The Role of Images in Astronomical Discovery Page 18
images of “nebulae” were loosely viewed as flattened whirlpools of condensing cosmic
matter. Furthermore, if they were proto-planetary systems, “nebulae” could only be rela-
tively nearby, thus part of the Milky Way, and not the island-universes Kant hypothesized
about.
Here is how Kant presented his hypothesis of an interstellar cloud of matter collapsing
under the action of gravity.
In the greatly expanded space in which the spread out elementary basic material prepares devel-
opments and systematic movements, the planets and comets are built up only out of those parts
of the elementary basic matter moving downward towards the central point of the force of attrac-
tion, which, through their fall and the reciprocal interaction of the particles collectively, were pre-
cisely adjusted for the velocity and direction required for orbital motion. . . . Now, because these
lighter and volatile parts are also the most effective at maintaining a fire, we see that, with their
addition the body at the central point of the system has the distinction of becoming a flaming
sphere, in a word, a sun. By contrast, the heavier and inert materials and those particles which are
poor fuel for a fire will make planets which are robbed of these properties merely cold and dead
clusters.15
I quote the extensive passage because many researchers and the “grand amateurs”
of the nineteenth and early twentieth centuries were bewitched by this impressive
image of the mind and imposed it on the “nebulae” they observed, sketched and pho-
tographed. The Nebular Hypothesis gained strength and popularity when, in 1796, Pierre
Simon Laplace presented a rigorous mathematical concept in Exposition du système
du monde.16
On the Road of Confusion
The observations conducted at Birr Castle constituted an observing program in the mod-
ern sense: first, a systematic search and examination of the objects found by Herschel and
catalogued in the General Catalogue of nebulae and clusters. Because of the controversy
surrounding the reality of several objects, the first step of the work was to confirm the exis-
tence of these objects; a few were found not to exist, to have been incorrectly located or to
be duplicates. Beyond the verification process, the Birr Castle observers paid meticulous
attention to the description of the objects and to drawing them. Importantly, William Par-
sons’ initial motivation was to resolve the “nebulae” into stars with the Leviathan; the plan
was soon derailed with the discovery of spirality in dozens of them.
In his long summary article of 1878, Lawrence, the Fourth Earl of Rosse, Parsons’ son,
never even hinted at the nature of these “nebulae.” The reporting was strictly factual. One
14 S. Schaffer, On Astronomical Drawing, in Picturing Science, Producing Art, C. A. Jones and P. Galison (editors), New York: Routledge, 1998, pp. 441–474.
15 I. Kant, Universal Natural History and Theory of the Heavens (translated by Ian Johnston), Arlington: Richer Resources Publications, 2008, pp. 105–106.
16 P. S. Laplace, Exposition du système du monde, Paris: Imprimerie du Cercle-Social, 1796.
14:05:49, subject to the Cambridge Core
.007
112
Part II – Images as Galaxy Discovery Engines
can, however, imagine the hours of discussion and speculation among the members of the
close-knit team. Outside, the “nebular” controversy and the confusion about what they were,
unresolved groupings of stars, some sort of shining fluid or such fluid mixed with stars,
continued to rage. Paradoxically, as imaging techniques became more sensitive, with larger
and better telescopes and the increased use of the photographic plates, the confusion seemed
to be amplified: some “nebulae” were shown or claimed to be resolved into stars, while
most remain “cloudy.” Proponents of the “local” hypothesis argued that if “nebulae” were
resolved into stars, they had to be relatively nearby, that is part of the Milky Way system.
The confusion of the end of the nineteenth century spilled over into the first decades of the
twentieth century, as swings of opinion favored first the local hypothesis, then the universe-
islands view, back and forth.
Resolving “nebulae” into stars became a battleground. The “resolution” craze was stim-
ulated by claims by the Fourth Earl of Rosse and George Bond that parts of the Orion Neb-
ulae split into numerous stars. “Resolving stars” reinforced the century-old argument put
forward first by Galileo: with sufficient optical power, all nebular forms can be resolved into
individual stars. Robert Smith quotes a triumphant passage from George Bond’s September
1847 observing notes on the Orion Nebula. An excited Bond, happy to show the power of
the new Harvard College 15-inch refractor, exclaimed: “Resolved. Mottled. Abundance of
Stars.”17
Claims of resolution of other “nebulae” into individual stars were heard repeatedly. Both
proponents and opponents of the extragalactic hypothesis used the partly fallacious argu-
ment of resolution to support either the local or external version of the nature of “nebulae.”
Current insights on how so many claimed to have resolved individual stars suggest that the
optics of the telescope used, especially metallic mirrors, combined with variable “seeing,”
could have generated an apparent granularity to the image as viewed in the eyepiece of
these instruments. Stated rather bluntly by the Pulkovo Observatory director Otto Wilhelm
Struve, “the alleged miracles of resolution are nothing but illusions.” He was right. We have
here a case of images turning out to be unreliable conveyors of reality.
At the top of the pile of nebular puzzles, observers claimed to have detected a variability
of brightness in a few nebulae, in some instances fading over time. Although there were
a few cases of true nebular variability, due to erupting dust-embedded protostars, most of
these were bogus. In particular, the several claims of the variability of the Orion Nebula were
all erroneous. Nevertheless, changes in apparent brightness were used as a strong argument
in favor of the local hypothesis. The argument, and it was logical, was that no giant systems
of stars could cohesively vary in unison. Simon Schaffer has recounted this fascinating
story, highlighting the debate among astronomers surrounding the elusive perception of
changes.18
17 R. W. Smith, The Expanding Universe, Astronomy’s Great Debate 1900–1931, Cambridge: Cambridge University Press, 1982, p. 45
18 S. Schaffer, On Astronomical Drawing, in Picturing Science, Producing Art, editors C. A. Jones and P. Galison, New York: Routledge, 1998, pp. 441–474.
14:05:49, subject to the Cambridge Core
.007
5. The One-Thousand-Year Journey
113
The Arrival of Spectroscopy
A transformational technique was introduced during the first decades of the nineteenth cen-
tury: spectroscopy was employed to spread the light of stars and “nebulae” as a function
of wavelength and to analyze it.19 German physicist and optician Joseph von Fraunhofer
(1787–1826) had invented the spectroscope in 1814. He explored the light from the flames
of several substances and analyzed sunlight, discovering 574 “dark lines” in its spectrum.
&nbs
p; Another “grand amateur,” British astronomer William Huggins (1824–1910) and his wife
Margaret Lindsay Huggins (1848–1915), pioneered stellar spectroscopy. They found that
the spectra of celestial objects, e.g. positions or wavelengths of the various “dark lines,”
resembled those of terrestrial substances; they concluded that stars contain some of the
same elements found on Earth.
Applying spectroscopy to the “nebulae” was difficult as the objects were extended and
very faint. Better optical designs finally made the avenue fruitful, and more telescopes
were equipped with spectrographs, and “nebulae” became prime targets for spectroscopists.
What the nebular spectra initially showed did not rally viewpoints but strengthened diver-
gence. In 1864, William Huggins obtained a spectrum of a planetary nebula in the constel-
lation of Draco. It was totally different from the spectra of the stars (Plate 5.1). A single
strong emission line dominated the spectrum, a clear indication that some sort of hot ten-
uous gas was the emitting source. Also differing from the Sun and stars that produced a
continuous range of colours broken by “dark lines,” all spectacular nebulae, like the Orion
Nebula, showed monochromatic emission: the light, dispersed by the spectroscope, was
concentrated over a very narrow range of wavelengths – called “emission lines,” as they
were bright, colourful features, the reverse of the dark lines seen in the solar or stellar spec-
tra. Observing other “nebulae,” it became obvious that many of them appeared to be made of
a gaseous and fluorescent substance. Huggins wrote with assurance “The riddle of the neb-
ulae was solved. . . . the light of this nebula had clearly been emitted by a luminous gas.”20
No one knew then that stars were also gaseous but of much higher density than nebulae.
The first spectroscopy results, which showed “nebulae” to be some sort of shining fluid,
appeared to support the local hypothesis. However, Huggins and Margaret Lindsay soon
found that “nebulae” displayed a range of spectral properties, with many others displaying
instead continuous spectra, similar to those of the Sun and of other stars. Hence, in the
end, they remained cautious about the nature of “nebulae.” It was wise. Too many were
committing again and again the sin of putting everything in the same bag.
How the Milky Way Became the Universe – For a While
William Herschel had already made the observation that the band of the Milky Way was
relatively devoid of “nebulae,” while large portions of the sky above and below the plane
19 J. B. Hearnshaw, The Analysis of Starlight: One Hundred and Fifty Years of Astronomical Spectroscopy, Cambridge: Cambridge University Press, 1986.
20 W. Huggins and Lady M. Huggins, The Scientific Papers of Sir William Huggins, London: Wesley & Son, 1909, p. 106.
14:05:49, subject to the Cambridge Core
.007
114
Part II – Images as Galaxy Discovery Engines
of the Milky Way were sprinkled with them. In the stubborn tradition of seeing humankind
at the center of all things, the apparent symmetry in the distribution of “nebulae” about the
Milky Way plane was hailed as an irrefutable argument for the local hypothesis. Nebulae
had to belong to the Milky Way in order to display such a symmetrical distribution. Favor-
ing this argument, English writer Richard Proctor (1837–1888) posited that this “zone of
avoidance” was an indication that spirals were part of the Milky Way: How could the “neb-
ulae” assemble this way around if they were totally independent of it? How would the outer
worlds know we were here if they were scattered across huge distances?
This galactocentric view gained support in the 1910s, when Shapley noted that globular
clusters of stars also obeyed the exclusion from the avoidance zone (Chapter 4). Globu-
lar clusters are giant, tight assemblies of hundreds of thousands, even millions, of stars
in systems of spherical shape. Shapley’s argument was that the zone of avoidance corre-
sponded to the high density of stars. Consequently, globular clusters went missing because
they were torn apart by tidal forces when they entered the region and were destroyed. A
convinced Milky Way centric, Shapley extended the same argument of destruction to spi-
rals, reinforcing his opinion that spirals were members of our galaxy. His inference for a
super Milky Way was based on the assumption that sidereal space is completely transpar-
ent: there is nothing to block or absorb starlight as it travels thousands of light-years of
interstellar space. Shapley’s explanation for the missing spirals in the band of the Milky
Way was completely false. At that time there was little recognition that interstellar dust
was severely blocking our view in several directions of the sky.
As the nineteenth century came to a close, the resolution of “nebulae” into stars, the
spectroscopic evidence of a gaseous composition and the discovery of variations in nebular
brightness appeared to carry the day for spirals, and hence all “nebulae,” to be local. The
enthusiasm for the local view grew in strength and boldness. Our old friend Isaac Roberts
encouraged the supporters of the Nebular Hypothesis to study his unique photographs of
the Andromeda Nebula “to see a new solar system in process of condensation from the neb-
ula.” No less enthusiastically, he envisioned the two small galaxy companions of M31 as
unique laboratories for cosmogony, as “the two [smaller] nebulae seem as though they were
already undergoing their transformation into planets.”21 Fanciful speculation was again get-
ting ahead of hard facts.
An unexpected phenomenon seemed to lend weight to “nebulae” being part of the Milky
Way. The “great nova” of 1885 in Andromeda became as bright as a tenth of the whole
Andromeda Nebula; most astronomers thought it inconceivable that a single star could
have outshone millions of stars, as would be the case for a distant object that consti-
tuted a multitude of stars. This puzzling new star just added to the quandary, and sug-
gested that the Andromeda Nebula had to be small and nearby, as argued by Shapley
during the “Great Debate.” We know today that S Andromedae or SN 1885 was not a
nova but a supernova, the first ever noted outside the Milky Way. As seen in Chapter 3,
21 I. Roberts, Photographs of the Nebulae in M32, h44, and h51 Andromedae, and M27 Vulpeculae, Monthly Notices of the Royal Astronomical Society, Vol. 49, p. 65.
14:05:49, subject to the Cambridge Core
.007
5. The One-Thousand-Year Journey
115
novae were well known because of their high frequency. Supernovae are much less fre-
quent events as they involve massive stars, which are sparser in population. When a mas-
sive star explodes, it releases, over a few weeks, as much light and mechanical energy
as the star ever did over its lifetime of millions of years. The luminosity of a supernova
can thus outshine an entire galaxy. This difference between novae and supernovae came to
be understood a few decades later, following the work of Walter Baade and Fritz Zwicky
in 1934.22
Robert Smith has nicely summarized the conflicting views and the triumphal mood of
the tenants of the local hypothesis at the turn of the century. “By
the late 1880s, then, an
astronomer could exploit the bright line spectrum of some nebulae, the peculiar distribution
of nebulae, the photographs of the Andromeda Nebula and the 1885 ‘nova’ to form the
seemingly overwhelming case against the island universe.”23
But things were about to change swiftly and dramatically in favor of the island-universe
concept. The early decades of the twentieth century saw the spectacular rebirth of the
externality of “nebulae” and the establishment of a world of galaxies. Oddly, the tri-
umphal change came first from spectroscopy, then later from imaging techniques that finally
allowed the measurement of accurate distances to the “nebulae.”
Moving “Nebulae” Out of the Milky Way
In 1899, the German astronomer Julius Scheiner (1858–1913) of the Potsdam Observatory
presented his spectroscopic results on the Andromeda Nebula in a succinct two-page paper.
“No traces of nebular lines are present, so that the interstellar space in the Andromeda neb-
ula, just as in our stellar system, is not appreciably occupied by gaseous matter.”24 Compar-
ing features of Andromeda with our own Milky Way, Scheiner suggested that the irregulari-
ties and “streams” of the Milky Way could be quite well accounted for if they were regarded
as a curl of “spirals.” He stated with certainty that spiral nebulae were giant star clusters;
the continuous spectrum of the decomposed nebular light was the synthesis of the light of a
multitude of unresolved stars. The island-universe proponents finally had spectroscopy on
their side. More spectroscopic evidence for the externality of most “nebulae” was to come
a decade later. Scheiner was not impressed with Roberts’ view that the Andromeda Nebula
was a forming protoplanetary system.
Nevertheless, to have irrefutable proof of the externality of most “nebulae” it was nec-
essary to overcome the barriers of determining accurate distances to them. Only this could
settle the enduring debate. A crucial observational wave developed to study and understand
variable stars. Until the twentieth century, variable stars were simply thought to be pairs
22 W. Baade and F. Zwicky, On Super-Novae, Proceedings of the National Academy of Sciences of the United States, 1934, Vol. 20, pp. 254–259.