The Arrow: Germany’s fastest piston fighter that never fought

How Dornier built the fastest propeller fighter of the war

Dornier Do 335 Pfeil — by the numbers

Baseline note: the figures below describe the Do 335 A-1 (single-seat fighter-bomber, two DB 603E-1, the only armed mark completed in more than token numbers). Deltas for the A-0 pre-production machines, the twin-cannon B-2 Zerstörer and the two-seaters are in grey. Three corrections before the tables. First, the explosive charges did not blow off the lower fin: they blew off the dorsal fin and the rear propeller so that the pilot could eject through the gap, and the ventral fin was a separate, separately jettisonable item meant to be dropped before a wheels-up landing so it would not dig in. Second, the surviving aeroplane was not restored by the Smithsonian in Washington; the Smithsonian shipped it back to its builder in Bavaria and Dornier restored it, several of the men on the job having built the type in 1944. Third, the popular claim that the Do 335 was the fastest piston-engined aircraft ever to reach production survives scrutiny, but only just, and the words "reach production" are carrying the weight — see the speed rows.

Armament & payload

The Pfeil carried a heavy centreline battery and a bomb bay, and never fired a shot in anger

The armament of the Do 335 is easy to describe and slightly unreal to write about, because no version of it was ever used against an enemy. The standard fit was a single 30 mm MK 103 firing through the hollow shaft of the front propeller and two 20 mm MG 151/20 synchronised over the nose engine — a genuinely heavy nose battery with no dispersion problem at all, since every barrel lay within centimetres of the aircraft’s centreline. Below the cockpit sat a weapons bay that could take one 500 kg bomb, two 250 kg bombs, cameras, or, most often in practice, a fuel tank; two wing hardpoints took another 250 kg apiece or the standard 300 l drop tank.

What the aeroplane conspicuously did not carry is as revealing as what it did. Germany in 1945 had guided anti-ship weapons, wire-guided air-to-air missiles and the R4M rocket in service or in trials, and none of them was ever integrated with the Do 335. The type had been ordered as a Schnellbomber, redefined as a multi-role fighter in autumn 1942, and then chased through fighter, Zerstörer, reconnaissance, night-fighter and trainer roles without ever being finished as any of them. Weapon fits below are those specified in the Baubeschreibung and the surviving variant documentation. Because so few airframes were completed, and because several were fitted out for testing rather than for a role, almost nothing here can be checked against an operational aircraft.

Sourcing caveat: gun and ammunition figures come from the Baubeschreibung Do 335 B-0 Zerstörer of 31 May 1944 and from Smith and Creek. Because the type never entered squadron service, there are no armourers’ returns, no expenditure records and no combat reports to check them against — unusually for a Second World War aircraft, every armament figure here is a specification rather than an observation.

Variants

Dornier spent thirty years arguing for tandem engines and got five months of production out of the argument

Claude Dornier began putting engines nose-to-tail in 1915, in the tandem nacelles of his wartime flying boats, and never stopped: the Do J Wal of 1922 and the twelve-engined Do X of 1929 both used the layout, because for a flying boat it keeps propellers out of spray and machinery on a single strut structure. In 1937 he patented the specific refinement that made the Do 335 possible — a pusher propeller driven by a remote engine through a long extension shaft — and in 1939 he had Ulrich Hütter and Schempp-Hirth build a validator for it. The Göppingen Gö 9 was a 40 per cent scale Do 17 fuselage and wing with no nacelles, a retractable tricycle undercarriage and an 80 hp Hirth HM 60 buried amidships turning a tail propeller through a tube. It first flew in June 1941, reached about 220 km/h, and showed no unexpected vices at all. The idea worked.

The aerodynamic case was and remains sound. Two engines in the fuselage present roughly the frontal area of one, so a twin-engined aircraft can have a single-engined aircraft’s drag. The mass sits on the centreline, so the roll inertia is low and the roll rate high. There is no torque couple in normal flight and no asymmetry when an engine quits. The costs are equally real, and every one of them showed up: a buried rear engine with no clean airflow is hard to cool; a rear propeller means an ejection seat, explosive bolts and a tall undercarriage; a 3 m driveshaft is another thing to build, balance and maintain; and the whole rear fuselage becomes an engine bay, which is where the Farnborough aeroplane’s fatal fire began. The layout is not wrong. It is simply expensive in exactly the ways that a country losing a war cannot afford.

Production tells the story. Hitler gave the Do 335 maximum priority under the Jägernotprogramm on 23 May 1944; Friedrichshafen was to deliver 120 pre-production machines by March 1946 and Munich over 2,000; a bombing raid destroyed the Manzell tooling and forced a new line at Oberpfaffenhofen; the He 219 was ordered cancelled to free DB 603 engines and capacity for it, an order Ernst Heinkel simply ignored. When the US Army reached Oberpfaffenhofen in late April 1945 the yard held rows of Do 335s, most of them without engines.

Prototypes V1–V12, M13–M17 (1943–45 / at least 16 flown)

W.Nr 230001–230017. V1 CP+UA first flew at Mengen-Hohentengen on 26 October 1943 with Hans Dieterle and made 27 flights. V2 followed on 31 December 1943 with DB 603A-2s, V3 on 20 January 1944 with pre-production DB 603G-0s and a revised canopy. V6 did the dive testing; V9 was still at Rechlin in April 1945 and flew one of the last recorded sorties of the type. Engine marks varied across the batch: DB 603A, A-2, G-0, E and E-1

Do 335 A-0 (from July 1944 / 10 aircraft, W.Nr from 240101)

Pre-production single-seaters, the first VG+PG delivered in July 1944. Roughly eleven A-0 airframes were later converted to A-11 trainer standard

Do 335 A-1 (deliveries from January 1945 / 11 completed)

The production single-seat fighter-bomber: DB 603E-1, MK 103 and two MG 151/20, internal bay, two wing hardpoints. Eleven were complete at Oberpfaffenhofen when the plant was overrun

Do 335 A-3 and A-4 (1944–45 / conversions and prototypes)

Reconnaissance. The A-3 was to be converted from A-1 airframes with cameras in the weapons bay; the A-4 was an unarmed long-range machine with a smaller camera pair. A high-speed unarmed reconnaissance aircraft was arguably the role the Do 335 was best suited to and least developed for

Do 335 A-6 (1944–45 / not completed)

Two-seat night fighter with a second cockpit raised behind and above the pilot’s, and FuG 217 or FuG 218 Neptun radar. The reason the He 219 cancellation order was issued in the first place; the reason it was a mistake is that the A-6 never appeared and the He 219 did

Do 335 A-10, A-11 and A-12 (1944–45 / roughly 11 A-11 conversions, 2 A-12 completed)

Two-seat trainers, the Ameisenbär. The A-10 and A-11 were conversions of prototype and A-0 airframes; the A-12 was the production trainer, of which exactly two were finished. That a type with barely three dozen airframes devoted a third of them to conversion training says everything about how hard it was thought to be to fly

Do 335 B-1 to B-6 (1944–45 / two airframes, both B-2)

The heavy series: B-1 fighter, B-2 Zerstörer with two extra wing MK 103 (M13 and M14, the only ones built), B-3 and B-4 with the long wing and DB 603L engines, B-5 and B-12 two-seat trainers, B-6 two-seat night fighter. None reached service

Do 435, Do 535 and P 256 (1944–45 / one possible airframe, otherwise paper)

The successor projects, and the arc of an idea being abandoned. The Do 435 is described in English sources as a Do 335 with the redesigned longer wing and in German ones as a lengthened fuselage with Jumo 213 engines; Allied intelligence reported sighting one at the Dornier field at Löwenthal in early May 1945, the only physical evidence either way. The Do 535 was the Dornier P 254 handed to Heinkel in October 1944 and properly called the He 535, with the rear piston engine replaced by a turbojet. The P 256, drawn against a requirement of 27 February 1945, gave up centreline thrust altogether for two podded HeS 011 turbojets and four MK 108 — the layout’s own house conceding the argument

Do 635 (1944–45 / mock-up and wind-tunnel model only)

Also called Do 335 Zwilling or Junkers Ju 8-635: two Do 335 fuselages on a common centre section, four DB 603s, two Rb 50 cameras in the port hull and provision for five 60 kg photoflash bombs. Junkers took the work over under Heinrich Hertel, who planned a first flight in late 1945; the Japanese delegation reviewing German projects at the end of 1944 was impressed by it. Landing wheels were shared with the Ju 352

Licence production and exports (none)

There were none, and there is no export designation to record. Second and third assembly sites were selected inside Germany — the Weingut II bunker in the Welfen barracks at Landsberg am Lech, and the Heinkel works at Oranienburg — but no Do 335 was ever built outside Dornier, and none was ever supplied to another air force

On totals and on combat, the honest position is a range and a qualified negative. The most-quoted production figure is 37 aircraft; German sources count 28 pre-series and 11 series machines complete at the surrender, with parts or advanced structures for about 50 more, which is where the round figure of roughly 90 airframes started comes from. English-language accounts give at least 16 prototypes flown and about 22 pre-production machines completed, and those counts do not reconcile cleanly with each other. As for combat: the Luftwaffe never used the Do 335 operationally, and no Do 335 is credited with a victory or recorded as lost to enemy action in the air. The single persistent story is Pierre Clostermann’s, who described leading four Tempests of No. 3 Squadron over northern Germany in April 1945 and meeting an aircraft answering the Pfeil’s description at treetop level and full speed; two of his pilots fired, the German reversed course, and Clostermann judged it too fast to chase. The encounter is not corroborated by any German record, and Clostermann’s memoir has been questioned elsewhere on points of detail. It is the best evidence there is, and it is not good evidence. Exactly one original Do 335 survives, and it is not a reproduction, a composite or a rebuild: Do 335 A-02, the second pre-production A-0, Werknummer 240102, Stammkennzeichen VG+PH, completed at Dornier’s Mengen plant on 30 September 1944 and flown through the winter of 1944–45. Hans-Werner Lerche flew it from Rechlin back to Oberpfaffenhofen by way of Prague and Lechfeld between 20 and 23 April 1945; Allied troops found it there on 29 April. In mid-June it was flown to Cherbourg by the German test pilot Hans Padell and shipped to the United States aboard HMS Reaper. The US Navy tested it at Patuxent River from 1945 to 1948, after which it lay outdoors at Norfolk; it was transferred to the Smithsonian in 1961 and continued to deteriorate. In October 1974 the Smithsonian sent it back to Oberpfaffenhofen — by then building Alpha Jets — and in 1975 Dornier restored it, many of the workers being men who had built the type in the war; they were startled to find the explosive charges for the dorsal fin and rear propeller still in place. It was shown at the Hannover air show from 1 to 9 May 1976 and then lent to the Deutsches Museum in Munich. Sources disagree on when it came home: the Smithsonian’s own record says 1986, Wikipedia and several German accounts say 1988. It is now on public display in the World War II German Aviation area at the Steven F. Udvar-Hazy Center in Chantilly, Virginia, inventory number A19610129000, alongside the only surviving Arado Ar 234 B-2 and the only complete Heinkel He 219 A. The other Do 335 shipped to the United States, FE-1012, went to the USAAF and was tested at Freeman Field, Indiana, in early 1946; its fate is unrecorded. The A-12 two-seater at Farnborough, W.Nr 240112, RP+UB, taken on charge as AM 223, was destroyed in January 1946 when its rear engine caught fire in flight, burned through the tail controls and dropped the aircraft onto the field behind Cove Junior School, killing Group Captain A. F. Hards of the Royal Aircraft Establishment on his first solo in the type and injuring people on the ground. Nothing else remains.

From an old patent to the last piston fighter

Eight chapters in the Arrow’s brief life

The Pfeil in pictures

‹›

See the Arrow in context

Was the Do 335 the best piston fighter of the war — or a brilliant answer to a question the jet age had already made obsolete? This short history traces the Arrow from Dornier’s drawing board to its single surviving airframe.

Video: “Dornier Do 335 Pfeil: The Best Plane of WWII?” by IHYLS (~331,000 views). Embedded via YouTube.

Where the Pfeil was built and flown

The Pfeil, by the numbers

The Do 335 has almost no combat record — and that is precisely its story. Here was one of the most advanced fighters of the war, faster than anything the Allies could put against it, that arrived far too late and in far too small a number to matter. Its record is a measure of lost potential rather than of victories won.

The Do 335 never reached operational service in strength, and its handful of flights were largely trials and ferry sorties; the figures above reflect a fighter that was captured almost before it could fight.

Dornier Do 335 Pfeil — questions & answers

Related aircraft

Where these facts come from

• Smithsonian NASM — Do 335 A-0 PfeilThe sole surviving Pfeil, restored by Dornier
• San Diego Air & Space MuseumArchive photography of the type
• WW2 Weapons — Do 335 ArrowTechnical data & variants
• Imperial War MuseumsCaptured-aircraft context

Figures are drawn from museum and reference sources and cross-checked; where sources differ, the most commonly cited value is given. Production totals for the Do 335 vary between sources because so many airframes were incomplete at war’s end.