[From Bruce Gregory (960517.1330 EDT)]
I relate this story because it illustrates the approach to science
that Bill P. champions so eloquently.
Each week during the academic year we have a colloquium on some
astrophysical topic. The talks range from the merely passable to the
very good, but yesterday's talk stood out in terms of its elegance
and message. For those of you you who are not quite up to speed on
astrophysics, I will lay a little groundwork.
Globular clusters are spherical collections of tens of thousands of
stars orbiting about the center of our galaxy and not confined to the
plain of the galaxy as are most the stars in the Milky Way. Mathematical
models of stellar evolution indicate that these clusters formed very
early in the history of the universe (so early that some estimations
of the age of the universe leads to a universe younger than the oldest
globular clusters -- a situation that causes astronomers distinct
unease.) The available evidence supports the notion that all the
stars in a cluster formed during a relatively short interval and are
thus of approximately the same age.
A star begins its life by converting hydrogen at its core into helium
(a process called hydrogen burning). The process is extremely
temperature sensitive and therefore is confined to a region near the
center of the star. Eventually the available hydrogen in the core is
consumed and the star adjusts it size and temperature by contracting
until it can burn helium in a shell surrounding the core. This
transformation causes the outer envelope of the star to expand and
cool and the star enters a "red giant" phase. The lifetime of a star
prior to its becoming a red giant is determined, according to our
models, solely by it mass (assuming a fixed chemical composition -- a
reasonable assumption for globular clusters that is consistent with
observations). The greater the mass, the shorter the lifetime.
According to our present understanding, the sun is approximately
halfway through its hydrogen-burning lifetime.
If a collection of stars was formed at the same time, the age of the
group can be determined by finding the most massive stars which are
not vet red giants and which therefore have not yet completed hydrogen
burning. The observations are then compared with theoretical
calculations to determine the age of the globular cluster. The
calculations have uncertainties linked largely to the compexities
of the absorption (and re-emission) of radiation as it makes its way
from the interior to the surface of the star. (The so-called "opacity
codes" used in these calculations employ a mysterious black art
originally nurtured by the desire to calculate the energy yields of
thermonuclear weapons.) The observed distributions of the luminosities
and temperatures of stars in globular clusters are consonant with the
calculated models of stellar evolution -- with a few anomalies. One
anomaly is the existence of "blue stragglers" -- a small (less than
200) stars, whose temperature and luminosity tell us they should have
entered their giant phase a long time ago. In other words, blue
stragglers appear to be much younger than their brethern. How have
these stars managed to resist the aging process? This question was the
focus of yesterday's talk.
The speaker identified two proposed solutions. The first involved
stellar collisions. The stars in globular clusters are separated by
average distances much smaller than those we find in the neighborhood
of the sun, and so collisions must be not as uncommon as they are in
most regions of the galaxy. It was conjectured that such
collisions might bring hydrogen to the core of the resulting star,
giving it a new lease on hydrogen-burning life and therefore making it
appear younger than its brethern. The second proposed mechanism
involved the merger of the stars in an extremely close ("contact")
binary, with similar redistribution of hydrogen. The question was
whether either proposed mechanism would work in a simulation. The
contact binary approach is extremely difficult to simulate because of
the complexities of rotation in the presence of magnetic fields,
leaving the collisional simulation much easier to do. The speaker
hoped the results would be persuasive enough that the binary
mechanism could be ignored. He show us the results of his
calculations of the outcome of collisions between stars.
Much to his surprise (and ours) the helium remained concentrated in
the core of the star emerging from the collision. No significant
redistribution of hydrogen occurred. The proposed mechanism simply
did not work and therefore could not explain the blue-straggler
phenomenon. (He was able to explain these findings quite straight-
forwardly in terms of the entropy associated with the helium cores.
As someone pointed out after the talk, studies in stellar evolution
seem often to follow the pattern of a very difficult and complex
calculation which leads to a result that it seems might have been
predicted on first principles, were we only smart enough to have
developed the argument before carrying out the simulation.)
The speaker showed us a transparency he had used when first talking
about these results some six months ago. On it he outlined several
possible "ways out" of the dilemma. Since the earlier talk he had
explored those mechanisms and found that they too did not work. In
the words of Rick's alter-ego T.H. Huxley, we were witness to "the
great tragedy of Science -- the slaying of a beautiful hypothesis by
an ugly fact." The speaker had set out to show that the collisional
hypothesis could explain the blue-straggler phenomenon. He wound up
showing in a very convincing manner that the hypothesis did not work
and had to be abandoned.
All colloquia end with a round of applause for the speaker. Often
this demonstration is polite, but yesterday it was enthusiastic. We had
just heard and seen an elegant demonstration of science working the
way it was supposed to. It was a pretty thing to see.
Bruce G.