This paper expands upon some of the arguments and issues surrounding object agency that have been discussed in this journal (Lindstrøm 2015; 2017; Ribeiro 2016a; 2016b; Sørensen 2016; 2018). More specifically, it challenges Sørensen’s support of object agency in his latest discussion on the topic (2018). The paper is divided into three parts: first, it questions the relevance of replacing the conventional usage of ‘agency’, generally attached to sociological studies and reserved to describe human action, with one supported by the New Materialists; second, it identifies a series of contradictions in how agency is defined according to the New Materialisms, namely how it can be very labile and scalable yet simultaneously universal and applicable across all cultures and time periods; and lastly, it questions the quality of the philosophical ideas supporting the New Materialist conception of agency, and its disadvantages in light of the current re-emergence and repopularization of processual archaeology.

This paper commends some ideas that Torill Christine Lindstrøm presented in this journal on the subject of ‘object agency’ (2015) and contends, against Tim Flohr Sørensen's reaction paper (2016), that object agency subsists on philosophically questionable premises. I argue that Sørensen relies on a definition of ‘agency’ which is too labile and virtually indistinguishable from ‘cause’. Moreover, certain premises that Sørensen supports lead to some contradictions with regard to social responsibility. Finally, I argue that object agency is part of a larger intellectual trend which perceives agency as a dynamic global force – and that a return to a more constrained understanding of agency would be of more benefit to archaeology.

In the following investigation the morphometric characteristics of the first two blastomeres of rainbow trout (Oncorhynchus mykiss) were determined. Embryos were incubated at 9°C and then fixed in a Stockard solution every 30 min starting from 8.5 to 12.5 h of incubation post fertilization. Embryonic discs were extracted and microphotographs were taken with Q Capture Pro 5.0 software using a stereomicroscope Olympus SZX7. The average size of the blastodiscs was 941.22 ± 160.42 μm. The first cleavage finished after approximately 12 h of incubation. The first two blastomeres were regularly symmetrical in their morphology. Blastomere 1 had an average length (L) of 942.68 ± 105.56 μm and width (W) of 467.34 ± 64.33 μm. Blastomere 2 had an average length of 887.60 ± 101.65 and width of 454.49 ± 47.25 μm (n = 91). Significant differences were found between the length and width of blastomeres 1 and 2. The proportion between the length of blastomeres 1 and 2 was 0.94 ± 0.07 (n = 91); between the width of blastomeres 1 and 2 it was 0.88 ± 0.11 (n = 91); and the width/length ratio was 0.51 ± 0.09 (n = 182). It was concluded that rainbow trout blastomeres tend to be asymmetrical in length with a higher dispersion of widths.