Carrier aircraft doctrine, the pre WWII production effects.

[Geoffrey Sinclair]

The period covered is 1935 to 1939 and mainly concerns the RN and USN as they have good available reports of aircraft production, concentrating on the carrier aircraft as well as those meant to be used by battleships and cruisers, the small float planes and amphibians.

As aircraft speeds increased after WWI the role of the carrier fighter declined as the chances for interception declined, fighters were still useful to eliminate enemy shadowers and spotters and, if they had the range, as escorts for outgoing strikes. When radar appeared that changed, then add the rise of the fighter bomber.

The nominal air groups of the USS Lexington and Enterprise as planned had a ratio of 1 fighter to 4 other types, the USS Essex had a 2 to 3 ratio, the USS Midway slightly better than 1 to 1.

One of the ways to track aviation doctrine is to see what aircraft were being built. It is understood when individual production runs occurred was often a function of finance and the need to work out problems found during test flights. This means the aircraft production totals can be a good guide but not an absolute one. Clearly the size of the carrier fleet determines the average production numbers, the evolution of aircraft and available development funds determines the arrival of new types.

Another factor is obviously when new aircraft carriers were being built if they were meant to expand the fleet, less so if they were meant to be replacements for existing hulls.

USN, 3 fleet carriers and 1 light fleet in commission in January 1935, with the light fleet carrier Langley decommissioned as a carrier in October 1936 and converted to a seaplane tender. Yorktown was commissioned in September 1937, Enterprise in May 1938. In September 1939 Wasp was under construction and on the 25th of the month Hornet was laid down.

RN, 4 fleet and 2 light fleet carriers in commission in January 1935, Ark Royal was commissioned in December 1938, with the light fleet carrier Argus in theory meant to be replaced by the new carrier. By September 1939 there were 6 fleet carriers under construction, the Illustrious class, plus the maintenance carrier Unicorn. At least some of the Illustrious class were meant as replacements for older ships.

IJN, 2 fleet and 2 light fleet carriers in commission in January 1935, later numbers are slightly complicated by the decision to build some hulls with the potential to later convert them to carriers. Soryu was commissioned in December 1937, Hiryu in July 1939. In September 1939 Shokaku and Zuikaku were under construction and Zuiho was being converted from the nominal submarine support ship it was officially built as. Rebuilds of the original fleet carriers, Kaga (1934 to 1935) and Akagi (1936 to 1938) upped their nominal air group sizes from 60 to 90.

In 1935 the IJNAF dive bomber was the Aichi D1A, the initial contract had been awarded in late 1934, production ending in 1937 or 1938. The replacement was the Aichi D3A, the Val, but the contract was only awarded in December 1939, when it arrived it meant an all the main IJN carrier aircraft were monoplanes. Biplanes survived longer as dive bombers due to the airframe strength factors needed plus the greater drag, keeping the dive speed down without the need for brakes.

The 1935 IJNAF fighter was the Nakajima A4N, the Mitsubishi A5M started production at the end of 1936, making it the first monoplane IJNAF fighter, staying in production until 1940 when it was replaced by the A6M.

The 1935 IJNAF torpedo bomber was the unreliable Yokosuka B3Y, the replacement was the Yokosuka B4Y, production contract awarded in November 1936. However the monoplane Nakajima B5N, the Kate, began production in November 1937. The competitor to the B5N was the Mitsubishi B5M, 125 were built before the decision was taken to concentrate on the B5N.

Essentially the IJN followed the USN lead in dive bombers but given the D3A 550 pound bomb size limit the torpedo bomber was the main strike aircraft.

USN and RN:

The pre war British military aircraft production report includes most built for export but excludes prototypes and some standard production types that were used for experimental purposes. The USN production report includes all prototypes and some aircraft for the Coast Guard but excludes export orders.

For some reason the British production report includes most Blackburn Sharks (the competitor to the Swordfish) under the General Reconnaissance category, similarly all Sea Gladiators are counted as RAF fighters.

The following analysis ignores trainers and the longer range patrol aircraft.

The USN production in the period was, year, fighters, carrier attack aircraft (dive and torpedo bombers), scouts

1935, 56, 76, 64
1936, 55, 63, 118
1937, 15, 156, 31
1938, 70, 213, 83
1939, 38, 141, 46

Clearly not as extreme a fighter to "attack" ratio as the RN, ended up with (see below) but the drift from parity in 1935 and 1936 to around 4 to 1 for the years 1937 to 1939 probably indicates a shift in the value of fighters but clearly designs entering or leaving production had an effect.

Ignoring prototypes all the 1935 to 1939 fighters were Grumman F2F (in 1935) or F3F (1936 on), except for 11 of the new monoplane Brewster F2A Buffalos in 1939.

The main attack aircraft in each year were
in 1935 the 49 Great Lakes BG-1, the 26 Vought SBU
in 1936 the 58 Vought SBU
in 1937 the 40 Vought SBU, the 82 Curtiss SBC and the 26 Douglas TBD Devastator, plus 5 of the new monoplane dive bomber the SB2U
in 1938 the 70 Vought SB2U, the 89 TBD Devastator, the 52 Northrop BT-1 (later developed into the 8A then the SBD)
in 1939 the 89 Curtiss SBC, the 36 Vought SB2U, the 15 TBD Devastator.

All dive bombers except for the Devastator, with it, the BT-1 and SB2U being monoplanes.

For the floatplanes it was 32 Vought O3U and 32 Curtiss SOC in 1935, the effectively all SOC for the remainder of the decade, including those built in 1938/39 as the SON by the Naval Aircraft Factory.

So 3 different types of fighters (2 closely related), 5 dive bomber designs and 1 torpedo bomber in series production, and essentially only 1 floatplane design. It seems clear the USN had a preference for dive bombers, both in terms of numbers built and the number of designs built. With the arrival of the Brewster F2A all main USN carrier types were monoplanes, with the exception of the SBC.

F4F Wildcat production started in July 1940.

As an aside, for the war years the annual production of USN fighters and carrier "attack" aircraft looks like this

1940, 267, 147
1941, 635, 516
1942, 1,678, 1,838
1943, 6,377, 6,947
1944, 14,723, 8,032

So initially back to the mid 1930's 1 to 1 or so, then the big shift in 1944 and the rise of the fighter bomber, plus the needs of the RN and USMC.

RN:

The 1930's naval aircraft orders were usually small, at 693 the Swordfish is by far the biggest for the 1930's, built over about 4 years from February 1936 on.

Principle types production in the period was, year, fighters, carrier attack aircraft (dive and torpedo bombers), scouts (as in official naval production figures)

1935, 56, 13, 17
1936, 0, 148, 37
1937, 0, 199, 100
1938, 0, 158, 128
1939, 81, 374, 52

Note 12 of the 1935 scouts were for the RAAF/RAN.

Adding in the Sharks classified as General Reconnaissance but excluding the 9 for the RCAF, plus the 60 Sea Gladiators (ignoring the extra 38 conversions not all of which ended up with the RN) amends the figures to,

1935, 56, 13, 17
1936, 0, 252, 37
1937, 0, 316, 100
1938, 0, 158, 128
1939, 141, 374, 52

The 1935 fighters were 16 Hawker Nimrod and 40 Osprey, the latter being classified fighter reconnaissance. The 1939 fighters were 60 Sea Gladiators and 81 Blackburn Roc, the latter being a turret fighter like the Boulton-Paul Defiant, and it seems even less successful.

The attack aircraft were
1935, 13 Blackburn Shark
1936, 104 Blackburn Shark, 148 Fairey Swordfish
1937, 117 Blackburn Shark, 198 Fairey Swordfish
1938, 14 Blackburn Skua, 144 Fairey Swordfish,
1939, 175 Blackburn Skua, 199 Fairey Swordfish,

Scouts were

1935 the 17 Fairey Seal, 1936 the 37 Supermarine Walrus or Seagull V, 1937 the 78 Supermarine Walrus or Seagull V, and 22 Fairey Seafox, in 1938 it was 86 Walrus, 42 Seafox and in 1939 the 52 Walrus.

Only the Skua and Roc were monoplanes, only the Skua was a dive bomber.

In one sense the system was going along well, new torpedo bomber in late 1934, new amphibian and torpedo bomber in 1936, then new float plane/spotter in 1937, then new dive bomber in 1938, then new interim fighter in 1939, then new fighter (Fulmar) and torpedo bomber (Albacore) in 1940.

However the new designs were falling behind the USN and IJN and also land based types, and that seems to be largely the result of RN decisions. With the situation worsened by the crisis of 1940 stopping any diversion of RAF types to the RN and slowing down the introduction of new naval aircraft types.

The Nimrod and Osprey had entered production in 1931 as fleet fighters, it does show the doctrine that the next fighter design appeared 8 years later, and why the RN was willing to take Sea Gladiators in 1939, proposals in 1939 and early 1940 for a naval version of the Spitfire at least were around but the initial proposal was cancelled in March 1940, while further attempts promptly ran into the effects of the crisis of mid 1940.

The RN emphasis on torpedo bombers is also obvious, tempered by the fact the Swordfish was meant as multi role, including gunnery spotting and reconnaissance. This in part explain the use of torpedo specialists as carrier commanders.

The short production run of the Skua seems to be mainly related to air group sizes, there was not enough room on the carriers for multiple types.

As of the end of the 1920's the RN was still thinking along the lines of the fleet fighters would be carried mainly on the battleships and cruisers, leaving the carriers to operate the heavier types. That had to change in the 1930's as aircraft became heavier.

Again air group sizes probably played a part in the number of fighters built for the RN, but clearly there was a naval version of the bomber would get through concept. Which probably feeds into the armoured hangar carrier ideas.

The decision to make RN fighters 2 man, seen in for example the 1940 Fairey Fulmar, is related to them having the range to escort strikes as well as being available for reconnaissance duties, as for example the operations from Victorious against Bismarck, coupled with the RN dislike of active radio homing beacons. The first attempts to use radar to intercept incoming strikes was in April 1940 off Norway, using radars mounted on escorting cruisers. This experiment later resulted in the RN keeping a dedicated communication channel for the radar plots, whereas the USN reportedly did not. Hence why the RN thought it was doing better at fighter control in the Pacific in 1945.

Compared to the USN and IJN the RN was receiving lower performance fighter and torpedo aircraft by the end of the 1930's, something that would become worse in 1940/41. The size of airgroups meant the next attack aircraft would try to be a dive and torpedo bomber, the Fairey Barracuda, the crisis of 1940 and the problems in the design meant the first production examples would not appear until 1942. The 1940 crisis also meant few high performance RAF fighters could be made available to the RN.

In fact no new RN aircraft designs entered production in 1941, it took until the second half 1942 for production versions of the Sea Hurricane, Seafire, Firefly and Barracuda to appear.

To complete things the Swordfish production line was shut down in early 1940, then revived at the Blackburn factory in late 1939, while between late 1939 and the end of 1942 there were 800 Fairey Albacore biplane torpedo bombers built. The Swordfish line was finally shut in August 1944.

In one sense the RN was fortunate when it came to US design carrier fighters, F4F production in 1940 was 103 of which 81 were for the RN, versus home production of 159 Fulmars.

In 1941 some 90 out of 324 F4F were for the RN or Greece. This goes a long way to explain why some USN fighter squadrons were still using Buffalos in December 1941. Then there are all the Buffalos produced for export, but their record is not as good. Later there was the initial USN rejection of the F4U Corsair for carrier operations, making it available for other users.

There has been much debate on the effect of RAF control over RN aviation. That the RAF had a bad effect on RN naval aviation is undoubted, but it seems to be subtle, things like the naval aviators leaving the RN thereby ensuring the WWII senior RN command had few men with aviation experience, also the dual control of aviation on board the carriers reducing innovation. The naval people had less ability to give orders to the aircrew. The RN regularly had a clear flight deck when landing aircraft, and fewer arrester wires than the USN, it meant an aircraft could more easily go round again if required, it also slowed down flight deck operations. When the RN introduced crash barriers (in around 1939?) the aircrew did not like them, correctly figuring out the barrier was a greater danger to the aircrew than trying a go round.

The fact the Air Ministry had to ultimately issue the specifications for new naval aircraft seems to have slowed the process down and generated some friction at times.

Whether the desires for great low speed handling, which seemingly inevitably compromised performance, were RAF or RN or both or simply habit is unclear. The two man fighter was an RN preference.

The use of flight deck round downs cut turbulence but also limited the amount of flight deck available to range and launch a strike. The Illustrious class had much of their round downs eliminated in refits.

When it comes to actual production there seems little RAF effect on the numbers ordered. To use the official figures, 1935 to 1940, naval aircraft production, the first figure as a percentage of total aircraft built, the second is the percentage after taking the trainers out, in other words combat types, given a lot of the trainers were common to both services.

1935, 9.97, 17.8
1936, 9.89, 20.8
1937, 13.6, 23.3
1938, 10.1, 20.5
1939, 6.4, 13.6
1940, 3.16, 5.5

Put the Blackburn Sharks and Sea Gladiators in as RN types and 1936 goes to 33.2%, 1937 to 32.6% and 1939 to 15.3% of combat types. So the RN was not being shortchanged when it came to percentages of aircraft built in the 1930's. Then the percentage went down as the RAF expansion plans kicked in. British military aircraft output in 1939 approached
3 times the 1938 output and 1940 production was about twice that of 1939. The RAF had about a nominal 60 or so active squadrons in early 1935, it was 158 in September 1939.

It also makes more financial sense in rivalry terms for the RAF to cut down RN production orders, so the money and capacity is freed for RAF orders, than it does to deny the RN a tender for a new aircraft design, the numbers suggest the RN was receiving a steady flow of naval aircraft at a time when Britain felt it was very short of combat aircraft.

The above figures makes it clear a claim the RAF starved the RN of aircraft in the mid/late 1930's is difficult to support. If anything naval aircraft are over represented. The 5 year totals for fighters, attack and floatplanes come to

RN, 197, 1113, 334
USN, 234, 649, 342

But of course the RN was moblising in a way the USN was not in the late 1930's and the RN fighter total masks the long gap in fighter production.

A final point, flight deck sizes in feet,

Enterprise 802 by 86

Ark Royal, 797 by 96.

Illustrious 747 by 95.

Implacable/Indefatigable 761 by 101/105 (on the same waterline beam as Illustrious, and the same standard tonnage as the Essex)

Short hull Essex 862 by 108 (second group were about 12 feet longer)

So if Enterprise flight deck area is 1 unit, Illustrious is 1.03, Ark Royal 1.11, Indefatigable 1.16 and Essex 1.35.

So on about 10% more displacement than Enterprise the Ark Royal had about an 11% bigger flight deck. Essex had around 23% more displacement than Ark Royal and around 22% more flight deck space. While 15% more displacement gave Illustrious lots of protection and the same flight deck area as Enterprise. Essex had around 16% more flight deck area than Indefatigable on a similar displacement. Nice rule of thumb on the cost of armouring the hangar but it masks a much more complex situation. Ark Royal apparently had more waterline beam, 95 versus an Essex at 93 feet

By the way Essex had two curved ramps fore and aft, 4 feet 9 inches long, not quite the round downs present in RN carrier designs.

 

[EwenS]

Geoffrey
A lot of information to digest but I've inserted some comments into your post.

The period covered is 1935 to 1939 and mainly concerns the RN and USN as they have good available reports of aircraft production, concentrating on the carrier aircraft as well as those meant to be used by battleships and cruisers, the small float planes and amphibians.

As aircraft speeds increased after WWI the role of the carrier fighter declined as the chances for interception declined, fighters were still useful to eliminate enemy shadowers and spotters and, if they had the range, as escorts for outgoing strikes. When radar appeared that changed, then add the rise of the fighter bomber.

The nominal air groups of the USS Lexington and Enterprise as planned had a ratio of 1 fighter to 4 other types, the USS Essex had a 2 to 3 ratio, the USS Midway slightly better than 1 to 1.

Those ar not the planned air groups for the Essex and Midway classes but rather the shape of the groups they entered service with in 1942/43 and 1945 respectively as altered in the light of war experience to date. The 1942 carrier battles revealed a need for more fighters, so they began to replace strike aircraft.

By 1939 aircraft had grown in size and the Yorktown class were no longer able to operate the 90 aircraft air group that they were initially designed for due to an increase in individual aircraft size. It was down to about 81. The Essex design was based around operating 90 of the forthcoming larger aircraft like the F4U, SB2C and TBF all of which were in the development phase of their lives. So the intended ratio was still to be 1 to 4 fighters to strike aircraft. By 1943 it was the 2 to 3 you indicated and by 1945 the ratio was 7 fighters for every 3 strike aircraft.

The Midway design signed off in March 1942 comprised 36 fighter, 48 dive bomber and 36 torpedo bomber. A ratio of 3 fighters for every 7 strike aircraft. It was the kamikaze threat that had the greatest impact on the ratio of fighters to strike aircraft carried from Dec 1944.

One of the ways to track aviation doctrine is to see what aircraft were being built. It is understood when individual production runs occurred was often a function of finance and the need to work out problems found during test flights. This means the aircraft production totals can be a good guide but not an absolute one. Clearly the size of the carrier fleet determines the average production numbers, the evolution of aircraft and available development funds determines the arrival of new types.

Another factor is obviously when new aircraft carriers were being built if they were meant to expand the fleet, less so if they were meant to be replacements for existing hulls.

USN, 3 fleet carriers and 1 light fleet in commission in January 1935, with the light fleet carrier Langley decommissioned as a carrier in October 1936 and converted to a seaplane tender. Yorktown was commissioned in September 1937, Enterprise in May 1938. In September 1939 Wasp was under construction and on the 25th of the month Hornet was laid down.

RN, 4 fleet and 2 light fleet carriers in commission in January 1935, Ark Royal was commissioned in December 1938, with the light fleet carrier Argus in theory meant to be replaced by the new carrier. By September 1939 there were 6 fleet carriers under construction, the Illustrious class, plus the maintenance carrier Unicorn. At least some of the Illustrious class were meant as replacements for older ships.

In 1935 Britain only had 5 carriers in service. Argus had reduced to reserve in Dec 1929 and didn't return to service until after her refit that ended in July 1938 when she emerged as a Queen Bee and training carrier. Hermes also spent a year in reserve in 1937/38 before again being brought forward as a training carrier.

When the RN in mid 1939, immediately after regaining full control of the FAA, sat down to look at proposed carrier deployments in 1942, the Argus and Eagle are missing from the lists suggesting that they would have been the first to be replaced while Furious is a training carrier in home waters and Hermes had a proposed trade protection role in the Atlantic.

IJN, 2 fleet and 2 light fleet carriers in commission in January 1935, later numbers are slightly complicated by the decision to build some hulls with the potential to later convert them to carriers. Soryu was commissioned in December 1937, Hiryu in July 1939. In September 1939 Shokaku and Zuikaku were under construction and Zuiho was being converted from the nominal submarine support ship it was officially built as. Rebuilds of the original fleet carriers, Kaga (1934 to 1935) and Akagi (1936 to 1938) upped their nominal air group sizes from 60 to 90.

Zuiho was laid up immediately after launch in 1936. AIUI her conversion to a carrier began in early1940 and completed at the end of the year. Her sister Shoho didn't begin conversion until Jan 1941 and was completed in Jan 1942.

As for the Japanese "shadow" carrier program the IJN subsidised the construction of about 10 vessels for conversion to carriers in the event of war. This was to allow incorporation of features that would suit conversion to carriers in the event of war. In the event not all were converted. However the first of these to be taken in hand for conversion were the hulls that became the Hiyo & Junyo while they were still on the slip in Oct 1940, although that move was kept secret until the acquisition was formally made in Feb 1941. The first of an eventual 5 escort carriers, Taiyo, began conversion in Feb 1941.

 

[EwenS]

The system wouldnt let me post everything in one go so here is part 2




In 1935 the IJNAF dive bomber was the Aichi D1A, the initial contract had been awarded in late 1934, production ending in 1937 or 1938. The replacement was the Aichi D3A, the Val, but the contract was only awarded in December 1939, when it arrived it meant an all the main IJN carrier aircraft were monoplanes. Biplanes survived longer as dive bombers due to the airframe strength factors needed plus the greater drag, keeping the dive speed down without the need for brakes.


The 1935 IJNAF fighter was the Nakajima A4N, the Mitsubishi A5M started production at the end of 1936, making it the first monoplane IJNAF fighter, staying in production until 1940 when it was replaced by the A6M.

The 1935 IJNAF torpedo bomber was the unreliable Yokosuka B3Y, the replacement was the Yokosuka B4Y, production contract awarded in November 1936. However the monoplane Nakajima B5N, the Kate, began production in November 1937. The competitor to the B5N was the Mitsubishi B5M, 125 were built before the decision was taken to concentrate on the B5N.

Essentially the IJN followed the USN lead in dive bombers but given the D3A 550 pound bomb size limit the torpedo bomber was the main strike aircraft.

USN and RN:

The pre war British military aircraft production report includes most built for export but excludes prototypes and some standard production types that were used for experimental purposes. The USN production report includes all prototypes and some aircraft for the Coast Guard but excludes export orders.

For some reason the British production report includes most Blackburn Sharks (the competitor to the Swordfish) under the General Reconnaissance category, similarly all Sea Gladiators are counted as RAF fighters.

The following analysis ignores trainers and the longer range patrol aircraft.

The USN production in the period was, year, fighters, carrier attack aircraft (dive and torpedo bombers), scouts

1935, 56, 76, 64
1936, 55, 63, 118
1937, 15, 156, 31
1938, 70, 213, 83
1939, 38, 141, 46

Clearly not as extreme a fighter to "attack" ratio as the RN, ended up with (see below) but the drift from parity in 1935 and 1936 to around 4 to 1 for the years 1937 to 1939 probably indicates a shift in the value of fighters but clearly designs entering or leaving production had an effect.

Ignoring prototypes all the 1935 to 1939 fighters were Grumman F2F (in 1935) or F3F (1936 on), except for 11 of the new monoplane Brewster F2A Buffalos in 1939.

The main attack aircraft in each year were
in 1935 the 49 Great Lakes BG-1, the 26 Vought SBU
in 1936 the 58 Vought SBU
in 1937 the 40 Vought SBU, the 82 Curtiss SBC and the 26 Douglas TBD Devastator, plus 5 of the new monoplane dive bomber the SB2U
in 1938 the 70 Vought SB2U, the 89 TBD Devastator, the 52 Northrop BT-1 (later developed into the 8A then the SBD)
in 1939 the 89 Curtiss SBC, the 36 Vought SB2U, the 15 TBD Devastator.

All dive bombers except for the Devastator, with it, the BT-1 and SB2U being monoplanes.

For the floatplanes it was 32 Vought O3U and 32 Curtiss SOC in 1935, the effectively all SOC for the remainder of the decade, including those built in 1938/39 as the SON by the Naval Aircraft Factory.

So 3 different types of fighters (2 closely related), 5 dive bomber designs and 1 torpedo bomber in series production, and essentially only 1 floatplane design. It seems clear the USN had a preference for dive bombers, both in terms of numbers built and the number of designs built. With the arrival of the Brewster F2A all main USN carrier types were monoplanes, with the exception of the SBC.

F4F Wildcat production started in July 1940.

As an aside, for the war years the annual production of USN fighters and carrier "attack" aircraft looks like this

1940, 267, 147
1941, 635, 516
1942, 1,678, 1,838
1943, 6,377, 6,947
1944, 14,723, 8,032

So initially back to the mid 1930's 1 to 1 or so, then the big shift in 1944 and the rise of the fighter bomber, plus the needs of the RN and USMC.

RN:

The 1930's naval aircraft orders were usually small, at 693 the Swordfish is by far the biggest for the 1930's, built over about 4 years from February 1936 on.

Principle types production in the period was, year, fighters, carrier attack aircraft (dive and torpedo bombers), scouts (as in official naval production figures)

1935, 56, 13, 17
1936, 0, 148, 37
1937, 0, 199, 100
1938, 0, 158, 128
1939, 81, 374, 52

Note 12 of the 1935 scouts were for the RAAF/RAN.

Adding in the Sharks classified as General Reconnaissance but excluding the 9 for the RCAF, plus the 60 Sea Gladiators (ignoring the extra 38 conversions not all of which ended up with the RN) amends the figures to,

1935, 56, 13, 17
1936, 0, 252, 37
1937, 0, 316, 100
1938, 0, 158, 128
1939, 141, 374, 52

The 1935 fighters were 16 Hawker Nimrod and 40 Osprey, the latter being classified fighter reconnaissance. The 1939 fighters were 60 Sea Gladiators and 81 Blackburn Roc, the latter being a turret fighter like the Boulton-Paul Defiant, and it seems even less successful.

The attack aircraft were
1935, 13 Blackburn Shark
1936, 104 Blackburn Shark, 148 Fairey Swordfish
1937, 117 Blackburn Shark, 198 Fairey Swordfish
1938, 14 Blackburn Skua, 144 Fairey Swordfish,
1939, 175 Blackburn Skua, 199 Fairey Swordfish,

Scouts were

1935 the 17 Fairey Seal, 1936 the 37 Supermarine Walrus or Seagull V, 1937 the 78 Supermarine Walrus or Seagull V, and 22 Fairey Seafox, in 1938 it was 86 Walrus, 42 Seafox and in 1939 the 52 Walrus.

Only the Skua and Roc were monoplanes, only the Skua was a dive bomber.

In one sense the system was going along well, new torpedo bomber in late 1934, new amphibian and torpedo bomber in 1936, then new float plane/spotter in 1937, then new dive bomber in 1938, then new interim fighter in 1939, then new fighter (Fulmar) and torpedo bomber (Albacore) in 1940.

However the new designs were falling behind the USN and IJN and also land based types, and that seems to be largely the result of RN decisions. With the situation worsened by the crisis of 1940 stopping any diversion of RAF types to the RN and slowing down the introduction of new naval aircraft types.

The Nimrod and Osprey had entered production in 1931 as fleet fighters, it does show the doctrine that the next fighter design appeared 8 years later, and why the RN was willing to take Sea Gladiators in 1939, proposals in 1939 and early 1940 for a naval version of the Spitfire at least were around but the initial proposal was cancelled in March 1940, while further attempts promptly ran into the effects of the crisis of mid 1940.

The RN emphasis on torpedo bombers is also obvious, tempered by the fact the Swordfish was meant as multi role, including gunnery spotting and reconnaissance. This in part explain the use of torpedo specialists as carrier commanders.

The short production run of the Skua seems to be mainly related to air group sizes, there was not enough room on the carriers for multiple types.

As of the end of the 1920's the RN was still thinking along the lines of the fleet fighters would be carried mainly on the battleships and cruisers, leaving the carriers to operate the heavier types. That had to change in the 1930's as aircraft became heavier.

Again air group sizes probably played a part in the number of fighters built for the RN, but clearly there was a naval version of the bomber would get through concept. Which probably feeds into the armoured hangar carrier ideas.

The decision to make RN fighters 2 man, seen in for example the 1940 Fairey Fulmar, is related to them having the range to escort strikes as well as being available for reconnaissance duties, as for example the operations from Victorious against Bismarck, coupled with the RN dislike of active radio homing beacons. The first attempts to use radar to intercept incoming strikes was in April 1940 off Norway, using radars mounted on escorting cruisers. This experiment later resulted in the RN keeping a dedicated communication channel for the radar plots, whereas the USN reportedly did not. Hence why the RN thought it was doing better at fighter control in the Pacific in 1945.


I've never heard of the dedicated communications channel you note. I wouldn't necessarily say that the RN was better at fighter control in the Pacific in 1945 but they had developed and did employ radar techniques that the USN did not, particularly in relation longer ranged height finding. The RN were using the various vertical lobes generated by the radar signals from their Type 279 and Type 281 sets to track aircraft and determine their height. Each set had different lobes and this was exploited in the fleet (some cruisers and battleships had 279 and some 281) and especially in the carriers that carried both. The USN were better at short range fighter control courtesy of the SM and SP fighter direction sets, of which the BPF only had one in Indomitable in 1945.


Compared to the USN and IJN the RN was receiving lower performance fighter and torpedo aircraft by the end of the 1930's, something that would become worse in 1940/41. The size of airgroups meant the next attack aircraft would try to be a dive and torpedo bomber, the Fairey Barracuda, the crisis of 1940 and the problems in the design meant the first production examples would not appear until 1942. The 1940 crisis also meant few high performance RAF fighters could be made available to the RN.

In fact no new RN aircraft designs entered production in 1941, it took until the second half 1942 for production versions of the Sea Hurricane, Seafire, Firefly and Barracuda to appear.

To complete things the Swordfish production line was shut down in early 1940, then revived at the Blackburn factory in late 1939, while between late 1939 and the end of 1942 there were 800 Fairey Albacore biplane torpedo bombers built. The Swordfish line was finally shut in August 1944.

The Swordfish production line shut at Fairey Hayes plant as planned pre-war to allow production to switch to the Albacore (first production Albacore Jan 1940 and last Swordfish Feb 1940). But teething problems with the Taurus engine and the need to expand the FAA saw new contracts for Swordfish issued to Blackburn in 1939 with the first aircraft not coming off that production line until Dec 1940.

Pre-war the plan for the Barracuda, produced to a 1937 spec, was for it to enter production in March 1941 following the Fulmar at the Fairey Stockport factory and production to be complete by April 1942. But between development delays (flight control issues and a partial redesign following cancellation of the RR Exe engine) and the need to maintain Fulmar production, the Barracuda production didnt start until April 1942, continuing until the end of the war. Its successor, the Fairey Spearfish didn't begin to be designed until 1943.


In one sense the RN was fortunate when it came to US design carrier fighters, F4F production in 1940 was 103 of which 81 were for the RN, versus home production of 159 Fulmars.

In 1941 some 90 out of 324 F4F were for the RN or Greece. This goes a long way to explain why some USN fighter squadrons were still using Buffalos in December 1941. Then there are all the Buffalos produced for export, but their record is not as good. Later there was the initial USN rejection of the F4U Corsair for carrier operations, making it available for other users.


The 71 Martlet I delivered in 1940 (another 10 were lost at sea) came from French orders taken over by Britain in June 1940. In March / April 1941 30 more were received in the Middle East being aircraft due to be delivered to the Greek Air Force. These all had fixed wings.

The first 10 Martlets from British orders were delivered in March 1941 again with fixed wings. Britain then chose to delay delivery of the remaining 90 aircraft in its order until the folding wing became available. The first of these aircraft came off the production line in in Oct 1941 with the last in April 1942. The first were used to equip 3 squadrons sent to the Indian Ocean on Formidable and Illustrious in early 1942 with most of the rest being sent there direct as attrition replacements.


There has been much debate on the effect of RAF control over RN aviation. That the RAF had a bad effect on RN naval aviation is undoubted, but it seems to be subtle, things like the naval aviators leaving the RN thereby ensuring the WWII senior RN command had few men with aviation experience, also the dual control of aviation on board the carriers reducing innovation. The naval people had less ability to give orders to the aircrew. The RN regularly had a clear flight deck when landing aircraft, and fewer arrester wires than the USN, it meant an aircraft could more easily go round again if required, it also slowed down flight deck operations. When the RN introduced crash barriers (in around 1939?) the aircrew did not like them, correctly figuring out the barrier was a greater danger to the aircrew than trying a go round.

The fact the Air Ministry had to ultimately issue the specifications for new naval aircraft seems to have slowed the process down and generated some friction at times.

Whether the desires for great low speed handling, which seemingly inevitably compromised performance, were RAF or RN or both or simply habit is unclear. The two man fighter was an RN preference.

The use of flight deck round downs cut turbulence but also limited the amount of flight deck available to range and launch a strike. The Illustrious class had much of their round downs eliminated in refits.

When it comes to actual production there seems little RAF effect on the numbers ordered. To use the official figures, 1935 to 1940, naval aircraft production, the first figure as a percentage of total aircraft built, the second is the percentage after taking the trainers out, in other words combat types, given a lot of the trainers were common to both services.

1935, 9.97, 17.8
1936, 9.89, 20.8
1937, 13.6, 23.3
1938, 10.1, 20.5
1939, 6.4, 13.6
1940, 3.16, 5.5

Put the Blackburn Sharks and Sea Gladiators in as RN types and 1936 goes to 33.2%, 1937 to 32.6% and 1939 to 15.3% of combat types. So the RN was not being shortchanged when it came to percentages of aircraft built in the 1930's. Then the percentage went down as the RAF expansion plans kicked in. British military aircraft output in 1939 approached
3 times the 1938 output and 1940 production was about twice that of 1939. The RAF had about a nominal 60 or so active squadrons in early 1935, it was 158 in September 1939.

It also makes more financial sense in rivalry terms for the RAF to cut down RN production orders, so the money and capacity is freed for RAF orders, than it does to deny the RN a tender for a new aircraft design, the numbers suggest the RN was receiving a steady flow of naval aircraft at a time when Britain felt it was very short of combat aircraft.

The above figures makes it clear a claim the RAF starved the RN of aircraft in the mid/late 1930's is difficult to support. If anything naval aircraft are over represented. The 5 year totals for fighters, attack and floatplanes come to

RN, 197, 1113, 334
USN, 234, 649, 342


Firstly it needs to be remembered that operating aircraft off a carrier deck is not like an onshore grass airfield with massive run offs in all directions. One mistake and the aircraft is damaged or written off. So direct comparisons of numbers do not always make sense. By WW2 the RN seems to have treated aircraft as a disposable item. So much so that while the RAF maintained its record cards to hit the archives the RN didn't bother.

Secondly the RAF v RN position is complicated and in a lot of ways makes no sense at all. For example the cost of aircraft for the FAA came out of RN budgets but yet the Air Ministry / RAF insisted that for many designs the FAA receive naval versions of RAF aircraft (see Nimtod and Osprey for example) rather than have new types designed that might have suited the RN better. They also until the early 1930s had ideas as to the numbers of aircraft that could be operated from carriers and wouldn't agree to buy more until the RN proved it. So the unit size in the 1920s was the 6 aircraft flight. It isn't until the early 1930s that they are amalgamated into 12 plane squadrons. Landing speeds and take off runs were being set by the RN but as they were not allowed contact with industry (an Air Ministry job) they couldn't develop any idea what was possible if they had relaxed some of the criteria. The only way that I can rationalise some of what went on is that the RAF was the new kid on the block and needed to establish its own authority over all aviation matters and therefore guarded its territory jealously. That suited the Army as policing the Empire then became a lot easier. More aircraft = fewer soldiers required on the ground coming out of Army budgets. It suits the RN a lot less.

After various experiments with arrester systems until the mid 1920s the RN removed it. But when you only have small numbers of relatively low powered aircraft aircraft that can stop easily why do you need it? It is only from around 1930 as aircraft weights grow and numbers grow that some way of stopping them needs to be installed.

As for barriers the need for these only becomes relevant if you are going to have deck parks or need to speed up the operating cycle as you have more aircraft in the air. And that only comes as the 1930s roll on. Ark was the first RN carrier to receive one. But none of the old carriers were so fitted prior to their loss or retirement. T shaped lifts allowed aircraft to land and be struck down into the hangar and have the wings folded there, all while the next aircraft was touching down.

 

[EwenS]

And part 3

But of course the RN was moblising in a way the USN was not in the late 1930's and the RN fighter total masks the long gap in fighter production.

A final point, flight deck sizes in feet,

Enterprise 802 by 86

Ark Royal, 797 by 96.

Illustrious 747 by 95.

Implacable/Indefatigable 761 by 101/105 (on the same waterline beam as Illustrious, and the same standard tonnage as the Essex)

Short hull Essex 862 by 108 (second group were about 12 feet longer)

So if Enterprise flight deck area is 1 unit, Illustrious is 1.03, Ark Royal 1.11, Indefatigable 1.16 and Essex 1.35.

So on about 10% more displacement than Enterprise the Ark Royal had about an 11% bigger flight deck. Essex had around 23% more displacement than Ark Royal and around 22% more flight deck space. While 15% more displacement gave Illustrious lots of protection and the same flight deck area as Enterprise. Essex had around 16% more flight deck area than Indefatigable on a similar displacement. Nice rule of thumb on the cost of armouring the hangar but it masks a much more complex situation. Ark Royal apparently had more waterline beam, 95 versus an Essex at 93 feet

By the way Essex had two curved ramps fore and aft, 4 feet 9 inches long, not quite the round downs present in RN carrier designs.


Comparisons are often made between USN and RN carrier practice, with the RN often being judged to come out unfavourably in the period up to WW2. But that hides a lot of differences in their operating environments and the roles that the carriers were to fulfill.

The RN was planning to operate in the rough waters of the North Sea and Atlantic. In the 1920s and 1930s aircraft would not have lasted long exposed to wind and wave on the flight deck. So it was necessary for the RN to keep them in the hangar and hangar capacity became the feature that drove air group size. Due to deck movements in such anticipated seas the landing spot on an RN carrier was placed much nearer midships where ship movement would be less making landing easier. The RN spent a lot of time with wind tunnel models of carriers shaping their flight decks and island structures with a view to making their generally smaller ships easier to land on (until the mid 1930s US experience was largely with the 33,000 ton Lexington and Saratoga). And their main weapon was to be the torpedo. Until the advent of the Skua, dive bombing is a secondary task for the torpedo aircraft.

Sea conditions can still be a limiting factor for the 100,000+ ton Nimitz class supercarriers with flight decks moving by metres at a time in heavy weather.

And this is a period of Treaties limiting carrier tonnage and individual ship size (27,000 tons to the end of 1936 and an overall tonnage of 135,000 tons. The US and Japan gained exemptions at the Washington Conference in 1922 to convert 2 battlecruisers each that were then building into carriers. Britain had no such ships then building. 23,000 tons after 1936 for individual carriers but no upper limit. That is why USS Wasp is so small, being designed to use up the balance of Treaty tonnage available to the USN.

So when it came to producing a new generation of RN carriers in the 1930s we get Ark Royal. Then we get the Abyssinian Crisis and a massive change of focus. The threat is seen to be the level bomber. The pre radar carrier can't hope to be able to launch fighters early enough to be at altitude to intercept the attackers nor can it carry enough fighters to maintain standing patrols. The RN isn't prepared to simply accept the ship being put out of action and so losing its strike capability. The RN concludes that the best route in the event of such an attack is to keep the aircraft in the hangar with an armoured lid and depend on the carriers' and the fleet's AA fire to protect the aircraft. So we get the Illustrious class. But with Treaty limits you can have either large numbers of aircraft (Ark Royal) or an armoured deck (Illustrious) but not both.

But the RN led the world in many ways in the early/mid 1930s. It began developing techniques to operate multiple carrier task groups and the aircraft in those multiple groups, something the Japanese did not begin until formation of the Kido Butai in April 1941 nor the USN until early 1942, which they then abandoned until mid 1943 due to lack of carriers. And it led the world in using radar for fighter direction from mid 1940.

The USN was planning to operate in the more benign waters of the Pacific. It too was limited to 135,000 tons of carriers but it was allowed the two 33,000 ton Lexingtons as part of that. Their purpose was to find the enemy and take out their carriers before they could get to US battleships. So their tactics developed around much larger strikes to swamp the enemy. So they carried many more aircraft to achieve that aim without the bad weather need to accomodate them all in the hangar. In order to operate those larger numbers of aircraft the USN decided it needed long decks so that half its air group could be parked before take off and after landing. That also drove the need to develop a barrier to protect aircraft parked forward. The USN virtually accepts that the carrier is a one shot deal. It must get the enemy carrier before being put out of action itself. And in the early 1930s the USN virtually abandons the torpedo bomber altogether (Ranger was built without any torpedo stowage for example and didn't operate any until 1941). It only makes a return with the TBD Devastator from 1935. Instead USN focus is on the dive bomber to break up opposing carrier decks. The WW2 dive bomber grows out of USN fighter types adapted to carry bombs. Even the F4F Wildcat was designed to carry 2x100lb bombs for just such a purpose.

Essex is not entirely a Treaty ship and therefore not directly comparable to any of the British carriers. Her design began before the beginning of WW2, (the Treaties ended with the outbreak of WW2 some would argue earlier) but was not completed until 1941. During that time the design grew in size from 20,400 tons of a modified Yorktown (which was the tonnage available to the USN in 1939 due to legislation passed by Congress) to an eventual 27,000 tons. So of course it achieved more than the Ark Royal or Illustrious.

The USN looked at an armored flight deck for the Essex but decided to continue its practice of favouring the larger air group. Having seen the RN experience in the Med in early 1941 and examined the Illustrious and Formidable at Norfolk Navy Yard, the USN concluded that the armoured flight deck was in fact a good idea. But to get it required a Midway class at 45,000 tons. Having a ship that size the larger air group almost came as a by product and one that was not welcomed in all corners of the fleet. The Japanese too concluded the armoured deck was a good idea and developed the Taiho design from 1939, but it checked in at about 30,000 tons.

I note that you highlight flight deck length. Those are the overall lengths. The RN round downs substantially reduced those. Ark Royal's usable fight deck was 720ft. That of Illustrious began at 620ft and grew to 740ft by 1946 as the round downs fore and aft were eliminated. Indomitable started at 680ft and grew to 745ft in 1943.

There a lot of other differences between RN and USN practice in carrier design of the period. Petrol stowage is one that comes immediately to mind. After a bad experience in WW1 the RN applied much more rigorous standards to protecting it than either the USN or the IJN. That was something that both those navies came to realise. The later Essex class had their petrol arrangements modified while building in the light of the loss of the Lexington in 1942. The USN also devoted more tonnage to carrying the ship's oil fuel needed for the fight expected across the broad expanse of the Pacific. The RN was expecting to operate closer to its system of world wide bases. Machinery design and practice also affected displacements. So direct comparison of displacements is not always instructive.

If you want to understand carrier design then I'd recommend
"British Aircraft Carriers Design Development and Service Histories" by David Hobbs
"British Carrier Aviation" by Norman Friedman
"US Aircraft Carriers An Illustrated Design History" by Norman Friedman

 

[SaparotRob]

Do we get college credit?

 

[EwenS]

Nah! Nor do I get commission for recommending the books!!

 

[nuuumannn]

A good, if not overly long synopsis and I'll leave Ewen to add the bigger details. a few things at odds with the situation that cannot be found in the presentation of figures alone.

There has been much debate on the effect of RAF control over RN aviation. That the RAF had a bad effect on RN naval aviation is undoubted, but it seems to be subtle,

It was far from subtle. It impacted the navy's type procurement for the rest of the war and had the navy had greater control it would have begun the war with more suitable types for its needs, particularly the lack of a modern single-seat fighter. As recounted in another thread, the navy desperately wanted and needed a single-seat carrier fighter and with the dawning realisation that the Skua was unsuitable as a fighter, even before it entered service led to many different paths being taken in attempts to get single-seat fighters. Hurricanes and Spitfires were requested but denied because of RAF priority, which led to the next step to either get someone else to build them outside of the normal channels, which led to the navy asking Richard Fairey to build Spitfires and when that failed, purchase from overseas sources was undertaken, the F4Fs, the only truly suitable option at that time.

In that time, the navy redrew the specifications for the Fulmar and Roc replacements, the spec for the Fulmar replacement was rewritten and reissued and became the Firefly and the Roc turret fighter replacement was dropped and a new specification was issued, to Blackburn for a single-seat fighter, which should have gone into service in 1942 as the Firebrand, but that was a disaster and again, it left the FAA high and dry, having to rely on second hand Hurricanes from the RAF, then fully carrier capable Sea Hurricanes, then Seafires. Added to this was the development of the Sea Typhoon as an alternative to the Firebrand, which the Admiralty stated wasn't needed because it would be too late (famous last words), and the hope that something could have been made of the wooden Miles M.20 stop gap fighter all point to the fact that the navy was desperate for fighters because of the decisions made by the Air Ministry when it had control over the FAA. Bear in mind this all happened between 1938 and 1942.

Whether the desires for great low speed handling, which seemingly inevitably compromised performance, were RAF or RN or both or simply habit is unclear. The two man fighter was an RN preference.

Initially when drawing up the specification that created the Skua back in 1934, but not once it became apparent it would not be a suitable fighter by the time the Spitfire and Hurricane prototypes appeared in 1936 and '35 respectively. The Fulmar was designed as an interim to fulfil a catapult and carrier fighter role until the turret operated Roc entered service, that this was woefully inadequate as a fighter meant that the FAA had to rely on the Fulmar as its main fighter, but that didn't mean the navy "preferred" two-seat fighters. That's all it had.

Yes, the Fulmar and Roc were designed with the intent of intercepting bombers far out to sea and therefore it was considered unlikely they would encounter single-seaters, but the decision not to pursue a modern single-seater in the between-the-wars years left the FAA woefully inadequate. After all, single-seaters had been a staple of the FAA since the end of the Great War and the RNAS days. The two-seat fighters were an aberration and were derived from the Air Ministry's concern that modern aircraft were getting bigger, therefore taking up space on carriers and therefore needed to be multi-role - the fighter/dive bomber for example, the Barracuda specification for a torpedo dive bomber/reconnaissance platform another example...

The above figures makes it clear a claim the RAF starved the RN of aircraft in the mid/late 1930's is difficult to support.

Perhaps not in terms of numbers, but definitely in terms of capability.

 

[nuuumannn]

An addition, in 1933 when the Director of Technical Development of the Air Ministry requested a two-seat fighter, the Director of the Naval Air Division (Air Ministry) wrote the following: "...while agreeing there would be occasions when fighter aircraft will be of value to the fleet, it is considered that the limited aircraft carrying capacity means we cannot devote much of this to have a purely defensive fighter". And with that, the FAA's future single-seat fighter hopes were dashed - in 1933.

In 1937 members of both the Air Ministry and the Admiralty held a meeting to discuss aircraft developments on 22 October and all present believed the Skua would be obsolete by 1939 and the following options were offered: 1) a conversion of an existing RAF type, 2) design and build a new 6 to 8 gun biplane (???) off the drawing board or 3) to struggle on with existing types until the time that conversion of existing RAF types could take place. It was recognised that 3) was out of the question, but what was definitely needed was a fighter with fixed forward firing guns, the Hurricane was deemed the best basis because of its wide undercarriage.

In September 1938 a letter from the Admiralty to the Air Ministry outlined the problem (dunno who from or who to specifically) with the Skua and Sea Gladiator as the only options for fighters at that time following the decision to discontinue with the Roc, made by the Air Ministry earlier that month, with the following:

"If the Skua fails there will be no suitable aircraft available to the FAA except for a small number of Sea Gladiators. The Admiralty has always strenuously objected to have to accept converted RAF types but we are deeply committed to this course. The Admiralty has been blown hither and thither, for the Air Ministry now have grave doubts about the Skua".

After this the Admiralty requested more Gladiators, but were told no and got 47 Skuas from reserves instead. There were more correspondence between the Admiralty and Air Staff following these and in May 1940 the First Sea Lord even wrote to Minister of Aircraft Production Lord Beaverbrook asking him to do something about the precarious lack of fighters for the FAA, requesting Spitfires specifically to replace Skuas because although Hurricanes were "...10 mph faster than the Ju 88 at 16,000 ft, it is barley fast enough".

It was at this time that the suggestion of the Miles M.20 as a naval fighter was proposed, by Beaverbrook. The First Sea Lord continued his letter war with MAP with this: "The whole history of the Fleet Air Arm is a succession of failures to be ready for the event of in the shape of having ready and proven a new aircraft type to replace old and unsuccessful ones when the requirement arises, including the Roc, Skua and 'stop-gap' Fulmar."

Admiral of the Fleet C M Forbes agreed and wrote: "Our Fleet Air Arm aircraft are hopelessly outclassed by everything that flies and the sooner we get some efficient aircraft the better."

 

[EwenS]

I forgot to add a link to another site for anyone interested in the development of the Armoured carriers and their use.

Armoured Aircraft Carriers in World War II

Action reports and battle damage for the armoured aircraft carriers HMS Illustrious, Formidable, Victorious, Indomitable, Indefatigable and Implacable in World War II, 1940-1942

www.armouredcarriers.com

 

[EwenS]

Returning to the relationship of a carrier and its aircraft, there is the issue of aircraft weight and landing/take off weights and speeds especially in the mid-late 1930s with significant increases in aircraft performance.

In the mid 1930s British arrester gear was designed to stop an 11,000lb aircraft arriving at 53 knots. In Ark Royal it got a bit more complicated with more arrester wires with different ratings being applied. But the most powerful could stop an 11,000lb at 55 knots pulling the wire out for up to 155 ft. The Illustrious class matched that.

I was recently looking at the specs that led to the Firefly and Firebrand initially issued in 1939 and about which there was a lot of confusion. See my post #10 here

FAA flying hardware, 1937-45?

Or, in other words, lets give to the FAA some attractive hardware that gets the job done and then some. Whether something later to be made earlier, or something that never flew, or a substantial mod to an existing design. Limitations of availability of engines, armament and electronics remain -...

www.secretprojects.co.uk

I've now tracked down the Specs N.8/39 and N.9/39 and also N.11/40 that became the final spec for the Firebrand as a fighter. What is striking is how in the space of a few months the RN decided it had to relax some of its criteria for a suitable carrier fighter. So while required physical dimensions remain the same (things like carrier lift sizes can't be easily changed although some small relaxation might have been possible) it is in things like aircraft weight and approach speed that are most striking. First figures relate to N.8/39 and N.9/39 and the second to N.11/40.

Aircraft all up weight - 9,000lb (preferable but must not exceed 10,500lb)/12,500lb
Max speed - in region of 275 knots at 15,000ft/ 350 knots at 15,000-20,000 ft
Stall speed in landing mode - 58 knots at full military load / 68 knots at max weight
Take off distance - 300 ft in a 20 knot wind / 350 ft against 20 knot wind.

The 10,500lb weight also happens to be the weight limit specified for the Barracuda in 1937 and not much higher than the 10,000lb limit set in 1936 for the Albacore.

The specification of stall speed at max weight is important since it represents a likely worst case scenario for an aircraft developing a problem and having to return to the flight deck immediately after take off. Rare but it does happen.

With these figures you can begin to see why the RN was so keen to limit weights and landing speeds. Exceeding the figures in the 1937/39 specs carries a cost penalty in terms of the work that then needs done across the entire aircraft carrier fleet to uprate the arrester gear if they are to take the new aircraft. During WW2 this was done to existing ships and with the new Colossus class light carriers arrester gear rated at 15,000lb at 60 knots and the Implacable arrester gear rated at 20,000lb at 60 knots. Postwar Eagle completed with gear rated at 30,000lb at 75 knots.

The US carriers began with the following arrester gear ratings (I don't currently have a figure for Yorktown):-
Saratoga - 8,000lb at 60 knots
Essex and Independence - 12,000lb at 75 knots
Midway - 30,000lb at 78 knots

Similar upgrades were required to British carrier accelerators (catapults) as the war went on.

Other examples of the problems this causes are aircraft dimensions. The Illustrious class was designed in 1936 with a hangar to hold 36 Swordfish / Skua aircraft each 36 ft long. By 1940 when they began to complete the hangar could only hold 33 of the new Albacore / Fulmar aircraft each 40ft long. And by 1938 the Admiralty had decided each carrier needed more aircraft so the Implacables gained an extra half hangar (increasing hangar stowage to 48 Albacore sized aircraft) at the design stage, while Indomitable, already building, was modified to get it (increasing her capacity to 45 Albacores).

Petrol consumption was another factor. Each new generation of aircraft brought more powerful and thirstier engines. So the number of sorties that could be generated from the limited supply on the carrier began to be a problem. Attempts were made in the 1930s to increase the petrol capacity of ships like Eagle and Hermes but the sacrifice was oil fuel so shortening further the carrier's own range.

 

[Geoffrey Sinclair]

I am using the Norman Friedman US Aircraft Carriers book. He notes as of late 1940 the Essex air group was in theory 27 F5F-1, 37 SB2A-1, 18 TBU-1, plus 3 observation and 2 utility aircraft, total 87, plus 21 spares. Things kept changing fast. As commissioned Essex had 36 fighters, 36 dive bombers, 18 torpedo bombers and 1 liaison plus 9 reserves. Agreed it is really a question of which date you pick.

Do you have a list of the ships the IJN earmarked for possible carrier conversion? Does it include Chitose and Chiyoda?

Sorry for the typo, as you note the Blackburn Swordfish production was end 1940. As far as I can tell beyond the problems with the next generation of naval aircraft the disruption of 1940 caused further delays, hence production starts in 1942 of the next generation of RN aircraft.

And yes I should have said the 1940 Martlets came from the French order which became RN, my main point was how generous the US was being.

The USN reports for the RN order for 100 Martlet II, 10 accepted in March 1941, which the usual references indicate had non folding wings, then 1 accepted in June, then from August onwards, 60 accepted for the year, the remaining 40 accepted February to April 1942, F4F-3/G-36B/Martlet II, contract A-1548.

I was trying to compare production for the USN with a similar carrier tonnage allowance to the RN, certainly the USN generally had better weather where they were expecting to operate aircraft carriers. The Illustrious class was an armoured hangar, Formidable in theory could have fought at Matapan, it had comparable side armour to the RN cruisers present, Midway was flight deck/hangar roof only, not sides. Agreed the RN devoted more tonnage to in particular fire protection, no British built carrier burned in WWII.

The point about flight deck sizes was to illustrate the costs of protection, the article was about what aircraft the navies ordered, not the doctrine itself. As another example of doctrine difference, USN ships as they burnt fuel allowed sea water into the tanks, keeping their displacement roughly constant, the RN did not, the USN system allowed more top weight.

Norman Friedman states the refitted Lexington was given mark 4 arrester gear, same as Ranger through the Essex class were built with. The Midway class had mark 5.

In April 1938 Britain presented Australia with a total cost of ownership of warships, that is build, running and modernisation/refit cost, and showed it as an annual cost, Battleship 706,800 pounds (Nelson class), large cruiser 323,600 pounds, small cruiser 225,400 pounds, Aircraft carrier with 36 aircraft, 894,000 pounds, with 15 aircraft 514,500 pounds, destroyer 66,000 pounds, 1,000 ton submarine 65,500 pounds. They were pushing the idea of an RAN battleship, annual cost of aircraft set at 11,500 pounds. Australian Archives A5954 1024/10.

Notes, aircraft lifetime is 5 years and cost includes relevant personnel costs, ship lifetimes, capital 26 years, carrier 20 years, cruiser 23.5 years, destroyer 22 years, submarine 14 years. Capital ships have 3 aircraft, cruisers 2. I do not have a figure for the Swordfish cost, but 11,500 pounds per year times 5 years suggests the initial cost of the aircraft was a minor part of total cost of ownership, pre war anyway.

Agreed the RAF was being heavy handed with the RN and the no squadrons edict seems to be control for its own sake, though the capacity of the Furious, Argus, Hermes and Eagle was 1or 2 squadrons, then came the Courageous and Glorious and then came RN squadrons from around 1933. It is clear in many ways the RAF was unaware of important parts of naval requirements, while the RN had little in house aviation expertise. The effect of its junior airmen joining the RAF when it formed.

I understood the carrier tonnage limit to end 1936 was 22,000 tons. Raised to 23,000 tons, hence the displacements of Ark Royal and Illustrious.

In 1935/36 the USN fighter was the Grumman F2F (2 seater) then F3F (single seater), not sure they carried bombs and the 1935 dive bomber in production was the Great Lakes BG-1, so what is the link WWII dive bomber to USN fighter bombers?

In January 1935 the USN was nominally 15 battleships, 4 carriers, 15 heavy and 10 light cruisers, 174 destroyers and 75 submarines, counting those in reserve, the RN had a nominal 2 more carriers, 2 more heavy and 19 more light cruisers, 43 less destroyers (but 18 ocean going anti submarine ships) and 34 fewer submarines, again counting reserves. Apart from a few cruisers and destroyers the cost of the RN was taken from the British economy plus colonial revenues which was half or less the size of the US economy, plus the British government had war debts.

I consider the effect subtle because of the hidden effects of the lack of aviation experience of senior officers in the RN, something that became worse as aviation became more complex. To have an RN equivalent of a Hurricane or Spitfire development needed to start about the same time, 1934/35. The Director Naval Air Division was Captain Henry C. Rawlings: March 1932-July 1934, using Wiki, so an RN officer.

How much did the pre war ideas of carriers staying out of at least land based fighter range play a part in a willingness to trade performance for other abilities. The pre war doctrine was carriers would normally lose against land based air, something that took war experience to change.

The 1933 ideas about 2 seater fighters and also in N8/39 showed a preference for two seaters, something the RAF shared, given the memories of the WWII Bristol fighter, cue Hawker Demon. Plus it allowed the aircraft to be a long range escort and do reconnaissance, a navigator and someone to operate the longer range Morse radio sets.

From the Stirling topic as well as this one.

Looks like you are also using the time line in Spitfire by Morgan and Shacklady. End 1933 two seater fighter and two seater strike/reconnaissance ideas, the navy noting as aircraft capacity was limited there was not much space available for carrying pure defensive fighters. In 1934 the Air Ministry said the planned performance was not required for about another 4 years, something Hawker agreed to. End 1934 Admiralty limited designs to a 40 foot wingspan. End 1937 all agreed the Skua and Roc were not good enough, obsolescent by end 1939, a committee was formed and made little progress over the next year. In May 1938 Richard Fairey was presented with the request to build Spitfires (which is 2 months before first Spitfire production), he came up with a number of reasons against the idea, killing the proposal. Things like factory capacity, the time it would take, delays in current Fairey designs, even throwing his current design team out of work.

The November 1938 request for more Gladiators and the reply talking about 47 Skua from reserves, what is interesting is it took until end February 1939 to build 36 Skua, end March for 61, so the reserves were theoretical at the time, it would be nice to know what the split was for the 136 on order was, front line, training, reserve.

A Spitfire had flown with an A frame arrester hook in October 1939. However the First Sea Lord, Winston Churchill cancelled the naval Spitfire project, the official letter arriving at Supermarine on 16 April 1940.

N5/38 was the single seater fighter, N6/38 the two seat turret fighter, then came N8/39, one or two seater, looks like Supermarine proposed a Merlin for the single seater, Griffon for the two seater.

How much of a catapult upgrade was being planned given the Swordfish was about 8,700 pounds loaded with a lower stalling speed versus the Fulmar I at 9,800 pounds loaded?

Blackburn Skua production began in October 1938, Roc in February 1939, Fairey Fulmar in April 1940. Sea Gladiator including interim from December 1938. Seems the RAF wanted the Roc out of the way so it could build Defiants, but the original order for 136 was completed (Order 632260/37 requisition 46/36), no additional orders were made.

Britain produced 1,438 fighters in 1939, including 320 Gladiator, 81 Roc and 16 Defiant, but not counting the 175 Skua. Ignoring transfers from RAF stocks, 24 Gladiator and 83 Hurricane were for export.

The G-36A/Martlet I were accepted July to October 1940. Two F4F-3 accepted in August, then production from November.

Sea Gladiators, the first Gladiator mark II were actually for Sweden, in July and August 1938 at the latest, mark I production for the RAF continued until October. RAF/RN mark II production began in December 1938, including some as Sea (interim) for the RN with 38 airframes allocated, 22 delivered. Production of the order for 60 Sea Gladiator began in February 1939.

N2265 to N2302, First set of Gladiator II serials, 38 Sea Gladiator II (interim) but 16 were retained by the RAF and sent to Aden, those to the RN being N2265 to N2276, N2281, N2282 and N2296 to N2302, the surviving RAF ones mostly to SAAF in April 1941. The relevant surviving interim models are marked transferred to Admiralty control 28 March 1940, or on loan as of April and June 1940, the Sea models marked transferred 24 May 1939.

 

[EwenS]

Do you have a list of the ships the IJN earmarked for possible carrier conversion? Does it include Chitose and Chiyoda?

When I made that post the conversions from IJN auxiliaries escaped my mind. I was only thinking about the mercantile conversions. Here is the list for both types that I have:

Sub depot ships
Takasaki - converted 1940 to Zuiho
Tsurugisaki - converted 1941 to Shoho
Taigei - converted 1941/42 to Ryuho

Seaplane carrier / mini sub carrier
Chitose - converted 1943
Chiyoda - converted 1943
Mizuho - lost May 1942 before conversion could begin.

I'm not sure whether the seaplane carrier Nissin was also part of this program or not.

Merchant ships earmarked for conversion and whose cost was subsidised by the IJN to include features making carrier conversion possible:-

Asama Maru class completed 1929/30 16,975 tons 20.7 knots (Asama Maru, Tatsuta Maru and Kamakura Maru ex Titibu Maru). All 3 were lost before they could be taken in hand for conversion. The last was requisitioned in 1941 as a transport and hospital ship and was due to go into the yard after Scharnhorst / Shinyo (see below) but was lost 28/4/43. Diesel engine Ines would have been replaced by destroyer turbines.

1937 Nitta Maru class 17,000 tons 21 knots
Nitta Maru, Kasuga Maru and Yawata Maru became escort carriers Chuyo (converted 1942), Taiyo (converted 1941) and Unyo (converted 1942).

1938 Argentina Maru class 13,000 tons 21 knots.
Brazil Maru lost before conversion could begin. Argentina Maru converted to Kaiyo 1942/43.

1938 Kashiwara Maru and Izumi Maru 27,500 tons 24 knots. Taken over while building in Aug 1940 and emerged as Junyo and Hiyo respectively. First Japanese carriers with combined island and funnel.

In addition the German liner Scharnhorst was trapped in Japan when war broke out in 1939. After Midway the IJN purchased her and converted her in 1942/43 along similar lines to the Nitta Maru class.

I was trying to compare production for the USN with a similar carrier tonnage allowance to the RN, certainly the USN generally had better weather where they were expecting to operate aircraft carriers. The Illustrious class was an armoured hangar, Formidable in theory could have fought at Matapan, it had comparable side armour to the RN cruisers present, Midway was flight deck/hangar roof only not sides. Agreed the RN devoted more tonnage to in particular fire protection, no British built carrier burned in WWII.

The point about flight deck sizes was to illustrate the costs of protection, the article was about what aircraft the navies ordered, not the doctrine itself. As another example of doctrine difference, USN ships as they burnt fuel allowed sea water into the tanks, keeping their displacement roughly constant, the RN did not, the USN system allowed more top weight.

But different doctrines and operating conditions is what drives the air group size and makes direct comparison of the USN carriers and British carriers difficult. That is why the best comparison for Illustrious is Ark Royal. On almost the same Treaty limited tonnage you get aircraft or armour but not both.

Another example. All US carriers through to the postwar United States (not built) were constructed with the hangar as a superstructure deck. The hangar deck was the main strength deck. From Ark Royal on RN carriers had the flight deck as the main strength deck. Even the Furious, Courageous and Glorious used the flight deck to add to the strength of the hull girder due to their relatively light cruiser construction when built.

In the Illustrious class the armour was structural not an add on the ships structure as in a cruiser. It was a very efficient way of utilising the weight available. Remove the armour plate and you need to add in a lot of steel to give the ship its strength back.

As for the hangar side armour, it was seen as necessary for two reasons. Firstly to keep out bombs that might strike the ship under the armoured flight deck thereby maintaining the all important protection for its main weapon, its aircraft. Secondly to provide protection from enemy cruiser gunfire, something that was seen as a threat by both navies. And so it transpired when Glorious encountered Scharnhorst and Gneisenau in June 1940 and when Formidable got caught up at Matapan. Incidentally Formidable is supposed to have loosed off a few 4.5" rounds before stopping to avoid drawing attention to herself.

Looking at Midway the only part of her flight deck that was armoured was the part between the centreline lifts. She also had an armoured hangar floor and second deck, and armoured transverse doors in the hangar. While there was no armour on the hangar sides, a fair degree of hangar protection was obtained from the pedestals on which the 18x5"/54 AA guns sat. And to support the weight of that armoured flight deck meant a lot of steelwork along her sides anyway. If you look at her there are far fewer openings in the hangar sides than any US carrier since the Lexingtons. And that portion of the hangar aft of the after lift was completely unprotected. No armoured flight deck or hangar floor. And her deck armour was not structural. It required a steel backing to support it. See the discussion here.

NavWeaps Forums-Midway class - flight deck armour layout

One (of many) criticisms of the Illustrious/Implacable class is that the 3" armour covered only that part of the flight deck over the hangar, with eve

www.tapatalk.com

I understood the carrier tonnage limit to end 1936 was 22,000 tons. Raised to 23,000 tons, hence the displacements of Ark Royal and Illustrious.

See Washington Treaty Article IX

CONFERENCE ON THE LIMITATION OF ARMAMENT, 1922.

The 1930 London Treaty changed the definition of carrier to include ships under 10,000 tons but otherwise the limits remained until the end of 1936. At that point the limit on individual carrier size dropped to 23,000 tons but there was no longer a total carrier allowance for each country. Britain had been seeking a limit of 22000 tons but had to settle for 23000 tons, which may be where you got the idea that the limit had increased when the Illustrious class were designed.

In the early 1930s the British carrier tonnage for Treaty purposes was made up of

Argus - 14000
Eagle - 21000
Hermes - 10500
Furious - 22450
Courageous - 22500
Glorious - 22500
Available for Ark Royal without scrapping earlier vessels - 22050
Total - 135000

Under the Washington Treaty the first 4 could be replaced at any time as being carrier tonnage built or building in Nov 1921. Furious was a prototype carrier in the eyes of the RN and had also been stripped to her main deck ready for reconstruction by the time the Conference started. Looking at Friedman"s British Carrier Aviation, the criteria set down by the Admiralty were able to be met on 22000 tons without resorting to scrapping one of the earlier vessels (Argus was in Reserve at this time).

For the USA (ignoring Langley converted to a seaplane carrier in 1937 to free up carrier tonnage on same principle as first 4 RN ships)

Lexington - 33000
Saratoga - 33000
Ranger - 14500
Yorktown - 20000
Enterprise - 20000
Wasp - 14500
Total - 135000

For Japan

Hosho - 7470
Akagi - 26900
Kaga - 26900
Ryujo - 7900
Unused - 11830
Total - 81000

Hosho could be replaced at any time.

Japan announced her withdrawal from the treaty system in Dec 1934 by which time the 15,900 ton Soryu was building and the Hiryu was planned as a follow on.

In 1935/36 the USN fighter was the Grumman F2F (2 seater) then F3F (single seater), not sure they carried bombs and the 1935 dive bomber in production was the Great Lakes BG-1, so what is the link WWII dive bomber to USN fighter bombers?

From Friedman "Carrier Air Power"
"...Dive-bombing in the US Navy began with fighter squadrons, some of which were specially designated bomber-fighters; the pre-war US Navy was unique in requiring its fighters to carry substantial bomb loads....."

That began in 1926 with trials with Chance Vought UO-1 Corsairs and later two seat fighters like the F8C-2 and the two seater scout Great Lakes BG. Then things get complicated with changes of designation and an alphabet of numbers O2C, O3C, F10C/S3C and F12C/S4C/SBC. That eventually led to the Northrop BT and Douglas SBD.

I consider the effect subtle because of the hidden effects of the lack of aviation experience of senior officers in the RN, something that became worse as aviation became more complex. To have an RN equivalent of a Hurricane or Spitfire development needed to start about the same time, 1934/35. The Director Naval Air Division was Captain Henry C. Rawlings: March 1932-July 1934, using Wiki, so an RN officer.

Per Hobbs "A Century of Carrier Aviation" in the 1920s the Air Ministry set up two committees, the Advisory Committee on Fleet Air Arm Aircraft and its Technical Sub-Committee. But most members were drawn from the RAF and Air Ministry who provided technical knowledge and direction. RN participation was limited to junior officers without staff experience. That is hardly a subtle way of exerting control.

It seems to have been the Air Ministry belief that naval aircraft fulfilled some secondary purpose that could be met with landplane conversions. That was then allowed to filter through to manufacturers who came to believe that developing naval aircraft was of little value. And of course all aircraft development had to be via the Air Ministry.

As for the Naval Air Division, that was a branch of the Admiralty, hence the naval officer in charge. It's responsibility was operational control of aircraft at sea. Nothing else. Anything connected to land operations and administration was the responsibility of the Air Ministry. So that included training and procurement of new aircraft.

Looks like you are also using the time line in Spitfire by Morgan and Shacklady.

It's in the library but never referenced it for this discussion.

How much of a catapult upgrade was being planned given the Swordfish was about 8,700 pounds loaded with a lower stalling speed versus the Fulmar I at 9,800 pounds loaded?

Carrier catapults were as follows:-

BH.I as in C&G, Argus and Ark Royal
Prototype 1935 8,000lb at 56 knots
Upgraded to 10,000lb at 52 knots
Upgraded to (1938) 12,000lb at 56 knots

BH.III Illustrious class, Colossus &early Majestic. (Illustrious was first fit)
Prototype 1940 11,000lb at 66 knots
Upgraded to 12,500lb at 66 knots with the trolley or 14,000lb at 66 knots using a bridle
Upgraded to 16,000lb at 66knots with trolley or 20,000lb at 56 knots (?) using a bridle on a twin track arrangement on later ships.

The trolley was time consuming to set the aircraft up on and the RN preferred free take offs whenever possible. Tail down launches with the bridle only began when US Lend Lease aircraft began to arrive around 1942 when BH.III were modified to do it. The first British aircraft so adapted was the prototype Seafire XV in 1944.

The last of those upgrades to the BH.III probably came in 1944.

 

[SaparotRob]

Very informative. Thanks, EwenS.

 

[EwenS]

In the US the Second Vinson Act of May 1938 authorised an increase in USN carrier tonnage from the Washington Treaty total of 135,000 tons to 175,000 tons.

When it came to deciding how to use that extra 40,000 tons the USN found itself constrained by a lack of design capacity (New Battleships had priority at that time). The increase is enough for 2 ships and neither could exceed 23,000 tons per 1936 London Treaty limits. It quickly concluded that on around 20,000 tons a repeat of the Yorktown design was the best carrier that could be achieved at that time. That led to the ordering of Hornet on 30 March 1939 with her construction starting on 25 Sept 1939 leading to completion on 20 October 1941.

The balance of available tonnage under the US legislation once everything had been scraped together was 20,400 tons and became the starting point in 1939 for the Essex design. Once war breaks out, Treaty limits end, and Congress finally passes the Two Ocean Navy Act in July 1940 authorising procurement of 18 new carriers. So the Essex design can finally grow to it's eventual 27,000 tons by the time the design is finalised in 1941 and the USN can begin to think of ordering more carriers.

Meanwhile CV-9 Essex is ordered in Feb 1940, with the next pair in May and 8 more in Sept, being CV-10 to -19. The last of the pre-Japanese war planned ships were ordered in Dec 1941, CV-20 & 21.

Essex is finally laid down in April 1941, 3 years after she was first being considered with planned completion in Q1 1944. CV-16 & 17 follow in July and Sept and a further pair, CV-10 & 11, at the beginning of Dec, just a week before Pearl Harbor (I have refrained from using names for these ships as there were a number of changes prior to launching partly to commemorate carriers lost in the first year of the Pacific war and it therefore gets confusing).

Meanwhile Britain orders two Illustrious class as part of her 1936 Programme and another pair as part of the 1937 Programme with all being laid down in 1937. For the 1938 Programme it decides it wants a faster ship with greater aircraft capacity and the Implacable design emerges with one ship laid down in March 1939, and a repeat ship in Nov 1939 as part of the 1939 Programme. Unfortunately, with the hiatus of the early war years these last two ships were severely delayed, not becoming operational until 1944. Meanwhile due to expected delays in the delivery of armour plate from Czechoslovakia, the design of Indomitable (1937 Programme ship) is recast in 1938 to incorporate as many features as possible from the 1938 carrier Implacable. The 4 Illustrious class ships complete between May 1940 and Oct 1941.

Over in a Japan now free of Treaty limits, Soryu (laid down in Nov 1934) is followed by a close but slightly larger sister, Hiryu (laid down July 1936) incorporating changes suggested as necessary by various Fleet Incidents which had revealed weaknesses in Japanese ship design. They are then followed by Shokaku and Zuikaku of 25,700 tons standard displacement in Dec 1937 and May 1938 for completion in Aug and Sept 1941, just in time for the outbreak of the Pacific war.

And between 1934 and 1938 Kaga and then Akagi are taken in hand for reconstruction which significantly improved their aviation capabilities.

In 1939 Japan begins the design of its next generation of carrier design, starting with an enlarged Shokaku design and then applying an armoured flight deck to the design before it was finally laid down in July 1941 and completed in Mar 1944 as Taiho.

In 1940 Japan triggers its shadow carrier programme with Zuiho, Shoho and Taiyo all in service before Pearl Harbor and more under conversion.

 

[Geoffrey Sinclair]

Thanks very much for the replies. A couple of points.

The early fighters, the Boeing ones do not seem to be bomb carriers, it does seem any early fighter that could carry bombs was redesignated as you note. The F8C becoming the OC and the O2C. The Grumman FF and 1930's successors were not bomb carriers, the BG-1 dates from mid 1933, the FF from late 1931. The last Boeing fighter from mid 1928. The F8C/OC/O2C 1928 to around 1930. It looks like in the 1920's there were attempts at multi purpose aircraft but that resolved itself in the 1930's to more single purpose types. The SBC dating from 1935.

I though the Hornet decision was based on time factors, an old design now versus a new one in a few year's time.

 

[EwenS]

I though the Hornet decision was based on time factors, an old design now versus a new one in a few year's time.

I think we are just picking up on different aspects from Friedman's US Carriers An Illustrated Design History. From p94.

"Although the Yorktown design was relatively old by 1938, the pressure of rearmament led to its duplication in the Hornet(CV 8) when additional carrier tonnage was authorized after the expiration of the Washington Treaty. In July 1938 C&R design priorities emphasised the new 45,000 ton battleship (which became the Iowa class); it was estimated that no new carrier design could be ready for as much as fifteen months, a year late. The alternatives were to duplicate the existing Yorktown with its known deficiencies or else to consign some of the designs to private firms, as had actually been done in 1933. The Yorktown design was repeated because it did not appear that much more could be achieved under 20,000 tons, and the Hornet (CV 8) was built to it under the FY 39 program....."

I suppose the other unstated pressure would have been political. To be seen doing something with that increased tonnage you have been requesting and now got.

But the 1938 decision to increase the permitted carrier tonnage by the specific amount of 40,000 tons is the bit that puzzles me. 1936 London Treaty limits individual carrier size to 23,000 tons. But you can now have as many as you like.

The 40,000 ton increase suggests 2 ships were what was in mind. And c20,000 was the size that they had been building to in the Yorktowns. But if you want to limit the size of your navy by only adding 2 carriers why not make the allowance 46,000 tons? Build them to the new Treaty limit. Maybe someone from the US can shed light on the reasoning for the limit being decided that way.