When to Send Your Telescope Aloft

The optical system

Perkin-Elmer initially proposed a refracting telescope using a 12-in doublet objective, an enlarging system, and photocell guidance, but Baker and others rejected the large lens as susceptible to mechanical distortion, temperature effects, and intrinsic astigmatism, all of which would reduce definition. ³⁸ Since mid-1954, Baker had been designing and testing quartz reflecting optical designs for the Geophysics Research Directorate in flights from Minneapolis, and the results were evidently highly promising. In addition, as Baker reported to Schwarzschild in June 1954, he had been developing a prototype quartz reflecting Cassegrainian system at Harvard that had an effective focal length of 400 in. He was also designing an optical enlarging system to provide an overall focal length of 200 ft, but he worried that it would be susceptible to vibration. He was waiting to speak to Harold Edgerton at MIT (Massachusetts Institute of Technology) to get advice on the mechanical design. ³⁹

Baker’s advice, in tune with other optical experts Schwarzschild consulted, like astronomer Robert Leighton at Caltech, led to a firm decision to have Perkin-Elmer construct a primary 12-in quartz optical mirror working as an F/8 Newtonian configuration to feed an optical enlarging system that increased it to F/200 and feed a 35-mm motion picture camera (Figure 1). This design provided a plate scale similar to Mount Wilson’s 60-ft solar tower, which Leighton had been using to explore high-resolution solar and planetary cinematic imaging, and was by then coming to some very tentative conclusions that differed with Richardson and Schwarzschild’s turbulence interpretation, preferring a convective granulation structure. ⁴⁰

OPEN IN VIEWER

Figure 1.

Stratoscope I optical telescope assembly. The 12-in mirror at the bottom of the tube fed an optical enlarging system at the Newtonian focus that produced high-resolution images recorded by the camera at the bottom of the tube.

Overcoming technical obstacles and the first flight season

By the mid-1950s, published capabilities for autonomous stabilization systems carried by balloons were not acceptable. An Air Force Geophysics Research Directorate team working with the “Hi-Altitude Instrument Company” of Denver, Colorado could point about 20 lb of payload to ±15 arc minutes, with a promise of reducing that to ±5 arc minutes. ⁴¹ This was nowhere in the range of Schwarzschild’s needs, which were sub-arc-second stabilization. Schwarzschild had indirect contact with that group through J. H. Rush at the High Altitude Observatory (HAO) of the University of Colorado and found that their published specifications were not what they could really do. The Denver group had in fact achieved much higher positioning and tracking for stationary ground-based telescopes at HAO using “highly refined electronic servo-guiders,” but he doubted that it was feasible from a balloon. The best results may come, he advised, by creating a balloon stabilizer that can damp its oscillations, and then take lots and lots of photographs at or near the stationary points of the oscillations. However, Rush later admitted that the Denver group had been refining a servo pointer for balloons, and that it was “a new Air Force device, and consequently should not be taken up by any other group without Air Force approval. I assume, however, that that goes without saying.” ⁴²

This was somewhat comforting news, which kept Schwarzschild hopeful that the means would be found. But throughout the rest of 1956 and well into 1957, Schwarzschild’s continued contact with Audoin Dollfus, Malcolm Ross, and Johns Hopkins physics professor John Strong, constantly raised the spectre of piloted flights. In March, Schwarzschild confessed to Dollfus: “The total number of technical complications of such an unmanned instrument appear very large indeed and I am worried and excited whether our instrument will work.” ⁴³ In April, even after Strong described an innovative design for stabilizing a telescope during a piloted flight, Schwarzschild remained sceptical, worrying about the “guiding difficulties which a manned gondola by the unavoidable motions of the personnel poses.” ⁴⁴

By the summer, however, Russ Nidey and others at the Research Services Laboratory of the University of Colorado had successfully adapted their servo-controlled biaxial pointing control system for Aerobee rockets to the needs of Stratoscope. Two sets of photoelectric diode pairs were placed at the foci: one for rough positioning and the second pair for fine positioning. These diode pairs communicated with servo motor-driven magnetic clutches and with a counter-rotating flywheel for corrections in azimuth, telling them which way to aim the telescope based upon the amount of sunlight they were sensing. ⁴⁵

By then, with some US$50,000 in total funding by ONR and Princeton, as well as the possibility of additional funding from the NSF, General Mills was contracted to supply the balloon and overall flight operations at their Flight Center near New Brighton, Minnesota. The 12-in reflecting telescope was nearing completion at Perkin-Elmer, and the pointing controls were being tested in the summer of 1957.

Schwarzschild had recruited one of his graduate students, John B. Rogerson, to the project. Rogerson was a 1954 Princeton PhD in astrophysics, who then spent about a year at Mount Wilson as a Carnegie Fellow, returning to Princeton in 1956 to work on Stratoscope, a decision that, Schwarzschild later recalled, “took fantastic courage.” ⁴⁶ Rogerson attended to the daily tasks, but both of them visited the contractors, including Perkin-Elmer in Norwalk, Connecticut, Russ Nidey and his team in Boulder, and the General Mills team in Minneapolis, to perform on-site testing and evaluation of their methods. One of the most difficult challenges in ballooning is a safe launch. The huge partially inflated gasbag, hundreds of feet long, needed to be rolled out on the General Mills airstrip and then partially filled with a controlled bubble of helium, causing the top of the balloon to rise off the ground. The gondola carrying the telescope, mounted on a flatbed truck, would be driven to a point downwind and attached to a long set of cables that connected to a recovery parachute and thence to the lower rim of the balloon. As the balloon slowly rose lifting the cables and parachute, the truck would then move downwind matching its speed to the wind. With all segments in equilibrium, the truck would then release the gondola, and if everything was correctly choreographed and no wind gusts came up, the balloon would be up and away in moments. Scenarios like these needed rehearsal, and though the General Mills team was well experienced with static payloads, the dynamic complexity of Stratoscope was a challenge.

Ready to fly by fall 1957, Schwarzschild and Rogerson packed their instrumentation in a trailer in Princeton and drove to Minneapolis to conduct final tests and integration, which took place in a large tent near the General Mills launch site (Figure 2). Members of the University of Minnesota physics department, especially Ed Ney, helped with final testing and with general operations. A test launch of the gondola in August to 82,500 ft confirmed that the system worked just fine, so the next step was to set the telescope within the open frame gondola. As Schwarzschild reported to ONR informally, “The test flight went off last Thursday, hair-raisingly jerky launching, soft landing in swampy ground, test results wonderful beyond expectation. Cross your fingers for full flight in 3 weeks.” ⁴⁷ Stratoscope was ready.

OPEN IN VIEWER

Figure 2.

Assembled payload at the launch site, Stratoscope I, General Mills, 1957.

The Operations Flight Plan prepared in September by General Mills called for launch near sunrise. The launch sequence started at 1:00 a.m., inflation by 6:00 a.m., an all-important break for coffee, and then launch at 7:00 a.m. The Plan called for 100 minutes of ascension, after which, with stabilization achieved, the telescope would come alive, seek out, and set on the sun and make recordings for an expected 3.5 hours.

On 25 September 1958, Stratoscope was launched just after 6:00 a.m. and reached its equilibrium altitude at 82,500 ft, floating there for almost 5 hours. The gondola was then released from the balloon, floating to earth under the parachute and landing some 150 miles east near Athens, Wisconsin, with little damage. The camera took an exposure a second continuously and there was so little damage that a subsequent flight took place on 17 October. After retrieval and inspection of the film back at Princeton, some 16,000 frames on 2000 ft of film yielded “about 10 frames [with] superior resolution for which we had hoped.” ⁴⁸ They observed solar granules clearly; some were as small as 180 miles in diameter. The granules also had very complicated geometries like irregular polygons with the dark intervening areas completely connective. The best Stratoscope images had resolutions of the order of one-third second of arc, far better than the average ground-based limits, and equal to the very best photographs ever taken from the ground.

In late October, in the wake of Sputnik, ONR held what it claimed was the first press conference in its history. “Press Conference on Operation STRATOSCOPE” was the title, featuring both Spitzer and Schwarzschild taking questions from the press (Figure 3). In some 20 pages of transcript, the press seemed more interested in Spitzer’s speculations about the future of astronomy in space than the results of the first flights. When finally asked about his results, Schwarzschild claimed that the new images thus far were,

quite bewildering and we do not understand them, and that, I think, in a way is the most positive sign that we have made a real scientific step. Isn’t that right? When one encounters something that one does not expect and does not understand, one knows that one has made something new.

OPEN IN VIEWER

Figure 3.

Martin Schwarzschild and Lyman Spitzer, 1958 ONR Press Conference, discussing Stratoscope concepts.

Indeed, they found that turbulence was more intense than expected but there was also more well-defined fine structure indicative of convection: “the hot out going elements go up and drop down on the edges.” ⁴⁹ They also did not find what the Richardson–Schwarzschild observations at Mount Wilson had predicted, tiny bright granules of the order a few hundred kilometres in extent. This lacunae had in fact been glimpsed by Leighton, using his cinematic techniques, and Schwarzschild ultimately accepted that fact, having for some time adorned his office with Leighton’s images puzzling over them as possibly due to uneven terrestrial atmospheric “seeing.” ⁵⁰

If there was a message in the press conference, it was Spitzer’s call for a telescope in space. Stratoscope was a “first step in this direction . . . We regard the balloon telescope as a stepping stone to the satellite observatory,” a large telescope in a 90-minute orbit. The next step, already in Spitzer’s mind, was a 36-in balloon-borne telescope that would outperform the 200-in from the ground in resolution. ⁵¹ Spitzer’s claim energized the press, who started asking many pointed questions regarding costs and impact to science and to national security. “If the race in intercontinental ballistic missiles continues,” Spitzer responded, “in a matter of years it will certainly be possible to get a rather large rocket without too great expense to launch a satellite orbit, and at that point we’d like to be ready for it.” When the questions moved on to balloon float time and satellite lifetime, a critical question for reconnaissance interests, Spitzer indicated that up to 8 hours might just be enough to examine the changing magnetic field activity in the granules. It could be done on multiple flights, he added, but satellite lifetimes of days, months, and years were preferable. Spitzer then pointed out that getting at the influence of magnetic fields on plasma turbulence is most important. And when asked why this was important and did it have practical application, he replied, provocatively, yes, in the field of controlled nuclear fusion. In “controlled nuclear reactors . . . one of the problems is how a gas holds together in a strong magnetic field.” This prompted questions about Spitzer’s involvement in Project Sherwood, which he dodged, but when pressed finally blurted out: “the subject of controlled thermo-nuclear reactors can be called, if you like, applied astrophysics.” ⁵²

Whatever ONR’s intentions were for the press conference, it was clear that Spitzer inspired the reporters. The next day the New York Times proclaimed that “huge ‘satellite observatories’ would ‘circle the earth in space for decades’.” And that the first step, Stratoscope, “has yielded data of possible use in harnessing hydrogen-bomb power for peaceful uses.” ⁵³

Modifications for the second season

When a traditional ground-based observatory develops a new piece of equipment, if it works to specification, usually it is used to death before it is cannibalized, modified, or placed in a museum. Not so with those who fly instruments on balloons or rockets. Their lives are one of constant modification, constant tweaking, and hopefully constant improvement. ⁵⁴ Once he found his stride, Van Allen was more interested in gathering data. He was one of the few who flew the same type of detector repeatedly to map out the electrical currents in the high atmosphere. Schwarzschild’s Stratoscope team fell somewhere in the middle. Schwarzschild wanted the best data; his team wanted to build the best instrument. They all agreed on one thing: future flights had to have a way to remotely control the telescope from the ground. Stratoscope was autonomous – once at altitude sun sensors acquired the solar image and centred it in the field, and kept it locked onto the solar disc throughout the observational phase of the flight. But there was no way to select single objects, like sunspots or filaments, and also there was no way to account for focal changes, especially in the position of the enlarging lens. This was the reason later given for so few images being acceptable in the first season. ⁵⁵ Thus, the puzzling results they obtained from those few clear images needed verification. Where were the tiny bright granules that Schwarzschild and Richardson had predicted?

Schwarzschild’s team was growing. After the first season of flights, recent PhDs J. D. R. Bahng from the University of Wisconsin and Robert E. Danielson, from the cosmic ray group at Minnesota, arrived as Research Associates and were assigned to analyse the data and assist in making modifications for the second season. Bahng, in fact, was immediately assigned the task of analysing the few useful images from the first season, and published the first results as a note in the Astrophysical Journal solely under his name. ⁵⁶ This prompted the Astrophysical Journal editor, Chandrasekhar, to caution Schwarzschild that “generosity of this kind is misplaced and contributes only to a falsification of history . . .” Nevertheless, Schwarzschild explained that he was only trying to avoid, in his mind, “. . . serious psychological set ups within our group . . .” and, moreover, that they agreed that scientific results would be by individuals in the group “so that each of us had an active personal incentive in the work, and in the scientific responsibility.” ⁵⁷ Clearly, Schwarzschild was finding the administration and management of people as complex, maybe even more so, than the telescope.

Above all, they all, including graduate students and technicians, revisited the issues that were raised from the first flights, assessing how a television monitoring system would aid tracking and object selection for later flights of the 12-in telescope. ⁵⁸ Their department secretary, as it turned out, was the granddaughter of RCA’s (Radio Corporation of America) Vladimir Zworykin, the developer of the iconoscope, the first practical television camera. ⁵⁹ Adding the television control system would allow observers to image sunspots and examine specific portions of the solar disc and limb, which greatly added to the versatility of the system and its usefulness to monitoring short time period changes. Getting familiar with various forms of electronic image formation was also another step into space. ⁶⁰

The team also re-examined the structure of the gondola and other components to find ways to reduce unwanted vibrations and a slight “pendulum-effect” that plagued the first season’s flights. In retrospect, Schwarzschild feels that his group was both very naive and very lucky in their first season out. Thinking that the whole system, composed of parts from many different contractors as well as from their own small shop, could be integrated in a few short weeks from a tent-laboratory, and then work quickly and properly in a manner never before attempted, seemed quite rash to Schwarzschild many years later. That it worked at all was impressive, although to be sure, the stability of the system did not have to be perfect because the white light exposures of the sun were only 1/1000th of a second duration “and a mild waving about on the sun was entirely tolerable and we never went beyond that.” ⁶¹

Their proposal to ONR in November 1957 for a second season of flights explicitly asked for modifications to the guidance and stabilization systems, the new TV equipment, and adequate laboratory and test facilities at Princeton. While unpacking back on campus, Rogerson complained about the lack of suitable facilities for processing and analysing the photographs. There was in fact no darkroom in the department, and he needed electronic imaging expertise, asking that one of Zworykin’s RCA staff, William A. Miller, be engaged. ⁶² The television system was deemed “essential, of course, for a satellite observatory.” ⁶³ It would take 6 months for the Princeton “Committee of Project Research and Inventions,” another administrative hurdle Schwarzschild was now keenly familiar with, to be able to report that ONR finally approved its subcontract with Zworykin’s group at RCA for both Stratoscope TV systems, applying RCA’s vidicon image tubes to the task.

The second season of flights in late summer 1959 from Lake Elmo Airport in Minnesota yielded far more usable images. Thanks to the TV control, they now were able to home in on particular features like sunspots. The first flight in July was very frustrating due to vibration problems, but hasty corrections were made for a second flight on 17 August, which was fully successful. Preliminary analysis mainly looking at short timescale variations in granules and in filaments promised much more information to come. ⁶⁴ Sky and Telescope and other news outlets quickly responded featuring images of sunspots, faculae, and granules of “unsurpassed quality.” A third flight in September was even more successful with improved TV control. Two series of time sequences, one on a large sunspot and another on granulation, proved especially valuable. ⁶⁵

The following April, Danielson reported on the structure of sunspot penumbrae based upon dozens of images with at least 1/3 arcsecond resolution, which resolved those wavy areas surrounding the darkest portion of the spot “into a complex array of predominantly radial filaments.” Finding they changed constantly “in a few minutes” but could remain intact for more than 40 minutes, longer than granulation lifetimes, they were best explained as elongated eddies or “convective rolls with the axis of the roll along the magnetic field” (Figure 4). ⁶⁶ The filaments therefore rose above the photosphere and were chromospheric structures above the sunspot. An International Astronomical Union symposium on “Aerodynamic phenomena in stellar atmospheres” took note of this finding as strong evidence of convection. Once again, Richardson and Schwarzschild’s predicted tiny bright granules were not seen. Bray and Loughhead noted this failure explicitly some years later. They regarded it as confirmation of the ground-based observations by Leighton and others, and so was “an ironic twist of fate,” which Schwarzschild “generously admitted.” ⁶⁷ Still, Schwarzschild had what he wanted: direct evidence to test his ever-changing models of stellar structure and energy transport. He also knew that Davidson’s conclusions strongly supported theoretical studies of convective rolls by Chandrasekhar. ⁶⁸

OPEN IN VIEWER

Figure 4.

Sunspot #14404 captured on the 4 September 1959 flight, one of a series of images showing the evolution of penumbral structure and granulation over a 90-minute span.

In late August, David Dewhirst of Cambridge announced lightheartedly that he was “now retiring” from solar granulation studies due to Schwarzschild’s “excellent work.” What Dewhirst probably did not know was that Schwarzschild had already decided to give over the entire Stratoscope gondola and pointing system to Gordon Newkirk of the HAO to fly an F/100 coronagraph with an external occultation disc, dubbed Project Coronascope. ⁶⁹ Schwarzschild well knew that Newkirk had been trying to fly sky brightness photometers with balloons and in fact had secured support from Malcolm Ross for piloted flights. Newkirk felt compelled to work with Ross. After all, he was “a young punk just out of the Army” and he “wasn’t Martin Schwarzschild . . . I didn’t have the reputation to bring in the money for an automatically guided gondola.” Given ONR’s priorities, piloted flights were more accessible for the scientist, and the piloted flights, he believed, were a “qualified success” not so much in scientific return, but in gaining experience for how to do it better next time. Indeed, vibration and movement of the pilot remained a problem through these years. Ross had made careful measurements of balloon motion but Jack Rogerson advised Schwarzschild that Ross’ study “doesn’t cover motions as small as we are interested in.” Newkirk, like Schwarzschild, also needed greater altitude. ⁷⁰

In late September 1959, Newkirk was in fact in the midst of one of his balloon ascents from the Stratobowl, near Rapid City, South Dakota, at the same time that Schwarzschild was flying from Lake Elmo. They had been in close contact during that time, and Newkirk had sent two of his students, Jack Eddy and Robert Cooper, to assist Schwarzschild and gain needed experience. He recalls getting a hasty call from Walt Roberts back in Colorado reporting that Schwarzschild offered them “unbidden” his gondola after the conclusion of this latest flight. ⁷¹ Stratoscope had done its job; Schwarzschild had the data he needed that “made me a little more confident on how to proceed” ⁷² and looked forward to the next step. Newkirk was elated.

Strengthening the infrastructure

Schwarzschild did not take these challenges lightly. His experience with General Mills’ behaviour prompted caution. At the beginning of the first launch season, General Mills sent their best people to get things rolling, but once they felt they were up to speed they brought in less competent people and things started going wrong. As a result, the second season was very troublesome logistically. Thus, Schwarzschild advised Newkirk and his own team to write contracts specifying the names of people who would be involved from those companies. ⁷⁶ Problems with the overall General Mills polyethylene balloon also stimulated Schwarzschild to look elsewhere. ONR had recommended that Schwarzschild engage the Vitro Corporation, a defence contractor that had overseen the design and construction of military test ranges with systems for guidance and control of missiles. ⁷⁷ In fact, the complexities of the larger balloon system mirrored those of the technical logistics Vitro had been dealing with, managing large-scale balloon activities for ONR for some years.

Under contract to Vitro, the G. T. Schjeldahl Company of Northfield, Minnesota developed and tested a new stronger Mylar and Dacron scrim balloon. Schjeldahl was at the time a leader in fabricating reinforced cells, contracted by ONR, and constructed both the main Stratoscope II balloon and appendages, as well as a smaller launch balloon that would allow for the gentlest possible takeoff. While their new design promised better performance and reliability, it also required rethinking how the balloons should be launched.

Spitzer took the lead planning for Stratoscope II. Almost immediately after the first season, he started contacting possible contractors, and in January 1958, sent out bids for the optical telescope assembly to Perkin-Elmer, J. W. Fecker, and Warner & Swasey. Stratoscope I was a Newtonian, but the much larger 36-in required mechanical compactness. The team chose to design around a Gregorian f/4 36-in primary mirror that would feed a complex optical transfer and guide star pickoff system sitting at right angles to the main optical axis (Figure 5). When Warner & Swasey bowed out, Schwarzschild and Rogerson visited Perkin-Elmer and Fecker, and after Spitzer created a numerical grading system to rate their proposals, the Princeton team met several times in May and June to deliberate. Although the two companies scored nearly equal, Fecker was deemed weaker in organization and optics, whereas Perkin-Elmer was weaker in the “guidance field.” They chose Perkin-Elmer because they knew there were other good sources for the latter, and of course, the company had “invaluable experience” in the field, building the first Stratoscope and a wide range of missile and satellite tracking cameras. ⁷⁸ It would take many months for the contract to be refined, approved, and signed, and there was still much to learn from the second season of Stratoscope flights that would factor into the next step.

OPEN IN VIEWER

Figure 5.

Stratoscope II optical system showing the gross and fine photoelectric searching and setting systems.

New patrons and priorities

At the ONR Press Conference, reporters asked Shirleigh Silverman, Director for the Physical Sciences, if there were any problems with financial support, to which he quipped: not yet “but we can certainly use more money. We always can.” ⁷⁹ NSF astronomy programme manager Geoffrey Keller heard this and contacted ONR to see how they might cooperate. Within several weeks, Spitzer and Schwarzschild were in Washington briefing the two agencies, speaking a bit about improvements to Stratoscope, but emphasizing their nascent plans for a new and enlarged “Proposed Program of High-Altitude Astronomy.” ⁸⁰

During these years, vocal advocates for piloted ballooning like Ross were calling for their use in all sorts of areas, including preparing astronauts for space travel. There was also increased interest from the military and an ever-widening circle of academic scientists. These trends prompted Walter Orr Roberts and Gordon Newkirk of the HAO to call for a conference under the aegis of ONR’s Project Strato-Lab to bring all parties together to establish priorities and boundaries. Participants would explore the many uses of balloons for astronomy, as well as for propulsion, fluid mechanics, upper atmospheric physics, communications, radiation, and space medicine. The Strato-Lab flights thus far had explored a wide range of topics, including atmospheric composition and energy transmission, the impact of high altitude on insect life, aeromedical studies, reconnaissance techniques with photographic and television equipment, and studies of stellar scintillation and sky brightness. ⁸¹ The contract to coordinate the conference went to Vitro and by September the goal was to debate the “strengths and weaknesses of a sealed gondola balloon laboratory” flying to 100,000 ft. What could be done better by humans, or by machines? ⁸²

The debate continued into the early 1960s as NASA started showing interest. At the time, military and government agencies were still competing for cognizance over the newly accessible territory of space. NASA did not respond to an October 1958 “Proposed Study Program for a Satellite Telescope” from Princeton’s Research Committee Executive Director R. J. Woodrow. Spitzer had prepared it and Woodrow sent it to six military and civilian organizations including ONR, NSF, the National Research Council, and NASA: “In view of our lack of definitive information as to the most appropriate agency in the government to whom such a proposal should be sent.” The proposal was to study the feasibility of building a spectroscopic telescope “of moderate aperture, in the range of 6″ to 24″” based upon the experience they were now gaining with Princeton’s “Flying Telescope project.” It made the strong point that “Many of the problems faced in this program, especially those of communication, command and data transmission, as well as temperature control and reliability, are similar to the problems that would be faced in a satellite telescope.” The AFCRC responded positively very quickly, in what was something of a preemptive move, laying claim to the territory.

By summer 1959, however, a new Chief of Astronomy Programmes arrived at NASA within the still fluid area known as the “Office of Space Flight Development.” ⁸³ Astronomer Nancy Grace Roman’s first challenge was to raise interest in NASA among astronomers, as well as to inform astronomers of ways to work with the new agency. One of her first targets was Lyman Spitzer. After a visit to Princeton, Spitzer learned that NASA would indeed be interested in both the “Flying Telescope” and the satellite, and “can well contribute of the order of $100,000 a year” especially “for the acquisition and pointing systems for use in a satellite.” NASA could not support “balloon research” directly, she advised, but evidently it wanted to get involved. She promised to follow-up with her bosses. ⁸⁴

Roman’s inquiries helped prompt astronomers to hold a “Conference on Astronomical Observations from Above the Earth’s Atmosphere,” sponsored by NSF and the American Astronomical Society (AAS) at the Case Institute of Technology, as a special symposium during the AAS annual meeting in December 1959. Six speakers covered a broad swath of possible applications including solar, stellar and interstellar surveys, celestial mechanics, and opening up the X-ray universe. Roman led off representing NASA’s vision of “Vehicles and Plans” followed by Lyman Spitzer, who provided a “complete and detailed design of a space telescope.” ⁸⁵

Spitzer’s contribution was in effect the highly detailed feasibility study that the AFCRC had funded. Covering a wide array of technical issues to be considered in the design of a “large telescope satellite for stellar observations,” he envisioned a 1000-kg structure some 2 m on a side, with two possible orbits: 800 or 36,000 km. A 24-in diameter all-reflecting system would feed a spectrograph equipped with a scanning high-resolution photoelectric detector. Although he did not say it here, some of the design elements described here mimicked those for Stratoscope II, save for the need for far tighter positioning, thermal controls, and power. ⁸⁶

By the time of the conference, NASA had joined ONR and NSF to fund Stratoscope at first as a junior partner. It would take about another year, once Spitzer concluded his AFCRC contract, to bring NASA fully on board. ⁸⁷ However, by then it was becoming clear that adding another funding organization to the mix, while critical to cover the costs, might make even worse what was becoming a public relations tussle. Differences in opinion existed between Princeton, the main contractors, such as Vitro, and the funders, Navy, Air Force, and NSF. It was all over language and style. Specifically, NSF objected that Navy press releases were too military in implication, and that Schwarzschild would be done a disservice if his efforts were connected to phrases like “missile weapon system,” “war plans officer,” and “war plans section.” NSF also objected that the scientific goals of the mission had been buried. ONR, coming to its collective senses, agreed with NSF at first. What was needed was a coordinator for news releases, and ONR assumed the task, as all public notices had to be cleared for classified information. ⁸⁸ On the other hand, nobody objected when a Corning news release highlighted that Stratoscope technology was a “technical forerunner” of satellite telescopes. This was possibly the least controversial aspect of the image all parties wished to create in what was, evidently, an early example of the social and political construction of space technology. ⁸⁹

Stratoscope II flight plans slipped almost immediately. In January 1961, the first fully operational flight had slipped to November. Slippage was due to several factors. According to Harold Glaser and Charles Steerman, ONR project officers, everyone had been worried about the accuracy of the guidance system and the readiness of the imaging system: could clear photographs be taken of faint celestial sources yet? ⁹⁰ There were also the questions of where to launch, how to launch, and even what to launch. Answering them would take two more years.

Design changes

By spring 1961, the weight of the payload increased: the telescope now weighed in at 2300 kg alone, with another 1000 kg or more for the gondola. And so once again, the balloon system had to be enlarged. The cost of meteorological analysis for the programme alone for five expected flights rose to over US$45,000, but this was handled by approaching the Weather Bureau for support, which was small in what was rapidly becoming a million dollar a year project. Just as William Blair a half-century earlier had made available his resources to the Smithsonian’s Charles Greeley Abbot, to gain needed experience in the logistics of ballooning, his descendants in the modern Weather Bureau were brought in to Stratoscope II to hone their methods of communication and control of the new balloon systems. ⁹¹ ONR had been worried that “no significant amount of thinking has been done by either Princeton or Vitro on operational analysis of the intended flights.” Thus, F. B. Isakson of ONR suggested that factoring in weather awareness via “war gaming techniques applied to the intended ‘live’ operational flights can shed much light on the present concepts of communications, control and logistics.” ⁹²

ONR’s continued advocacy for piloted flights led the HAO astronomers to seek alternatives. Emboldened by NSF’s support for his National Center for Atmospheric Research (NCAR) in Boulder, Roberts started campaigning for “some sort of a national balloon flight facility.” In October 1960, he told Schwarzschild that both the Air Force and the Navy were committed to the idea, and even NASA was showing interest. Offering his “Thanks for all your help and moral support!” Roberts asked that it extend to this latest plan. ⁹³ Schwarzschild was sympathetic, but based upon his experience with companies like Raven and Schjeldahl, he was not convinced that they were “capable of thinking through, in advance, technical complications which they are not used to.” Specialists from Vitro made all the difference to keep the subcontractors under control and responsive. Because of this, Schwarzschild was convinced that balloon firms have to be managed and supervised by “really good outside engineers.” ⁹⁴ NCAR, he advised, would do better to build up a competent engineering balloon group at NCAR rather than supporting a dedicated launch facility. Roberts heeded this advice, but did both. After yet another meeting of government, university and military parties convened at the University Corporation for Atmospheric Research (UCAR) in December 1960, Roberts was able to first establish a balloon-engineering group, and then create a standing committee on the scientific uses of balloons to advise it on the need for a national facility.

A new national flight facility

Roberts, Newkirk, and Schwarzschild were among other scientists on the new NCAR standing committee. One of its first actions was to recommend a national balloon facility. The Weather Bureau had identified a specific region in western Arkansas and eastern Texas as prime for balloon flight drift patterns. Initial tests in Arkansas were promising, so NCAR asked the Weather Bureau to do a full meteorological analysis of the surrounding area, as well as check with the Federal Aviation Administration (FAA) about possible interference with known flight patterns. The FAA rejected Arkansas but approved a site in northeastern Texas near Palestine in March 1962. By then, two Stratoscope II test flights from Goodfellow Air Force Base, San Angelo, Texas were made. Even though they both failed, in both cases, the problems were quickly found and corrected. ⁹⁵ Thus, after negotiations between the city of Palestine and the NSF, and several months preparing the site, the first successful test flight took place on 12 December 1962, the third test of Stratoscope II. ⁹⁶

More slippages and switch to spectroscopy

By then, however, the continued launch slippages and Glaser and Steerman’s growing concerns over guiding accuracy heightened debate over the best instrumentation for the initial fully instrumented flights. There had been another conference in October 1962 that outlined the various balloon stabilization efforts. It reviewed some six different systems, including piloted and automated options for a variety of astronomical applications. Stratoscope II was singled out as the most sophisticated and the most demanding of them all, “designed for an accuracy of .02 second RMS error over a period of 5 to 10 minutes of time.” This design goal, however, was far from being met. ⁹⁷ Schwarzschild by now was growing very frustrated. The July flight that had slipped to October was slipped again to 1963. After the second slip, reeling from the October conference, and embarrassed by “friendly publicity” from the New York Times, Schwarzschild confided in Chandrasekhar that he was living on “borrowed credit” and that at least another half-year was needed to “even get it off the ground, much less be certain of real scientific results.” ⁹⁸

In addition to the guiding problems, through 1962, optical quality assessments of the main mirror indicated that its surface was still plagued by light scattering and from distortion due to thermal changes. The Perkin-Elmer contract specified that it acquire two fused quartz primary mirrors and by the fall of 1961, both were nominally ready to permit “all-up” testing of the optical system. One of them was committed to the task, while the second one continued to be refined. Through 1962, the optical imperfections in the committed primary prompted Perkin-Elmer to refine its methods on the second mirror, which it was still testing. ⁹⁹ Until this mirror was declared worthy, diffraction-limited imaging was not guaranteed.

NASA, now in the game, argued that spectroscopic observations demanded less stringent guidance as well as optical quality, and that Stratoscope should, in the first instance, carry a spectrophotometer using newly available germanium bolometers as infrared detectors. In the interim, with the end of the AFCRC contract, NASA had in fact become the primary backer of Stratoscope, contributing some 40 percent of the cost, compared to 30 percent each for NSF and ONR. Since there was also some question about the stability of the mirror after launch, spectroscopy became the new immediate goal, even though imaging was always the ultimate goal. ¹⁰⁰

By this time, it was clear that adding NASA very much changed the political dynamics of the enterprise. NASA’s demand was part of an emerging policy developed in its Space Sciences Steering Committee that pushed for planetary studies favouring infrared spectrophotometry studies of planetary atmospheres and reflectance photometry of planetary surfaces. Balloon-borne experiments were a part of this scenario, itself derived largely from an earlier National Academy of Sciences Space Science Board study by its “Ad Hoc Panel on Planetary Atmospheres,” which had held deliberations in 1960 and 1961. ¹⁰¹ The SSB study acknowledged Schwarzschild’s primary motive for Stratoscope II as high-resolution photography. But it added that

. . . its existence also suggests that in the near future we may be able to obtain from the Earth infrared spectra of the nearer planets with better than 0.1 [microns] resolution . . . Such spectra would give a great deal more quantitative scientific information about the composition and temperature structure of planetary atmospheres. ¹⁰²

The far infrared region was also becoming more interesting, especially after thousands of objects were found in an all-sky survey from Mount Wilson. This change was noted in a later SSB study in 1965 at Woods Hole, which opted to include the infrared in its recommendations for non-solar and non-planetary optical astronomy. ¹⁰³

Carl Sagan, one of the authors of the 1961 Ad Hoc Panel study, later became one of the primary players in developing an infrared spectrophotometer for Stratoscope II. Both the SSB study and the NASA committee were heavily influenced by scientists from the Jet Propulsion Laboratory (JPL) and by Sagan, then at Berkeley. JPL was becoming established as a national centre to develop planetary programmes for NASA and its scientists advocated all forms of observations in support of eventual space missions. One of the most critical was improved knowledge of planetary atmospheres.

The decision to switch to infrared spectrophotometry was made just prior to a 10 September 1962, Stratoscope II press briefing. In fact, the timing was so close that not all players were fully informed of the change. Preliminary press releases by ONR, NSF, Vitro, and RCA all failed to mention it. Only the new player, the University of California at Berkeley, discussed the shift to spectroscopy, with an instrument to be built by Harold Weaver and Carl Sagan. Weaver, a radio astronomer, was identified as “faculty investigator,” while Sagan, then Miller Research Fellow at Berkeley and a vocal force in the SSB study, was project director for the construction and operation of the US$336,000 infrared spectrophotometer, funded by NASA. ¹⁰⁴

Schwarzschild and Danielson later pointed out that the shift was due to three factors. In agreement with Glaser and the SSB study, there was “intrinsic scientific value” to the infrared observations. The overall system had to be tested with a less demanding payload, “its simpler infrared configuration,” and more time was needed to correct “remaining engineering problems for the photographic flight.” ¹⁰⁵ Mars was the target for a spectroscopic flight, now slated for early 1963. An infrared spectrum of Mars taken above the earth’s own atmosphere might show if water vapour existed on Mars. Canals still existed in many astronomers’ minds, and seasonal changes on Mars were still thought to be due to a water cycle. Mars was the only planet left, it was thought at the time, where life of some sort might exist in the solar system beyond Earth. Finding water vapour, and possibly even the infrared emission of organic molecules, would go a long way to solving this age-old astronomical quest. ¹⁰⁶

The infrared spectrophotometer used a fluorite prism with reflection optics by Perkin-Elmer. The detectors were three Texas Instruments germanium bolometers cooled by liquid helium in cryostats to an operational temperature of 1.7 °K. The low temperature was required to reduce thermal noise so that the instrument could detect radiation between 1 and 7 µm. Its theoretical resolution was 0.02 µm at 3 µm.

While the spectrophotometer was built and tested, balloon tests continued, but not without considerable problems. Neville J. Woolf, a Manchester PhD and postdoctoral fellow at Princeton, took over the responsibility of visiting Perkin-Elmer because they were worried about guidance problems. There were still issues with testing the primary mirrors and creating a suitable “crash pad” that would protect the payload upon landing. Schwarzschild remained very concerned about avoiding confusion at the launch site, mulling about Vitro employees somehow overstepping their responsibilities and making changes to the telescope. It was during this time that Schwarzschild darkly observed that he was living on “borrowed credit.” ¹⁰⁷

When the much-delayed test took place from Palestine, the results were far from satisfying. The long-awaited full-up test flight, on 12–14 December 1962, ended when ground commands to bring it down failed. The balloon hard landed first in Louisiana, causing large pieces of the payload to break away. The balloon ascended again and entered commercial airspace. It then had to be brought down by a Sparrow missile fired from a scrambled F4D aircraft from Boca Chica Naval Air Station. The missile detonated within the balloon and “an immediate increase in descent rate was noted . . . it appeared likely [to ONR] that a kill had been achieved.” ¹⁰⁸

The next flight met with delay after delay due to weather and to balloon problems. After 3 weeks and 15 aborts, the launch crew was more than anxious and, predictably, the press caught on. When the first successful flight took place on 1 March, the National Observer proudly claimed “Stratoscope II Beats a Jinx.” ¹⁰⁹ The night flight went well, but there had to be an emergency parachute landing in Tennessee; although the telescope was not seriously damaged, the gondola was bent a bit upon hitting the Tennessee mud.

Throughout all this, Schwarzschild remained steadfast in his determination to see the project through. Working with Danielson, Woolf, and one of his graduate students, John Gaustad, Schwarzschild had followed all aspects of the programme, conducted many site visits, and, with his wife Barbara, attended all the launches (Figure 6). But the strain was evident. Schwarzschild had already delegated more of the work to Danielson and Woolf, but in the wake of the first successful light on 1 March, when at last everything worked to plan, Schwarzschild reported to Chandrasekhar that the ballooning had taken a toll: “the unexpectedly long activity in Palestine has come closer to the limits of my energies than I had thought.” Even before they could catch their breaths, with Danielson, Woolf, and Gaustad scrutinizing the data, Schwarzschild had to secure the contractual arrangements for the following year “in a fair hurry so as not to lose the good engineers working with us.” ¹¹⁰

OPEN IN VIEWER

Figure 6.

Martin Schwarzschild standing with Stratoscope II.

The first science from Stratoscope II

Labelled a “preliminary infra-red flight,” it was set to trace the infrared spectrum of Mars from 1 to 7.5 µm, searching for water vapour and carbon dioxide bands and possibly evidence of organic matter on Mars. Instrumental problems limited the performance. The new bolometers had blind spots and uneven spectral response, so only a fraction of the planned observations were possible. ¹¹¹ Even so, spectra were obtained from 2 to 2.7 µm that recorded a carbon dioxide band and very faint hint of a water vapour band at 2.7 µm. Weaver’s initial excited reaction, as recorded by Business Week, was that “it is highly likely that some form of life exists on Mars.” ¹¹² But more detailed analysis of the extreme weakness of the band later indicated that a mere 10–40 µm of perceptible water vapour existed in the Martian atmosphere, or less than 1/1000th–4/1000th of that in the earth’s atmosphere. Mars was a dry place indeed. ¹¹³

This landmark observation was confirmed in April 1963 from Mount Wilson Observatory’s 100-in reflector. Hyron Spinrad, Guido Műnch, and Lewis D. Kaplan, of JPL, Caltech, and Mount Wilson, examined Mars using a high-dispersion spectrograph at the Coudé focus of the 100-in reflector. They revived an old technique developed by Percival Lowell to use the relative velocity of the earth and Mars to separate the Martian spectrum from the obscuring terrestrial spectrum by the Doppler effect. ¹¹⁴ The three astronomers photographed the Martian spectrum on 12/13 April, about 2 months past the February 1963 opposition when the Earth was receding from the red planet at maximum velocity (+15 km/s). They were able to identify 11 faint lines due to water vapour in the 8200-Å region after a 4.5-hour exposure. ¹¹⁵ Spinrad, Műnch, and Kaplan found 14 ± 7 µm for water vapour, in good agreement with Stratoscope’s lower end, but in “severe disagreement” with all previous measures, the most recent being that of the French balloonist/astronomer A. Dollfus’ observations of the 1.4 µm band from the Jungfraujoch, where he concluded that that band alone indicated some 200 µm of water vapour. ¹¹⁶ Not only was the amount of water vapour on Mars far less than previously proposed – the whole history of the subject was one of diminishing upper limits – but the overall atmospheric pressure on Mars was also tremendously reduced by the Mount Wilson observations, now to the range of 25 ± 15 mb, which would eventually require rethinking how a spacecraft might be able to land there.

Chronologically, the observations of Stratoscope II came first, but the Mount Wilson observations carried great weight. The priority of the ground-based observations was most likely because the Stratoscope II detection was marginal at best. While some agreement for water vapour was argued, on the other hand, Stratoscope’s observations of carbon dioxide differed from the Mount Wilson results. A conservative evaluation would be therefore that Stratoscope in fact did not actually detect water vapour bands in the infrared, but set a new low value for an upper limit on visibility of the bands. If they were there, knowing the instrumental limitations of the telescope and detectors but not definitely seeing the bands set an upper limit to how much water vapour could actually be present. This is far different from actually detecting spectral features due to water vapour, which is what Spinrad, Műnch, and Kaplan achieved. Spinrad was explicit about this in an earlier paper, concluding,

The practical limits for detection of Martian water vapor by Earth-bound, balloon, and space probe techniques indicate that spectroscopic observations from the Earth can be refined to a point where they are at least as sensitive as present infrared space experiments. ¹¹⁷

Stratoscope II’s apparent confirmation of the Mount Wilson observations of water vapour was still a most worthy result of its maiden voyage. By then, some US$5,000,000 had been spent, but this was not much for a major “space mission.” Business Week was happy to point this out, highlighting Schwarzschild’s opinion that Stratoscope is a “relatively low-cost method . . . [that] does work.” ¹¹⁸ That was what was important, not scientific credit nor priority for an observation. A few months after the flight Schwarzschild testified in Washington before the Senate Committee on Aeronautical and Space Sciences, calling out the wonderful promise and work by satellites that are best “executed in close conjunction with experiments on the ground which they never replace but which they complement in a decisive manner.” ¹¹⁹

A second flight took place on 26 November 1963, again from Palestine. The 11,000-lb payload was the heaviest yet flown and the target this time was Jupiter and bright, cool, red giant stars, another Schwarzschild interest. Stratoscope II observed 10 objects in 12 hours of stable flight. As calibration, observations of the Moon and the very bright blue star Sirius revealed no large absorption bands, indicating that the telescope was above virtually all detectable terrestrial water vapour. Stratoscope II did detect a weak carbon dioxide band at 2.7 µm, which was probably vestiges of the terrestrial atmosphere, but the payoff was detection of molecular bands in bright, late-type red stars such as Betelgeuse in Orion. The second flight carried improved Texas Instruments indium arsenide photoconductors as detectors, replacing the bolometer detectors of the first flight. The output of the detectors was stored onboard on magnetic tape. ¹²⁰ The telescope landed near Newport, Mississippi, and required a bulldozer crew to cut out a road to get to the 3.5-ton telescope and gondola. ¹²¹

Harold Glaser of ONR claimed that the second flight “was such an unqualified success that it is regarded as an important landmark in astronomy.” ¹²² Glaser’s advocacy helped Schwarzschild’s team, primarily Danielson, argue for an extension to the programme as well as increased funding, asking for an additional US$300,000 over the US$1,100,000 already requested for FY1964. The additional funding was required to retrofit Stratoscope for its original purpose: high-resolution imaging, as well as to keep together the engineering and technical staffs at Princeton, Vitro, and Perkin-Elmer. ¹²³ Finally, Schwarzschild certainly must have been happy too that the results showed strong evidence of molecular bands in the outer atmospheres of older red giants, supporting the theoretical work he was then doing with H. Härm showing that mixing took place subsequent to the helium flash in these evolved stars. ¹²⁴

Turning back to imaging, more frustration, and finally success

After two successful flights of the infrared spectrophotometer, the entire system seemed ready, politically, scientifically, logistically, and technically, for direct high-resolution imaging. Princeton’s scientific agenda was now far stronger. Everyone shared the age-old desire for high-resolution images of planetary features, both permanent (and possibly artificial) on Mars, and dynamic, as on Jupiter, and Schwarzschild certainly sought out observational evidence that might inform his stellar models. But there were now many more problems Princeton might tackle, such as Spitzer’s interests in gaseous nebulae and the interstellar medium, to look for dynamical processes that might hint at how stars are born and die. Dense inner portions of globular clusters might be resolved to yield knowledge about the types of stars found there, and how they move through those systems. And, most provocative, there were the sizes and structures of nuclei of spiral galaxies, which might finally be detectable and tell much about how galaxies form and what maintains them. During their many discussions preparing reports to ONR, NSF, and NASA, the Princeton scientists also pondered Stratoscope’s potential to provide high-resolution images of the lunar surface in service to the Apollo programme. If NASA’s Lunar Orbiter missions were held up or for some reason did not meet requirements, Stratoscope would be available to aid in lunar mapping in the search for safe landing sites. ¹²⁵ Thus, while they had lost the best Martian oppositions in the early 1960s, there was a wealth of other things to do, both pure and applied, justifying the effort.

The first photographic flight of Stratoscope II was planned for 24 November 1964. It was delayed until 5 December, when a takeoff was attempted, but this was aborted when high winds damaged the gondola at takeoff: it was “a complete launch failure.” The decision to fly was controversial. Harold Glaser criticized the decision-making process and the launch operations facility, including the site at Palestine. Glaser argued that weather forecasting is difficult there due to the proximity to the Gulf. But his primary contention was that unnamed “forces at work were in the direction of accepting marginal conditions.” ¹²⁶

The “forces” Glaser darkly contended to exist were at least in part a result of economies Schwarzschild had been insisting upon. In January 1965, back from Palestine, Schwarzschild wrote to Chandrasekhar lamenting about the “very strenuous period in Texas at the Balloon Station . . .” It all ended with a bad launch failure, which was a “severe shock to everyone.” But “What made it particularly depressing for me,” Schwarzschild admitted, “was that the failure was clearly a consequence of a couple of wrong decisions directly by myself. It takes a bit of time to recover from such a experience, but I think I have my optimism and energy entirely back by now.” ¹²⁷ Indeed, despite Schwarzschild’s apparent optimism, it was coming clear that there had been serious management problems thus far, especially with so many players, from Perkin-Elmer, Vitro, General Electric, RCA, Sylvania, and other subcontractors to the patrons, now ONR, NSF, and NASA. John Dearden, an administrative specialist brought on a year later to “keep all the loose ends tied down,” recalls that “the sponsors had experienced what they perceived to be management shortcomings in the Stratoscope program as it grew larger and more complex.” He sensed that

Martin was always driven to do things in the least expensive way possible . . . [but] . . . after we had gone through a couple of flights, he became convinced that that had been the wrong way to proceed, that in fact it would have been more productive to have optimized, almost maximized, progress when money was freely available, rather than try to operate more or less on a shoestring, to just apply resources in an optimal fashion and try to get the job done as well as it could be done, with everything you could bring to bear on it, rather than try to do it sort of like a back yard physics experiment. ¹²⁸

The telescope had survived damaged but intact, but the balloon was shredded and had to be scrapped. This, along with similar failures in other flights at Palestine, and in the light of continuing criticisms raised by Glaser and a growing concern over the management of Palestine by NCAR from Schwarzschild as well as Roberts, especially over the apparent lack of “skilled engineering talent,” led to a deep reassessment in June 1966. ¹²⁹ The sponsors formed a review committee early in the year and met in June, where both ONR and NSF asked NASA to take over responsibility for all management, design and construction, testing, and operations. Over the rest of the year, the Committee decided that Stratoscope’s basic design was acceptable, but “the system was necessarily of such complexity that modern aerospace reliability engineering was required . . .” ¹³⁰ This meant all-up testing of the entire system under flight conditions, which required a large thermal-vacuum test chamber. General Electric had chambers, but they were “booked,” so NASA’s Marshall Spaceflight Center (MSFC) in Huntsville, similarly equipped, was made available along with a team of engineers. Schwarzschild had been apprehensive about the review, but was much relieved, according to a later Report, by “the stern, but smiling, guidance” of the project manager from MSFC who reduced friction overall. The Review was complete by December, and called for continued flights of Stratoscope, but not before a far more comprehensive check list and controls were in force. ¹³¹

By then, Vitro reported that it had completed a new tandem balloon system to improve launches for very heavy payloads. There would now be a 5,250,000 ft³ main balloon as well as a new 300,000 ft³ launch balloon to carry the 7350-lb payload. ¹³² The tandem balloon design would make launches smoother and reduce the possibility that the main balloon was contaminated or torn in the process since the large balloon now did not have to be spread out on the tarmac, but could be fed out from the truck. The small balloon would lift the main balloon and payload to some 14,000 ft where the main balloon, now fully extended by the reduced atmospheric pressure, would take over.

However, later flights of Stratoscope II were still a study in frustration. Schwarzschild recalls three successive failures – one launch failure and two failures of the mechanism that would release the telescope from its stowed position and allow it to point properly at its operating altitude. Each time the telescope landed, there was some damage, so turnaround time for a next flight was slow. On one of its last flights, however, during the night of 18/19 May 1968, from Palestine, the telescope floated for 8 hours at 80,000 ft before landing near Cushing, Texas. All went well, except that atmospheric turbulence degraded the focus. ¹³³ Still, images of the nucleus of the Seyfert galaxy NGC 4151 revealed a structure as small as 0.3 seconds of arc, certainly far higher resolution than earth-bound telescopes could achieve, but not even close to the theoretical resolving power of Stratoscope II’s 36-in diffraction-limited mirror. ¹³⁴ Even though there seemed to be thermal problems, the guiding and pointing accuracy was claimed to be accurate to within 0.05 seconds of arc, which was generally recognized as a highly impressive technical achievement. ¹³⁵

As with the goal of Stratoscope I, the observations obtained here reflected another facet of Schwarzschild’s scientific interests, though not as central. He had been interested in the mass distribution in galaxies since the mid-1950s. By the early 1970s, the most puzzling aspect of this subfield was the nature of galaxies that possessed anomalously bright cores, commonly referred to as Seyfert galaxies. Another flight, the seventh, in March 1970, lasting over 14 hours, yielded an even higher resolution of NGC 4151, which after densitometry analysis of the images yielded a “safe upper limit to the half-intensity diameter of the nucleus of 0″.08.” The distance to the galaxy was hardly certain, but with the latest estimate of the Hubble Constant by Sandage of 55 km/s/Mpc, this meant that the bright object in the centre was no larger than some 7 pc. The long flight also yielded images of Uranus, Jupiter, and Io as well as a host of comparison stars. ¹³⁶

In his report to patrons in October 1970, Princeton’s R. J. Woodrow pronounced this last flight an “unqualified” success, the first in the seven flights, and many more aborted attempts, that had taken place since March 1963. Despite this success, funding and organization issues were plaguing the project. The NASA/Marshall testing and evaluation switch had quickly resulted in large cost overruns. As Schwarzschild reported to Chandrasekhar in July 1969, “we have gotten into a very difficult situation in the Stratoscope project, largely caused by so severe fund limitations that make the project hang on a very thin hair.” ¹³⁷ The situation deteriorated even more in the spring of 1970 when the Mansfield Amendment required that ONR withdraw support “in consequence of new government regulations.” NSF, unfortunately, was unable to support everything, and there were also cuts in the NASA budget as Apollo funding had long passed its peak. As a result, Danielson and Schwarzschild turned to the MSFC to take it all over, except that Princeton would still provide the “astronomical manpower.” The funding shifts were now making it very hard for the contractors to maintain their teams, and changes at Princeton, mainly the fact that Schwarzschild was now President of the AAS, a responsibility he had not sought out, required that he relinquish primary responsibility for Stratoscope to Danielson. ¹³⁸

NSF was also wondering about its future involvement. As astronomy programme officer Robert Fleischer observed, NASA and NSF had to decide “whether the Stratoscope project is space astronomy or ground-based astronomy.” He objected to further funding until this was made clear: “This is so because it is not a trivial project.” Indeed, by then, Stratoscope required a US$2 million operating budget, which was equivalent to NSF’s larger national ground-based facilities. ¹³⁹ And even NASA was concerned about costs. In June, Jesse Mitchell, of NASA’s Office of Space Science and Applications, calculated that Stratoscope cost them 38 man-years/year of effort at Marshall alone. ¹⁴⁰ NASA was particularly concerned that an external prime contractor, Princeton, administered this multimillion dollar project. All this led to a “momentary standstill” after the seventh flight. At first, if Princeton agreed to transfer management and real property to Marshall, NASA would be willing to support continued data analysis for a year or two. ¹⁴¹ But soon NASA’s position softened and funds were found for continued flights, as long as the project was completely reorganized. By 1971, all property and management was shifted to NASA and balloon operations to NCAR. An eighth flight would concentrate on planets, as well as the nuclei of the nearest galaxies, M31 and M32.

That flight took place in September 1971 and yielded “superior” images “characteristic of the diffraction pattern of the 36-inch aperture.” M31’s nucleus was easily resolved. While these results were certainly valuable, a later analysis of the telescope’s performance on the seventh and eighth flights, trying to image Uranus, showed that there was a lack of thermal uniformity. This later detailed analysis of the images showed that Stratoscope’s overall “Modulation Transfer Function,” its ability to distinguish contrast differences as a function of spatial frequency, was less than ideal, but was still quite an achievement. ¹⁴²

One notable characteristic of this latest report on galactic nuclei, received by the Astrophysical Journal in June 1972 and published in December, was that the opening sentence placed Stratoscope II in the past tense. After flight #8, Danielson and his team started to plan for a Stratoscope III while hoping for yet another flight of Stratoscope II some time in 1973, to image Pluto and the nucleus of the giant elliptical galaxy M87. In January 1972, however, NASA decided that Stratoscope II had done its job and “maximum emphasis and resources be placed on the Stratoscope III program” which would test systems thought to be identical to the eventual LST. Accordingly, in June, Danielson submitted a new proposal to NASA for “Stratoscope III” specifying that its “optical system be as identical to the LST as is possible.” This in effect aligned the programme completely with the goals of what was then called the “LST Steering Committee,” which was a constantly changing set of review and analysis groups. Spitzer himself had been chair of an early “Ad Hoc Committee on the Large Space Telescope” formed in 1965. ¹⁴³ By 1973, the LST, what was to become the Hubble Space Telescope, was still in preliminary analysis, or Phase A, studies, and Spitzer designed Princeton’s proposal to fit that stage, focusing on “long lead problems” like object selection, data reduction, and quantitative imagery with an “integrating TV.” Spitzer identified Danielson as the key player. ¹⁴⁴ Schwarzschild was no longer the lead, though he remained active and available as needed, as we note below.