Page 5 - i1052-5173-32-10_FY22_new

to see things for which one is not looking.
For example, quartz xenocrysts are com- 4 cm
mon in andesites and are now recognized as
clear evidence of magma mixing, but this
was widely ignored for decades because
fractional crystallization was the paradigm
under which volcanic rocks were inter-
preted. The emergence of Sr-isotopic stud-
ies in the 1970s showed that this was incor-
rect. Similarly, for decades low-angle normal
faults were mapped as thrust faults, uncon-
formities, or gravity slides, because low-
angle extensional faults were regarded as
mechanically impossible. Although the
mechanics of low-angle normal faults are
still problematic, abundant and compelling
geologic evidence has led to broad accep-
tance of their existence (e.g., Collettini, 2011).
Therefore, as also advocated by Chamber-
lin, keeping alternative hypotheses—that is,
alternative explanations—in mind is impor- Figure 1. In granites worldwide, accumulations of K-feldspar such as this have been
interpreted variously as slurries deposited on a magma chamber floor, as concentra-
tant in fieldwork. A particular formation tions produced by shear sorting during magma flow, and as masses that rose buoy-
may be used as a benchmark to assign adja- antly within a magma chamber. These interpretations can be ruled out on the basis of
phase equilibria, mineral chemistry, volcanic petrology, and basic physics. Are the
cent strata to other formations and work out rocks lying? Photo courtesy of Bryan Law.
the geologic structure. If evidence arises
indicating that the rocks have been over-
turned, the stratigraphic assignments and been taught about plutonic systems and to Cleland (2001, p. 988) stated, “The famous
structure must be revised. If one’s mind is develop new explanations for how they work. Popperian directive to bite the bullet and
closed to this possibility, then “all hope of Bateman and Chappell (1979) had proposed a reject the hypothesis in the face of a failed
progress is gone.” model, widely reproduced in textbooks, in prediction has no logical force,” owing to
which the TIS was intruded in several distinct auxiliary conditions on the test. In our
Tests of Field-Based Hypotheses: pulses, each of which shoved aside older, but case, we concluded that our data were not
Are the Rocks Lying? still partially molten, material. This hypothe- precise enough to show the true small age
Data for testing field-based hypotheses sis makes several predictions, including that range that had been predicted.
can come from multiple sources, including (1) construction should have taken <~1 m.y.; Meanwhile, other conflicts with the
laboratory analyses, remote sensing, and (2) ages within a single pulse, and therefore a standard model arose. We set out to mea-
geophysical imaging. This paper focuses on single map unit, should cluster even more sure vertical variations in the Half Dome
conflicts that can arise between field-based tightly; and (3) such large magma chambers Granodiorite of the TIS over its 1800 m of
observations and information from these should show vertical gradients in composi- local relief, expecting to find gradients in
other sources. In cases of disagreement, a tion. However, predictions 1 and 2 were con- geochemistry, mineralogy, and xenolith
field-oriented geologist might insist that “the tradicted by a spread of 10 m.y. in low-preci- abundance consistent with processes in a
rocks don’t lie” and, on that basis, dismiss sion ages for the TIS (Kistler and Fleck, 1994, magma chamber the size of the mapped plu-
inconsistencies with the field-based hypoth- their fig. 14), far longer than predicted by the ton. This effort failed; we found none of the
esis (Fig. 1). Nevertheless, what the rocks Bateman and Chappell model. predicted vertical gradients (Gray et al.,
indicate (what they “have to say to us”) may In 1994 we collected samples from the 2008), nor did Putnam et al. (2015) find
be misunderstood. Field-based interpreta- western side of the TIS for analysis using them in the 1-km-tall southeast face of El
tions that are inconsistent with results from more advanced U-Pb techniques. In keep- Capitan. Mahan et al. (2003) concluded that
other disciplines must be questioned, and ing with the nested-construction hypothe- the McDoogle pluton south of Yosemite
inconsistencies should be used to drive the sis, we predicted that the ages would reveal formed by amalgamation of vertical sheets,
development of new hypotheses. a duration of ~1 m.y., but our results instead rather than having been intruded in one
matched the eastward-younging 10-m.y. large pulse. Contacts between sheets are
FAILED HYPOTHESES range of the earlier ages (Coleman and only noticeable where marked by screens of
ABOUT THE CONSTRUCTION Glazner, 1997). We had been taught that wall rock, and this observation planted a
OF PLUTONIC SYSTEMS science works by falsifying hypotheses, seed: Might there also be indistinct contacts
We begin by summarizing how new data but rather than rejecting the standard in plutons that lacked wall-rock screens to
collected during work on the Late Cretaceous model in light of these data, we sought mark them?
Tuolumne Intrusive Suite (TIS; Fig. 2) in other explanations. This is standard prac- In 2000–2001, we used yet higher-preci-
Yosemite National Park forced three of the tice, although it conflicts with the scien- sion analytical methods to date new sam-
authors to abandon much of what they had tific method as commonly understood; ples from the western side of the TIS. These

www.geosociety.org/gsatoday 5