Alaria Schrank 1788

REMARKS ON ALARIA

The members of Alaria in the two phylogenies based on 28S had only slight differences in topology (Figs 1, 2). At the same time, the phylogenies of 28S and cox1 limited to members of Alaria showed more pronounced differences in branch topology (Figs 2, 3). Alaria mustelae was positioned as a sister taxon to the other Alaria spp. in the second 28S analysis (Fig. 2), while in the cox1 phylogeny, A. ovalis and A. procyonis formed an unsupported clade that was placed as a sister group to the other members of Alaria (Fig. 3). The positions of A. alata + Alaria sp. 1 and A. marcianae + Alaria sp. 3 varied between the two analyses as well (Figs 2, 3). Discordance between phylogenies based on ribosomal and mitochondrial data has been well documented among other diplostomoideans (e.g. Brabec et al., 2015; Heneberg et al., 2020; Hoogendoorn et al., 2020; Achatz et al., In press). Faster mutating genes, such as cox1, are more reliable for distinguishing between closely related diplostomoidean species/species-level lineages (Table 2; Supporting Information, Table S1), but slower mutating genes, such as 28S, remain more suitable for phylogenetic inference at taxonomic levels above genus.

All Alaria spp. in the present study, except for A. alata, were collected from North America. The nested phylogenetic position of A. alata clearly suggests a geographic expansion from the Nearctic into the Palaearctic (Figs 1–3).

It is difficult to address questions related to host switching of Alaria spp., considering that many species have been historically reported in a diversity of mammalian hosts (e.g. see Dubois, 1968 and references therein). The accuracy of Alaria spp. identifications in previous reports is questionable considering that most publications lack DNA sequence data and many Alaria spp. are morphologically similar. Some Alaria spp., such as A. arisaemoides, are also known to have substantial morphological variation (e.g. Hall & Wigdor, 1918; Dubois, 1968). The topology of our molecular phylogeny based on the 28S of Alaria spp. (Fig. 2) is not well enough supported to confidently infer evolutionary patterns of definitive host associations; the discordance between topologies of 28S (Fig. 2) and cox1 (Fig. 3) further complicates the situation. Our specimen of Alaria sp. 3 from the cougar Puma concolor (Linnaeus, 1758) is immature; hence, additional collection of well-fixed, mature specimens of Alaria sp. 3 is crucial for accurate species identification and confirmation of its definitive host.

It is worth noting that our specimens of A. arisaemoides (Fig. 4B) conform closely to the original description of the species and subsequent descriptions of the species (e.g. Augustine & Uribe, 1927; Dubois, 1968). However, the cox1 sequences of our specimens are only 1.9–2.6% different from material identified as Alaria americana Hall & Wigdor, 1918 by Locke et al. (2018) (Supporting Information, Table S1). The material described by Locke et al. (2018) is somewhat different to the original description of A. americana described by Hall & Wigdor (1918). For instance, A. americana was originally described with vitellarium that does not extend anteriorly beyond the level of the ventral sucker. The vitellarium of A. americana from Locke et al. (2018) extends anteriorly to the level of the ventral sucker, similar to the condition in A. arisaemoides. In our opinion, the specimens identified as A. americana by Locke et al. (2018) are likely misidentified specimens of A. arisaemoides.