The Granville Brothers, the Gee Bee Racers, and the Most Beautiful Dangerous Thing That Ever Flew Over Cleveland

The Granville Brothers’ Gee Bee Model R was the fastest piston-engine aircraft on the American racing circuit in 1932, winning the Thompson Trophy at an average speed of 252 miles per hour - and killing two accomplished pilots along the way. Built by five self-taught brothers from Springfield, Massachusetts, the Gee Bee remains one of aviation’s most compelling and contested machines: a triumph of pure aerodynamic intent that operated on the narrowest possible margin between record and disaster.

The National Air Races and What Was at Stake

The National Air Races of the late 1920s and 1930s were not a sideshow. They were the central proving ground of American aviation. The Thompson Trophy - a closed-course pylon race flown at full throttle, wings nearly vertical in the turns - and the Bendix Trophy - a transcontinental speed contest - drew crowds of 100,000 people to Cleveland Municipal Airport on September afternoons. Winners became national celebrities. The military paid close attention. Every fast aircraft in America traced some lineage to what happened at those races.

Into that world stepped the Granville brothers: Zantford, Edward, Robert, Mark, and Tom. Their father had run a hardware store in Springfield. None held formal engineering degrees. What they had was a machine shop, an obsession with flight, and a clear-eyed understanding of what winning a pylon race required.

How the Granvilles Thought About Speed

The Granville Brothers Aircraft Company was founded around 1928. Their early work was conventional - biplanes, monoplanes, the output of a thousand small shops that sprang up in the years after Lindbergh’s Atlantic crossing. But they were watching the races carefully, and they had arrived at a conclusion that many established manufacturers were slow to admit.

Reduce drag to almost nothing. Install the largest engine available in the smallest airframe that will contain it. Accept every consequence that follows from those two decisions.

That philosophy produced the Gee Bee - the nickname derived directly from the company’s initials, G.B. for Granville Brothers.

What the Gee Bee Model R Actually Looked Like

The Model R did not look like a conventional airplane. The fuselage was nearly spherical at the front - genuinely barrel-shaped - housing a Pratt & Whitney Wasp engine driving an enormous propeller that was nearly as wide as the fuselage was tall. Behind that engine section, the fuselage tapered sharply into a cone ending at a tail that looked impossibly small to exert authority over the whole aircraft.

The wings were proportionally tiny, attached near the widest point of the fuselage. Their span barely seemed adequate to bank the machine hard around a pylon at competition speeds. The cockpit sat so far aft that the pilot’s position was nearly at the base of the vertical stabilizer, with forward visibility largely blocked by the massive cowling ahead.

It was an extraordinary-looking machine. It was also extraordinarily dangerous.

Why the Gee Bee Was So Difficult to Fly

The weight distribution was the core problem. Most of the aircraft’s mass was concentrated forward in the engine. The small wings generated lift near the middle of the fuselage. The tail had very little mechanical advantage over the whole system. The result was longitudinal instability - the airplane wanted to swap ends.

In straight and level flight, a skilled pilot paying constant attention could manage it. Around a pylon at competition speed, in turbulence and prop wash from other aircraft, the Gee Bee was always searching for an opportunity to overcome its pilot. That was not a figure of speech. It was an accurate description of what the aircraft would do if attention lapsed.

Lowell Bayles and the December 1931 Speed Record Attempt

Before the Thompson Trophy victory came a death that might have ended the program.

Lowell Bayles was one of the finest racing pilots in America - experienced, careful by the standards of that profession, and widely respected by his peers. In December 1931, Bayles was flying the R-One in an attempt on the world landplane speed record, making low-altitude runs at full power. On one of those passes, something failed. Witnesses described the aircraft entering an outside snap roll at low altitude with no recovery. Bayles died in the wreckage. He was 31 years old.

The Granville brothers kept building.

The 1932 National Air Races at Cleveland

At the 1932 National Air Races in Cleveland, the Gee Bee R-One and R-Two were the fastest piston-engine aircraft on the American circuit. The R-One won the Thompson Trophy at an average closed-course speed of 252 miles per hour - in an open cockpit, around a pylon course, with other aircraft in the air around it. The pilot who won said afterward it was the most demanding aircraft he had ever flown. The R-Two finished second in the Bendix Trophy.

That sentence - from a man who was among the most experienced high-speed pilots alive - is worth sitting with.

Florence Klingensmith and the 1933 Chicago Races

In August 1933, at the National Air Races in Chicago, Florence Klingensmith was competing in the Frank Phillips Trophy race. She was among the most accomplished women pilots in the country - a veteran of cross-country competition with thousands of logged hours. She was flying a modified Gee Bee variant. During the race, the aircraft broke apart in flight. Klingensmith did not survive.

By the end of 1933, the Granville Brothers Aircraft Company was bankrupt. The Depression had contracted available capital, and the financial and legal aftermath of the accidents had done the rest. The five brothers who had built the fastest landplanes in America were out of the aircraft business.

A Design Debate That Was Never Fully Settled

Aviation historians have argued about the Gee Bee for decades, and the argument is worth understanding on its own terms.

One position holds that the aircraft was poorly designed - that the Granvilles were self-taught engineers working beyond their competence, and that the instability was a flaw that formally trained aeronautical engineers would have caught and corrected. The deaths were a predictable outcome, not an acceptable cost. That argument has merit.

The opposing position holds that the Gee Bee was exactly what it was intended to be. It was not a trainer or a transport. It was a purpose-built racing weapon optimized for a single metric - maximum speed around a closed pylon course - and on those terms it succeeded completely. The problem was not bad design. The problem was that the margin between success and catastrophe was so narrow that only a handful of pilots alive could work near it without crossing it.

Both arguments hold. That tension is what the golden age of air racing actually was: breathtaking capability produced by people learning as they went, demonstrated in the only laboratory that worked, at a casualty rate that would terminate any equivalent program before its second season today.

The Flying Replicas and What They Proved

For roughly fifty years, the Gee Bee existed mainly in photographs and in the memories of people who had stood at the Cleveland fence in the 1930s.

In the 1980s and 1990s, a serious engineering effort produced flying replicas of the R-One and R-Two, built from original drawings with one critical addition: the stability corrections the originals never received. The team discovered something significant. With the longitudinal instability addressed and the weight distribution corrected, the basic concept was not only flyable but manageable. The airplane that had killed people in the hands of the finest pilots of the 1930s turned out to be a reasonable aircraft once the specific design flaw was fixed.

Steve Wolf flew the replica R-Two at airshows for years. That yellow-and-black barrel shape banking low and fast still stops people who have never heard of the Granvilles. Because even without knowing a single name - not Bayles, not Klingensmith, not Zantford Granville - when you watch a Gee Bee fly, you understand immediately what it was built to do and what it cost.

Why This Story Matters for Aviation

The Gee Bee is a case study in the relationship between performance optimization, stability margin, and human cost - one that flight safety professionals and aeronautical engineers still reference.

The aircraft demonstrated that extreme aerodynamic optimization is achievable outside formal institutions, with limited resources and hand tools. It also demonstrated that a design operating with insufficient stability margins will eventually encounter a situation its pilot cannot recover from. The replica program then confirmed that the original concept was fundamentally sound - the problem was a specific, correctable flaw, not a failed idea.

That distinction is not obvious in real time, under competitive pressure, when the machine is already winning races. It almost never is.

Key Takeaways

The Granville Brothers Aircraft Company (est. ~1928) was founded by five self-taught brothers from Springfield, Massachusetts - none held formal engineering degrees
The Gee Bee Model R won the 1932 Thompson Trophy at an average speed of 252 mph, the fastest piston-engine aircraft on the American racing circuit
The aircraft’s longitudinal instability - caused by its massive forward engine, tiny wings, and undersized tail - made it manageable only for the most skilled pilots under near-ideal conditions
Lowell Bayles (killed December 1931) and Florence Klingensmith (killed August 1933) both died in Gee Bee variants; the company declared bankruptcy by the end of 1933
Flying replicas built in the 1980s–1990s with stability corrections proved the core concept was sound - it was a specific, fixable design flaw, not a fundamentally failed aircraft