THE VANISHING STAR ENIGMA AND THE 1952 WASHINGTON D.C. UFO WAVE
BEATRIZ VILLARROEL
·JANUARY 21, 2024
As we look up at the starry sky, countless celestial bodies silently peer down upon us. Most of these have been there for billions of years as stellar processes slowly unfold, starting from their birth until their final demise. Light from other celestial objects, though long vanished, has only recently reached us. In other instances, swift changes in the sky occur at timescales as short as seconds or minutes, like when a dwarf star momentarily flares up or when a human satellite crosses the field of view.
My team has been searching for objects that may have vanished. As an unexpected result of our searches, we found cases where multiple star-like objects (transients) appeared and vanished in a small image within an hour, and even more peculiarly, two of our brightest cases happened in July 1952, coinciding in time with the 1952 Washington D.C. UFO flyovers. But what have we actually found, and how do these two events potentially link to one another?
In the Vanishing & Appearing Sources during a Century of Observations (VASCO) project, our team has been dedicated to the search for celestial objects that vanished over the span of 70 years. In the grand scheme of cosmic time and the billions of years needed for a low-mass star to turn into a white dwarf, seventy years is only a fleeting moment in cosmic time. But 70 years is also much longer than the time needed for a satellite to pass through the telescope’s field of view. Our original objective was to search for a star that had vanished, with the hope of detecting instances where a star directly collapses into a black hole (failed supernova), an event predicted by supernova theoreticians. Alternatively, we were intrigued by the prospect of finding a star that vanishes entirely without a trace or explanation; a signature of a highly advanced civilization.
However, this task was far from straightforward. My colleague spent two years developing powerful methods [5] for sifting through the vast terabytes of image data involved. In parallel, we were (and still are) running a citizen science project together with scientists, amateur astronomers, and students primarily in Algeria and Nigeria, to search for these vanishing stars.
For our searches, we employed an object catalog sourced from the US Naval Observatory (USNO) together with archival images dating back to the early 1950s, captured at the Palomar Observatory in California. The images from Palomar predate the dawn of human space exploration. This night sky was pristine, and a far cry from today’s sky that is littered with tens of thousands of debris pieces from human satellites in orbits around the Earth, many producing flashes lasting fractions of a second as they reflect sunlight and tumble through space. These images we compared to the modern databases from Palomar Sky Survey, PanSTARRS, and the Gaia satellite in our quest to find disappearing objects.
We still haven’t found a single failed supernova candidate. However, our exploration has led us to a more intriguing discovery: several images where multiple star-like objects appear in a single snapshot of the sky, never to be seen again. In a specific instance [1], nine faint objects looking like stars were visible in an image captured on April 12, 1950, during a 50-minute exposure. However, they were absent in the image taken just 30 minutes earlier and in another image from six days later. We searched through all available archives in an attempt to locate the nine objects. We directed the world’s largest optical telescope, the Gran Telescopio Canarias, with its 10.4-meter aperture, to the locations where the transients had been. Nothing was found. The objects had simply vanished.
Given the faint nature of these objects and how close to the detection limit they were, we wondered whether these were spurious objects caused by some rare contamination with coincidentally star-like shapes (plate defects), possibly caused by secret atomic bomb tests, or if these phenomena were authentic observations. Artificial objects situated in high-altitude orbits, tens of thousands of kilometers above the Earth’s surface could potentially generate similar flashes as they tumble and reflect sunlight. But a flash could also be produced by the intrinsic light of an object.
To find further evidence of artificial objects outside Earth’s atmosphere, one can search for multiple transients that, additionally, are also aligned [2]. Satellite reflections can manifest as either a single flash or a sequence of consecutive flashes falling on a line, depending on variables like geometry, rotational speed, and the dimensions of our field of view. The exclusive use of pre-Sputnik catalogs further ensures that only non-human objects are included. A white paper describing this search for technosignatures in detail, including the accompanying statistical analysis, has undergone peer review and was published in Acta Astronautica in 2022.
Then we executed the search. We identified two candidates that were statistically significant and had unlikely alignments of transients and an additional three statistically weaker candidates (in total 5 candidates). Our best candidate, Candidate 5, had a probability of p ~ 0.0001 to exist by chance (see Figure 1 below). Regrettably, one journal after another declined to send our paper for peer review, informing us that the topic of the paper consistently fell ‘outside the scope of the journal’. Only one journal sent the paper to referees, which ended up being rejected after several confusing rounds. The paper remains on the arXiv preprint server.
As we look up at the starry sky, countless celestial bodies silently peer down upon us. Most of these have been there for billions of years as stellar processes slowly unfold, starting from their birth until their final demise. Light from other celestial objects, though long vanished, has only recently reached us. In other instances, swift changes in the sky occur at timescales as short as seconds or minutes, like when a dwarf star momentarily flares up or when a human satellite crosses the field of view.
My team has been searching for objects that may have vanished. As an unexpected result of our searches, we found cases where multiple star-like objects (transients) appeared and vanished in a small image within an hour, and even more peculiarly, two of our brightest cases happened in July 1952, coinciding in time with the 1952 Washington D.C. UFO flyovers. But what have we actually found, and how do these two events potentially link to one another?
In the Vanishing & Appearing Sources during a Century of Observations (VASCO) project, our team has been dedicated to the search for celestial objects that vanished over the span of 70 years. In the grand scheme of cosmic time and the billions of years needed for a low-mass star to turn into a white dwarf, seventy years is only a fleeting moment in cosmic time. But 70 years is also much longer than the time needed for a satellite to pass through the telescope’s field of view. Our original objective was to search for a star that had vanished, with the hope of detecting instances where a star directly collapses into a black hole (failed supernova), an event predicted by supernova theoreticians. Alternatively, we were intrigued by the prospect of finding a star that vanishes entirely without a trace or explanation; a signature of a highly advanced civilization.
However, this task was far from straightforward. My colleague spent two years developing powerful methods [5] for sifting through the vast terabytes of image data involved. In parallel, we were (and still are) running a citizen science project together with scientists, amateur astronomers, and students primarily in Algeria and Nigeria, to search for these vanishing stars.
For our searches, we employed an object catalog sourced from the US Naval Observatory (USNO) together with archival images dating back to the early 1950s, captured at the Palomar Observatory in California. The images from Palomar predate the dawn of human space exploration. This night sky was pristine, and a far cry from today’s sky that is littered with tens of thousands of debris pieces from human satellites in orbits around the Earth, many producing flashes lasting fractions of a second as they reflect sunlight and tumble through space. These images we compared to the modern databases from Palomar Sky Survey, PanSTARRS, and the Gaia satellite in our quest to find disappearing objects.
We still haven’t found a single failed supernova candidate. However, our exploration has led us to a more intriguing discovery: several images where multiple star-like objects appear in a single snapshot of the sky, never to be seen again. In a specific instance [1], nine faint objects looking like stars were visible in an image captured on April 12, 1950, during a 50-minute exposure. However, they were absent in the image taken just 30 minutes earlier and in another image from six days later. We searched through all available archives in an attempt to locate the nine objects. We directed the world’s largest optical telescope, the Gran Telescopio Canarias, with its 10.4-meter aperture, to the locations where the transients had been. Nothing was found. The objects had simply vanished.
Given the faint nature of these objects and how close to the detection limit they were, we wondered whether these were spurious objects caused by some rare contamination with coincidentally star-like shapes (plate defects), possibly caused by secret atomic bomb tests, or if these phenomena were authentic observations. Artificial objects situated in high-altitude orbits, tens of thousands of kilometers above the Earth’s surface could potentially generate similar flashes as they tumble and reflect sunlight. But a flash could also be produced by the intrinsic light of an object.
To find further evidence of artificial objects outside Earth’s atmosphere, one can search for multiple transients that, additionally, are also aligned [2]. Satellite reflections can manifest as either a single flash or a sequence of consecutive flashes falling on a line, depending on variables like geometry, rotational speed, and the dimensions of our field of view. The exclusive use of pre-Sputnik catalogs further ensures that only non-human objects are included. A white paper describing this search for technosignatures in detail, including the accompanying statistical analysis, has undergone peer review and was published in Acta Astronautica in 2022.
Then we executed the search. We identified two candidates that were statistically significant and had unlikely alignments of transients and an additional three statistically weaker candidates (in total 5 candidates). Our best candidate, Candidate 5, had a probability of p ~ 0.0001 to exist by chance (see Figure 1 below). Regrettably, one journal after another declined to send our paper for peer review, informing us that the topic of the paper consistently fell ‘outside the scope of the journal’. Only one journal sent the paper to referees, which ended up being rejected after several confusing rounds. The paper remains on the arXiv preprint server.
Figure 1. The left image shows five transients on the 27th of July 1952 in the First Palomar Sky Survey. The right image shows the same star field in the Second Palomar Sky Survey, about 30 years later. From Villarroel, Solano et al., 2022, arXiv. (Note: In the article, the date is wrongly stated to be the 28th of July 1952.)
Meanwhile, our team continued its search efforts. One year ago, my colleague presented the team with a case that he had uncovered during the automated searches [5]. The image showed three bright and beautiful objects looking just like stars in a POSS-I image from the 19th of July 1952 that appeared and vanished within a plate exposure [3] (see Figure 2 below). The three bright objects seemed as real as Betelgeuse itself. We explored a gravitational lensing hypothesis, considering the possibility that a massive foreground object could bend the passing light so that three images appear only to vanish moments later. Perhaps, a supermassive black hole located just a few light-years from Earth, with a mass ten times that of the black hole at the center of the Milky Way, could explain the triple transient? We were not convinced. The article was recently published in Monthly Notices of the Royal Astronomical Society.
Meanwhile, our team continued its search efforts. One year ago, my colleague presented the team with a case that he had uncovered during the automated searches [5]. The image showed three bright and beautiful objects looking just like stars in a POSS-I image from the 19th of July 1952 that appeared and vanished within a plate exposure [3] (see Figure 2 below). The three bright objects seemed as real as Betelgeuse itself. We explored a gravitational lensing hypothesis, considering the possibility that a massive foreground object could bend the passing light so that three images appear only to vanish moments later. Perhaps, a supermassive black hole located just a few light-years from Earth, with a mass ten times that of the black hole at the center of the Milky Way, could explain the triple transient? We were not convinced. The article was recently published in Monthly Notices of the Royal Astronomical Society.
Figure 2. The left image shows three transients on the 19th of July 1952 in the First Palomar Sky Survey. The right image shows the same star field in the Second Palomar Sky Survey, about 30 years later. This image is an adaption of the images in Solano, Marcy et al., 2023.
Let us have a closer look at the time period of the observations. July 1952 was a very special month in Washington D.C. Between July 12th and July 29th, 1952, a multitude of UFO sightings occurred over Washington D.C. in the United States involving radar-visual sightings in simultaneous observations, and even circumstances where a Navy pilot was authorized to shoot down an object (with some debris collected). The highest number of sightings took place during two weekends: July 19th – 20th and the weekend of July 26th – 27th, 1952.
An official record from the National Archives of Australia reveals that while the average monthly number of UFO reports of sightings between 1948 and 1951 was 15 per month, there were approximately 536 reports in July 1952. The US Air Force held a large press conference at the Pentagon (the largest press conference since the end of World War II), claiming that the radar observations were caused by temperature inversions, citing a theory produced and promoted [6,7] by the astronomer and UFO skeptic Donald Menzel, who had the highest clearances within the National Security Agency in the United States. Further work cast serious doubts on Menzel’s explanation. Also, the US Air Force failed to explain the numerous eyewitness reports.
This is where things get fun. The three transients we found were, indeed, observed on the night of July 19th, 1952, which coincided with the first weekend of the Washington UFO flyovers – a coincidence first noticed by my friend David Altman. As scientists, we recognize that coincidences will occur from time to time, no matter how remarkable they may seem. Yet, I couldn’t help but wonder about whether any of the five top candidates in our previous 2022 arXiv paper could have been observed during the Washington flyovers.
Indeed, our absolute top candidate (Candidate 5, with a mere 0.0001% probability) was observed on July 27th, 1952, aligning with the second weekend (Author’s Note: In the arXiv paper and all presentations, including one I presented at the Sol Foundation event at Stanford University in October, 2023, and up until December 2023, I have quoted the wrong date of 28th of July 1952. The correct date for the XE141 plate, according to the STScI plate archive, is the 27th of July 1952). Ironically, our two most prominent and brightest cases of multiple transients coincided in time with the two weekends of the renowned Washington UFO flyovers.
See Also
MOST OF THE UNIVERSE IS INVISIBLE: HERE’S WHAT YOU NEED TO KNOW ABOUT WHAT YOU CAN’T SEE
What types of events might lead to the detection of multiple transients in plates from the same period? One possible explanation is that these transients are indeed UFOs. Another hypothesis is that high-altitude atomic bomb tests could have generated aurora-like phenomena over Washington. These events might have produced nuclear fallout in other regions of the vast country, which–perhaps–could be detected as false stars on the photographic plates. Perhaps, in this instance, Occam’s razor suggests that the first hypothesis requires less of a stretch.
A dataset that possibly could help to solve the mystery is the Digital Access to a Sky Century @ Harvard (DASCH) project, which comprises digitized portions of Harvard College Observatory’s photographic plate collection. The observatory’s plate collection consists of 550,000+ plates that played a pivotal role in cutting-edge astronomy research spanning over a century. Given that the Harvard Observatory is considerably closer to Washington D.C. than Palomar, it might possess valuable records from the time of the Washington D.C. UFO flyovers. As recently as a few weeks ago, the DASCH project came back online after a several years-long hiatus; a blessing to many astronomers.
Despite its numerous successes, Harvard’s photographic plate collection faced many challenges during its history. In the early 1950s, Harvard University chose to destroy a portion of its own photographic plate collection under the directive of its Director Donald Menzel who entered office in 1952. This story is carefully retold in the astronomer Dorrit Hoffleit’s autobiography, Misfortunes as Blessings in Disguise. Furthermore, Donald Menzel stopped Harvard’s observatory from conducting further photographic plate surveys of the sky in 1953. The latter event is now commonly referred to as the ‘Menzel gap‘. Constraints related to storage space and budgetary limitations were cited. Only fifteen years later, Harvard resumed its surveillance of the sky upon Menzel’s retirement.
This remarkable sequence of unusual events suggests that we investigate more photographic plates from the summer of 1952 to see whether there was a higher incidence of anomalous transients than in previous summers. We can also examine the sky as it looks today. With our new endeavor, the ExoProbe project, we will search for similar types of transient events in the modern sky; the hope is to find a case of such anomalous transients that can be carefully studied with today’s instrumentation. We will use a network of small telescopes equipped with high-resolution cameras that permit us to validate the finding in multiple telescopes immediately, locate the object in 3D (if inside the Solar System), and characterize it with a spectrum. Discovering such anomalous transients in modern data helps to circumvent the challenges posed by photographic plate surveys, including the inherent difficulty of tracking and locating these objects once they vanish.
There is undoubtedly a compelling necessity for exploring this mystery. With a bit of luck, maybe we can find statistical support for a connection between historical UFO sightings and multiple transients in photographic plates. If not, these peculiar coincidences may have to remain as intriguing anecdotes in our documentation of stellar history… and maybe that is just fine.
Beatriz Villarroel is the leader of the VASCO project, which incorporates more than 40 members in different countries. She is a researcher at the Nordic Institute for Theoretical Physics (Nordita) in Stockholm.
ReferencesBeatriz Villarroel, Geoffrey Marcy, Stefan Geier, Alina Streblyanska, Enrique Solano, Vitaly
Andruk, Matthew E. Shultz, Alok C. Gupta, Lars Mattsson,“Exploring nine simultaneously occuring transients on April 12th 1950”, 2021, Scientific Reports, 11, 12794.
Beatriz Villarroel, Enrique Solano, Hichem Guergouri, Alina Streblyanska, Lars Mattsson, Rudolf Bär, JamalMimouni, Stefan Geier, Alok C. Gupta, Vanessa Okororie, Khaoula Laggoune, Matthew E. Shultz, Robert A. Freitas Jr., Martin Ward,“Is there a background population of high- albedo objects in geosynchronous orbits around Earth?”, 2022, arXiv: 2204.06091Enrique Solano, Geoffrey Marcy, Beatriz Villarroel, Stefan Geier, Alina Streblyanska, Gianluca Lombardi,Rudolf E. Bär, Vitaly N. Andruk, “A Bright Triple Transient that Vanished within 50 Minutes”, 2023, Monthly Notices of the Royal Astronomical Society, 527, 6312Beatriz Villarroel, Kristiaan Pelckmans, Enrique Solano, Mikael Laaksoharju, Abel Souza, Onyeuwaoma Nnaemeka Dom, Khaoula Laggoune, Jamal Mimouni, Hichem Guergouri, Lars Mattsson, Johan Soodla, DiegoCastillo, Matthew Shultz, Rubby Aworka, Sébastien Comerón, Stefan Geier, Geoffrey Marcy, Alok C. Gupta, Josefine Bergstedt, Rudolf E. Bär, Bart Buelens, Christopher K. Mellon, M. Almudena Prieto, Dismas Simiyu Wamalwa, Martin J. Ward, “Launching the VASCO citizen science project”, 2022, MDPI’s Universe, 8, 561Enrique Solano, Beatriz Villarroel, Carlos Rodrigo, “Discovering vanishing objects in POSS I red images using the Virtual Observatory”, 2022, Monthly Notices of the Royal Astronomical Society, 515, 1380Gilgoff D., 2001, Washington City Paper, https://washingtoncitypaper.com/article/260860/saucers-full-of-secrets/Klass, Phillip , Skeptics UFO Newsletter, 1998, 52
Hoffleit , ”Misfortunes as blessings in disguise”, https://www.aavso.org/dorrit-hoffleit- autobiography-misfortunes-blessings-disguise
Let us have a closer look at the time period of the observations. July 1952 was a very special month in Washington D.C. Between July 12th and July 29th, 1952, a multitude of UFO sightings occurred over Washington D.C. in the United States involving radar-visual sightings in simultaneous observations, and even circumstances where a Navy pilot was authorized to shoot down an object (with some debris collected). The highest number of sightings took place during two weekends: July 19th – 20th and the weekend of July 26th – 27th, 1952.
An official record from the National Archives of Australia reveals that while the average monthly number of UFO reports of sightings between 1948 and 1951 was 15 per month, there were approximately 536 reports in July 1952. The US Air Force held a large press conference at the Pentagon (the largest press conference since the end of World War II), claiming that the radar observations were caused by temperature inversions, citing a theory produced and promoted [6,7] by the astronomer and UFO skeptic Donald Menzel, who had the highest clearances within the National Security Agency in the United States. Further work cast serious doubts on Menzel’s explanation. Also, the US Air Force failed to explain the numerous eyewitness reports.
This is where things get fun. The three transients we found were, indeed, observed on the night of July 19th, 1952, which coincided with the first weekend of the Washington UFO flyovers – a coincidence first noticed by my friend David Altman. As scientists, we recognize that coincidences will occur from time to time, no matter how remarkable they may seem. Yet, I couldn’t help but wonder about whether any of the five top candidates in our previous 2022 arXiv paper could have been observed during the Washington flyovers.
Indeed, our absolute top candidate (Candidate 5, with a mere 0.0001% probability) was observed on July 27th, 1952, aligning with the second weekend (Author’s Note: In the arXiv paper and all presentations, including one I presented at the Sol Foundation event at Stanford University in October, 2023, and up until December 2023, I have quoted the wrong date of 28th of July 1952. The correct date for the XE141 plate, according to the STScI plate archive, is the 27th of July 1952). Ironically, our two most prominent and brightest cases of multiple transients coincided in time with the two weekends of the renowned Washington UFO flyovers.
See Also
MOST OF THE UNIVERSE IS INVISIBLE: HERE’S WHAT YOU NEED TO KNOW ABOUT WHAT YOU CAN’T SEE
What types of events might lead to the detection of multiple transients in plates from the same period? One possible explanation is that these transients are indeed UFOs. Another hypothesis is that high-altitude atomic bomb tests could have generated aurora-like phenomena over Washington. These events might have produced nuclear fallout in other regions of the vast country, which–perhaps–could be detected as false stars on the photographic plates. Perhaps, in this instance, Occam’s razor suggests that the first hypothesis requires less of a stretch.
A dataset that possibly could help to solve the mystery is the Digital Access to a Sky Century @ Harvard (DASCH) project, which comprises digitized portions of Harvard College Observatory’s photographic plate collection. The observatory’s plate collection consists of 550,000+ plates that played a pivotal role in cutting-edge astronomy research spanning over a century. Given that the Harvard Observatory is considerably closer to Washington D.C. than Palomar, it might possess valuable records from the time of the Washington D.C. UFO flyovers. As recently as a few weeks ago, the DASCH project came back online after a several years-long hiatus; a blessing to many astronomers.
Despite its numerous successes, Harvard’s photographic plate collection faced many challenges during its history. In the early 1950s, Harvard University chose to destroy a portion of its own photographic plate collection under the directive of its Director Donald Menzel who entered office in 1952. This story is carefully retold in the astronomer Dorrit Hoffleit’s autobiography, Misfortunes as Blessings in Disguise. Furthermore, Donald Menzel stopped Harvard’s observatory from conducting further photographic plate surveys of the sky in 1953. The latter event is now commonly referred to as the ‘Menzel gap‘. Constraints related to storage space and budgetary limitations were cited. Only fifteen years later, Harvard resumed its surveillance of the sky upon Menzel’s retirement.
This remarkable sequence of unusual events suggests that we investigate more photographic plates from the summer of 1952 to see whether there was a higher incidence of anomalous transients than in previous summers. We can also examine the sky as it looks today. With our new endeavor, the ExoProbe project, we will search for similar types of transient events in the modern sky; the hope is to find a case of such anomalous transients that can be carefully studied with today’s instrumentation. We will use a network of small telescopes equipped with high-resolution cameras that permit us to validate the finding in multiple telescopes immediately, locate the object in 3D (if inside the Solar System), and characterize it with a spectrum. Discovering such anomalous transients in modern data helps to circumvent the challenges posed by photographic plate surveys, including the inherent difficulty of tracking and locating these objects once they vanish.
There is undoubtedly a compelling necessity for exploring this mystery. With a bit of luck, maybe we can find statistical support for a connection between historical UFO sightings and multiple transients in photographic plates. If not, these peculiar coincidences may have to remain as intriguing anecdotes in our documentation of stellar history… and maybe that is just fine.
Beatriz Villarroel is the leader of the VASCO project, which incorporates more than 40 members in different countries. She is a researcher at the Nordic Institute for Theoretical Physics (Nordita) in Stockholm.
ReferencesBeatriz Villarroel, Geoffrey Marcy, Stefan Geier, Alina Streblyanska, Enrique Solano, Vitaly
Andruk, Matthew E. Shultz, Alok C. Gupta, Lars Mattsson,“Exploring nine simultaneously occuring transients on April 12th 1950”, 2021, Scientific Reports, 11, 12794.
Beatriz Villarroel, Enrique Solano, Hichem Guergouri, Alina Streblyanska, Lars Mattsson, Rudolf Bär, JamalMimouni, Stefan Geier, Alok C. Gupta, Vanessa Okororie, Khaoula Laggoune, Matthew E. Shultz, Robert A. Freitas Jr., Martin Ward,“Is there a background population of high- albedo objects in geosynchronous orbits around Earth?”, 2022, arXiv: 2204.06091Enrique Solano, Geoffrey Marcy, Beatriz Villarroel, Stefan Geier, Alina Streblyanska, Gianluca Lombardi,Rudolf E. Bär, Vitaly N. Andruk, “A Bright Triple Transient that Vanished within 50 Minutes”, 2023, Monthly Notices of the Royal Astronomical Society, 527, 6312Beatriz Villarroel, Kristiaan Pelckmans, Enrique Solano, Mikael Laaksoharju, Abel Souza, Onyeuwaoma Nnaemeka Dom, Khaoula Laggoune, Jamal Mimouni, Hichem Guergouri, Lars Mattsson, Johan Soodla, DiegoCastillo, Matthew Shultz, Rubby Aworka, Sébastien Comerón, Stefan Geier, Geoffrey Marcy, Alok C. Gupta, Josefine Bergstedt, Rudolf E. Bär, Bart Buelens, Christopher K. Mellon, M. Almudena Prieto, Dismas Simiyu Wamalwa, Martin J. Ward, “Launching the VASCO citizen science project”, 2022, MDPI’s Universe, 8, 561Enrique Solano, Beatriz Villarroel, Carlos Rodrigo, “Discovering vanishing objects in POSS I red images using the Virtual Observatory”, 2022, Monthly Notices of the Royal Astronomical Society, 515, 1380Gilgoff D., 2001, Washington City Paper, https://washingtoncitypaper.com/article/260860/saucers-full-of-secrets/Klass, Phillip , Skeptics UFO Newsletter, 1998, 52
Hoffleit , ”Misfortunes as blessings in disguise”, https://www.aavso.org/dorrit-hoffleit- autobiography-misfortunes-blessings-disguise
