The Role of Images in Astronomical Discovery Page 19
23 R. W. Smith, The Expanding Universe, Astronomy’s Great Debate 1900–1931, Cambridge: Cambridge University Press, 1982, p. 5.
24 J. Scheiner, On the Spectrum of the Great Nebula in Andromeda, The Astrophysical Journal, 1899, Vol. 9, pp. 149–150. The spectrum of Messier 31 had required a 7.5-hour exposure.
14:05:49, subject to the Cambridge Core
.007
116
Part II – Images as Galaxy Discovery Engines
of stars, i.e. double stars, eclipsing each other, with one star passing in front of the other
and dimming its light. However, it was possible that the variability in brightness could
have other causes. It took time for astronomers to realize and demonstrate that many stars
underwent radial pulsations: that is periodic expansion and contraction of the whole star. A
particularly important type of variable pointed towards the light at the end of the tunnel.
Exploding stars such as novae and supernovae are variable, but they are not pulsating.
Even as irregular variables, novae can be used as “standard candles.” These objects belong
to the same class; they have a known brightness and variable behavior, which are a function
of time (light curves), as based on direct measurements; for example, a nearby object of
the same family whose distance has been established by the parallax method. In particular,
they appeared to attain roughly the same maximum absolute magnitude. By comparing this
known luminosity to a distant object’s observed brightness, the distance of the latter can be
derived using the inverse square law. Heber Curtis, George Ritchey and Knut Lundmark,
a young Swedish astronomer who was then working in the United States, studied novae
in “nebulae,” as detailed in Chapter 3. The three astronomers published independently on
novae they found in “nebulae.” Novae were recognized in Messier 31 as clones of galactic
ones. By 1917, Ritchey, Curtis and Lundmark had estimated the distances of those faint
novae to be about 500,000 to 650,000 light-years, well outside the limit of even the super
Milky Way of Shapley. These conclusions were based on the superb images provided by
the refurbished 36-inch Crossley at the Lick Observatory and the new 60-inch on Mount
Wilson.
Speeding “Nebulae”
With studies on novae paving the way, two other independent directions of study led to
the determinations of distance that put most “nebulae” well outside the boundaries of the
Milky Way. In a splendid paper published in 1917, Vesto Slipher reported on the large
radial velocities he had derived from the spectroscopy of several “nebulae.” Slipher’s spec-
troscopic observations were the results of a painstaking and relentless program using the
24-inch refractor at the Lowell Observatory, Arizona. Slipher was as patient as he was quiet,
consistently compiling multiple nights of exposure to get 20 to 40 hours of photons trickling
onto each photographic plate. To obtain the velocities of the “nebulae,” he used the Doppler
effect, which shifts the position of spectral lines in proportion to their relative radial veloc-
ities, along the line of sight. He derived the movements of the “nebulae” with respect to the
Milky Way. What Slipher found was phenomenal.
Although he measured a few objects like the Andromeda Nebula to be moving towards
us, the majority (17 out of 21) were receding at high velocities, in some cases reaching
1,100 km/s: Slipher surmised most correctly that at such speeds, these objects could not
be retained by the gravity of the Milky Way. At the end of his short note, he expressed his
opinion: “It has for a long time been suggested that the spiral nebulae are stellar systems
seen at great distances. This is the so-called ‘island-universe’ theory, which regards our
14:05:49, subject to the Cambridge Core
.007
5. The One-Thousand-Year Journey
117
stellar system and the Milky Way as a great spiral nebula that we see from within. This
theory, it seems to me, gains favor in the present observations.”25
Also leading the charge in distance determination, the brilliant Estonian astronomer
Ernst Öpik (1893–1985) undertook a different and very original approach, the results of
which he published in 1922 (Fig. 5.3). He was one of the most visionary astrophysicists
of the first half of the twentieth century. Öpik analyzed the internal rotational motion of
Messier 31, the “great nebula” in Andromeda, using spectroscopic data. He inferred its
luminous mass from its surface brightness, assuming that the energy radiated by stars had
the same behavior as in the Milky Way and that the velocities of stars in the system obeyed
Newtonian dynamics. Employing simple and clever assumptions, he derived the distance
to the Andromeda Nebula to be about 1.5 million light-years, his value being the closest
of its day to the modern one of 2.6 million light-years. Öpik’s method, known as the virial
theorem (Chapter 8), is still in use today by researchers studying distant galaxies and galaxy
clusters. Slipher and Öpik had confirmed the extragalactic scenario based on solid physical
arguments. Now, let us return to variable stars.
How Henrietta Leavitt Gave the Key to Edwin Hubble
By the late eighteenth century, hundreds of regular variable stars had been identified and
monitored. Harvard College Observatory, under its energetic directors Edward Charles
Pickering (1848–1919) and his successor, Shapley, led the effort. Some variables were dou-
ble stars, true eclipsing binary systems due to the orientation of their orbital plane with
respect to us, but most were of a different nature. In the 1920s, as stellar interiors and
atmospheres became better understood, regular variables were explained as being caused
by radial pulsations: the star’s volume and temperature change in a steady cycle of ups
and downs, causing variable luminosity. This turned out to be especially true for a type
of variable star called a Cepheid, the name coming from the first known representative in
the constellation of Cepheus. The British amateurs Edward Pigott (1753–1825) and John
Goodricke (1764–1786) discovered this special class of variables in 1784; that same year,
Goodricke measured that δ Cephei varied by one magnitude over a period of 5.37 days.
Sadly, the young Goodricke died from pneumonia, which he contracted during the numer-
ous nights of observing variable stars in cold England; he was only 22.
Cepheid variable stars had thus been known for more than a century, but their obser-
vational properties were not fully understood. A detailed study of Cepheid variables by
Harvard College Observatory astronomer Henrietta Swan Leavitt (1868–1921) changed
everything regarding the astrophysical importance and use of these stars (Fig. 5.4).26 Hen-
rietta Leavitt worked as a “computer” under Edward Pickering at $10.50 per week. She was
assigned to count images on photographic plates, and to make photometric measurements
of variable stars.
25 V. M. Slipher, Nebulae, Proceedings of the American Philosophical Society, 1917, Vol. 56, pp. 403–409.
26 G. Johnson, Miss Leavitt’s Stars: The Untold Story of the Woman Who Discovered How to Measure the Universe, New York: W. W. Norton & Company, 2005.
14:05:49, subject to the Cambridge Core
.007
118
Fig. 5.3 Ernst Öpik observing with the 8-inch Zeiss refractor of Tartu Observatory, Estonia. Credit:
Tartu Observatory (University of Tartu Museum Collections, UAMF 211_15) and Armagh Observa-
tory and Planetarium.27
27 This image is very similar to a photograph published in Tartu Tahetomi Kalendar XXI, 1944.
14:05:49, subject to the Cambridge Core
.007
5. The One-Thousand-Year Journey
119
Fig. 5.4 Henrietta Leavitt. Courtesy of the American Association of Variable Star Observers
(AAVSO).28
In 1908, Leavitt published her findings of the analysis of many photographic plates of
the stars in the Magellanic Clouds, which had been obtained with the Bruce 24-inch tele-
scope at Arequipa, Peru. Among 1,777 variables, she identified a class of variable stars that
obeyed a surprisingly close relation between the luminosity and the period; while appar-
ently fainter, they had similar features to the well-known galactic Cepheids.29 The most
luminous variable stars had the longest periods of variability, and the fainter one, the short-
est. As these stars are on average very luminous, they could be seen to great distances and
used as reliable standard candles. By simply measuring the period of a cycle of pulsation,
the luminosity of the star could be derived in a straightforward manner (Fig. 5.5). It was
magnificent work co-authored by Pickering.30 Leavitt had provided the magic key to unlock
the safe for measuring large cosmic distances. Unfortunately, Leavitt received little recog-
nition in her time.
28 Photograph of Henrietta Leavitt as it appears in Popular Astronomy, 1922, Vol. 30, p. 197.
29 H. S. Leavitt, 1777 Variables in the Magellanic Clouds, Annals of the Harvard College Observatory, 1908, Vol. 60, pp. 87–108.
30 H. S. Leavitt and E. C. Pickering, Periods of 25 Variable Stars in the Small Magellanic Cloud, Harvard College Observatory Circular, 1912, Vol. 173, pp. 1–3.
14:05:49, subject to the Cambridge Core
.007
120
Part II – Images as Galaxy Discovery Engines
Fig. 5.5 Light curves of four Cepheid variables in the nearby galaxy Messier 33. From Hubble
(1926), Contributions from the Mount Wilson Observatory.
Leavitt’s work fundamentally changed the paradigm of distance determination by pro-
viding a very reliable ruler with which to measure the universe. Her work made huge impact.
The game of the day became measuring distances using Cepheids as standard candles.31
The triumphal act was for someone other than Leavitt. The American astronomer Edwin
Powell Hubble (1889–1953) turned the Cepheid “key” in a majestic way and unlocked the
distances to the galaxies. These were indeed staggering.
Hubble Cuts the Gordian Knot
Appointed at Mount Wilson Observatory in 1919, the young Hubble rapidly gained access
to the newest and most powerful telescopes, the 60-inch and 100-inch of Mount Wilson
Observatory (Chapter 3) (Fig. 5.6). The powerful telescopes were located on the dry moun-
tains of the San Gabriel range just northeast of the city of Los Angeles. The night sky
was then extremely dark and the stable atmosphere provided telescopic images of great
definition. Hubble achieved his epoch-marking determination of extragalactic distances by
using the newly calibrated variable stars, pulsating Cepheids. As stated by authors Harry
Nussbaumer and Lydia Bieri, Hubble “cut the Gordian knot.”32
Hubble’s Ph.D. thesis topic had been the photographic investigations of faint nebulae
based on work with the Yerkes Observatory 24-inch reflector, designed and built by Ritchey
31 H. S. Leavitt, 1777 Variables in the Magellanic Clouds, Annals of the Harvard College Observatory, 1908, Vol. 60, pp. 87–108.
32 H. Nussbaumer and L. Bieri, Discovering the Expanding Universe, Cambridge: Cambridge University Press, 2009.
14:05:49, subject to the Cambridge Core
.007
5. The One-Thousand-Year Journey
121
Fig. 5.6 Edwin Powell Hubble scanning a photographic plate. Credit: Image courtesy of the Obser-
vatories of the Carnegie Institution for Science Collection at the Huntington Library, San Marino,
California.
(Chapter 3). Returning from World War I work, Hubble started using the recently commis-
sioned 100-inch Hooker reflector, initiating his formative program to observe “nebulae” in a
systematic way. Following the works of Ritchey, Lundmark and Curtis, Hubble was expect-
ing to find lots of new novae in the “nebulae” he was photographing. He obtained several
photographic plates and compared them with archival ones, especially those acquired with
the 60-inch, as this telescope had been in operation since December 1908.
Novae he found, but one peculiar object in Messier 31 caught his attention, as the same
object was visible on plates taken a few years earlier (e.g. 1913). It could not be a nova
since novae are only very rarely recurrent, and when they are, it is after periods of many
years of dormancy. Initially, he erroneously marked the object as a nova with “N” for nova;
he subsequently crossed it out with an X and correctly identified the star as a Cepheid
variable with a period of 31.4 days (Fig. 5.7). Soon Hubble found several other Cepheids.
Using Leavitt’s period–luminosity relation, he derived a distance of 680,000 light-years to
the Andromeda Nebula. He prepared his work for publication; it was near the end of 1924.
14:05:49, subject to the Cambridge Core
.007
122
Part II – Images as Galaxy Discovery Engines
Fig. 5.7 Transformational Image: Cepheid V1 and Novae in Andromeda (Messier 31). Plate
H335H of Messier 31 obtained by Edwin Hubble on the night of October 5–6, 1923. The letters N
indicated novae. The top right N was crossed out and marked instead “VAR!”. Hubble originally
thought the star was a nova, but eventually discovered that it varied in brightness like a Cepheid. This
is one of the most memorable images in the whole history of astronomy. Credit: Image courtesy of
the Carnegie Observatories.
By then, Hubble had identified 36 variables in Messier 31; 12 of them were Cepheids (see
Plate 5.2).
There are two interesting side anecdotes to the detection of Cepheids in “nebulae.”
Allan Sandage has recounted how a young observatory assistant, Milton Lassell Humason
14:05:49, subject to the Cambridge Core
.007
5. The One-Thousand-Year Journey
123
Fig. 5.8 Milton Humason. Credit: Image courtesy of the Observatories of the Carnegie Institution
for Science Collection at the Huntington Library, San Marino, California.
(1891–1972), who had started as a mule driver at Mount Wilson, became involved in the
observatory scientific work (Fig. 5.8). Humason was doing photographic plate measure-
ments for Harlow Shapley, who was then with the Mount Wilson Observatory. In the late
1910s, the curious Humason approached Shapley asking him if some of the variable stars he
had noticed in “nebulae” could be Cepheids. An incredulous Shapley exclaimed: “Impossi-
ble, the nebulae are part of the big Milky Way.”33 For Shapley, nothing existed outside his
Milky Way. Later, in 1924, in a letter to Hubble, who had informed him about the
Cepheids
he had found in Messier 31, the skeptical and almost satirical Shapley wrote back: “Your
letter telling of the . . . variable stars in the direction of the Andromeda nebula is the most
entertaining piece of literature I have seen for a long time.” As Owen Gingerich pointed out,
Shapley did not say “in” the Andromeda nebula, but “in the direction of . . . ”34 These sto-
ries, as several others, show how Shapley had a “talent” for missing the train, and as Walter
33 Story told in A. R. Sandage, Centennial History of the Carnegie Institution, Volume 1: The Mount Wilson Observatory, Cambridge: Cambridge University Press, 2004, pp. 495–598.
34 O. Gingerich, Through Rugged Ways to the Galaxies, Journal for the History of Astronomy, 1990, Vol. XXI, p. 78.
14:05:49, subject to the Cambridge Core
.007
124
Part II – Images as Galaxy Discovery Engines
Baade harshly commented, to produce works that “never went beyond trivialities.” Never-
theless, Shapley did produce some solid pieces of astronomical work. For many years, he
remained a staunch supporter of the local hypothesis, seeing everything as part of a super
Milky Way; this was in part due to the work of his friend at Mount Wilson, Adriaan van
Maanen (1884–1946), a colleague of Hubble, who claimed – wrongly – to have measured
the rotational motion of “nebulae” over short periods of time. If real, such motions would
have implied velocities close to that of light. Van Maanen’s measurements were erroneous
and his interpretation fallacious.
For puzzling reasons, Hubble did not attend the 1924 December meeting of the American
Astronomical Society to present his groundbreaking results. Instead, it was Henry Norris
Russell who read a communication of Hubble’s finding of “Cepheids in spiral nebulae” at
the session of 1 January 1925. Historian Marcia Bartusiak described that moment emphat-
ically as “the day we discovered the universe.”35 As has been described and emphasized
throughout this book, the “day” had been a very long day.
Continuing his work with the 100-inch of Mount Wilson Observatory, Hubble identified
and measured several more Cepheids in other “nebulae.” From observations of Messier
33 and NGC 6822, he quickly derived accurate distances for these systems and published