The Role of Images in Astronomical Discovery

by Rene Roy

assumptions which underlie the description and attempted explanations of the phenomena.

  It is my opinion that what appear to be brilliant intuitions about ejection in galaxies, the

  role of nuclei, associations of spiral arms, blue objects, and quasars are really the result of

  Ambartsumian’s considering the problem in great generality and, above all, of his careful

  reasoning coupled with the visual inspection and study of a very large number of actual

  galaxy forms.”19

  Arp was not the first to raise the issue of diversity and to emphasize chaos in the world

  of galaxy forms. For his atlas, Arp had selected objects from Vorontsov-Velyaminov’s

  1959 catalogue of 355 peculiar and interacting galaxies. The Russian astronomer and his

  colleagues had painstakingly identified weirdly shaped galaxies from the photographic

  National Geographic Society Palomar Sky Survey (PSS), which was completed with the

  Mount Palomar 48-inch Schmidt Telescope in 1956.20 Re-inspecting the plates of the PSS

  used by Vorontsov-Velyaminov for his catalogue and the notes of A. G. Wilson, who

  obtained the original plates, Arp put together a new sample of galaxies. He then spent four

  years photographing the objects with the more powerful Palomar 200-inch; he also added to

  his sample objects from other observers. Of the 338 photographs shown in the atlas, most

  are from plates taken with the 200-inch telescope. Arp was making a point of using the

  same telescopes as Sandage and applying the best photographic processing techniques of

  the day.21

  Arp’s purpose was to produce his own atlas and present galaxies that had been ignored

  by Hubble and Sandage. It was almost an “anti-atlas.” The work showed multiple examples

  of the various kinds of peculiar galaxies: perturbed, interacting, deformed and many others

  with various appendages or tidal tails (Fig. 10.5). The objects were grouped – based on a

  “rough, initial classification.” It was an empirical approach, and the suspicious Arp wrote

  “the physical processes pictured are not understood.” What Arp had in mind, then and

  19 H. C. Arp, Book Review: The Structure and Evolution of Galaxies, Science, 1966, Vol. 154, pp. 1439.

  20 B. A. Vorontosov-Velyaminov, R. I. Noskova and V. P. Arkhipova, Atlas and Catalogue of Interacting Galaxies, Sternberg Astronomical Institute, Moscow State University, Moscow, 1959. A second part was published in the 1970s.

  21 H. C. Arp and B. F. Madore later published A Catalogue of Southern Peculiar Galaxies and Associations, Vols. 1 & 2, Cambridge: Cambridge University Press, 1987.

  14:11:09, subject to the Cambridge Core

  .012

  214

  Part III – Organizing the World of Galaxies

  Fig. 10.5 Peculiar galaxies Arp 81 and Arp 82 from a page of Arp (1966), Atlas of Peculiar Galaxies.

  Courtesy of Carnegie Science.

  thereafter, was that weirdo galaxies demonstrated the need for some novel kinds of physics.

  However, as the brilliant work of the brothers Alar and Juri Toomre showed in the 1970s,

  the unusual or peculiar galaxies had simpler explanations as interacting or merging systems.

  Out of enthusiasm or frustration, Arp may have gone too far in seeing abnormality. As

  Sidney van den Bergh remarked, “some of the objects in Arp’s catalogue are dwarfs that

  are not peculiar at all.”22 The large spiral Messier 101 (Arp 26) is asymmetrical at large

  galactocentric distances, but one may indeed wonder on how peculiar it is.

  Notwithstanding it being a bit of a mixed bag, Arp’s atlas became a fundamental contri-

  bution to galaxy structure. It highlighted a much broader range of morphologies than The

  Hubble Atlas of Galaxies. “Arp’s atlas turned out to be an invaluable source of material

  during the subsequent development of the subject of galactic interactions and mergers.”23

  The collection of peculiar galaxies had a tremendous impact as it presented galaxies as

  real dynamic systems, with the environment playing a major role in their evolution. As the

  Canadian astronomer Pierre Martin commented, “galaxies are just complicated beasts and

  not simple objects to be put in the box.”24

  22 S. van den Bergh, Galaxy Morphology and Classification, Cambridge: Cambridge University Press, 1998.

  23 F. Schweizer, Observational Evidence for Interactions and Mergers, in Galaxies: Interactions and Induced Star Formation, R. C. Kennicutt Jr. et al. (editors), Saas-Fee Advanced Course 26, Swiss Society for Astrophysics and Astronomy, 1996, pp. 105–274.

  24 P. Martin, personal communication to the author, 2015.

  14:11:09, subject to the Cambridge Core

  .012

  10. Atlases of Galaxies, Picturing “Island-Universes”

  215

  Daston and Galison wrote: “the atlas aims to make nature safe for science; to replace

  raw experience – the accidental, contingent experience of specific individual objects – with

  digested experience.”25 This was Hubble’s vision. Arp’s vision was iconoclastic and he

  wished to destabilize established science. Today, the puzzle of galaxy shapes still shows

  gaps and cracks. Alar Toomre, one of the pioneers of computer simulations of galaxies and

  of their interaction, wrote recently: “Even today I cannot pretend that our 1972 work or its

  numerous and much more realistic sequels by others have reasonably explained any more

  than perhaps one-half of the strange images in Arp’s atlas but, sad to say, we have collec-

  tively poured tons of cold water onto Chip’s own hopes that perhaps/maybe/just-possibly

  he was on to some really new physics.”26

  As a working object, the Atlas of Peculiar Galaxies and its images illustrated a rival cos-

  mological viewpoint. The morphology of galaxies was meaningful but could betray pro-

  cesses more fundamental and perhaps hide unknown physics. Arp’s provocative approach

  invited the researcher and the student to keep a sharp eye out for unusual features as they

  might provide better insights into galaxy formation and evolution than the regular sym-

  metrical normal galaxies. Peculiar objects might provide better clues to the physical forces

  responsible for shaping the observed forms of galaxies. Presenting exquisite photographical

  material he had obtained with the most powerful telescope of the time, Arp also conveyed

  a sense that he had access to rare and inaccessible categories of objects.

  To understand the diverging approaches of Sandage/Hubble and Arp to galaxy structure,

  it is again useful to refer to another aphorism of Francis Bacon. “The greatest, and, perhaps,

  radical distinction between different men’s dispositions for philosophy and the sciences is

  this; that some are more vigorous and active in observing the differences of things, oth-

  ers in observing their resemblances . . . Each of them readily falls into excess, by catching

  either as nice distinctions or shadows of resemblance.”27 As I will show, Arp and Sérsic

  were focusing on the differences, while Hubble, Sandage, de Vaucouleurs and their fol-

  lowers concentrated on resemblances. Today, we understand galaxy evolution much better

  following the pioneering work of the Toomre brothers and their followers. Most “peculiar”

  cases may be considered as transitional or temporary states in the complex lives of galaxies.

  Historically, such an epistemic clash is not new. Here’s just one other example: the French

  naturalist Georges-François Leclerc (1707–1788), Comte de Buffon, w
as an ardent oppo-

  nent of Carl Linnaeus’ structured binomial classification system. A bit like Arp on galaxies,

  Buffon thought Linnaeus’ approach tried to force disparate groups of species into “artificial

  assemblages,” and that nature was richer than our systems.28

  1968. Atlas de Galaxias Australes , Illustrating the Normal and Abnormal

  Ready support for Arp’s contrarian view soon came, from an astronomer working in the

  southern hemisphere. Until the mid twentieth century, all large telescopes were located

  25 L. Daston and P. Galison, The Image of Objectivity, Representations, No. 40, Special Issue: Seeing Science, 1992, p. 85.

  26 A. Toomre, private e-mail to the author (October 2014).

  27 F. Bacon, Aphorism 45, Novum Organum, 1620. See Hanover Historical Texts Project on-line.

  28 J. Elphick, Birds, The Art of Ornithology, New York: Rizzoli, 2014, p. 47.

  14:11:09, subject to the Cambridge Core

  .012

  .012

  14:11:09

  Fig. 10.6 Interior page of the Atlas de galaxias australes, showing the radial brightness profile (left) and the isophote map (right) of the barred galaxies NGC 1365. From Sérsic (1968), Atlas de Galaxias Australes. With permission of Universidad de Córdoba, Observatorio Astronómico.

  , subject to the Cambridge Core

  10. Atlases of Galaxies, Picturing “Island-Universes”

  217

  mostly in the northern hemisphere. This disparity biased the observations as well as the

  sample of galaxies studied by researchers. The imbalance needed to be addressed, consid-

  ering that the southern hemisphere hosts several archetypal objects, such as the Magellanic

  Clouds. The American astronomer Charles Perrine (1867–1951), whom we saw at Lick

  Observatory (Chapter 3), had moved to Argentina. While at Lick, Perrine had overseen the

  complete reconstruction of the Crossley 36-inch and had continued James Keeler’s obser-

  vational program on bright “nebulae.” As director of the Observatorio Astronómico de Cór-

  doba from 1918 to 1922, Perrine conducted a research program on galaxies and pioneered

  galaxy photography in the southern hemisphere using a small 76-cm telescope. Understand-

  ing what the instrument required for galaxy research, Perrine pushed for a larger reflector.

  On his initiative, a 1.54-m telescope was completed at Bosque Alegre, Córdoba, in 1942.

  José Luis Sérsic followed brilliantly in Perrine’s tracks, using the new and larger reflec-

  tor. Initially, Sérsic intended to concentrate on studying the morphology and on the photom-

  etry of the galaxies brighter than magnitude 11.0 (Harvard scale) and south of declination

  –35° in the sky. He added more objects to assemble a sample large enough to produce a

  meaningful atlas. Sérsic’s intention and ambition was to replicate for the southern hemi-

  sphere what Sandage had done so magnificently for galaxies observable from California.

  Atlas de galaxias australes appeared in the Spanish language in 1968.29 It was a fine

  work, rigorously produced. There were interesting similarities between Sérsic’s atlas and

  Sandage’s The Hubble Atlas of Galaxies: for example the format, the presentation and the

  use of the Hubble classification scheme were very similar. There were also some significant

  differences. For example, Sérsic insisted on obtaining and including in his atlas quantitative

  information, such as surface brightness (isophotal) maps and photometric profiles. John

  Reynolds had pioneered the measurement of a light-brightness profile across the bulges of

  spirals as early as 1913.30 Sérsic did not have access to a very large telescope or a good

  site; hence, his photographs were not of the same quality, angular resolution and depth as

  the material of Sandage and Arp. Interestingly, Sérsic augmented the atlas with objects that

  revealed new phenomena or did not fit under the orderly scheme of Hubble and Sandage,

  hence the affinity with Arp’s approach.

  For Sérsic, morphology was important because of its “alto valor heurístico” (high heuris-

  tic value), which is a great aid for learning and discovering. His comments emphasize the

  nature of his atlas as a working object. He meant that images could reveal more than the

  schematic models, which were then “limited by our inability to resolve complex systems

  of equations.” Nevertheless, he insisted that it is important to explore galaxies by employ-

  ing more quantitative tools. In the footsteps of the 1920 seminal work of Reynolds,31 Sérsic

  promoted and illustrated his use of photometry by plotting brightness profiles across galaxy

  images (Fig. 10.6), a prelude to the more systematic works by Japanese astronomers that

  29 J. L. Sérsic, Atlas de galaxias australes, Cordoba: Observatorio Astronómico, 1968.

  30 D. L. Block and K. Freeman, Shrouds of the Night, Masks of the Milky Way and Our Awesome New View of Galaxies, New York: Springer, 2008, p. 211.

  31 J. H. Reynolds, Photometric Measures of the Nuclei of some Typical Spiral Nebulae, Monthly Notices of the Royal Astronomical Society, 1920, Vol. 80, pp. 746–753.

  14:11:09, subject to the Cambridge Core

  .012

  218

  Part III – Organizing the World of Galaxies

  were published later (see Fig. 10.8). To highlight the photometric material, Sérsic divided

  the atlas into two parts: (a) images taken with the 1.54-m of the Estación Astrofísica de

  Bosque Alegre, and (b) photometric maps, profiles and tabular data for several of the objects

  presented. Consequently Sérsic’s work had a much more quantitative base than Sandage’s

  atlas. This made the work unique and innovative for its time, and insured its durable impact.

  Following the works of Arp (1966), Zwicky (1959) and Vorontsov-Velyaminov (1951),

  Sérsic proposed that galaxies that appear “abnormal” are an indication of some unusual and

  important processes. Thus, a distinctive aspect of the atlas was the evidence that galaxies

  undergo a much more complex and disturbed formation and evolution process than was

  apparently inferred by the Hubble–Sandage scheme. Illustrating this perspective, Sérsic’s

  atlas included several examples of distorted and asymmetrical galaxies. Since anomalous

  galaxies are infrequent, these disturbances probably involved short timescales, e.g. 107 to

  108 years. Commenting on Arp’s term, “peculiar” galaxies, Sérsic asked: “What is a normal

  galaxy?” Perhaps wishing to be more constructive than Arp, Sérsic stated that “it is possible

  to clearly define the extreme of ‘normality’ and ‘peculiarity,’ leaving intermediary cases for

  discussion later.” Sérsic aimed to be helpful and to advance the debate for the sake of better

  understanding galaxies and their evolution. As we will see in the next chapter, the surprising

  popularity of his atlas is testimony to the success of his broad vision and approach.

  1981. The Revised Shapley–Ames Catalog of Bright Galaxies , the Guide

  to All Galaxy Atlases

  In 1981 (a second edition appeared in 1987), Allan Sandage and Swiss astronomer Gustav

  A. Tammann published The Revised Shapley-Ames Catalog of Bright Galaxies ( RSA).32 The

  RSA represented a complete overhaul of the original work by Shapley and Ames of 1932

  and retained some of the format and the main content of their original work. However, this

  time, Sandage and Tammann added an important set of reference images of the main mor-

&nb
sp; phological types of galaxies, with new material on southern hemisphere objects obtained

  with the wide-field Irénée du Pont 2.5-m telescope of Las Campanas Observatory, Chile,

  that had been in operation since 1977. Harold Babcock played a leading role in having this

  telescope built, following a gift from Mr. and Mrs. Crawford H. Greenewalt to the Carnegie

  Institution of Washington. In their introduction to the atlas, Sandage and Tammann stressed

  the importance of galaxy surveys, first started by William Herschel, and they outlined the

  history of successive works. “The listing is meant as an aid in planning various observing

  programs based on knowledge of types as it existed in 1979.” The addition of high quality

  images representative of the main galaxy types made the revised work a sort of mini atlas

  (Fig. 10.7).

  An additional classification criterion was introduced. Sidney van den Bergh had shown

  that galaxies with the highest luminosities have the longest and most highly developed

  arms, whereas fainter systems show poorly developed arms. The luminosity classes range

  32 A. R. Sandage and G. A. Tammann, The Revised Shapley–Ames Catalog of Bright Galaxies, Washington: Carnegie Institution of Washington Publications, 1981.

  14:11:09, subject to the Cambridge Core

  .012

  10. Atlases of Galaxies, Picturing “Island-Universes”

  219

  Fig. 10.7 Interior pages of The Revised Shapley-Ames Catalog of Bright Galaxies, Sandage and

  Tammann (1981). With permission of Carnegie Science.

  from I to V, in decreasing luminosity; for example, a luminous spiral galaxy would have

  the extended descriptor Sbc I-II.33 To make Hubble morphological types more inclusive

  descriptors, the RSA listing incorporated van den Bergh luminosity classes, called by

  Sandage the “beauty” criterion. American astronomers Debra Meloy Elmegreen and Bruce

  Gordon Elmegreen, who later developed a spiral-arm classification system of their own,

  highlighted this aspect, by introducing very visual descriptors such as “grand design” ver-

  sus “flocculent” galaxies.34

  As an important add-on of the RSA, the authors included a list of 822 new galaxies with

  photographic magnitudes of less than 13.2, which expanded on the original Shapley–Ames