Steam and smoke. Not a secondary explosion, but “steam and smoke rising,” as one would expect from a train station. The Japanese reports state only one bomb impact was witnessed and recorded, yet in all American accounts provided by York and Emmens, the entire load was dropped. Nolan Herndon clearly stated, “We went up to 1,500 feet and dropped the bombs.”
Bombs, which would seem to mean all three five-hundred-pounders and the incendiary.
From a tactical standpoint, this makes perfect sense: extra bombs were extra weight, and since there were no other targets of opportunity in the area, York would have wanted to immediately proceed across Honshu as fast as possible to get out of Japan and on to the Soviet Union. There are two solutions to this puzzle.
First, because of the low altitude, the other three bomb fuses did not function, so there were no explosions. In 1942 the area was more forested than it is today, and there was a canal, so if the other bombs were dropped and did not fuse, they may have landed in the water or in the soft dirt of the surrounding fields. In any event, if anything was recovered, it was never reported. Second, there is incontrovertible evidence that Niitsu, a larger town in the Niigata Prefecture on the northwest coast, also suffered an attack. The Imperial Army Aeronautical Technical Unit investigated an air raid that occurred on April 18, 1942, and subsequently filed a report outlining an attempt to bomb a bridge over the Agano River about 1.5 miles northeast of Niitsu. An extract from a wartime log available in the Japanese Center for Asian Historical Records preserves a telephone message from Niigata Prefecture stating, “aircraft came from upstream [southeast] of Agano river, dropped bomb and flew away towards direction of Soviet Union.”
One bomb landed on a sandbar in the river, and the other two in farm fields on the west bank of the river. The bridge was unharmed, and a nearby farmhouse suffered minor damage from “5 roof tiles and 3 screen windows.”
On that same day, the prefectural governor, Shohei Doi, arrived to inspect the damage, and the Niitsu town record for April 1942 documents the following:
On 18th April, at 1:30 P.M. one american [sic] aircraft arrived and dropped 3 bombs, but damage was very minimal at the farmland of Ohaza, Naka-Shinden. No human / livestock casualties. At the same time bombs were dropped, the aircraft made a strafing as well as dropping incendiaries, but it was lucky that all of them fell into Agano River.
The basic problems with the Mitchell continuing on to Niitsu were fuel consumption due to extra weight and, most important, rationale. The distances involved are relatively equal. From Utsunomiya, where York believed they were, to the Niigata coast below Sado Island, was approximately ninety miles, and to the Agano River bridge the distance was right at eighty miles. This is the same distance from Nishi Nasuno to the coast, or to the bridge in the Niigata Prefecture. Interestingly, the bomber was seen over Niitsu from 1330 to 1340, depending on the source, and this is perfectly within the nineteen-to-twenty-minute flight time required from Nishi Nasuno to the Agano River area at the 250 miles per hour York maintained over enemy territory.
Now, if Plane 8 dropped only a single bomb on Nishi Nasuno, then it carried fifteen hundred pounds of ordnance eighty additional miles, which, combined with higher fuel flows for the increased airspeed, would have cost an extra twenty to thirty gallons of gasoline. Not a huge amount, unless you still have five hundred miles to fly across the Sea of Japan to the Soviet Union. Also, flying to Niitsu and then heading directly toward Russia would place the bomber north of Sado Island, yet both pilots later emphatically stated they passed south of the island on their way across the Sea of Japan.
The final difficulty with this scenario lies with rationale. The real mission was to get to the Soviet Union, so why risk another attack on a city far to the north, expending precious gasoline and critical time to do so? The point had been made by dropping on the first target, so there was no need to hit another. York, in fact, states unequivocally, “We passed within sight of it [Sado Island]—to the southwest of it. As soon as we got down in this flat, we let down near a bay northwest of Central Honshu south of Sado Island, and continued in a northwesterly direction toward Vladivostok.” Forty-two years later in his United States Air Force oral history interview, Ski again affirmed that “after we got rid of our bombs, we made a beeline to the northwest.” There has never been a mention, from any member of the crew, that Plane 8 struck two targets in Japan—but given the detailed Japanese records compiled from independent sources, this second attack is now a certainty. Only the crew knows why it was never mentioned. Possibly the secrecy surrounding the mission, or a fear of facing questions about bombing unidentified and possibly civilian targets.
There are a few possibilities that must be considered for the Niitsu story.
First, the incident was a product of Japanese propaganda put forward to further stoke war fever, control the populace, and encourage greater efforts from northern prefectures fairly far removed from the industrial south. Second, it was an accident. The Imperial Japanese Army Air Force (JAAF) had at least twelve airfields from which training was conducted scattered throughout Honshu alone. According to Makoto Morimoto, a piece of metal was recovered with CHICAGO ACME STEEL stamped onto the surface. While this is no doubt accurate, by itself it does not guarantee the metal was part of a bomb delivered from a U.S. aircraft. A significant number of Japanese munitions, including aerial bombs, were made from the millions of tons of American scrap metal, particularly steel, imported during the previous decade.* On the other hand, if this incident was an accidental bombing by the JAAF, then they would hardly admit it given the other events of that day.
The third, and most likely, possibility is that York, a consummately professional military pilot, dropped only a single bomb on the train station at Nishi Nasuno because he did not see the factory he planned to hit—obviously because Plane 8 was not over Utsunomiya. This left him with two five-hundred-pounders and the incendiary bomb. His only real option would be to get to the coast and find a target of opportunity, such as a bridge, or simply release the remaining weapons over the water. If one takes their planned egress heading from Utsunomiya to the coast and shifts it north to commence from Nishi Nasuno, the Mitchell’s flight path would be directly over Niitsu: the same heading and the same distance—310 degrees for approximately 80 miles. After exiting the hills and dropping down on the coastal plain, both pilots and the navigator would have clearly seen Sado Island, which is forty miles long and barely twenty miles offshore. The Agano River bridge is practically the first target of significance as one approaches Niitsu, so upon seeing this it is logical that the remaining bombs were dropped immediately. A due west heading for sixteen miles would take the Mitchell over the Shinano River and out to sea toward the south tip of Sado Island, where they could now pick up their planned route to Russia.
In this context, the final scenario fits the events recorded at Nishi Nasuno and Niigata on April 18, 1942. There were witnesses to both attacks, and the Japanese naval attaché in Niigata clearly reported that military police identified a “two engine, single wing type” bomber with “a circle with star” marking on the underside of the wing and on the fuselage “similar to those insignia found on Imperial Japanese Army soldier’s field caps.” There were no witnesses from the downtown area, but the Agano River bridge is on the eastern edge of the city, and if Plane 8 dropped back down to fifty feet and turned immediately toward Sado Island, then it would have missed most of the populated area. There were also no photographs of any other targets from the bomber’s tail camera, but this is hardly definitive. The film could have run out, or been lost in Russia, or simply not passed on to the American military attaché in Moscow, as was the first batch of pictures from Nishi Nasuno. The fact that York went on to bomb another target near Niitsu on the northwestern coast did not affect the rest of the mission, but it does clear up yet another historical mystery.
One hour out from the Russian coast, Plane 8 would have had less than 250 gallons of gasoline, and when York made landfall an hour later after 470 uneventful miles across the Sea of Japan, there would not have been fuel remaining to loiter anyplace very long, nor did he. “I didn’t want to land in Korea,” Ski recounted. “I knew Korea was part of Japan in those days. When we made landfall, I just turned up the coast and flew on until our fuel started getting pretty low.”
In fact, he flew about twenty miles northwest along the coast, later claiming he wanted to make certain it was Russia, not the Korean Peninsula, but this also is obfuscation. The extreme northern edge of Korea was another seventy to eighty miles farther west—at least another half hour of flying. Also, with Sado Island as an exact landmark, Nolan Herndon was able to precisely calculate the heading to Cape Povorotnyy, the closest piece of Soviet territory they would encounter. The heading difference between Korea and the cape would have been twenty-five to thirty degrees—an error that York, Herndon, or Emmens would not make.
Far more certain is that York had planned on landfall at Cape Povorotnyy, but when he arrived it wasn’t clear if the spit of land off his nose was the correct spot, so he turned right, to the northwest. This made perfect sense from the Korean point of view, if one accepts that two pilots and a navigator followed a thirty-degree incorrect heading for eighty miles farther than planned, but that theory holds no water. However, it is logical that from one hundred feet over the water, a pilot getting his first look at such a landmark could not absolutely determine if the rocky shoreline was Cape Povorotnyy or Cape Sysoyeva, thirty-five miles farther west. A pilot would turn right, figuring that if it was the latter, the bay he was looking for would quickly appear, and if the former, he would know in twenty miles or so, because there would be no bay.
In this case, it was Cape Povorotnyy.
There was no large bay after York turned right, so he turned around and headed back toward Povorotnyy. But why did this matter? They were plainly not in the area close to Vladivostok, where York later insisted he was heading, so what were they looking for? What bay in that area, sixty-five miles east of Vladivostok, was he trying to locate, and why?
Nakhodka. Also known as American Bay.
In fact, another obscure sentence buried in the U.S. Army intelligence memorandum unambiguously admits that Plane 8 “landed at an airfield on American Bay.” York never intended to land near Vladivostok any more than he intended to bomb Tokyo. Short on fuel as they were, Ski would not have flown up and down the coast unless he was looking for a specific area, which he found. York reveals that he landed the bomber “at an airfield on American Bay,” which is Nakhodka Bay, “about forty miles from Vladivostok.” In fact, American Bay is closer to sixty miles east of the city, and the only airfield that fits the description was Unashi Aerodrome, which is eleven miles north of the bay on the Partizanskaya River.* Incidentally, this neatly matches the cryptic description given of a “fair sized river” at the head of a bay, which is not an obvious feature anywhere around Vladivostok. Unashi also would fit York’s account of an airfield, not a runway, in a valley, as would the description of Russians dressed in black coats, with “flat black caps with ribbons in the back and blue lettering on the bands.”
This was the uniform of the Russian Naval Infantry, and the field was therefore a Russian naval airfield. Unashi’s location makes this a certainty, as it was closer to the Sea of Japan than the bases near Vladivostok. York himself specifies that the Soviet officer who visited them later that night was not an army colonel but rather a “naval colonel.” American Bay was selected because from the maps he procured in Washington, from Hap Arnold or Army G-2, York knew there was a base less than fifteen miles up the valley. Had this not been known in advance, there was no reason to turn north, ninety degrees away from Vladivostok, and fly to Unashi. Verifying its condition, its infrastructure or lack thereof, and its suitability for USAAF use against the Japanese islands was a central objective of Plane 8’s mission. Unashi was also far enough from Vladivostok and the prying eyes of Soviet officials that Ski believed he could bluff the surprised local Russians into simply refueling the bomber and letting it continue. In any event, he certainly would not have risked fuel at such a critical juncture had he not already known of the airfield.
Unashi is not a place one finds by accident—especially from a few hundred feet up over unfamiliar, marginally mapped terrain. Gasoline was now very short, and a competent pilot would never arbitrarily head off in a random direction hoping to discover a place to land, especially when there were known airfields fifty miles farther near Vladivostok. York explicitly names American Bay, and this is corroborated by Russian historian Boris Egorov, who writes of the incident, “It was only at half past five in the afternoon, after the American warplane had appeared over Unashi military airfield a few dozen kilometers from the port of Nakhodka.”
A final photo was taken of York’s bomber by the Russians before it was confiscated by the Red Air Force. According to Walter Kurilchyk, an amateur sleuth who made the Mitchell’s fate a sixteen-year hobby, “After landing in Russia, it was flown 107 hours. It was used to shuttle mail and military personnel [in Russia]. Then it was overhauled with two new engines and flown [back] to a station in Unashi, in eastern Russia.”
As with everything regarding Plane 8’s mission, deception and obfuscation were vital components, not only during the original flight but also during the decades following the flight. No one has satisfactorily answered why one of the spare crews that made the trip west to California did not substitute in for Bates and Roloson, nor is there an explanation except that York had to go on the mission. This is also why the sixteenth Mitchell was added at Alameda. Doolittle always said he was not certain how many bombers would fit on the Hornet’s flight deck, but this is nonsensical. With a doctorate from MIT, Jimmy was quite capable of figuring out the flight deck’s available square footage, and since he required 350 feet for his takeoff, he would know how many B-25s could fit in the remaining space, angled outboard and parked staggered nose to tail.* Like York’s crew, the sixteenth plane was added, seemingly by chance at the last minute, because Doolittle knew he was one bomber short; York would not be over Tokyo, and Jimmy needed five flights of three aircraft to cover the listed targets.
Deception.
The Mitchell’s sudden appearance over the Maritime Territory brings up yet another series of opaque details and obfuscation from this extraordinary flight. These elements are amplified by two contradictory reports later made to Army intelligence and York’s own oral history. The first document is the combined debriefing from the entire crew under the auspices of the assistant chief of air staff, intelligence, on June 3, 1943 (see appendix one). This lengthy interview was conducted by ten Army officers, the majority with operations or intelligence backgrounds, though two men represented the historical division, so this account was quite obviously being preserved for posterity.
Interestingly, in this document the McClellan carburetor incident was covered first, before any other facet of the crew’s mission or internment. York stuck to his story, emphatically making the point that the only reason he had to divert was due to the increased fuel consumption and that the first time he knew about the switch was aboard the Hornet. When queried about crossing into Japanese airspace, he disingenuously said, “We made our landfall and should have been in the area of our target in about 20 to 25 minutes … we still hadn’t spotted Tokyo itself.” Which was true; he had not spotted Tokyo because Plane 8 was fifty miles north of the city going in the other direction.
Only York and Emmens were in on the airborne portion of the plan, and they did not reveal anything more than the other three crew members had experienced. In fact, the entire document is a recitation of the party-line story that would be told again and again over the decades. The gunner and engineer would not have been able to piece together what truly happened, and they believed what they were told. In any event, sergeants don’t question captains and aircraft commanders, so even if they had doubts nothing would have been said. The navigator, Lieutenant Nolan Herndon, was a different matter. He was not part of Arnold’s plan, and was never told about it. Yet he could read a map and was quite aware the pilots were flying nowhere near the planned targets. He would also have known that the tip of Primorsky Krai was not the Muravyov-Amursky Peninsula containing Vladivostok.
What transpired between the navigator and the pilots has never been revealed, but Herndon figured out fairly quickly that York was executing his own plan independent of the raid on Tokyo. Years later, the navigator put the question directly to Jimmy Doolittle, who answered, “I’ll tell you one thing, Herndon: I didn’t send you there.” Lieutenant Tom Griffin also stated in later years, “I guess we will never find out if the State Department or the Secret Service set up ‘Ski’ York for a special secret mission. It is my belief that his clandestine mission instructions were never put in writing.”
The second document is an interview compiled in memorandum form for the Army Air Forces director of intelligence, and the interviewee is Major Edward York—alone (see appendix one). There is no discussion of carburetors or not being able to find Tokyo. In this document, York specifically mentions Utsunomiya, American Bay, and the nearby airfield. Most of this memorandum is an assessment of the Russians, both collectively and individually, the conditions of the country, Stalin, and the crew’s itinerary over their eighteen-month internment. This document contains the truth, not the cover story, and deals with some of the questions Ski was sent to answer regarding Soviet ability and capabilities. This needed to be in the official record, and it is quite revealing that the interview is conducted solely with York, and not for one of the lesser departments, but for the director of Army intelligence.
With the luxury of examining all the available documents, and with a combat pilot’s experience as to what is probable and improbable, a clear pattern emerges. It is important to note that even if the deception was penetrated, after 1943 it was a moot point. Still, the stories published for public consumption were, in modern eyes, contradictory and woefully unbelievable. In the first interview with military intelligence, York specifically states he landed at a base “40 kilometers north” of Vladivostok, after overflying one field. The only two fields that fit this description within forty kilometers of the city are Uglovoye and Knevichi. Yet this begs the same question, if fuel was so tight, why would he overfly a suitable place to land? He would not, because he was never close to Vladivostok, nor would an American pilot in 1942 measure distance in kilometers.
Deception.
At first it seemed possible that York and Emmens simply misidentified their location. This would be all too easy after their extremely long day and the poor quality of their maps. Yet the subsequent comments by York identifying American Bay and the very vague descriptions of their actual location negate that assertion. Emmens writes that they were flying over an “inlet, or bay,” but even with their inaccurate maps the pilots would have been aware of the obvious size difference between this bay and those bordering Vladivostok and the Muravyov-Amursky Peninsula. Nakhodka, or American Bay, is approximately six miles wide and ten miles inland. Ussuriyski Bay, east of Vladivostok, is at least twenty miles wide and thirty miles deep, while Amurskiy Bay to the west is nine miles wide and some twenty miles deep. Even one thousand feet up, York and Emmens would have recognized that they were not close to Vladivostok.
Obfuscation.
Emmens also claimed in Guests of the Kremlin that York had never made a simulated carrier takeoff, yet that was not true. “I had done two or three of those short takeoffs,” Ski recounted, “and that is enough. You don’t have to do 500 to be proficient. It is not all that big a deal to tell the truth.” Emmens knew this, of course, but the conversation took place over the bomber’s intercom, so the rest of the crew heard it. This also was another layer of the cover story that, if ever repeated, would add substance to York’s “last minute” inclusion in the raid.
Deception.
Unquestionably, landing near American Bay was the plan, so why was this important to Hap Arnold? Undoubtably, his reasons were based on pressure from Roosevelt regarding a second air front against the Japanese and the level of expected Soviet cooperation. Hap’s brother Tom Arnold had worked as a field engineer in the Soviet Union during the 1920s, when Stalin desperately required foreign expertise to build infrastructure—especially airfields. Convinced the Russians would never let him return to the United States, Tom escaped from the Soviet Union one dark winter’s night by walking across frozen Lake Ladoga into Finland. What he had seen, and what he knew of Russia, was passed along to his brother, including the existence of aerodromes and rail and port facilities near Vladivostok.
Yet this data was more than fifteen years old by April 1942, and Arnold would need to know the exact field location north of American Bay, and whether there was an actual concrete runway or simply grass. Was there room for additional runways to be built, as well as hangars, with support buildings and billets? What type of roads and/or railways ran into the area from Vostochny Port on American Bay? Were any deep-draft vessels visible in the bay that might give an indication of how deep it was? Were there wharfs and dockyard facilities, and how close was the railhead? Answering these questions was the reason the Mitchell purposely came in over the bay rather than simply hopping the ridgeline into the valley. Both Ed York and Bob Emmens were capable of gathering viable intelligence from a brief look at the area.
The remote location was another consideration in choosing Unashi. Such a base would be easier for the Americans to secure and, if the port was suitable, to resupply without having to depend on Vladivostok. It would also be easier for Plane 8 to fly out of as they planned to do. The sudden appearance of an American bomber would take the Soviets by surprise, so York would have a better chance of capitalizing on the confusion to get fuel and simply depart again from Unashi rather than the center of power in Vladivostok. In fact, this was likely the plan, as Bob Emmens related Ed York wanted to continue on that very night, even after flying nine hours. Both men realized, or had been briefed, that the longer they remained in the Soviet Union the greater the risks—of detainment or even being turned over to the Japanese. A quick drop in, with a fast refueling and departure, stood a good chance of success as long as the local Russians were caught off guard and remained so.
