<p>In the eclogite facies shear zone of Bottarello, Monte Rosa, Western Alps [1], fault recrystallization around 600 &#176;C gives concordant Lu-Hf (garnet) and <sup>39</sup>Ar-<sup>40</sup>Ar (white mica, WM) 47 Ma ages whereas <100 m from the fault the unsheared rock at the same T preserves Mesozoic inheritance. The Ar retentivity of WM is not accurately predicted by hydrothermal laboratory experiments, because the latter are plagued by massive dissolution artefacts [2]. Independent field observations confirm that WM only starts losing Ar in dry rocks above 600 &#176;C [3-8], but when retrograde reactions occur, WM can recrystallize and be totally reset below 230 &#176;C [9]. The Bottarello fault obliterated all relict WM from the protolith; the neoformed WM records its own formation age.</p><p>The island of Naxos (Cyclades, Greece) is the classic example of multiple, coexisting WM generations [10]: relict pre-eclogitic basement WM, and eclogitic phengite retrograded to muscovite. Electron microprobe element maps demonstrate intergrowths at a scale <5 &#181;m, which makes laser microprobe dating useless. Bulk mica dissolution for Rb-Sr gives Eocene ages [11], which agree with bulk K-Ar ages. This is paradoxical, as Ar diffusivity is c. 4 orders of magnitude higher than that of Sr [12]; the only explanation is that both chronometric systems record formation ages around 500-600 &#176;C. The WM generations can be unravelled by their Ca/Cl/K signatures; coarse and fine sieve fractions are never isomineralic. Ages of pure mica generations are obtained by extrapolating Ca/Cl/K-vs-age trends.</p><p>The in-sequence thrusts of the Garhwal Himalaya add one complication: thrusting was long-lived. Microstructures combined with chemical microanalysis distinguish three monazite generations (dated by U-Pb) and three WM generations: relicts in microlithons, foliation-defining mica, and static coronas. As in the previous examples, intergrowths are <<10 &#181;m and only combining Ca/Cl/K systematics with the observed differences in structural breakdown temperatures can assign the different WM ages in the same sample to chemically distinct generations [13]. WM formation ages overlap with Mnz ages and date the onset of faulting, the kinematic peak, and the post-faulting corona formation.</p><p>There is no free lunch: dating deformation is extremely labor-intensive and requires, always, establishing the context between microtextural, microchemical, petrological and multichronometric analyses. Whenever one of these four is missing, the tectonic reconstruction is invariably faulty [14].</p><p>&#160;</p><p>[1] Villa &al, J Petrol 55 (2014) 803-830</p><p>[2] Villa, Geol Soc London Spec Pub 332 (2010) 1-15</p><p>[3] Di Vincenzo &al, J Petrol 45 (2004) 1013-1043</p><p>[4] Itaya &al, Island Arc 18 (2009) 293-305</p><p>[5] Heri &al, Geol Soc London Spec Pub 378 (2014) 69-78</p><p>[6] Laurent &al, Lithos, 272-273 (2017) 315-335</p><p>[7] Airaghi &al, J Metam Geol 36 (2018) 933-958</p><p>[8] Imayama &al, Geol Soc London Spec Pub 481 (2019) 147-173</p><p>[9] Maineri &al, Mineralium Deposita 38 (2003) 67-86</p><p>[10 Wijbrans & McDougall, Contrib Min Petr 93 (1986) 187-194</p><p>[11] Peillod &al, J Metam Geol 35 (2017) 805-830</p><p>[12] Cherniak & Watson, EPSL 113 (1992) 411-425</p><p>[13] Montemagni &al, Geol Soc London Spec Pub 481 (2019) 127-146</p><p>[14] Bosse & Villa, Gondwana Res 71 (2019) 76-90</p>