Geoscience Reference
In-Depth Information
Gilbert and his assistant conducted careful geological, magnetic, and topograph-
ic surveys of Coon Mountain. Contrary to his expectation, the volume of materi-
al ejected from the crater turned out to be roughly the same as the volume of the
crater itself, apparently satisfying his first test and suggesting a steam eruption. He
found no anomalous magnetism, satisfying the second test. But neither did he find
any actual evidence of volcanism. Now we are about to be reminded that to err is
human, even for the greatest among us.
Most geologists, having found no evidence of volcanism but plenty of evidence
of impact in the form of the abundant meteorite fragments, would have logically
concludedthattheoriginofCoonMountainwasatleastadraw,ifmeteoriteimpact
were not the outright winner. Few would have chosen the cause for which there
was no direct evidence while ignoring the alternative for which the evidence was
unmistakable. But Gilbert had set up an inexorable train of logic that demanded
that very choice. This forced him to endorse a process for which he had estimated
the odds as eight hundred to one against: “The shower of meteoric iron was sub-
attributed the near-equal volume of crater and ejecta to “coincidence,” but he did
not.
In 1896, Gilbert published a paper titled “The Origin of Hypotheses, Illustrated
Gilbert had delivered the paper in 1895 from the bully pulpit of the president of
the Geological Society of Washington, D.C., and had given talks on the subject to
theNationalGeographicSociety,theNationalAcademyofSciences,theAmerican
AssociationofScience,andseveraluniversities.Thushewasonrecord,repeatedly
and in distinguished company, with his claim that Coon Mountain is volcanic. As
William Morris Davis sagely observed in his memorial: “A reader might indeed
contend with good reason that [Gilbert] seemed more interested in the abstract dis-
cussion of his problem than in any concrete result.”
18
Near the end of his 1896 paper, Gilbert opened the door to meteorite impact as
the cause of Coon Mountain, reporting that a colleague had found that the iron
meteorite fragments differed slightly from one another. Thus, Gilbert wrote, the
meteorites might have been embedded in some larger nonmagnetic object, “like
plums in an astral pudding,” leaving the hypothetical buried star with much lower
magnetism thanhehadoriginally supposed.Inthatcase,theabsence ofmagnetism
at the crater site might not disprove the impact theory after all. Not only that, but
perhaps some of the target rocks had been “condensed by shock so as to occupy
less space,” which would mean that the test based on the volume of ejecta could
also have been misleading. Gilbert noted that these ideas were “eminently pertin-
