ATLAS OF THE UNIVERSE
Different Types of Stars
A
sk any professional astronomer to name the most
valuable scientific instrument at his or her disposal, and
the reply will probably be: ‘The spectroscope.’ Without
them our knowledge of the stars would indeed be meagre.
Since the stars are suns, it is logical to expect them to
show spectra of the same type as our Sun. This is true, but
there are very marked differences in detail. For example, the
spectra of white stars such as Sirius are dominated by lines
due to hydrogen, while the cool orange-red stars produce
very complex spectra with many bands due to molecules.
Pioneering efforts were made during the 19th century to
classify the stars into various spectral types. The system
adopted was that drawn up at Harvard, where each type of
star was given a letter according to its spectrum. In order of
decreasing surface temperature, the accepted types are W,
O, B, A, F, G, K, M and then R, N and S, whose surface
temperatures are much the same (nowadays types R and N
are often classed together as type C). In 1998 two new
types, L and T, were added to accommodate very cool red
stars (‘brown dwarfs’) of mass no more than^1 / 20 that of the
Sun. The original scheme began A, B...but complications
meant that the final sequence was alphabetically chaotic.
There is a mnemonic to help in getting the order right:
‘Wow! O, Be A Fine Girl Kiss Me Right Now Sweetie.’
Each type is again subdivided; thus a star of type A5 is
intermediate between A0 and F0. Our Sun is of type G2.
In 1908, the Danish astronomer Ejnar Hertzsprung
drew up a diagram in which he plotted the luminosities of
the stars against their spectral types (plotting absolute
magnitude against surface temperature comes to the same
thing). Similar work was carried out in America by Henry
Norris Russell, and diagrams of this sort are now known
as Hertzsprung–Russell or HR Diagrams. Their impor-
tance in astrophysics cannot be overestimated. You can
see at once that most of the stars lie along a band running
from the top left to the bottom right of the diagram; this
makes up what is termed the Main Sequence. Our Sun is a
typical Main Sequence star. To the upper right lie giants
and supergiants of tremendous luminosity, while to the
lower left there are the white dwarfs, which are in a differ-
ent category and were not known when HR Diagrams
were introduced. Note also that most of the stars belong to
types B to M. The very hottest types (W and O) and the
very coolest (R, N and S) are relatively rare.
It is also obvious that the red and orange stars (conven-
tionally, though misleadingly, referred to as ‘late’ type) are
of two definite kinds; very powerful giants and very feeble
dwarfs, with virtually no examples of intermediate lumin-osity. The giant-and-dwarf separation is less marked for
the yellow stars, though it is still perceptible; thus Capella
and the Sun are both of type G, but Capella is a giant,
while the Sun is ranked as a dwarf. The distinction does
not apply to the white or bluish stars, those of ‘early’ type.
The stars show a tremendous range in size, temperature
and luminosity. The very hottest stars are of type W; they
are often called Wolf–Rayet stars, after the two French
astronomers who made careful studies of them over a centu-
ry ago, and have surface temperatures of up to 80,000
degrees C. Their spectra show many bright emission lines,
and they are unstable, with expanding shells moving out-
wards at up to 3000 kilometres (over 1800 miles) per sec-
ond. O-type stars show both emission and dark lines, and
have temperatures of up to 40,000 degrees C. At the other
end of the scale we have cool red giants of types R, N and S,
where surface temperatures are no more than 2600 degrees
C; almost all stars of this type are variable in output.
Some supergiants are powerful by any standards; S
Doradûs, in the Large Cloud of Magellan – one of the clos-
est of the external galaxies, at a distance of 169,000 light-
years – is at least a million times as luminous as the Sun,
though it is too far away to be seen with the naked eye.
Even more powerful is the erratic variable Eta Carinae,
which may equal 6 million Suns and has a peculiar spec-
trum which cannot be put into any regular type. On the
other hand a dim star known as MH 18, identified in 1990
by M. H. Hawkins at the Royal Observatory Edinburgh,
has only^1 /20,000the luminosity of the Sun. The range in
mass is not so great; the present holder of the ‘heavy-
weight’ record seems to be Plaskett’s Star in Monoceros,
which is a binary system with two O7-type components,
each of which is about 55 times as massive as the Sun.
Direct measurements of star diameters are very diffi-
cult. The stars with the greatest apparent diameters are
probably Betelgeux in Orion (at 0.044 arc second) and the
red variable R Doradûs (0.057 arc second). New direct
measurements are being made by a team led by John Davis,
who built SUSI, the Sydney University Stellar
Interferometer; this is made up of a number of relatively
small telescopes working together, and can measure the
width of a human hair from a distance of some 100 kilome-
tres (over 60 miles). It has even become possible to detect
surface details on a few stars. At present the largest known
stars are the red supergiants KW Sagittarii (distance 9800
light-years), KY Cygni (5200 light-years) and V354 Cephei
(9000 light-years), each with a diameter of 1.5 thousand
million km. Fourth comes the ‘Garnet Star’, Mu Cephei.SUSI:the Sydney
University Stellar
Interferometer, set up
at Narrabri in New South
Wales under the direction
of John Davis. (Photograph
by the author, January 1994.)
This is designed to measure
the apparent diameters
of stars, and is amazingly
sensitive.152-191 Atl of Univ Phil'05 7/6/05 1:22 pm Page 172