patterns can occur with little warning.  PNG weather is subject to the vagaries of the "inter-tropic convergence zone", the boundary between airstreams originating in the northern and southern hemispheres.  When convergence is extreme, cumulo-nimbus cloud can build to 40,000 feet or more.  The day to day movement of the inter-tropic zone is erratic, but it would normally be well to the north of Papua New Guinea in July when the flight took place.

To describe the reason for the flight in question requires retracing the events of the previous flight a month earlier.


The Lae crew of VH-AGX consisted of Jack Tierney navigator, Bob Jones camera operator and myself as pilot. (Bob Jones had been recruited in Lae to replace Tony Burgess who, in recent weeks, had returned to Australia to be married.  Tony later became a senior airline check captain.) Brian Smith, the maintenance engineer, completed the field crew.  Part of Brian's task was to perform a daily inspection of the aircraft preflight, to meet the aircraft on our return, and to attend to any unserviceabilities.

On the morning of 27 June 1962 we took off from Lae on a survey flight over central western New Guinea.  We climbed to 25,000 feet and completed several photographic runs.  Because the weather was reasonably clear further to the north-west, I headed for another survey area towards Wewak.

Cruising at 25,000 feet, about 30 nautical miles inland and midway between Madang and Wewak, the starboard engine stopped suddenly.  There was no warning.  The aircraft had been operating normally when suddenly it lurched violently; I glanced towards the starboard engine and the propeller was stopped.

This was an unusual engine failure, unlike anything I had previously experienced. My first impression was that the engine had seized and I was cursing myself for having missed the drop in oil pressure.  I glanced at the oil pressure gauge and the needle was dropping rapidly through 50 pounds per square inch (normally 75-80 psi) and I knew then that it hadn't been an oil supply problem.

From a cruising rpm of about 2100 the propeller stopped suddenly and completely.  Had it been a problem other than a mechanical failure the propeller would have "windmilled".  There was no windmilling; the propeller was locked solid and the blades of the three bladed propeller were in a normal cruising pitch.

(I learned months later that the master connecting rod "big end', had separated as a result of a fatigue failure of one of the connecting rod bolts.  The whole of the master and articulating rod assembly was a mangled mass.  From normal cruising rpm to stop, the engine had rotated only 2/3 to 3/4 of a turn.  With such a sudden stop the inertia of the propeller could have torsionally failed the propeller shaft with a complete separation of the propeller from the aircraft.  Fortunately this didn't happen. 2100 engine rpm might seem slow in comparison to today's high performance motor vehicle engines, but these were 9 cylinder radial engines, of 1200 horse power; slow revving but with a lot of torque.)

The violent lurch of the aircraft at the time of the engine failure was from a combination of a sudden absence of engine power on one side, normal cruising power on the other, and a rolling motion from the significant rotational inertia of the stopped propeller and engine mass being translated into a rolling force.

It was not a difficult situation to control.  Being about midway between Wewak and Madang I turned towards Madang and went through the engine shut down procedure.  It was unusual feathering a stopped propeller in flight and watching the blades rotate from cruising pitch to one of minimum drag. Once pressed, the feathering button should have remained depressed and, by means of a limiting mechanism, should have popped out automatically on completion of the feathering cycle. On this occasion the feathering button remained depressed; the cycle continued beyond the full feathered position and the blades began to unfeather.  I pulled out the feathering button to stop the progression of the blades.  Even though the blades were then slightly out of full feather, I decided to accept this partial drag rather than risk a further cycle with perhaps a similar, or worse, result.

Asymmetric flying in multi-engined aircraft requires the aircraft to be as "clean" as possible so that aerodynamic drag is reduced to a minimum (propeller of the failed engine feathered, landing gear and flaps retracted and, if applicable, oil cooler flaps and cowl gills all closed or streamlined).  Adequate control of the situation also requires full power availability on the "good" engine, full rudder control capability, careful control of airspeed and the more height the better.  If one has to experience an engine failure in a Lockheed Hudson, an altitude of 25,000 feet or more is ideal!

There was ample time to ensure the aircraft was "cleaned up" aerodynamically.  I did not attempt to maintain height because of our proximity to Madang.  The Hudson would not maintain height on one engine above about 13,000 feet anyway.  I reduced power for the descent.

Jack and Bob at their positions in the nose of the aircraft were well aware of the engine failure.  I advised them on intercom that we would land at Madang.  The Hudson was an unpressurised aircraft, so they needed to remain at their positions and continue using their oxygen masks until we descended to a lower altitude.

With a gradual descent from 25,000 feet it would not have been over-taxing the aircraft to have landed at Lae.  This would have been more convenient, but it was inadvisable for a couple of reasons.

First, there was a regulation stipulating that under such circumstances the aircraft was to land at the nearest suitable aerodrome.

Secondly, the port engine was on an "extended life" and was burning more oil than usual for Wright Cyclones, renowned for their oil burning characteristics at the best of times.  With a normal maximum fuel endurance of six and a half hours flight time, VH-AGX was limited to four hours because of the oil consumption of our "good" engine!

Better not to push our luck, Madang was the logical choice for the landing.  I called "Madang Tower" on VHF radio.  I advised them of the engine failure, that we would land at Madang and would arrive overhead in about 30 minutes.  To the tower controller's query "Are your operations normal?" I responded: "I anticipate a normal asymmetric approach." The tower controller told me later he didn't appreciate that the situation could deteriorate rapidly until he mentioned the impending asymmetric landing to the duty fire officer.

When he advised the fire officer his response was: "Shit!  Those bloody Hudsons have difficulty flying on two engines let alone one, we'll have everything standing by!" His appreciation of the flying performance of the Hudson was in error, but I was grateful for his positive reaction.

In 1962 the then Australian Department of Civil Aviation (DCA) had a well deserved reputation for professional excellence.  It had the responsibility of providing a number of services for the safety of aircraft operations.  DCA units provided services throughout PNG until after its independence.  Air Traffic Control and the Fire Service were two of these services.

I descended with sufficient power to keep the good engine warm, applying a higher burst of power periodically to ensure the engine was at a safe operating temperature, important if a single engined "go-around" became necessary.  During the descent I described to Madang Tower the type of engine failure which had occurred.  This was unnecessary and was low on my priority list.  But there was ample time during the descent and such information could be helpful for the investigation if the landing was unsuccessful.

The equipment referred to by the fire officer ("everything standing by") was an antique fire tender and a converted wartime Jeep.  The fire officer was preparing for the worst happening - always a sound aviation philosophy, and he undoubtedly had in mind two previous Adastra accidents involving Hudson aircraft in recent years, one at Horn Island and one at Lae.  Both were making single engined approaches and both rolled over and crashed, killing all on board.  There were different circumstances leading to each occurrence, but history of this kind has the effect of sharpening the concentration when making an asymmetric landing approach, especially in a Lockheed Hudson for which the optimum approach speed limits are narrow and critical.

Nevertheless I enjoyed flying the Hudson.  To become familiar with the handling characteristics of any aircraft takes time and the Hudson had a few idiosyncrasies which needed to be respected.  It was an excellent aircraft for survey work but, when most operations were at 25,000 feet and the engines were on high power for repeated climbs to this altitude, engine life sometimes was less than advertised.

Jack and Bob vacated the nose of the aircraft during the descent and Jack stood at the right hand control seat position. (For survey flying the right hand control seat was removed to allow the navigator and camera operator access to their operating positions in the aircraft's nose.  The Hudson was operated on survey work as a single pilot aircraft and was flown from the left hand control seat.) It was always supportive to have one of the other crew members up front and Jack was particularly helpful on this occasion.  We had exchanged a few comments about the quantity of oil being burned by the port engine on its "extended life" before Jack went back to the cabin area to strap himself in for the landing.

I have no doubt that Jack and Bob were thinking about the earlier Horn Island and Lae accidents.  They were well aware of the possibility that the landing might not be normal, but both displayed a resigned "Oh well, you can handle it" confidence which my gastric juices at the time might well have denied. On our single engined descent we arrived over Madang at about 3000 feet altitude and I planned the landing approach towards the east-north-east, runway 07.  The Madang runway length was 4,500 feet, adequate but not generous for a Hudson asymmetric landing.  The touchdown was smooth; it was further into the runway than I would have liked, but far better than undershooting and having to apply high power on the landing approach.

The aircraft stopped on the over-run and the fire tender, which had been positioned adjacent to the runway, was now following closely.  Fortunately Jack was able to get out of the aircraft before the tender arrived and he managed to stop the fire crew covering our "good" engine with foam.  The "normal" exhaust smoke trailing from the port engine led them to believe it was on fire!



PORT MORESBY: A plane carrying 35 passengers overshot the runway at Madang Airport on Papua New Guinea's north coast and slid into shallow water on the edge of the harbour on Wednesday.  None of the passengers or the four crew members was injured. - AAP"

("The runway was not generous" - a recent news article)

VH-AGX was towed to the tarmac and I then needed to make arrangements for a replacement engine.  Communication with Sydney was only by cable/telegram.  Jack McDonald, in Sydney, on receipt of the telegram had difficulty in accepting that the starboard engine had failed.  He was well aware of the high oil consumption of the port engine and he believed that the port engine had probably seized.  In any event he made arrangements for a replacement engine to be shipped to Madang, but it would be a month before it arrived.

After the landing at Madang on 27 June 1962 Jack Tierney, Bob and I went back to Lae by Mandated Airlines to await the arrival of the replacement engine.

My wife, Margaret, and our two children, Wendy and Robin, had been with me in Lae since February of that year. (Cameron, our third child, was not born until 1965.) Adastra Head Office decided to relocate VH-AGX for another contract in Australia after the engine change, so I made arrangements for Margaret, Wendy and Robin to return to Sydney by the TAA DC-6 service before the replacement engine arrived.  It was a wrench to see them off at Lae Airport.


When the replacement engine was due to arrive in Madang by ship, Jack McDonald flew to Lae from Sydney by the scheduled TAA DC6 aircraft. On his arrival we (Jack McDonald, Brian Smith and I) flew to Madang by Mandated Airlines to change the engine in VH-AGX.  Jack Tierney and Bob Jones remained in Lae.

Jack McDonald arranged to use one of the Mandated hangars on Madang aerodrome for the engine change, which took the three of us the best part of two days.  Jack noticed a slight bend in one of the push-rod covers of the failed engine.  There was no doubt about the mechanical nature of the engine failure from the immovable propeller, but it was further confirmed by the bent push rod.

The engine change proceeded without difficulty, although it needed Jack's engineering expertise and practical flair to provide the answers to a couple of discrepancies between the control linkages of the old and new engine installations.

Brian Smith was a competent young engineer, but Jack's vast experience knowledge, ability, his understanding of DCA requirements and his determination of what was reasonable and safe in circumstances when it was impossible to "go by the book" on every occasion in remote locations was masterful.  As one who spent most of his time in the field I was unaware of most of what occurred at the Sydney base.  I hope Jack's many engineering talents and his willingness were fully appreciated by Adastra management.

On the morning of 31 July 1962 the installation of our replacement starboard engine was almost complete.  We checked out of the hotel expecting that we would be on our way back to Lae about mid morning.  I would air test the engine at the same time.  The final split pin and piece of locking wire secured, VH-AGX was wheeled out onto the tarmac and the ground run of the replacement engine was satisfactory.  And this is where the story really begins.


I completed weight and balance calculations and prepared a load sheet.  Jack decided to run the port engine while I submitted a flight plan.

It was mid-morning when I went to Madang Tower to obtain a weather briefing and to submit a flight plan for the flight to Lae.

On my way to the Tower, I spoke to a Mandated Airlines DC3 crew who had landed at Madang only about 15 minutes previously on completion of a flight from Lae.  As is usual in PNG, pilots invariably ask other crews about weather conditions they have encountered.  The Mandated crew advised that they had flown up the Markham Valley without any difficulty and that the weather at Lae and throughout the flight had been fine with no problems for visual flight.

Visual Flight Rules (VFR) applied to all Adastra flights.  VH-AGX was minimally equipped for instrument flight and was not approved for flight under the Instrument Flight Rules (IFR).  All survey flights required clear weather for photography.  When one task was completed and the aircraft needed to be repositioned for another survey area, if the weather was unsuitable it was more economical to wait for a day of good weather than for the company to have the ongoing expense of IFR equipment and its servicing as well as flight time costs for pilot training and periodic currency checks.

The weather forecasts en route and at Lae were consistent with the information I learned from the DC3 crew.  Madang weather was fine with normal cumulus build up and a reasonably high base.  The Markham Valley was the shortest, most direct route and was the appropriate way to plan.  Allowing for a "dog-leg" to enter the Markham Valley, flight time in the Hudson would be about 45 minutes.  The Finisterre Mountain Range ran along the northern coast of PNG from a position north of Lae to a position south of Madang and the range, up to some 13,500 feet elevation, formed the northern boundary of the Markham Valley.  From Madang aerodrome, weather towards this range appeared reasonably clear.  If flight through the Markham became inadvisable because of cloud buildup, the alternative was a longer route coastal via Finschhafen, almost double the distance.

There was no problem with fuel reserves if the longer route became necessary, or to return to Madang via the same longer route if Lae weather deteriorated.  Based on hours flown since last refuelling and a check of the fuel gauges, in theory there was fuel sufficient for 4 hours - 3 hours 15 minutes flight plus 45 minutes reserve.  In fact there would have been something less than this, mainly because of evaporation during the period the aircraft had been standing at Madang.  I considered refuelling at Madang but decided, in view of the existing fuel state and the relatively short flight time, to delay refuelling until after we landed at Lae.

Aircraft fuel tanks are normally kept filled to capacity, especially in the tropics, because of condensation.  Water in fuel is to be avoided at all costs and our fuel draining after VH-AGX had stood for so long was thorough.  With the current weather forecasts the minimum fuel requirement would have been 90 minutes - 45 minutes flight fuel plus 45 minutes reserve.  I was satisfied we had sufficient for Madang - Lae with a return to Madang if necessary and with generous reserves.  The port engine oil tank was topped up to ensure we had a safe oil endurance.

I planned via the Markham Valley with an estimated time of departure (ETD) 30 minutes from the time of lodgement of the flight plan.  This would allow a comfortable period for Jack to run the port engine and for final loading of tools and equipment.  If the aircraft did not taxy out within 30 minutes of the ETD, the flight plan would be invalid and submission of another flight plan would be required.

Walking back to the tarmac area I could see that Jack was having difficulty in starting the port engine.  The engines were normally started with the aircraft's batteries.  If one engine was difficult to start, starting the other and using generator power was the next step.  On this occasion the starboard engine was stationary and Jack had arranged to use a battery cart.  I knew there was a problem.

The port engine, having been subjected to tropical moisture for a month without being run, was being more than ordinarily difficult.

Ted McKenzie, Adastra Operations Manager and Chief Pilot, had briefed me thoroughly for the New Guinea operation and had commented on this moisture problem.  He advised that, in the event that weather was unsuitable for survey for lengthy periods, I should fly the aircraft every week for an hour to keep the ignition system from becoming saturated.  If the weather was unsuitable for flying, then to run the engines.  Weather was rarely suitable for survey and it was essential for the aircraft to be in readiness for those rare occasions.  I made a practice of flying the aircraft for 30 minutes twice each week.

Ted was an exceptional pilot and an excellent check and training captain.  When some action was needed in flight which was a little out of the ordinary, I was quietly grateful to Ted on more than one occasion for his sound advice and for his white knuckle, perspiration producing, fly-to-the-limit flight checks.  Ted did everything he could to ensure that all of his pilots would be able to handle the Hudson competently under all circumstances, and particularly after the unfortunate accidents at Horn Island and Lae.  My six monthly flight checks often concluded with my clothing absolutely saturated with perspiration and without a dry pocket in which to put my flight check certificate of competency.

So, having not been run for a month, the port engine ignition system was breaking down with tropical moisture and was difficult to start.  How long it would take to dry the system sufficiently to get the engine started was uncertain. (There was no WD40 in those days!) When the port engine continued its cantankerous behaviour I cancelled the flight Plan, explaining to the tower controller that I would re-submit the plan after the port engine was started.

There was yet another setback for our flight to Lae.  On my return from flight planning, the period of validity for HF radio servicing was found to be expired and an approach to DCA by Brian for a concession was not approved.  Jack McDonald made a second approach, explaining that radio servicing facilities were at Lae and were unavailable at Madang.  It would be expensive to fly a radio serviceman from Lae to Madang for what would probably amount to a radio check.  DCA granted a concession for a single flight, Madang to Lae, subject to a satisfactory preflight HF radio check.

The radio ground check proved satisfactory, but the HF radio was to be a problem later in the day.  Communications between Air Traffic Control units were often difficult in 1962 and en route communications relied primarily on HF radio.  VHF was far better quality in terms of clarity, but was limited to “line of sight" (no benefit of satellites!).

It was after 1600 hours EST (4 pm) before the port engine started, roughly at first but it was soon firing evenly after it began to warm up. I had almost decided to abandon the idea of flying to Lae that afternoon, but there was still time, and after all, the weather was fine throughout - or so I thought.

If a diversion became necessary I would need to return to Madang. It would have been stretching it to get to Port Moresby.  This would have involved a climb of some 15,000 feet over the Owen Stanley Ranges and in any event there was insufficient time before last light to get to Moresby.  It is always desirable to have plan B in mind when planning a flight, and perhaps plan C as well.  For example there is always a possibility that a runway can be closed just prior to landing because of a preceding landing aircraft with a blown tyre.  Anticipating such events helps avoid that “thinking feeling" if circumstances change suddenly.

As soon as the port engine was operating satisfactorily on its ground run I went to the tower to submit another flight plan.  There was no weather forecaster at Madang but from the tower I was able to speak by radio directly to the forecaster at Lae.  I jotted down the forecast which hadn't altered significantly from the morning forecast.  It was satisfactory for a VFR flight with light and variable winds throughout below 5000 feet.

I requested an actual observation at Lae and this was also satisfactory, visibility 15 miles, main cloud base 5000 feet with 1/8 of scud at 1000. 1/8 scud at 1000?  Unusual, must be the low scuddy cloud which often formed over the Markham River, just south of Lae.  I asked the forecaster which route appeared clearer, coastal or via the Markham Valley.  He indicated that via the Markham should be the clearer route.

Night flying was not permitted in PNG and all flights were to be planned with an ETA of at least 10 minutes before last light.  At this stage, last light was becoming a consideration.  The end of daylight in the tropics is significantly different from middle and high latitudes.  The end of daylight charts take into account a certain amount of twilight after sunset, depending upon latitude.  The higher the latitude, the more twilight until, above the arctic and antarctic circles there is "land of the midnight sun".  In the tropics there is very little twilight; soon after sunset darkness falls like a blind.

Last light at Lae was shown to be 0836 hours GMT (1836 EST or 6.36pm). If I set course by 0640 GMT, (4.40 pm) and flew to Lae via the Markham. I could still return and land at Madang about 25 minutes before last light if Lae weather deteriorated.  The requirement to plan with an ETA at least 10 minutes before last light could be met comfortably.

It looked good.  Why did I have this lousy feeling that I ought to wait until tomorrow?


I walked briskly back to the tarmac and Jack and Brian were ready to go. The battery cart had been wheeled away.  Jack was in the control seat with both engines operating.  Brian boarded the aircraft behind me and closed the door.  Jack vacated the control seat and I strapped myself in.  I wasted no time in completing a pre-taxy check, taxying to the holding point and conducting an engine run-up.  We took off from runway 07 and I obtained approval for a right turn towards the Markham Valley.  No problems with the replacement engine; it was performing beautifully with all temperatures and pressures well within their normal range. Our departure time was 0640 hours GMT.

Soon after departure and with a better appreciation of general weather conditions with height, it became apparent that we would be unable to enter the Markham Valley because of cloud build up, even with a significant diversion to the north western end of the valley.  From a base of about 3000 feet, towering cumulus cloud covered the rising Finisterres completely, cloud tops being 20,000 feet or more.  Most IFR aircraft avoided the direct track Madang - Lae.  It required a cruising level of at least the minimum safe altitude of 16,500 feet because of the high terrain.  I altered heading coastal and calculated an ETA Lae of 0800 GMT, 36 minutes before last light.

The ETA for Lae was closer to last light than I liked even though it was 26 minutes more than the 10 minute minimum required.  Last light graphs do not take into account factors such as the terrain surrounding a location, a cloudy sky or visibility less than unlimited.  Significant cloud cover, poor visibility or high ground to the west can result in the earlier onset of last light.  However the Hudson cruised at 170 knots and, with light and variable winds forecast, the ETA was likely to be accurate.  At least I could continue for awhile and still return to Madang before last light if the weather deteriorated.

Full reporting procedure is required for all flights in PNG.  I called Madang Tower on VHF and advised of our change of plan, that we would track coastal to Lae via Finschhafen.  I gave my ETA Saidor (on the coast about 50 nautical miles (nm) from Madang and 65 track miles after taking into account our initial track towards the Markham Valley) of 0703 and Lae of 0800 (1800 - 6.00pm). This was acknowledged.  I reported to Madang approaching Saidor and we were still within VHF range at our cruising level of 4,000 feet.  My next report would be a scheduled reporting time, approximately midway between Madang and Lae at 0720 GMT when we would be about 100 nm from Madang with about 55 nm to run to Finschhafen.

To remain “legal” relative to last light, we could stay airborne until 0826, 10 minutes before last light.  That is, with a departure time of 0640 we could fly for 106 minutes.  So our "point of no return" based on last light considerations and light and variable wind conditions was 0733, being 53 minutes after our departure time.  If we needed to return to Madang this would give a couple of minutes margin because last light at Madang was a little later than at Lae.

Why did I have this uncomfortable feeling that I ought to be returning to Madang?

In PNG particularly, I always liked to “look over my shoulder" with time to divert and land at an aerodrome I had overflown or where I knew the weather to be suitable.  Runway lengths and surfaces in PNG in 1962 limited the operation of Hudson aircraft to only five aerodromes; Port Moresby, Lae, Madang, Wewak and Rabaul.

My practical options were to continue to Lae or return to Madang.

I called on HF at the scheduled reporting time of 0720 GMT and received no reply.  I called again and asked for actual weather conditions at Lae.  No response.  I called on VHF but we were then too far from Madang for VHF transmissions to be possible and we were certainly too far from Lae, especially with the Finisterres blocking our line of sight.  Tracking along the coast we were cruising at 4000 feet below a broken cloud cover.

I called “any aircraft" on VHF hoping for someone to relay our position but there were apparently no other aircraft in our area or on our frequency.  I tried another HF frequency, but still no success.  The fuse for the HF equipment looked satisfactory but I changed it anyway.  Try another microphone.  Another radio check, no response.

Damn!  We were half way to Lae, our groundspeed was consistent with the flight plan and the weather ahead looked quite reasonable.  Continue to Lae or return to Madang?  Radio failure requirements were to proceed to the nearest suitable aerodrome, land, and telephone air traffic control.  The nearest suitable aerodromes were either Lae or Madang because we were midway between the two.

Radio servicing facilities were at Lae but not Madang.  The weather en route and at Lae was fine (so I thought), we needed the HF radio serviced at Lae, Jack needed to return to Sydney without too much delay, and our base accommodation was at Lae, in the DCA mess.  The replacement engine was performing well and all information available indicated that logically we should go to Lae.  Why this strong gut feeling to go back to Madang?

I called Lae again, advising that we were not receiving on HF and we would continue to Lae, ETA 0800 and, “if receiving this transmission, please advise the Lae weather conditions".  No reply.  I had no way of knowing whether my HF transmitter was functioning, but radio fail procedure requires this assumption so that air traffic control may have some information to work on.

We were approaching our point of no return (PNR) based on last light, 0733.  I descended to 3000 feet and the weather ahead looked good, hazy, but the visibility was about fifteen miles.  Our PNR came and went and, with an uncomfortable feeling of resignation based on nothing but a lousy gut feeling, I committed the flight to continue to Lae.  Soon the cloud layer, about 500 feet above us, seemed to be lowering and I descended.  The weather still looked satisfactory ahead with no observable deterioration.  I continued the descent down to 1000 feet and, looking down at the kunai grass it was bent over at right angles towards us. Obviously a strong headwind at the surface and certainly not the light and variable winds forecast, at least not in this area.  We could be a minute or two later at Lae because of this and our last light margin would be reduced.  My gastric juices acknowledged.

Soon after passing our PNR, a general weather deterioration was apparent in the distance with widespread areas of rain.  At that stage I could almost see Finschhafen near Cape Cretin, the most prominent northern "bump" of PNG, but Finschhafen appeared to be almost obscured by rain.  It was too late to return and land at Madang before last light.  Lae it had to be.

What about a landing at Finschhafen?  A natural surface grass strip, probably saturated with water and unsuitable for Hudson operations even in dry conditions.  Finschhafen was not on the list of approved aerodromes.  I would have considered a precautionary landing there, but getting closer to Finschhafen and flying now well below 500 feet to remain in visual contact with the ground, rain squalls obscured Finschhafen airstrip.

Blast.  We were about three minutes behind our ETA Finschhafen of 0741.  I flew slightly inland, across the corner of Cape Cretin, abeam Finschhafen.  There was low lying land for a few miles inland at this point and I flew towards the Huon Gulf coastline which I could see, with Finschhafen, or where I assessed Finschhafen to be, about three or four miles to our left.  This "short cut" would reduce our flight time by a minute or two and would help make good our ETA at Lae.

The foothills of the Finisterre Ranges to our right were totally obscured by cloud and it was black ahead. On a previous flight from Rabaul to Lae there were large areas of localised deteriorated weather off the coast from Finschhafen and I was hoping the Lae aerodrome forecast of clear weather was accurate.

Perhaps the weather would open up after I reached the coast and got closer to Lae; after all, the Lae actual weather just before we departed from Madang was good.  But all I could see ahead were worsening conditions. My decision to continue the flight beyond our PNR was not made lightly but overflying Cape Cretin I thought "cretin" was an accurate description of my self image at that time.

I needed to descend well below 500 feet to maintain forward visibility.  We were running into dense rain squalls.  The cloud was dark grey with an indistinct base, not far above us.  As well, daylight was fading; the cloud mass over the ranges to the west of the Huon Gulf must have been very dense and deep.  The sun would have been low on the western horizon somewhere above this cloud mass, and it would have been also behind the ranges to the west, contributing to the earlier onset of last light.

Apart from being illegal, climbing into cloud could have been disastrous.  The lowest safe altitude for an IFR flight approaching Lae from the Huon Gulf was 3400 feet.  If I climbed into cloud, a descent in clear weather to seawards off Lae looked extremely unlikely.  If there was little clearance between the cloud base and water surface, a descent in darkening conditions could result in the aircraft crashing into the water before the water surface became discernible.  It was imperative to maintain visual contact with the ground or water.

VH-AGX was equipped with basic attitude instruments and a single ADF, an instrument which, when tuned to a ground station (NDB), would indicate the direction of the station from the aircraft.  It was a useful navigation instrument, but it had its limitations and was known to point to thunderstorms at times instead of to the station to which it was tuned. (When flying in the RAAF I had experienced this.)

I tuned to Lae NDB and received a positive needle indication.  We were too far from Lae to identify its transmitted signal but its steady indication was supporting.

I needed to stay below the amorphous cloud base to maintain forward visibility. But forward vision was difficult in the Hudson with its flat, sloping windscreen panels.  Rain would strike and spread on these flat surfaces, virtually obliterating forward vision. (Curved windscreens on modern light aircraft help overcome this problem.)

There were no windscreen wipers.  I opened the port sliding cockpit window for better external vision.

At this stage, flying very low across swampy mangroves and with no indication of improvement in any direction, I fleetingly considered the possibility of putting the aircraft down on any reasonably flat area.  No, it would be better to go back and try Finschhafen if I could find it, or put the aircraft down on an area of kunai grass - a bit better than swamp crocodiles; but what if we were to overturn on touchdown?  The whole area was pockmarked with craters from the World War Two bombardment, too lumpy for a safe precautionary landing.

Climb through the murk and go to Moresby?  Moresby would have runway lighting, and I had by then put behind me the niceties of trying to stay “legal".  That is, the aircraft's instrumentation was not approved for instrument flight and I had not held an instrument rating since leaving the RAAF some 3 years previously.  But this was a question of survival; we were far from legal as it was regarding specified visibility and distance from cloud and ground for a VFR flight but there was little option.  And if there was a softer option than rock-filled cloud I would take it, so stuff the regulations.

But Moresby was too far; we would arrive there at least an hour after last light.  And, assuming our radio problem was not common to both VHF and HF and our VHF was serviceable, the air traffic controllers would all be at their favourite watering holes long before we had even cleared the Owen Stanleys and were within VHF range.  As well, the fuel quantity remaining was doubtful for the distance, and I didn't want to check the serviceability of the aircraft's internal lighting which probably hadn't been used for years.  If it worked it would wreck my external vision.  It had to be Lae.

Fleeting thoughts, pelting rain, obscured windscreen, about 50 nm to run to Lae, flying about 100 feet above the ground, getting darker, bloody hell Bowlesie, you've stuffed up this time, looks like you've written yourself off with Jack and Brian and written off the aircraft as well.  I felt a twinge for AGX, the fastest aircraft of the fleet.  I accepted that we had next to no chance of recovering the situation I had placed us in.  I believed our chances of flying into the real estate were very high indeed.  I accepted death.

And then IT happened.

My impression at the time was that something, something intangible outside myself enveloped me and permeated my entire being.  It was instantaneous; the moment I accepted death as virtually inevitable there was this all pervasive feeling of unimaginable calmness.  And a second or two later: "But we are not dead yet - perhaps we have a 1% chance of getting away with this.  Let's see what I can do with this 1%.

I felt like someone with his back to the wall, intense concentration, calculated efficient actions, yet all the time this unusual, unbelievable, peaceful serenity without a word from my gastric juices.  Perhaps they had all been used up.

I reached the coastline and paralleled the coast towards Lae, about 50 nautical miles ahead.  At this stage we were flying below 100 feet and probably below 30 feet for much of the time. Wind squalls were visible on the sea surface which was flattened by heavy rain.  The dark grey metallic colour of the sea merged with the low, indeterminate cloud base.  In the fading light the torrent on the windscreen totally obscured forward visibility.  We were about half a mile to seaward.

At about this time Jack and Brian came up front, viewing the weather with some concern.  I asked Jack to open the starboard side window.  By now the open side windows were our only means of visual reference.  To our right, through the restricted opening of the starboard window, the dark grey of the water merged with the darkness of the vegetation along the coast.  I concentrated on staying out of the water and getting some idea of attitude from an occasional glance at the artificial horizon, but mainly from the appearance of the terrain on our right.  From my position in the left hand control seat the starboard wing obscured my view of the coastline for much of the time.  By banking the aircraft occasionally I tried to identify a coastal feature, but I found this impossible.

The colour of the dark coastal vegetation on rising hills mingled with the indistinct cloud base, which seemed less than 100 feet above the water with heavy rain and scud beneath.

There was no possibility of obtaining a visual fix; to look from the ground to the map was risking inadvertent water contact.  It required all my concentration to stay clear of the water and clear of the ground when we passed over a small headland.  Yet all the time there was this incredible feeling of peacefulness, totally inconsistent with the situation.

I continued to call Lae Tower on HF and VHF and finally, about 30 miles from Lae by estimation, there was a welcoming response: "Alpha Golf Xray, Lae Tower, we've been calling you for some time, go ahead."

I recognised the tower controller's voice, a somewhat dour ex-RAAF Wing Commander, Phil Graham.  I replied: "Alpha Golf Xray now approximately three zero miles from Lae, coastal, estimate Lae at zero three, what are your weather conditions?" There was concern in Phil’s usual monotone when he replied simply: "The weather is poor."

This was a particularly unusual Air Traffic Control transmission.  Inbound aircraft at 30 miles usually receive landing instructions such as: "AGX runway --, the wind --, QNH ---, report one zero miles." "The weather is poor" meant the weather was bloody awful.  I asked Phil: "What’s the visibility?" He replied: "About a mile".  If my heart hadn't dropped to my boots about 20 minutes previously it would have sank; yet there was still this extraordinary, almost detached feeling of tranquillity.  There was obviously no improvement in the weather at Lae; perhaps it was even worse than that which we'd been experiencing since before passing abeam Finschhafen.

“Lae Tower this as Alpha Golf Xray. What is the landing direction?" Phil was remarkably quiet.  "The wind currently favours 32, and the QNH 1008”. I learned later that Phil had telephoned Jack Tierney and invited him to the tower "to see these galahs prang.  There’s no way they are going to be able to land in this."

So Phil didn't waste words.  If the Hudson was going to prang, why waste breath on landing instructions?  But the weather would have closed the aerodrome and Phil couldn't legally issue landing instructions.  However there is no way he would have denied assistance to an aircraft in our circumstances.

Soon after this communication about 30 miles from Lae (I had no idea of precisely where we were) I selected the landing gear down.  Gear down lowers the nose position slightly for level flight and is supposed to give a slightly better forward visibility.  Handy if we happened to encounter a break in the rain to give us some forward visibility!  Because of what happened a little later I was pleased I took this action.

I obtained an identifiable signal from Lae NDB and the ADF needle was pointing ahead steadily.  Was it pointing to Lae or towards a thunderstorm cell?

At least there was no visible lightning to suggest thunderstorm activity; it was more than enough to cope with the torrential rain and dense nimbo-stratus cloud almost to the ground and water surface without further complications.

I figured our ETA Lae would be a little later than planned because of the headwinds at low level along the north coast, shown by our loss of a few minutes abeam Finschhafen.  However our heading change from south easterly to westerly after Finschhafen could have resulted in a tailwind component which would make up some of our loss.  Again, the reduction in speed for gear down would make us a little later, but not more than a few minutes.  Impossible to identify any landmarks in this torrential downpour, let alone calculate a revised ETA.

Jack remained at the starboard window.  He had a slightly better view of the terrain than I and, even though nothing was said, I knew that he was looking out for possible obstructions.  Saying anything was pointless anyway - nothing could be heard above the noise of the engines with the side windows open.  I was grateful for Jack's presence.  From my familiarity with the general area I knew there were no obstructions as long as we hugged the coast.  If we missed seeing Lae and overflew the Markham River, we would be faced with the spine of the Owen Stanleys rising to over 9000 feet just beyond the Markham.  We would use up our final 1% if I missed Lae.  I needed to keep the aircraft low to get what visibility I could from the side windows.  But even at a reduced cruising speed, necessary with the landing gear down, there would be insufficient manoeuvring distance to avoid rising terrain if it loomed out of the murk ahead.  Slowing the aircraft further could help a little, but with darkness fast approaching I was reluctant to reduce speed, so while I could see the coastline and remain just to seaward of it I would be making the most of our 1%.

Brian went back to the cabin, put on a pair of headphones and listened to the radio transmissions.

As a minimum runway length, 4000 feet was required for the Hudson.  This requirement could vary depending on surface wind, temperature and other atmospheric conditions.  The landing charts were based on normal flapped approaches and the 4000 feet minimum length was accepted as a safe landing distance for Hudson operations.  Lae runway, aligned 320/140 degrees was 4000 feet in length with jungle at the western end and a drop of about ten feet to the sea at the eastern end.  The Lae landing chart showed the aerodrome elevation as eleven feet above mean sea level.

The "Tenyo Maru” was a Japanese vessel which sank in the Huon Gulf during World War Two.  It was run aground on the edge of a steeply sloping reef or shelf just a few hundred metres off the eastern end of the Lae runway.  Its bow, angled up steeply, was perhaps 30 feet above the water.  It was an excellent point of reference when approaching Lae, especially in conditions of haze or reduced visibility.  On final approach to Lae runway from the sea the Tenyo Maru was about 100 feet to the left of the approach path.

But where was Lae?  The ETA of 1803 I had passed to Lae Tower was to accommodate the time loss observed near Finschhafen, even though I felt we may have picked up a little, both from "cutting the corner" and from a possible tailwind component after the heading change near Finschhafen.  I needed to reduce the indicated airspeed to 125 knots to select the gear down, which I had done at an estimated 30 miles from Lae, and this speed reduction over that distance would account for a loss of a few minutes.

My expectation was that we would reach Lae no later than about 1802.  It was now 1802 and no sign of Lae.  In this miserable weather and fading light I was feeling a little doubtful whether we would be able to see Lae, even from 100 feet.  Let's hope we might see a few lights in some of the homes and buildings on the higher ground which I knew to be to the north of Lae aerodrome.  No doubt the Mandated Airlines Club on the hill above the aerodrome would have its normal patronage and its lights might be visible, providing it wasn't enshrouded in low cloud.

1803, nothing.  And now 1804.  Where the hell was Lae?  Nothing except continuous torrential rain, an obliterated windscreen, low cloud and scud, an indistinct landscape to the west and darkening conditions.  Jack and I were oblivious to the rain through the open windows; not all of it was whipped away by the slipstream.  At least the deafening roar of the engines was comforting.  By this time I began to wonder whether we had passed Lae and were heading for the ranges across the Markham.  No, not yet anyway because we were still following the coastline and heading generally west.  The Huon Gulf coastline swung to the south and then south-east just beyond Lae.  My mental alertness was acute.  We didn't need the 9000 feet ranges across the Markham to suddenly terminate the flight, tracking over coastal projections, a tall tree might achieve a similar result.

I began to alter my concentration from the starboard window and to scan the water to the left.  Perhaps I would spot the Tenyo Maru, or less palatable, terrain too close to avoid.  But the ADF needle maintained its steady indication ahead.  The ADF in VH-AGX had always been reliable and Lae NDB was transmitting a strong signal so, believe it, Bowlesie!  I mentally put on "hold" the possibility that the ADF needle may be pointing to a thunderstorm cell.

ETA updates are required if the calculated ETA varies by more than two minutes.  We were a minute over our revised ETA of 1803, so I called the Tower again and gave another revision to 1806, purely by guesstimation.  We were still hugging the coastline but precisely where was anyone's guess.

1806 and still no sign of Lae.  This was crazy, as crazy as this feeling of incredible calmness in this situation.  At the speed of the Hudson, wind conditions need to be significantly different from forecast to alter the ETA to any extent.  On one occasion I flew for about 5 hours from Mackay in Queensland to Sydney and arrived within a minute of the original ETA calculated on departure.  A six minute variation over the relatively short flight from Madang just didn't make sense.  At least the ADF needle indication was still pointing ahead and we still had the Huon Gulf coastline on our right.  I revised my guesstimated ETA to Phil: "Zero eight".

But had we passed Lae without seeing it?  Were we now heading for the 9000 feet mountain range?  Through the deluge I tried to pick up the first indication of a change in direction of the coastline towards the south.  I would know then that Lae was behind us.  But then...

Jack saw it first.  He looked towards me and pointed ahead.  I leaned across to see better through the starboard window and there it was: the Bureau of Meteorology's cloud searchlight.  And - runway lights!

Unbelievable!  And the time? 1810!!

Because night flying was not permitted in PNG and because Adastra aircraft generally only flew in excellent weather and always landed several hours before last light, I didn't consider the possibility that Lae was equipped with runway lighting.  It seems ridiculous that I had operated from Lae for about six months without being aware of this installation.

What an unexpected bonus!  We were almost over the runway at a height of 50 feet or less when it was first visible.  Jack and I exchanged a glance of relieved understanding and he then returned to the cabin to strap himself in.  Thanks, Jack.  A big sigh.  Perhaps we'll live to fly another day!

At about this time, there was a change in my feeling of detached calm.  It was still there to the time we landed; but there was a gradual change - difficult to explain - a change back to my normal perception though my "back to the wall" mental concentration remained throughout.

We crossed the runway at right angles close to the 32 landing threshold.  I immediately reduced power and commenced to turn left.  Phil was right.  Poor weather was an understatement.  It was way below the circuit minimum required for VFR aircraft - 1500 feet cloud base and visibility 3 nautical miles.  No good for IFR either.

There was no way I was going to lose those runway lights.  This would be very much a modified circuit pattern.  I reduced speed for initial flap extension, 107 knots, and selected flaps down.  They began to extend and then stopped at only 5 percent!  Oh shit.  A hydraulic failure.  I could do without this.

It meant virtually a flapless landing on a runway with minimum length for the Hudson for a normal flapped landing.  No time or inclination to try to sort this one out in flight under the conditions of lousy weather and approaching darkness.  It probably couldn't be sorted out in flight anyway and I needed to concentrate totally on staying out of the water and not losing sight of the runway.  It was a relief that I had selected the landing gear down much earlier and there was a safe (green) gear indication.  The landing gear was operated by the same hydraulic system.

Propeller pitch full fine - a normal pre-landing procedure for a possible go-around.  There’d be no go-around but, without flap, I would need whatever drag I could get to stop in the available runway length and fine pitch would help.  Selected early it would reduce actions necessary on final approach.

The port wingtip seemed almost to skim the water as I continued the left turn, looking back over my left shoulder through the open side window to keep the runway lights in sight.  It was a continuous turn through 270 degrees and I was then aligned with the 32 runway, with the Tenyo Maru ahead to our left.

At some stage Phil cleared us to land.  An academic instruction at this stage - we would be landing ready or not!  To land, I needed to look ahead through the rain splattered windscreen.  It was like looking through frosted glass.  The misshapen appearance of the runway lights through the heavy rain on the windscreen was far from their normal crispness; the lights at the far end of the runway were barely visible. Oh for some windscreen wipers right now!  But at least the runway lighting array was in reasonable perspective for landing.

I kept the indicated airspeed at 105 knots.  Anything less without flap, and with anything other than gentle control movements, could result in a stall.  The approach of necessity was very shallow but we were stabilised in the slot for landing with engine power a little above idle and we would be on the ground in about 8 seconds.

And then - Holy bloody catfish!!  The runway lights disappeared and I was looking up at the dark shape of the Tenyo Maru beside us!  It felt as if a sudden downwash of wind off the end of the runway was forcing us down and we were below the level of the runway!

My right hand was on both throttles and I slammed them forward.  Fortunately both engines responded; an engine failure now would be disastrous.  It was a "go around" situation for a second approach but this was out of the question.  I knew there were cloud covered hills to the west which were obscured by heavy rain.  Going around just wasn't on.  The burst of power was for only 3 or 4 seconds, enough to lift the aircraft above the level of the embankment at the approach end of the runway.

I closed both throttles almost immediately because of the need for a low approach over the runway threshold without flaps.  We cleared the embankment at little more than flare height; we were perhaps 5 feet above the runway threshold.  With throttles closed, I eased the control wheel back slightly to flare the aircraft and we were on.

Surprisingly it was a smooth touchdown.  We ran through to the far end of the runway, quite easily done without flaps!  But at least we stopped in the available length.  The wind, which almost had us in the water from the eddy spilling over the end of the runway, was a bonus during the landing run and allowed us to slow from the higher landing speed necessary for a flapless approach. We landed about 25 minutes before the end of official daylight but, by the end of the landing run, it was too dark to read the instruments. I taxied the aircraft along the taxiway through the downpour in the semi darkness to our usual parking position and shut down the engines.


By this time the feeling of total calm had left me.  I was having a controlled attack of the heebie-jeebies.

I went to the control tower to thank Phil for his help, in particular for the runway lighting, without which we would not have seen Lae. I was surprised to see Jack Tierney there.  He shook my now trembling hand and said: "Am I glad to see you!  We thought you were gone.  When you climbed up out of the water you disappeared into cloud."

Interestingly, Lae Tower was the only "tower" I have seen which was at ground level.  It was situated to the north of the runway, about midway along.  From memory it was built on housebuilding piers; the floor may have been 3 feet above the ground.

Even though it appeared to Jack Tierney that the aircraft entered cloud, it didn't, but his comment gives an indication of the local weather conditions.  A flapless landing requires a higher than normal approach speed and the aircraft needs to be lower over the threshold than normal even when landing on a runway of adequate length.  And when the runway is around the minimum length needed for a normal flapped landing, there is little margin for error when landing without flaps.  That is, the least height over the threshold with safety is making the most of the available landing length.

I lifted the aircraft just enough to clear the embankment.  I believe we cleared it by about five feet.  Jack McDonald said later: "I thought we went through it!" (It's the only time I have ever needed to climb an aircraft on a landing approach to flare height!).  Jack Tierney's comment "you disappeared into cloud', was somewhat remarkable, especially when his viewpoint was near ground level and only about 2,000 feet horizontally from us.  His view was probably obstructed by some intervening low scud.

After Jack's generous greeting, Phil, who was then off duty and about to leave the tower, was not so affable.  He said: “I suppose you think you are a bloody hero." His comment was like a body blow.

I said: "Phil, the forecast looked good and so was the actual weather at Lae when I filed my plan at Madang.  There's no way I would have come to Lae had I known what the weather's like."

Phil seemed a little unsettled from his usual aloof self.  I think he was unwinding after preparing for a possible prang on or close to the aerodrome.  It was unusual that he had asked Jack Tierney to come to the tower.  He probably had in mind Jack's usefulness in the event of an accident - knowledge of crew etc.

Phil softened, and when we discussed the situation a little more, he indicated that he had been calling us on HF from the time we reported at the half-way position, telling us to go back to Madang - that Lae was closed because of adverse weather.  He was receiving our HF transmissions and he was unaware that our HF receiver was faulty.  Even though I had transmitted on several occasions that we were not receiving on HF, he somehow had the idea that we were.

Phil also mentioned that the Mandated DC3 aircraft which had left Lae for Madang early in the day was the only aircraft to depart from Lae all day.  The low cloud and heavy rain had moved in soon after its departure and there had been no significant improvement.

When I'd spoken to the Lae forecaster and obtained the actual weather conditions at Lae before taking off from Madang, it was during the only 15 minute break in the weather all day.  The weather closed in again soon afterwards with persistent torrential rain.

There is often a reluctance on the part of weather forecasters to change their original forecast and there was no mention of anything other than suitable conditions when I spoke to him.  He probably believed the break in the weather was consistent with his original forecast and that the improvement would continue.

We must have been very close to the sea surface when we encountered the downwash on final approach.  Had the flight terminated in the sea near the Tenyo Maru we would have joined the earlier Hudson when it had rolled over and crashed in almost precisely the same position some years before.

The height of the Hudson is 11 feet 10˝ inches from the ground to the top of the pilot's compartment.  The pilot's head would be about 18 inches below the highest point of the compartment.  That is, sitting in the control seat with the aircraft on the ground my eye height would be about ten feet above the ground.

On final approach when my eye level was below the level of the runway lights, the wheels would have been ten feet or more below the level of the runway.  When the runway reference point was eleven feet above mean sea level it didn't leave much margin.  Tidal movement at Lae is only a few feet, but perhaps it was low tide, or perhaps the end of the runway was slightly higher than the runway reference point.  In any event we were bloody close!

Jack Tierney drove us to the DCA mess in the Adastra Jeep.  We were the last to arrive for the evening meal and, still unwinding from the recent events, my appetite was somewhat subdued.  I mention the evening meal only because of something Brian Smith did which was unusual.

We had eaten our main course and dessert was being served.  Brian was sitting next to me and he received his plate of dessert first.  He passed it to me - as if to say: "Thank you." Totally unnecessary, my sense of self-preservation had been working overtime.  It was a simple, unexpected, gesture.  I didn't feel like eating dessert (unusual for me!) but I did anyway.

Also during the meal, Brian, having done some Private Pilot training, said he thought we were in quite a difficult situation.  In the cabin he had been listening to the radio transmissions with headphones.  He said he was surprised at how calm my voice sounded and, because of this, he thought perhaps it wasn't as bad as it looked.  His comment about the calmness of my transmissions was confirmation to me of the strength of this strange tranquillity I had experienced.  I didn't have the heart to tell him that at that stage I had written us all off.  I made some inane response that I wouldn't want to repeat the performance before breakfast every morning and left it at that.

But this unusual calm feeling, change of perception, change of consciousness, serene peacefulness, call it what you will, needs to be experienced to be believed.  It was about six months before I felt comfortable about mentioning it to anyone, even though the experience was rarely far from my thoughts.  It was so unexpected, so inconsistent with the need for intense concentration and rapid, positive responses, and it was just suddenly there.  Where did it come from?  Outside, inside?  It was certainly an internal experience which I found to be a distinct advantage in the circumstances.

I have since heard of other people experiencing something similar in life threatening circumstances when death seems imminent and inescapable.

I wonder whether it's a kind of internal reaction, common to all life forms in near-death circumstances.  For example when a mouse "freezes" before being swallowed by a snake, is it suffering from shock, has it given up on its future existence, or is its state of immobility an indication of some internally produced anaesthesia?

My experience seemed to be beyond panic and, apart from the acceptance of a remote possibility of survival, I had virtually given up on our future existence.  However the feeling was by no means immobilising.  As well as being grateful that we survived, I am grateful to have had the experience.

Many changes have taken place in Papua New Guinea since 1962.  In recent years the Tenyo Maru slid into deep water and pilots were deprived of a valuable approach aid.  But Lae no longer exists as an aerodrome either.  It was disbanded in favour of Nadzab some years ago, Nadzab being about 20 miles inland, in the Markham Valley.  On recent maps Cape Cretin, near Finschhafen, has been renamed "Schollenbruch Point."

Jack McDonald’s "fright" in VH-AGX was not the reason he didn’t fly in Lockheed Hudson aircraft again.  The company insured aircrews against the possibility of accident but they did not insure Jack.  Had he been involved in an aircraft accident there would have been only workers, compensation.  So Jack's decision not to fly applied to all Adastra aircraft.  Jack flew often in airline aircraft and, being Jack, he did in fact fly, uninsured, in the company's DC3 when he was called upon for other technical problems in the field.

And now that there is only an occasional Lockheed Hudson flight anywhere in the world by carefully qualified enthusiasts who have temporarily brushed the museum dust off its wings, I believe I am safe in claiming, with doubtful honour, that Jack's final Hudson flight was with me.



This manuscript can be read in different ways.

To a school teacher it could be read as:  an essay
To my sister Laurel: a cliff-hanging piece of creative writing
To a publishing house: a feature
To the Reader's Digest: an unforgettable experience
To the Department of Civil Aviation (as it was): a detailed Air Safety Incident Report (CA225)
To Phil Graham, Lae air traffic controller: a potential accident going somewhere to happen
To fellow members of the "RA Club":  a relief that this topic finally can be put to rest
To Jack, Brian and myself:   a  20 minute adrenalin rush



Sincere thanks to Jack McDonald and Brian Smith, who calmly and helpfully shared the experience with me.

Sincere thanks also to my sister Laurel Dumbrell, to Mike Wood, to Jack McDonald and to Margaret for proof reading several drafts, to Dean Darcey and Gordon Phipps for clarifying the accuracy of Papua New Guinea aeronautical information, to the unknown Madang photographer, to Phil Graham whose illuminating actions saved the day, to fellow members of the "RA Club" for their encouragement, and to Margaret for her helpful comments and for cheerfully accommodating disturbed nights while I wrote and re-lived this episode from many years past.

And thank you Jack for getting our saturated engine started.  Had it taken you 20 minutes longer there would have been no story to tell because we wouldn't have flown that day.  But please don't be so darned efficient next time!

lst Australian rights only. (c) Copyright W.H. Bowles June 1995