The loss of complementation (LOC) hypothesis argues that an important barrier to the evolution of automictic forms of asexual reproduction is the loss of fitness due to a loss of complementation between alleles. Automixis can be considered a form of inbreeding and LOC seems to be related to the older literature on the causes of heterosis achieved by outbreeding. The so-called ‘dominance’ hypothesis of heterosis posits that the reduced fitness of inbreds is due to a loss of complementation (see, for example, Plant Cell 22: 2105). It might be valuable to briefly refer to this older literature.

Archetti (2020; JEB 33: 460) argued that one reason for the relative rarity of “inverted meiosis” in which sister centromeres separate at meiosis I rather than meiosis II is that automixis with inverted meiosis is associated with little loss of complementation, facilitating the loss of sex, and the increased tendency of lineages with inverted meiosis to go extinct.

Most forms of automixis involve central fusion (between non-sister nuclei of meiotic tetrads) rather than terminal fusion (between sister nuclei). Central fusion preserves heterozygosity for centromeres and loci linked to centromeres. The oribatid mites are an exception to this generalization and have automixis by terminal fusion. This manuscript argues that the apparent anomaly is explained because oribatid mites have “inverted meiosis.” As a result, terminal fusion is genetically similar (conserving complementation at centromeres) to central fusion in taxa with conventional meiosis.

Two forms of segregation, x and z, are posited to occur in inverted meiosis in Figure 2. The reference for these two forms of segregation is to Stern (1936), a paper about mitotic, not meiotic, crossing-over. I have thought about this and I think the author is correct. Segregation at first division of inverted meiosis would be like segregation of the products of mitotic crossing-over.

Figure 2 shows chromosomes with localized centromeres but Wrensch et al. (1994) hypothesize that oribatid mites have holocentric chromosomes. If they do, then I am unclear how one distinguishes between ‘standard’ and ‘inverted’ meiosis unless there is no crossing-over.

I am reminded of another Genetics paper from 1936 by Sturtevant and Beedle (Genetics 21: 554). They proposed that in paracentric inversion heterozygotes in Drosophila oocytes the cross-over products within inversions (dicentrics and acentrics) are eliminated in the polar bodie so that the oocyte always receives a non-recombinant chromosome. I have looked at a paper on oribatid cytogenetics (Soil Organisms 80: 223-232) and it is difficult to tell what is happening for these tiny chromosomes in oogenesis (including whether there are crossovers). LOC would be completely abolished in orabatids is meiosis were organized such that recombinant chromatids segregate together to the polar body.

The author has responded appropriately to my comments.