Astrophotography and the Founding of the Astronomical Institute in Amsterdam

After more than a decade of working in Berlin and Bremen as one of the main theorists of the German Social Democratic Party, Anton Pannekoek (1873-1960) was excited when, in 1918, he was offered the position of assistant director of the Leiden Observatory, where he could pursue his old passion of astronomy. The conservative Dutch government was less enthusiastic about the prospect of a prominent Marxist working at a state university, however, and they intervened to prevent his hiring.² This allowed the University of Amsterdam, a municipal rather than a state university, to swoop in and hire Pannekoek instead. There was one big problem, however: the University of Amsterdam had neither an observatory nor any telescopes.

Pannekoek was encouraged to pursue astronomical research by the University, but his initial plan to establish an observatory in an abandoned water tower was rejected by the municipality as being too costly. Pannekoek had to pivot and decided to build an astronomical laboratory instead. Rather than installing a telescope, he ordered a variety of specialized instruments for measuring photographic plates that could be imported from other observatories . He also hired two 15-year-old secondary-school graduates to operate the instruments and perform calculations, the so-called human computers (rekenaars). With them joining in 1921, the Astronomical Institute in Amsterdam was effectively founded.³

Photography in Astronomy

The Astronomical Institute in Amsterdam that Pannekoek founded is exemplary for a new type of institution that emerged in astronomy in the late nineteenth and early twentieth century due to the growing professional implementation of astrophotography. The work done there illustrates the significant impact of photography on the daily practice of astronomy during this period.

Astronomy was one of the first sciences to embrace the potential of photography. Soon after the invention of the first photographic techniques in the mid-nineteenth century, photographic researchers and astronomers (often these were one and the same) started experimenting with capturing photographic images of astronomical objects.⁶ Initially, these efforts were concentrated on luminous objects like the sun, the moon, and even comets. By the late nineteenth century, photographic techniques had improved to the point that capturing the stars was also possible.

Despite these initial successes, a large number of astronomers remained skeptical. They had genuine epistemic concerns about the value of photography for scientific research. Photographic emulsions had varying light sensitivity, were difficult to make homogeneous, and were typically sensitive to different wavelengths than the human eye. Moreover, it was unknown how photographic emulsions would react to prolonged storage.⁷ There was, however, one major advantage of photography over visual observations that made it especially valuable for astronomy. The possibility of recording observations that could then be measured at a later time by different people.⁸ Pannekoek wanted to use that property by requesting photographic plates from foreign observatories for measurement.

By founding an astronomical laboratory, Pannekoek explicitly followed in the footsteps of Groningen astronomer Jacobus C. Kapteyn, who like Pannekoek, had failed to build an observatory of his own. Instead, from 1884 onward, Kapteyn began collaborating with David Gill of the Royal Observatory at Cape of Good Hope on a comprehensive photographic catalogue of the southern sky. This plan entailed that hundreds of photographic plates were taken at the Cape and then shipped to Groningen where Kapteyn would measure them and compute the stellar coordinates.⁹ This collaboration resulted in the Cape Photographic Durchmusterung, a three-volume, 1992-page star catalogue that provided the coordinates of 454.875 stars in the southern hemisphere.¹⁰

Photographic Milky Way

Whereas Kapteyn’s research focused on positional astronomy, Pannekoek used photographic plates to determine the surface brightness of the Milky Way and measure the absorption lines of stellar spectra.¹² The appearance of the Milky Way, the faint milky band of light visible in the night sky, was a lifelong obsession of Pannekoek. His first recorded observations of the Milky Way date back to 1888, while his final publication on the subject appeared almost seven decades later in 1957. Pannekoek also elaborated a theory on what caused this appearance. He argued that the Milky Way is essentially an optical illusion created by the combined light of countless stars hitting the retinal elements in our eyes, whose signals were then processed by the brain. Despite this illusionary nature, Pannekoek still thought it was worthwhile to create an accurate representation of the Milky Way. He argued that such a representation showed the general distribution of stars in the galaxy and enabled this distribution to be tracked over time.

The appearance of the Milky Way was notoriously difficult to capture on the photographic plate, however.¹³ The first attempts at photographic images of the Milky Way, made in the late nineteenth century, were met with a mixed reaction from Pannekoek. On the one hand, he considered them a ‘revelation’ because they convincingly showed that the Milky Way consisted of the light of countless faint stars.¹⁴ Yet, he also cautioned that they failed to show its large-scale structure. In the faint areas, the light of the Milky Way was entirely dissolved into individual stars, while in the more densely packed areas, the combined effect was exaggerated and overexposed.¹⁵

Pannekoek’s initial solution for representing the Milky Way was to rely on drawings and diagrams made according to naked-eye observations. He quickly realized, however, that different observers each made very different drawings. To filter out these variations between observers and thus provide a generalized image of the Milky Way, Pannekoek decided to combine the drawings and calculate the average surface brightness for each part of the Milky Way to produce what he called the ‘mean subjective image’ (Fig. 3).¹⁶ This image was represented through both numerical tables and isophotic diagrams.

Despite the success of his drawings, Pannekoek remained interested in producing a photographic representation of the Milky Way as well. He eventually decided on the method of extrafocal photometry. This required photographic plates to be exposed out of focus so that the light of each star would be spread out over a small circle, instead of being concentrated into a point. These circles could then overlap, which recreated the illusion of Milky Way clouds (Fig. 4).¹⁸

To capture the entire Milky Way, Pannekoek required over a hundred photographic plates taken at multiple observatories on different locations. For the northern hemisphere, the plates were taken by Max Wolf in Heidelberg, Germany. Wolf sent various batches of plates between 1921 and 1928, each time adjusting the setup according to Pannekoek’s instructions. For the southern hemisphere, the plates were initially taken by Joan Voute in Lembang, then Netherlands Indies, between 1933 and 1939 and subsequently by John Paraskevopoulous in Maselspoort, South Africa, in 1942 and 1946. Here too, multiple batches of plates were needed, mainly to replace plates found to be flawed when examined in Amsterdam.¹⁹

Getting the plates was only the first step in the process, however. A lot of work was still required to turn them into suitable results. Each individual plate had to be measured meticulously with a microphotometer to measure the blacking on set parts of the plate. Work that was done by his longtime computer David Koelbloed (Fig. 1). These measurements were then calibrated by comparing the measurements from parts of the plates where they overlapped with one other. Next, the calibrated measurements were copied onto glass slides (Fig. 4), which were then projected onto large sheets of paper. The projections made it possible to combine the numbers in the same drawing. Finally, Pannekoek drew lines on the sheets to indicate the shape of the Milky Way clouds (Fig. 5, top).²¹

Using these sheets, Pannekoek could produce numerical tables and isophotic diagrams (Fig. 5, middle) that gave a quantitative representation of the surface brightness of the Milky Way. These were the main result of his photographic research. However, Pannekoek did not stop there. He also decided to make drawings based on these results to illustrate how the photographic Milky Way would look if we could see it with the human eye (Fig. 5, bottom). He remarked that using this method led to a Milky Way image that showed a ‘far greater wealth of detail’ than the drawings based on visual observations while smoothing out all the sharp detail of focal photographic images, ‘thus gaining a true representation of the surface intensity which is lacking there.’²²

Pannekoek’s photographic research on the Milky Way illustrates two important things. First, that there was a continuity in what he aimed to achieve with visual observations and photographic research. In both cases, we see that Pannekoek desired to capture the surface brightness of the Milky Way’s appearance. When regular photography was incapable of doing so, he did not adapt his goals but changed his methods. Second, Pannekoek had no desire to let photographic images ‘speak for themselves’. Photographic plates had to be manipulated, measured, and calibrated before they could reveal a true presentation of the Milky Way.

Spectrophotography

Although the Astronomical Institute started out focusing on the structure and the appearance of the Milky Way, Pannekoek soon added another research subject: the astrophysics of stellar atmospheres.²⁶ This subject aimed to determine the physical conditions and processes in the outer layers of stars. Here, the spectra of stars played a vital role. The darkened lines in these spectra could be compared with the spectral lines that were produced in laboratories on Earth, which makes it possible to deduce information like the temperature and chemical composition of stars.

Photography played a vital role in stellar astrophysics because photographic spectra could be measured to a very high precision using specialized instruments. Again, measuring the photographic spectra and then calibrating these measurements was a very labour-intensive process, taking much more time than was needed to record them. For the most part, this labour was performed by human computers, often women or young men who had received no formal education in astronomy.²⁷ But even with their labour, many photographic observatories produced more photographic spectra than they could viably measure themselves. This surplus of photographic plates enabled an international collaboration among astronomical institutions from which Pannekoek profited.

Pannekoek made use of the opportunities provided by spectral astrophotography in various ways. Among other things, he borrowed plates that other observatories had used in previous investigations and went on eclipse expeditions to record the spectra of the solar corona. He also requested specifically-made plates from observatories with available observation time. The most prominent supplier of these plates was the Dominion Astrophysical Observatory (DAO) in Victoria, British Colombia.

Collaboration at a distance came with distinct problems, however. Transporting the plates by train and boat took weeks, severely limiting Pannekoek’s ability to adjust the observation setup when he found out the spectra were not made to his liking. At the same time, the DAO director John S. Plaskett grew frustrated that Pannekoek did not seem to realize the practical difficulties of some of his requests.²⁹ Finally, Plaskett invited Pannekoek to come to Victoria to make the plates himself. Pannekoek gladly accepted the invitation. He spent six months at the DOA, during which he took some 50 spectral photographs (Fig. 7). At the time, he wrote to his wife that he expected this would be enough to work on for several years.³⁰ That turned out to be a gross underestimation: the main publication based on these plates only appeared in 1950, more than two decades after they were made.³¹

The photographic spectra ordered and made by Pannekoek were used for various investigations. First, he published catalogues containing the wavelength and strength of every line in a photographic spectrum. In some cases, this meant more than a thousand lines, all of which had to be measured in excruciating detail (Fig. 8). Pannekoek believed that this level of detail was a unique strength of Amsterdam. Where photographic observatories simply had too many photographic plates to measure, the lack of direct access to a telescope forced the Amsterdam astronomers to get the most out of their limited number of plates.³³

The measurements of spectral lines could then be used to determine the physical properties of the outer layers of stars. For instance, the temperature could be determined by looking at the presence and strength of certain lines, while the pressure could be derived from the width of the lines.³⁴ It was even possible to investigate what physical processes occur within stars by comparing the observed spectra with the outcome of complex theoretical models of stars. Here again, Pannekoek tried to get the most out of his limited data by introducing numerical computations into his calculations.³⁵

The Practice of Astrophotography

Pannekoek’s photographic research provides a clear impression of the role and impact of photography in early twentieth-century astronomy. Professional astrophotography was hardly ever used to depict the night sky directly. Photographic catalogues of stellar coordinates or spectral lines seldom contained images. Rather, they presented pages and pages of measurements neatly sorted into tables. Even in the case of the visual appearance of the Milky Way, Pannekoek actively avoided direct depiction because it would give a false impression. Instead, he relied on out-of-focus photographs that were then measured, calibrated, and combined into numerical tables, diagrams, and drawings.

Rather than using photography for illustration, astronomers used photography to record observations – observations that could be stored, shared, transported, measured, and analysed. As a result, photography had a profound impact on the daily practice of many astronomers. Instead of looking through a telescope all night, they now looked through measurement instruments during the day. The amount of work involved in the photographic process also enabled and encouraged an international division of labour among astronomical institutions.

With the rise of digital techniques, photographic plates have been replaced with digital files, measurement instruments with computers, and ships with internet cables. But the general principles remain the same. Even though the Astronomical Institute in Amsterdam now has telescopes, most of their astronomers still primarily work on observations made elsewhere.

---

1. Archive of the Anton Pannekoek Institute for Astronomy (API Archive) ↩︎
2. David Baneke, “Pannekoek’s One Revolution: Anton Pannekoek and the Modernization of the Dutch Astronomical Community,” in Anton Pannekoek: Ways of Viewing Science and Society, ed. Chaokang Tai, Bart van der Steen, and Jeroen van Dongen (Amsterdam University Press, 2019), 87–108, https://doi.org/10.2307/j.ctvp7d57c.7. ↩︎
3. Anton Pannekoek, Herinneringen: herinneringen uit de arbeidersbeweging; sterrenkundige herinneringen, ed. B. A. Sijes, J. M. Welcker, and J. R. van der Leeuw (Amsterdam: Van Gennep, 1982), 246–49. ↩︎
4. Image by curtosy of Laurence A. Marschal ↩︎
5. API Archive ↩︎
6. Kelley E. Wilder, Photography and Science (London: Reaktion, 2009), 18–31. ↩︎
7. For more on these concerns, see e.g. Jimena Canales, “Photogenic Venus: The ‘Cinematographic Turn’ and Its Alternatives in Nineteenth‐Century France,” Isis 93, no. 4 (2002): 585–613, https://doi.org/10.1086/375953; Omar W. Nasim, “Hybrid Photography in the History of Science: The Case of Astronomical Practice,” in Hybrid Photography: Intermedial Practices in Science and Humanities, ed. Sara Hillnhütter, Stefanie Klamm, and Friedrich Tietjen (Abingdon: Routledge, 2021), 11–27. ↩︎
8. John Lankford, “The Impact of Photography on Astronomy,” in The General History of Astronomy: Volume 4, Astrophysics and Twentieth-Century Astronomy to 1950: Part A, ed. Owen Gingerich (Cambridge: Cambridge University Press, 1984), 16–39; Charlotte Bigg, “Photography and Labour History of Astrometry: The Carte Du Ciel,” in The Role of Visual Representations in Astronomy: History and Research Practice, ed. Klaus Hentschel and Axel D. Wittmann (Thun: Verlag Harri Deutsch, 2000), 90–106. ↩︎
9. Michael Feast, “Kapteyn and South Africa,” in The Legacy of J.C. Kapteyn: Studies on Kapteyn and the Development of Modern Astronomy, ed. Pieter C. van der Kruit and Klaas van Berkel (Dordrecht: Kluwer, 2000); Pieter C. van der Kruit, Jacobus Cornelius Kapteyn: Born Investigator of the Heavens (Cham: Springer, 2015), 159–204, https://doi.org/10.1007/978-3-319-10876-6. ↩︎
10. The number of pages is without the introduction pages. David Gill and Jacobus C. Kapteyn, The Cape Photographic Durchmusterung for the Equinox 1875, in 3 parts, Annals of the Cape Observatory, 3-5, (London: H.M. Stationary Office, 1896-1900). ↩︎
11. Image credit: South African Astronomical Observatory via westlicht.com. ↩︎
12. This section is based on Chaokang Tai, “The Milky Way as Optical Phenomenon: Perception and Photography in the Drawings of Anton Pannekoek,” in Anton Pannekoek: Ways of Viewing Science and Society, ed. Chaokang Tai, Bart van der Steen, and Jeroen van Dongen (Amsterdam University Press, 2019), 219–47, https://doi.org/10.2307/j.ctvp7d57c.13. ↩︎
13. Anton Pannekoek, Die Nördliche Milchstrasse, Annalen van de Sterrewacht te Leiden, 11:3 (Haarlem: Joh. Enschedé en Zonen, 1920), 15–16. ↩︎
14. Anton Pannekoek, A History of Astronomy (New York: Interscience, 1961), 475. ↩︎
15. Anton Pannekoek and David Koelbloed, Photographic Photometry of the Southern Milky Way, Publications of the Astronomical Institute of the University of Amsterdam 9 (Amsterdam: Stadsdrukkerij, 1949), 4. ↩︎
16. Pannekoek, Die Nördliche Milchstrasse, 16. ↩︎
17. Pannekoek, Die Nördliche Milchstrasse plate 7. ↩︎
18. Anton Pannekoek, “Photographic Photometry of the Milky Way and the Colour of the Scutum Cloud,” Bulletin of the Astronomical Institutes of the Netherlands 2, no. 44 (1923): 19–24. ↩︎
19. Anton Pannekoek, Photographische Photometrie der nördlichen Milchstrasse, Publications of the Astronomical Institute of the University of Amsterdam 3 (Amsterdam: Stadsdrukkerij, 1933); Pannekoek and Koelbloed, Photographic Photometry of the Southern Milky Way. ↩︎
20. API Archive ↩︎
21. Pannekoek and Koelbloed, Photographic Photometry of the Southern Milky Way, 4–28. ↩︎
22. Pannekoek and Koelbloed, 28. ↩︎
23. Worksheet, ca. 1948, Archive of Anton Pannekoek (UBA195), Allard Pierson, inv. nr. 11. ↩︎
24. Pannekoek and Koelbloed, Photographic Photometry of the Southern Milky Way chart 13. ↩︎
25. Pannekoek and Koelbloed plate 3. ↩︎
26. This section is based on chapter 3 of Chaokang Tai, “Anton Pannekoek, Marxist Astronomer: Photography, Epistemic Virtues, and Political Philosophy in Early Twentieth-Century Astronomy” (PhD Thesis, Amsterdam, University of Amsterdam, 2021). ↩︎
27. Pamela E. Mack, “Strategies and Compromises: Women in Astronomy at Harvard College Observatory, 1870–1920,” Journal for the History of Astronomy 21, no. 1 (February 1990): 65–76, https://doi.org/10.1177/002182869002100108; Eun-Joo Ahn, “Finding the Invisible Workers in Astronomy: The Case of Mount Wilson Observatory, 1900–1930,” Historical Studies in the Natural Sciences 52, no. 5 (November 1, 2022): 555–88, https://doi.org/10.1525/hsns.2022.52.5.555. ↩︎
28. Worksheet, ca. 1948, Archive of Anton Pannekoek (UBA195), Allard Pierson, inv. nr. 11. ↩︎
29. R. Peter Broughton, Northern Star: J.S. Plaskett (Toronto ; Buffalo: University of Toronto Press, 2018), 271. ↩︎
30. Pannekoek to Anna Pannekoek-Nassau Noordewier, Persoonlijk archief van Antonie Pannekoek, Museum Boerhaave, box 2. ↩︎
31. Anton Pannekoek, Line Intensities in Spectra of Advanced Type, Publications of the Dominion Astrophysical Observatory, 8:5 (Ottawa: Edmond Cloutier, 1950). ↩︎
32. Image credit: National Archives of Canada via Wikimedia Commons ↩︎
33. Elsa van Dien-van Albada, Niet-letterlijke weergave van toespraakje over Pannekoek, 18 May 1982, Archive of the Anton Pannekoek Institute for Astronomy. ↩︎
34. Anton Pannekoek, “The Influence of Collisions on the Formation of the Fraunhofer Lines,” Proceedings of the Section of Sciences, Koninklijke Akademie van Wetenschappen Te Amsterdam 34 (1931): 755–63. ↩︎
35. Anton Pannekoek, The Theoretical Intensities of Absorption Lines in Stellar Spectra, Publications of the Astronomical Institute of the University of Amsterdam 4 (Amsterdam: Stadsdrukkerij, 1935). ↩︎
36. Pannekoek, “The Influence of Collisions on the Formation of the Fraunhofer Lines” between 758 and 759. ↩︎