The first jet fighters introduced near the end of World War 2 were notorious fuel hogs, and they all promised to have insufficient range to escort the long-range B-35 and B-36 bombers then on the drawing boards. On January 29, 1944, the Army Air Forces invited the industry to submit concept proposals for jet fighters capable of escorting its long-range heavy bombers. As one possible solution to this range problem, the USAAF revived the parasite fighter idea of the early 1930s, and proposed that the long-range bombers carry their protective fighters right along with them.
The McDonnell Aircraft Corporation of St Louis was the only company to respond to the proposal. McDonnell proposed a small fighter aircraft to be carried partially inside a parent B-29, B-36, or B-35 heavy bomber. However, the AAF rejected this plan in January of 1945, concluding that the fighter would have to be carried ENTIRELY inside the B-35 or B-36.
On March 19, 1945, McDonnell submitted a revised proposal--a plan for a tiny aircraft with an egg-shaped fuselage, a triple vertical tail, a tailplane with pronounced anhedral, and vertically-folding swept-back wings. The engine was to be a 3000 lb.st. Westinghouse J34-WE-7 axial-flow turbojet with a nose intake and a straight-through exhaust. The aircraft was to be fitted with a pressurized cockpit and an ejector seat. Armament was to be four 0.50-cal machine guns in the forward fuselage sides. It would be launched and recovered from a trapeze-like structure which would be extended from its parent aircraft.
The USAAF liked the McDonnell proposal, and on October 9, 1945 they ordered two prototypes (plus one static test article) under the designation XP-85. As a parallel development, the USAAF specified that the 24th and subsequent B-36s to be accepted by the service would be capable of carrying one P-85 in addition to the usual bomb load. It was even planned that some B-36s would be modified so that they could carry THREE P-85 fighters and no bomb load.
Conditional upon the results of flight trials with the XP-85, the AAF had intended to order an initial batch of 30 production examples, but before the completion of the first prototype this plan was shelved in favor of a more cautious approach in which only the two experimental aircraft would be acquired. If flight tests were favorable, more could be ordered later.
Since the XP-85 was to be launched and recovered from a retractable trapeze underneath its parent bomber, no conventional landing gear was fitted. A retractable hook was fitted to the fuselage in front of the cockpit. During recovery, the XP-85 would approach its parent bomber from underneath, and the hook would gently engage the trapeze. Once securely attached, the aircraft would be pulled up into the belly of the bomber. If an emergency landing were necessary, the aircraft was provided with a retractable steel skid underneath the fuselage, and the wingtips were protected by steel runners.
Since no B-36 could be spared as yet for the project, a Bell-Atlanta-built Boeing B-29B-65-BA (Ser No 44-84111) was specially modified for use as the mothership in the initial testing. Redesignated EB-29B, it was fitted with a special launch-and-recovery trapeze that would be used for the first test flights of the XP-85. A few test flights were made with the XP-85, but the recovery operation proved to be much more difficult than expected, forcing several emergency landings using the retractable steel skid. The Air Force reluctantly concluded that since the recovery operation was so difficult a job for even experienced test pilots, it would probably be far beyond the capabilities of the average squadron pilot. In addition, it was projected that the performance of the XF-85 would likely be inferior to that of foreign interceptors that would soon enter service. Furthermore, a budget crunch in the autumn of 1949 led to a severe shortage of funds for developmental projects. Consequently, the Air Force terminated the XF-85 program on October 24, 1949.
I have never found any references which state that the XP-85 ever made any flights from a B-36 mothership.
Although the XF-85 Goblin was ultimately unsuccessful, it did provide some valuable data that was of use in the 1950s when the Republic RF-84F Thunderflash reconnaissance aircraft was adapted for launch and recovery beneath a B-36 bomber.