On the morning of July 28, 1943, a Japanese destroyer captain watched through his binoculars as an American twin-engine bomber approached at mast height and opened fire—not with bombs or rockets, but with a genuine 75mm artillery cannon. The same type of field gun his own army used on land was now firing from the nose of an aircraft at 2,000 yards. That evening, he filed a report stating his ship had been attacked by an American warship hidden beyond the horizon, because no other explanation existed in his manuals. There was no warship.
There was only a modified B-25 Mitchell carrying the same 75mm M4 cannon being installed at that moment in the turrets of every Sherman tank rolling off assembly lines in Detroit. This is the story of the strangest weapon of the Pacific War. A weapon genuinely feared by the men it was used against, and despised by nearly every American pilot who flew it. It was quietly withdrawn from frontline service within seven weeks of its first combat mission.
But before it was pulled, it did something no aircraft in the history of warfare had ever done before. And a 43-year-old former Navy chief petty officer named Paul Gunn, with four children held in a Japanese internment camp across the water, was the man who pulled the trigger. The sea north of New Britain was perfectly calm that morning when 12 B-25 Mitchells flew west at wave-top height, close enough for sea spray to occasionally touch their wing edges. Leading the formation was Lieutenant Colonel Paul Gunn.
A naval enlisted man who had returned from retirement after Pearl Harbor. A mechanic by trade, with a 9-and-a-half-foot artillery barrel mounted in the nose of his aircraft, called “Lil Fox. ” It was the first B-25 in the Southwest Pacific theater to carry the new 75mm cannon. The lookout in the top turret reported two destroyers at 10 degrees off the starboard bow, sailing west.
They turned to intercept. What happened in the next 90 seconds was recorded in after-action reports in two different ways. One account, which rose through the chain of command to generals in Washington, stated the cannon hit the destroyer four times out of four shots. A perfect score.
The other account, kept in the private diaries of the men who flew that mission, said something quieter and more honest. Only one shot hit the target. But that shot did not hit the hull or the deck guns. It struck the fire control system—the most vulnerable part of the entire ship, the single component that directed every gun on that destroyer toward its target.
Without it, the ship’s anti-aircraft batteries could still fire, but they could no longer coordinate. Each gun crew worked alone, relying on eyesight, guessing distance, and wasting ammunition at an empty sky toward a target that had already vanished. The destroyer kept sailing. It did not sink.
It simply lost its ability to respond accurately for the rest of the evening. Here is the part never included in any official account. The destroyer was finished later that day, not by the 75mm cannon at all, but by a standard 1,000-pound bomb dropped by another aircraft in the same formation. Paul Gunn knew this.
Yet he submitted the official paperwork, crediting the cannon. The entire program depended on showing the weapon’s success, and Gunn had been in the Navy long enough to understand exactly how those matters worked. The formation turned east toward Cape Gloucester. This could have been where the story ended.
But as “Lil Fox” flew low over the airstrip, still carrying ammunition, a Japanese transport plane moved along the runway below. Gunn saw it through the windshield. He did not climb. He did not fire a warning.
He aimed directly at the aircraft like a tank gunner engaging another tank in an open field. Then he pulled the trigger. The first round tore through the transport’s engine. The second exploded somewhere inside the cockpit.
At that moment, for the first time in the history of air warfare, a moving aircraft destroyed another aircraft using direct field artillery fire—not a machine gun, not a rocket, not a bomb, but an artillery piece, the same class of weapon that had thundered behind World War I trenches. Word of the kill reached high command within hours. That evening, still in his flight suit, Gunn sat in a tent at the forward airstrip and wrote his report. He did not write that the cannon was a triumph.
He wrote, in the driest language possible, that the cannon was not worth the trouble. He was not complaining. He was warning. There was a reason Gunn understood how these things worked.
Two years earlier, in December 1941, he was running a small airline in Manila. He was 42, retired from the U. S. Navy after twenty years of service, married to a woman named Clara—called “Polly” by everyone in the family—and raising four children.
When the Japanese bombed Manila the week after Pearl Harbor, the U. S. Army in the Philippines enlisted Gunn immediately. He was made a captain within days and spent weeks flying supplies into Bataan under Japanese air cover.
When Manila fell, the Army evacuated him to Australia. His wife and four children were left behind. They were interned with all other American civilians at Santo Tomas Internment Camp on the grounds of the old university, where they spent the next three years suffering hunger and disease, cut off from any news of the husband and father they had last seen climbing into a cockpit outside Manila. Gunn did not talk about it much.
But every man who served under him understood that when Colonel Gunn walked to his aircraft at dawn, he was not on an ordinary mission. He carried a special record in his chest with four names on it, and he worked on completing it day after day. In some ways, the cannon in the nose of “Lil Fox” was the least strange thing Gunn had installed on an aircraft. He had spent most of the year welding extra .
50-caliber machine guns into the noses of twenty light bombers, working from a hangar at Eagle Farm outside Brisbane with a North American Aviation field representative named Jack Fox. Together, they turned the medium bomber into an entirely new kind of aircraft—not by requesting permission from Washington, but by picking up a wrench and starting work at 4 a. m. when the coral runway was still empty.
To understand why the man who just made history with a 75mm cannon was writing against it that same night, you have to go back more than a year to a problem quietly killing American pilots in the Pacific. In early 1942, the standard American approach to attacking Japanese ships was to send medium bombers over the target at 10,000, 12,000, or 15,000 feet and drop conventional bombs. The theory was clear. The application was catastrophic.
A ship moving at 15 knots travels about 25 feet per second. A bomb dropped from 15,000 feet takes almost half a minute to reach the water. In that time, the ship has moved, changed course, and done everything a competent captain does when he hears aircraft overhead. Most estimates put accuracy against moving ships from high altitude at less than 1%.
One hit per 100 bombs dropped. The crews flying those missions felt it on every sortie. Long approach, anti-aircraft fire closing in, a bomb tumbling through the air, and 30 seconds later, a column of white water rising harmlessly beside a Japanese transport still heading for New Guinea. In August 1942, a man named George Kenney arrived in Australia to take command of the Fifth Air Force.
He was 53 years old. He had flown combat missions in World War I, when aircraft were still made of wood and fabric. He looked at his crews’ accuracy numbers and their casualty rates, then asked a different question: “What would happen if no one in the Fifth Air Force ever bombed from 15,000 feet again? ” The answer came to be known as the low-level attack.
Flying at mast height and skipping bombs across the water like flat stones, letting them slam into the side of a ship instead of missing from above. They called it “skip bombing. ” In early March 1943, in the Bismarck Sea, skip bombing proved itself with devastating results. A Japanese convoy moving south from Rabaul toward Lae—eight transports, eight escorting destroyers, and nearly 7,000 troops—was destroyed over 48 hours.
Transports were torn apart. Half the destroyers sank. Thousands of Japanese soldiers drowned in the Bismarck Sea. The Fifth Air Force had a victory.
But the men who flew those missions knew the cost. To skip a bomb into a ship’s side, a pilot had to maintain a perfectly straight, perfectly level course at low altitude, directly into the ship’s defensive fire for long seconds. There was no jinking, no evasive maneuvering. Crews were dying in those sorties in numbers no one wanted to say aloud in the briefing tent.
And even with skip bombing working as planned, there remained a target type the Fifth Air Force could not reliably destroy in a single pass: the destroyer. A skipping bomb hitting a destroyer amidships might penetrate the outer hull, but destroyers were built from keel to top to absorb exactly that kind of localized damage. A single ricochet bomb would not sink a destroyer outright. It would wound it, slow it, maybe set it on fire, but the ship would keep sailing and keep firing.
What Kenney needed was something that could knock out a destroyer in one pass. One approach, one weapon, one hit that guaranteed sinking the ship before its gun crews found time to range in on the second aircraft coming behind. The solution the Army settled on was direct: take the M4 75mm cannon—the same gun being mounted in every M4 Sherman tank production line in Detroit—and mount it in the nose of a B-25 Mitchell. If a bomb dropped from above could not reliably destroy a destroyer, fly straight at the destroyer and shoot it, like a tank gunner firing at an enemy tank across open ground.
North American Aviation won the contract. The co-pilot’s seat was removed entirely. A 9-and-a-half-foot barrel extended through the nose, its muzzle protruding from a reinforced steel housing where the glass bombardier’s compartment had been. The breech sat directly behind the pilot’s left shoulder, close enough for him to reach back and place his palm on cold metal without leaving his seat.
Four hundred new aircraft were ordered directly from the production line. Sixty-three existing B-25Cs were pulled from service and converted. In total, 463 cannon-armed B-25Gs entered service. The Army was confident.
The engineers who designed the modification were confident. The generals who signed the purchase papers were confident. The men who actually had to fly the aircraft were far less confident, and they did not take long to say so. Ground crews at forward bases in New Guinea saw a plane that looked lopsided.
The nose was heavier than it should be. The familiar glass bombardier’s compartment was gone, replaced by a steel structure extending into the fuselage like the barrel of something that did not belong on an aircraft at all. They walked around it slowly before its first flight, the way one walks around a strange dog in a strange yard, unsure whether it should be trusted. To understand why the men who flew the B-25G hated it, you have to sit inside it.
The 75mm M4 cannon was 9-and-a-half feet long from breech to muzzle, weighing nearly 900 pounds unloaded. The ammunition rack was mounted directly behind the pilot’s left shoulder, close enough for the man in the left seat to feel cold metal against his flight jacket when he leaned back on a long transit flight over open water. The man responsible for loading that gun was still called “navigator” on paper. In practice, everyone in the squadron began calling him something else: the gunner.
His workspace was a narrow passage squeezed between the pilot’s seat and the top turret gunner’s position behind him. A space never designed for a grown man in a flight suit and flak vest to twist, turn, and reload a gun while under fire. He could not stand fully. He could not sit fully.
He worked on his knees or in an uncomfortable half-crouch, bracing one boot against a support strut to keep his balance when the aircraft banked. Each brass cannon round weighed about 15 pounds—a full bag of potatoes, a large housecat squirming to escape your arms. Now imagine lifting that weight in a space smaller than a closet while the aircraft bounces in turbulent air, while the target rushes toward you through the windshield, while every crew member depends on you to move faster than the Japanese gunners on the deck below can aim. The gunner picked up the round.
Twisted in a space not meant for twisting. Pushed 15 pounds of brass and steel into the open breach. Slammed it shut. Then reached forward and tapped the pilot’s shoulder once.
“Ready? ” That tap was the complete signal. No radio call. No verbal report.
Just one gloved hand on leather. Then the pilot pulled the trigger. For a full second, the aircraft tried to tear itself apart around every man inside it. On the ground in daylight, with the plane sitting perfectly still on the runway and an instructor standing behind the shoulder with a stopwatch, a trained gunner could load a round in under half a minute.
In combat at 230 miles per hour, with the aircraft pitching through turbulence and the target rushing through the windscreen, that half-minute felt like a lifetime compressed into thirty seconds. And that was only loading. Firing was a different experience entirely. When the pilot pulled the trigger, the cannon recoiled so violently that engineers in California had to design a special cradle to absorb it—a rail system allowing the 900-pound gun to slide back 21 inches inside the airframe before hitting the recoil stop.
Twenty-one inches of controlled recoil built into the aircraft structure. Still, it was not enough to make firing feel easy. The entire aircraft shuddered with every round. Speed dropped noticeably at the moment of firing, as if the plane had hit something solid hanging in the air.
Rivets popped routinely, and ground crews learned to expect a handful of loose rivets after any mission involving the cannon. Then there was the smoke. The 75mm round is fixed artillery ammunition, complete with propellant, brass case, and projectile fired together in a violent chemical reaction. When that reaction occurs inside a closed metal tube barely wider than the round itself, combustion gases have nowhere to go.
Some go forward through the muzzle. A great deal rushes back directly into the cockpit after two or three consecutive shots. The compartment was flooded with a thick cloud of cordite smoke, sharp enough to sting a man’s eyes until they watered, bitter enough to settle in the back of his throat for hours after the mission. Pilots completed the rest of their attacks half-blind, squinting to hold their aim through a haze that would not clear, hoping the top turret gunner could still see the destroyer they could no longer distinguish through the glass.
Add another difficulty on top of all that. In practice, most crews could not fire more than two to four rounds in any single attack. Not because ammunition ran out—the plane carried 21 rounds—but because the window for attacking a moving ship at wave-top level was extremely short, and the loader could not reload faster in that narrow, bouncing passageway. After those first few rounds, the target had already slipped under the nose, or the pilot had to yank hard to avoid colliding directly with the ship’s masts.
Then came the part every man in that compartment dreaded far more than the firing itself. The pilot had to climb. He had to turn, and he had to come around for a second pass. During that turn—low, slow, and painfully predictable—every anti-aircraft gun on every ship in that convoy had a clear, stable view of a target that no longer surprised anyone.
The first pass had been made in silence. The first pass was over before the deck gunners found time to react. The second pass was a different matter entirely. The second pass was flying into fire that was waiting for you.
Based on dozens of after-battle reports from gun crews over those 49 days, here is what one of those second passes looked like. Somewhere along the north coast of New Guinea near Huon Gulf. A squadron of four B-25s, two of them cannon-armed G models, spotted a small Japanese convoy of three transports and two escorts heading east in pale dawn light. The first pass had already been made.
The lead cannon-armed aircraft fired three rounds at the largest transport. Two missed. One hit somewhere below the bridge, and a column of black smoke was rising from its superstructure. It was still moving.
It was not sinking. It had not even slowed noticeably. The crews on deck were now fully alert, tracking every aircraft in the formation with every weapon they had. The pilot of the second cannon-armed B-25 banked hard right to come around again.
His airspeed dropped below 200 miles per hour as he entered the turn. He lost altitude with it because he needed to get back to wave-top level before crossing the transport’s bow. For the next 22 seconds, his aircraft became the slowest, most exposed, most predictable target in that entire stretch of ocean. Every gun on every ship in that convoy tracked him through his turn.
Japanese anti-aircraft crews did not have to guess his destination. They had watched American aircraft execute the same maneuver in the same order over the same water for months. They pre-aimed their barrels through the arc. They led the turn.
They opened fire. The first tracers passed the right wing of the B-25. The pilot felt the aircraft shudder as a round hit somewhere behind the cockpit. The gunner in the passage behind him had already loaded the next round and was bracing himself against the strut.
Sweat ran into his eyes, and his ears rang from the previous shot. The pilot came out of the turn. The transport was dead ahead now, growing in the windscreen. Two thousand yards.
Fifteen hundred. Twelve hundred. He let the stream of . 50-caliber tracers climb along the hull until it reached the waterline.
He fired the cannon. The aircraft shuddered. Airspeed dropped. Powder smoke filled the cockpit.
And through that smoke, for one long second before the pilot had to pull away over the ship’s superstructure, he watched a small round hole appear in the transport’s side just above the waterline, followed by a much larger bloom of orange flame exiting the exit hole on the far side of the hull. The transport did not survive the day. The B-25 returned to base with 14 bullet holes in the fuselage, a bullet through the vertical stabilizer, and a gunner in the back who did not speak to anyone for the first hours after landing. That was what a good day looked like.
One way to describe what those men were doing: it was like approaching a man holding a knife, stopping six inches away, and calmly reloading your pistol, hoping he did not move while you did it. Crews filed their post-mission reports. In the evening interrogation tent, they compared notes and began a conversation with their commanders that would not stop for months. The conversation followed the same pattern, mission after mission: “We want the cannon removed.
We want six . 50-caliber machine guns in its place. Eight if the airframe can take them. ” Commanders listened, nodded, and told the crews to keep flying the aircraft they had.
There were still hundreds of these B-25Gs in the theater waiting to be used, and the men who ordered them from North American Aviation were not yet ready to admit the men in the cockpits knew something the procurement officers in Washington did not. Meanwhile, on the receiving end of this cannon, the men on Japanese ships were filing a completely different kind of report. Their side of the story tells us something American papers alone could not fully explain. Japanese convoy escort ships in 1943 carried anti-aircraft guns arranged around a specific set of expected threats: high-altitude level bombers, low-altitude torpedo planes, and dive bombers attacking at steep angles.
Every gun platform on every deck, every firing arc they trained on, every drill crews had practiced hundreds of times, was designed for one of those three patterns of attack—and nothing else. What none of those guns were designed to deal with was an aircraft approaching level at mast height, flying straight toward the bow in a straight and level line, closing at over 200 miles per hour. The geometry of the problem was simply wrong for the men trying to defend against it. A gun crew trained for years to track a dive bomber descending at a steep angle from above suddenly had to deal with a target coming straight at them along the water’s surface, hugging the waves, closing the distance at terrifying speed.
From the moment a lookout spotted a low-flying B-25 on the horizon to the moment it passed the ship and flew away, the time window for doing anything was measured in seconds, not minutes. Most gun crews simply could not respond in time. The crews that did respond often fired at a target flying so low that the mechanical limits of their gun mounts would not allow the barrels to depress far enough to catch it. The guns physically could not aim down that far.
American pilots chose to fly that low for exactly that reason. It was the logic of the entire attack, built deliberately around a weakness the Japanese gunnery could not train its way past. But the cannon added something low-level strafing had never provided: range. A B-25G could open fire on a target from nearly 2,000 yards away—more than a mile.
At that distance, a Japanese lookout was often still trying to identify what kind of aircraft he was looking at. No one expected artillery fire to come from that direction. There was no visible bomb falling from above to warn of danger. No torpedo wake cutting through the sea.
Only a quick flash from the nose of a distant aircraft. A fraction of a second of silence. Then the round hit its target. The first time Japanese crews encountered this, they did not realize what had struck them.
The entry hole in the hull was clean, almost surgical, with no trace of the fire or explosion caused by a bomb hitting above the waterline. Damage was concentrated deep inside the ship where the round had come to rest before losing energy and detonating. Early Japanese damage assessments described the injuries as consistent with a new type of naval gun fired from a nearby surface ship—not from an aircraft overhead. Re-read that for a moment.
The reports said the ships had been hit by another warship. Lookouts scanned the horizon for an American cruiser that did not actually exist. It took several weeks and the recovery of a handful of empty brass casings for the picture to begin clearing on the Japanese side. American crews had received early orders to collect their empty casings during exercises rather than simply jettisoning them over the target area—precisely so Japanese intelligence would not have an easy opportunity to identify the weapon before the element of surprise was exhausted.
Even when the picture began to clear, the response from the Japanese side revealed something telling about how institutional thinking bends—or fails to bend. In the middle of a war that had already outlived its written manuals, Japanese naval doctrine classified enemy air threats into three ordered categories: bombers, fighters, and torpedo planes. An aircraft flying low and firing artillery shells directly did not fit comfortably into any of those categories. Reports that eventually rose through the chain used a phrase translated roughly as “aircraft artillery”—not spoken as a compliment.
It was a classification problem and an admission from the men writing the report that they had no existing category under which to file this threat. There were no fixed tactics in the manuals to counter it, no convoy crew drills designed specifically for this danger. Their response, understandably, was to ask for more: more escorting destroyers per convoy, more fighter cover, tighter ship formations to ensure mutual protection. On paper, each of these requests made sense.
In practice, each one made the problem measurably worse. More escorting destroyers simply meant more individual targets within range of the cannon-armed B-25s and the skip bombers flying beside them. A tighter convoy formation meant that once one ship was hit and began slowing, the surrounding ships had far less open water to maneuver away from the attack path before the next round arrived. The truth that emerged during those seven weeks was more complex than the documents of either side indicated at the time.
The cannon was ultimately not the reliable ship-killer its designers at home had promised. Against destroyers, the round penetrated almost every time, punching through outer hull and armor at combat range, just as the specification sheets claimed. But destroyers were compartmented ships designed deck by deck specifically to absorb that kind of localized damage and keep fighting. The hits wounded, but rarely finished the ship outright.
Most often, the damaged destroyer had to be finished off by skip bombs dropped from the other aircraft in the same formation—aircraft carrying no cannons at all. Against unprotected transports, the story reversed completely. One 75mm round striking the waterline of a fully loaded troop transport—a thin-hulled ship carrying fuel, ammunition, and hundreds of men crowded below deck—could cause flooding no damage-control team on earth could contain in time. These were exactly the kind of ships the cannon had been designed to kill, even if no one in Washington wrote that distinction into the original proposal.
Crews in the field learned to sort their targets accordingly without waiting for an official order. Destroyers received skip bombs. Transports received cannon fire. This was not doctrine drawn on a blackboard in Australia.
It was doctrine produced by seven straight weeks of violent combat over open ocean, mission after mission, written not in ink on a briefing room wall but in quiet post-battle conversations among exhausted men in their tents, trading notes over lukewarm coffee at the end of another long day. The clearest evidence of what the cannon could really do came not against warships, but against something far more modest. The Japanese supply network along the New Guinea coast relied heavily on small motor barges—wooden-hulled vessels that moved troops and supplies at night, then hid in narrow bays and river mouths by day. They were numerous, and larger conventional bombers had long struggled to find them hiding along the shoreline.
The 75mm cannon performed exactly as its designers at home had promised. One round penetrating a wooden hull loaded with fuel or ammunition did not need a long after-action report to describe its result. Crews began actively hunting barge traffic along that coast, where the cannon worked without exception. One round, confirmed kill.
No frustrating second pass required at all. The coastal supply routes that had been operating quietly at night, during late summer 1943, became a place where Japanese barge crews had real reason to be afraid. During August of that year, B-25Gs flew as part of larger, carefully coordinated formations striking Japanese convoy routes along the coast from Rabaul toward New Guinea. By this stage of the campaign, the standard attack sequence had become precise, drilled, and brutally efficient.
Strafing aircraft entered first at treetop height, neutralizing exposed gun crews on the escort ships’ decks to suppress their defensive fire before the main formation arrived. Behind them came the B-25s themselves—a mix of cannon-armed G models and traditional skip bombers. They divided their attention between destroyers and transports according to the pattern crews had already learned the hard way. The cannon-armed planes lined up on transports from a distance of nearly 2,000 yards, walking .
50-caliber fire toward the hull to adjust range, then fired the cannon the moment the stream of fire reached its target. The skip bombers went after the destroyers in one pass. It did not work perfectly on every mission, but it worked well enough that Japanese convoy commanders began reporting to Rabaul that their current defensive formations could no longer counter the attack pattern they were facing. Convoys demanded more air cover, then more of it, then specifically low-altitude air cover to intercept this new threat of mast-height attacks.
Japanese fighter pilots assigned to protect convoys found the mast-height attack pattern as difficult to counter from the air as the ships’ gun crews found it from the deck. A Japanese fighter diving on a B-25 flying at 50 feet had almost no room to pull out safely. A pilot committed to the intercept risked hitting the sea directly. A pilot who hesitated simply gave the B-25 time to finish its attack and escape toward open water.
Above, American fighters flew cover, occupying the Japanese escorts at altitude, while the B-25Gs below continued their difficult, grinding work of hunting ships, convoy by convoy. By mid-September 1943, the theater’s entire stockpile of 75mm ammunition was effectively exhausted. Seven weeks of intense combat had consumed supplies faster than anyone at Fifth Air Force headquarters had planned. The B-25Gs continued flying missions even when ammunition ran out completely, but they now flew as ordinary skip bombers.
The cannon in the nose fell silent. The gunner sat alone in his narrow passage with nothing to load. Not a single 15-pound brass round remained in the racks that would ever find its way into the breech before that campaign ended. The crews did not complain about this turn of events.
On the contrary, many felt a quiet relief. They submitted their request again, this time formally through proper channels. The same request they had been making informally since the first week of combat. This time, with ammunition stockpiles empty and no clear resupply date, the request carried a weight it had never carried before.
None of this was happening in a vacuum at headquarters either. General George Kenney read every combat report that reached his desk himself—a habit that made him an unusually effective commander and unusually hard to fool with a weapons program that looked good on paper but did not perform in the air. During July and August, the reports stacking up in front of him told two different stories at once. Both stories were true.
One story was of genuine success against barges and transports: ships sinking that would never have sunk from a high-altitude bomb. A weapon doing real work in a real war. The other story was the rising chorus of exhausted, smoke-stained pilots demanding on every mission to get their aircraft back the way they had been. Kenney did not answer either stack of paperwork immediately.
He let seven weeks of combat run their course. He let the records pile up and waited to see which of the two stories the ammunition supply would settle first, the way a man lets a fire burn low before deciding whether it deserves more fuel or is worth letting die out on its own. The empty ammunition racks settled the question that the two stacks of paperwork on Kenney’s desk could not settle by themselves. He had a genuinely useful weapon against soft targets in one column of his ledger.
He had a consensus of exhausted pilots demanding the weapon’s removal in the other column. Now, in mid-September 1943, a third fact emerged that made the decision easier than either column could have done alone. There was simply nothing left to load. Supply ships arriving from Australia carried food, fuel, and letters from home.
They did not carry 75mm rounds in any meaningful quantity. The rounds were spent. The barges along the New Guinea coast had consumed a great many of them. The transports crossing the Bismarck Sea had consumed more.
And in warehouses in Brisbane, the pallets that should have held the next shipment sat nearly empty. Priority in the American manufacturing pipeline had been given to other things: Sherman tanks, field artillery batteries for the North Africa campaign, and every other place a 75mm round could go and do concrete work. Kenney approved the move away from the cannon. Not immediately, and not without some friction working its way up the supply chain toward Washington.
The Army had already paid for 463 of these aircraft. No one in the procurement office was eager to write off that expense without at least a formal argument. But in the field, the debate had already been settled in the skies over New Guinea and the Bismarck Sea long before any memo in Washington caught up with what the crews already knew from experience. And here the path takes a turn rarely mentioned in official history.
Ground crews and armament technicians at forward bases—the same inventive men who had helped Paul Gunn install extra machine guns in the noses of bombers at Eagle Farm the year before—began quietly removing the cannons from the G models on their own initiative. No paperwork, no requests up the chain of command for approval. They simply unbolted the cannon, crated it, and re-rigged the empty nose with extra . 50-caliber mounts using whatever equipment was available in the field that week.
A pair of machine guns salvaged from a wrecked P-40. A gunsight pulled from an out-of-service B-25 sitting at the end of the runway. A steel mounting plate cut and drilled by a mechanic who had learned his trade in a tool-and-die shop in Milwaukee before the war and had never lost the habit of solving problems with his hands. When commanders discovered what their mechanics had done without permission, most of them looked at the modification.
They looked at the growing stack of combat results on their desks. They quietly signed off on work that had already been completed days earlier. That is how the 75mm cannon actually left the B-25Gs of the Southwest Pacific. Not by order from Washington, not by an official announcement from Fifth Air Force headquarters, but by the decision of a mechanic in a canvas tent, wrench in hand and dog tag around his neck, deciding that the pilots he cared about would not fly another mission with a weapon they had been demanding to remove since July.
The official answer came eventually, from North American Aviation, in the form of the B-25H. It carried a lighter cannon, the T3E1, designed from the start for aerial use rather than adapted from a tank turret. Four . 50-caliber machine guns mounted in the nose flanked it on both sides.
Extra gun blisters appeared along the fuselage. The top turret moved forward to add more forward firepower on every attack run. About a thousand of the new model were built quietly, without any real celebration and without official acknowledgment from those who had approved the original program that the men who flew those aircraft had been right since its first week and that the procurement system at home had been wrong. The B-25G was withdrawn from frontline service.
The cannons were removed for good. Machine guns took their place. The powder smoke cleared from those compartments for the last time and never returned. The pilots finally got what they had been demanding since the first week of combat.
It had taken seven weeks of fighting, a steady stream of identical complaints, and the loss of hundreds of expensive aircraft for the Army to finally agree with what its own crews had known since the first moment they pulled the trigger and felt the entire aircraft shudder around them. There is a particular kind of quiet that settles over a squadron when a piece of equipment everyone hated disappears. No parade was held on the day the last cannon was removed from the last B-25G in any squadron. No general came from Australia to give a speech.
No one wrote a memo declaring the argument officially over. Ground crews simply worked through the night on airframe after airframe. The next morning, a pilot climbed into a cockpit that felt different in his hands—lighter in the nose, quieter behind his shoulder—and flew a mission in which no one on board had to time their breathing to the rhythm of reloads. Some of the men who spent seven weeks flying with that cannon behind them said later that the strangest part was not the relief.
It was the speed with which the fear associated with it faded once it was gone. The way a scar stops hurting long before its trace disappears from the skin. Think for a moment about the men who actually lived through those seven weeks. Because the story of any weapon is not really just the story of a weapon.
It is the story of a gunner folded into a narrow passageway smaller than a closet, timing his breathing to the rhythm of the aircraft, learning how to brace his entire body half a second before the trigger was pulled so the recoil would not crack a rib against the turret ring behind him. It is the story of a pilot flying half-blind through powder smoke on a second pass he never wanted to make, holding a straight line through enemy fire because the mission required it and there was no other way to get the job done. Most of the names connected to those seven weeks were never written anywhere beyond the squadron log and a few triplicate after-action reports, eventually stored in boxes in a records depot somewhere in Missouri, where they remain today on shelves in climate-controlled rooms, waiting for a historian with the time to open that box. They did not ask for recognition, and in most cases they never received it.
What they left behind was a set of numbers in a supply ledger, a pile of complaints that turned out to be entirely correct, and a coastline in New Guinea where Japanese barge crews learned the hard way not to move except in complete darkness. Now, here is the final irony of the entire 75mm cannon program. The part that makes this story stranger than a pure triumph or a simple failure. Japan never solved the riddle of the B-25G while the riddle still mattered.
In late 1943, after the cannon had been withdrawn from frontline units, Japanese forces recovered a crashed B-25G somewhere in the Southwest Pacific. The airframe was largely intact after its forced landing. Technical intelligence officers examined the wreckage carefully, and what they found in the nose stunned them. Not a bombsight, not just a row of machine guns like the strafer models they had already learned to identify and fear.
A complete artillery piece extending from the nose to a mounting base beneath the former pilot’s seat. The inspection report that followed used language translating again to something like “aircraft artillery”—the same troubling phrase circulating through Japanese naval channels for months by then, not as praise but as an ongoing classification dilemma no one above them had solved. The Japanese framework for understanding enemy air threats still recognized only three categories: bombers, fighters, torpedo planes. And what sat before these officers, half-buried in the wreckage, fit none of them perfectly.
It was bomber-sized. It carried a weapon for surface engagements—the kind of gun a warship might carry, not an aircraft. There was no item in Japanese Imperial Navy military doctrine dedicated to countering it. No fixed defensive formation designed from the ground up to defeat an aircraft that could open fire from 2,000 yards with an armor-piercing projectile.
Their response, as it had been since the whole affair began, was to request more escorting destroyers and more air cover. The same answer to a question whose nature had quietly changed while their report was still making its way up the chain of command. By the time Japanese intelligence finished processing the wreckage, presented their findings, and began serious discussion of what an appropriate countermeasure might look like, the B-25G no longer existed in the form that had caused them so much confusion in the first place. The cannons had already been removed from every remaining aircraft.
The machine guns had already been installed in their place. The weapon that produced all those damage reports, all that doctrinal confusion, and all those months of genuine tactical anxiety along the convoy routes was already sitting disassembled in a supply depot in Australia, its parts crated and largely forgotten. There is something almost absurd at the core of this timing. An entire intelligence operation, recovered wreckage, meticulous technical examination, weeks of damage report analysis, successive interrogations after every convoy.
All aimed at understanding a weapons system the men who operated it had already been trying to get rid of since the first week they flew it in combat. The pilots who despised the B-25G had won their argument before Japan even finished understanding what the argument was about. Two more threads deserve to be followed to the end. The first concerns Paul Gunn himself.
After the cannon program was withdrawn from frontline service, he returned briefly to the United States to work with North American Aviation on the later B-25H model. He toured the factory floor in California. He sat with the engineers who had designed the original G model and told them, in the direct language of a man who had flown the aircraft in combat, exactly what needed to change and why. When he finished, he flew back to the Southwest Pacific and returned to the war.
In late 1944, on Leyte Island in the Philippines, a Japanese air raid struck the airstrip where he was working. Shrapnel from that attack pierced his body. He survived, but his wounds were serious enough that his combat career was effectively over. He spent the rest of the war recovering—first in a field hospital, then back in the United States, unable to fly the missions his men were still flying without him.
The Philippines were liberated in spring 1945. When American forces reached the Santo Tomas University grounds in Manila, they found more than 3,000 civilians who had been held captive there throughout the war. Among them were Paul Gunn’s wife, Polly, and their four children. They were thin.
They were sick. They were alive. Gunn was reunited with them within weeks. The special record he had carried in his chest for three years was finally closed.
He retired from the Army with the rank of full colonel in 1948. He returned to the Philippines. He rebuilt the small airline he had started before the war. He kept flying.
On October 11, 1957, Paul Gunn was piloting a chartered plane in bad weather in the Philippines when it crashed. He was 57 years old. He died in the crash. He was buried in the country he had first come to as a young Navy pilot more than three decades earlier.
The second thread concerns the weapon itself, because the 75mm cannon in the nose of the B-25G was not the end of that idea. It was the beginning. Less than a decade after Gunn’s death, in the skies over Vietnam, the U. S.
Air Force introduced an aircraft called the AC-47 Spooky—a modified transport mounting multiple side-firing machine guns, flying long orbits over jungle bases at night. The men who flew it were solving the same problem Gunn had solved differently more than 20 years earlier: how to bring direct, sustained fire onto a ground target from the air, using a platform that could carry more ammunition and stay over the target longer than any fighter could. The AC-47 gave way to the AC-119 Shadow. The AC-119 gave way to the AC-130 Spectre and eventually the AC-130 Ghostrider.
These are the great side-firing gunship aircraft of the modern Air Force, still flying today. Among their weapons, they carry a 105mm howitzer, adapted from a U. S. Army field gun, loaded by a crew member standing beside the breech in the back of the aircraft.
Exactly the way Paul Gunn’s gunner loaded a 75mm round into a B-25G 80 years earlier. The recoil is absorbed by a specialized recoil base. Everything Gunn and Jack Fox worked on in an aircraft hangar at Eagle Farm in 1942 is still in service, still doing the job. The weapon that seven weeks of combat convinced its own crews to abandon in 1943 turned out to be the ancestor of the greatest line of gunship aircraft in the history of American military aviation.
Gunn did not live to see it, but every AC-130 that has flown a mission in the past fifty years has been, in a real sense, the aircraft he tried to build in a canvas tent outside Brisbane in 1942. Bigger, heavier, better engineered. But the same idea, the same instinct, the same refusal to accept the calculations of unfair combat and let the enemy go untouched. Some ideas outlive the men who first applied them.
Some ideas outlive the aircraft they were first mounted on. Some ideas outlive the wars they were built to win. And there is one more thing worth saying honestly, because a story like this deserves to be told in full and not left in the shadows. The identity of the destroyer Gunn hit on July 28, 1943, has never been definitively established.
American after-action reports recorded that location, result, and ammunition expended, but they did not—and could not—identify the specific Japanese ship by name. Japanese naval records from the same period, when cross-referenced with every known engagement in that area on that date, have not yielded an exact match either. That happens in Pacific War history more often than most people realize. Some engagements occurred at dusk.
Some involved ships that sank themselves within weeks, taking their logbooks to the bottom with them. Some were never recorded on the Japanese side because the officer responsible for filing that report was killed before he had a chance to write it. Historians are still working on it 80 years later in the archives of Tokyo and Washington. It is entirely possible a name will surface one day.
Until then, the destroyer remains as it was that morning: a gray shape on gray water, hit once by a weapon its crew did not fully understand, finished off by a thousand-pound bomb, with the record since then careful to assign the credit properly. Sailed by men about whom we know even less than we know about the ship they served on. That is the story of the B-25G and the 75mm cannon it carried for seven short weeks in the summer of 1943. Not a miracle weapon that changed the course of the war on its own.
Not a failed experiment that never worked at all. But something stranger and more honest than either of those simple stories. A weapon genuinely feared by the men it was used against, described in their reports as something new, something nameless in their doctrine, something that demanded a response they never finished building in time. And in that same season, a weapon despised by nearly every man who flew it, officially retired from frontline service before the war it was built for even reached its midpoint.
The Army built 463 of these aircraft, flew them through seven weeks of fierce combat, then removed the cannons, replaced them with machine guns, and quietly moved on to the next problem in the queue. Japan kept filing reports of “aircraft artillery” for months afterward, trying to understand a threat that—by the time those reports reached anyone who could act on them—had already been disassembled and shipped south to a depot in Australia. That is the Pacific War in miniature. Not clean victories and tidy lessons learned from a single attempt, but complex machines flown by exhausted men.
Decisions made at thirty feet above open water, with people at home still writing the manuals trying to catch up to what those men already knew, long after the shooting stopped in any given week.