Horacek et al. (2021) commented on our publication arguing that we used an incorrect biochronology to define the Permian-Triassic (PT) boundary and that this inaccurate definition resulted in an erroneous interpretation of the oxygen isotope record in the studied Chanakhchi (former Sovestashen) section. Their comment gives us the opportunity to discuss in depth the identification of the PT boundary and to address some of the flawed arguments of Horacek et al.

Horacek et al. argue that Joachimski et al. (2020) “base their stratigraphy on the conodont investigation of the Chanakhchi section by Grigoryan (1990)….” This statement is incorrect since conodont biostratigraphy in Joachimski et al. (2020) is based on a new taxonomic study of a total of 90 conodont samples (see “Conodont Biostratigraphy” section and fig. 6 in Joachimski et al., 2020) with the highest sampling density of all studies performed so far in Chanakhchi. According to this new study, Hindeodus parvus first occurs two meters above the base of the microbial limestones and not at the base of the microbial limestones as suggested by Horacek et al. (2021).

Horacek et al. claim that Grigoryan (1990) identified “the first Hindeodus parvus on top the basal microbialites.” Again, this statement is false, since Grigoryan (1990) reported the first occurrence of H. parvus from the uppermost reddish marls (“Boundary Clay”) below the basal microbialites (see as well figure 1 in Zhakarov et al., 2005). The apparent discrepancy between the taxonomic determinations of A. Grigoryan (base of the H. parvus Zone in the topmost “Boundary Clay”) and our determinations (base of H. parvus Zone 2 m above the base of the microbial limestones) is explained by the fact that in 1990, when Grigoryan's thesis was completed, Hindeodus praeparvus was not yet identified. The first description of H. praeparvus was published by Kozur (1996), and before this, all forms with a relatively high cusp were assigned to H. parvus. H. praeparvus was initially established as a subspecies of H. latidentatus, though in our opinion H. praeparvus has very little in common with this latter species. It differs from H. parvus by a relatively shorter and wider cusp. The holotype of H. parvus (Kozur, 1975, plate 1, fig. 22; Kozur et al., 1975, plate 7, fig. 7; Kozur, 1996, plate 2, fig. 7) shows a cusp ∼3 times higher than the next denticle, which is greatly reduced in size. Instead, the cusp of the holotype of H. praeparvus (Kozur, 1996, plate 2, fig. 2) is only 1.5 time higher than the next denticle that is not small. H. Kozur did not indicate how the height of the denticles should be measured, thus, the correct identification of these two very close morphotypes is difficult. As a consequence, Kozur (2004, p. 56) distinguished some advanced forms of H. praeparvus as “transitional to H. parvus.

Horacek et al. justify their criticism by arguing that the taxonomic determinations of A. Grigoryan were revised by H. Kozur (in Zakharov et al., 2005). However, Kozur never visited Yerevan (Armenia) and thus never studied Grigoryan's conodont collection. Zakharov et al. (2005, p. 141) mention that “we were able to copy some parts of his thesis.” In our opinion, it is inadequate to revise taxonomic determinations based on a few specimens of the genus Hindeodus illustrated in a photocopied thesis.

In most sections studied in Iran, Nakhichevan, and Armenia, the typical elements of H. parvus occur at various levels in the limestones overlying the “Boundary Clay” (e.g., Kozur 2004, 2005, 2007). For example, Kozur (2004) illustrated three rather typical H. parvus specimens (Kozur, 2004, plate 1, figs. 6–8) from sample Z 1 in the Zal 1 section, taken ∼2.3 m above the “Boundary clay,” in the upper part of the H. parvus Zone (Kozur, 2005, fig. 9). The first occurrence of H. parvus (Kozur, 2005, table 5) is shown lower in sample Z 3, ∼1 m above the top of the “Boundary Clay.” In the “Boundary clay” and up to the level of sample Z 3, H. praeparvus occurs. H. parvus was also illustrated from the Kuh-e-Hambast (Abadeh) section (Kozur, 2004, plate 1, fig. 5). This specimen comes from sample VI 68c, a microbial limestone ∼1.1 m above the “Boundary Clay” (Kozur, 2005, Fig. 3), and represents the lowest occurrence of H. parvus in the section. Instead, Richoz et al. (2010; p. 241) claim that the first occurrence of H. parvus is from “….silty marl containing limestone concretions and thin limey beds” (sample 94/264). However, the illustrated specimen of H. parvus (sample 94/264; Richoz et al., 2010, plate 2, fig. 6) is a fragmentary element with broken tips of denticles and also a broken and strongly curved cusp, which to our evaluation can neither be attributed to H. praeparvus nor to H. parvus. In summary, the first occurrences of H. parvus in limestones overlying the “Boundary Clay” in Iranian sections are consistent with what we observe in the Chanakhchi section.

Horacek et al. (2021) argue that the wrong placement of the PTB is mirrored in the δ13Ccarb record with the minimum in δ13C expected the upper H. parvus and I. isarcica Zone. The δ13C minimum in the Chanakhchi is registered at the PT boundary and in the H. parvus Zone, in agreement with Korte and Kozur (2010). However, we acknowledge that Horacek et al. might have been irritated by the low δ13Ccarb values measured in the H. praeparvus Zone. These low values are interpreted as a diagenetic artifact which is underlined by the unusual large scatter in δ13Ccarb of neighboring samples.

We agree with Horacek et al. that the PT boundary should be positioned as precisely as possible. However, in our opinion, the available biostratigraphic data from the Chanakhchi as well as Iranian sections argue for a position of the boundary within the overlying microbial limestones. Most important, our publication is focusing on the temperature evolution across the PT boundary with the interpretation not depending on whether the boundary is placed 2 m higher or lower in the section.

Science Editor: Brad S. Singer
Gold Open Access: This paper is published under the terms of the CC-BY license.