Obviously, if Austin wanted a sample that only represented the material that solidified during the 1986 eruption, he would have had to remove ALL of the plagioclase and other phenocrysts from the glass component.
I've attempted to separate very fined-grained minerals from glass in coal ashes by using magnetic separation and hydrofluoric and other acids. Austin even admits that the dacite is 45% phenocrysts and 'lithic (rock) inclusions.'Although Austin claimed that he took precautions to avoid laboratory contamination and that he and his team removed the obvious xenoliths from the dacite sample, Austin's own words refute Swenson's illusions that the dacite mineral/glass 'fractions' were suitably 'pure' enough for testing the validity of the K-Ar method.
Because intrusive rocks solidify deep within the Earth away from cool water and air, volcanic glass is absent and the grains may be fairly large (that is, easily reaching lengths of one centimeter or more).
In other cases, such as Austin's dacite, a partially crystallized melt erupts on the Earth's surface and produces a volcanic rock, which may be a mixture of rapidly quenched volcanic glass and coarser phenocrysts (Hyndman, 1985, p. The zoning appears as a series of concentric rings of various shades of gray within the grains (see the two obvious examples in the middle of Figure 4).
The validity of either hypothesis #2 or #3 would provide additional evidence that Austin's application of the K-Ar method is flawed and that he has failed to prove that the K-Ar method is universally invalid. Zoned crystals also may show Carlsbad twinning, which is typical of feldspars (Perkins and Henke, 2000, Plate 10; Klein and Hurlbut, 1999, p. Well-established laboratory studies (Klein and Hurlbut, 1999, p. 86-89) indicate that calcium-rich plagioclases crystallize in melts at higher temperatures than albite or sodium-rich plagioclases. Because of their crystalline and chemical differences, the calcium-rich plagioclase cores have somewhat different optical properties than the sodium-rich rims, which produce the noticeable concentric zoning in the grains in Austin's thin section photograph. However, rather than dealing with this issue and critically evaluating Austin's other procedures (including the unacceptable mineral and glass impurities in his 'fractions'), YECs loudly proclaim that the results are discrepant with the 1986 AD eruption. Considering that the dacite probably erupted in 1986 AD, Austin should have known that at least some of the samples would have given dates that were younger than 2 million years old and that Geochron Laboratories could not have provided reliable answers.