Genus Hyalella (Amphipoda: Hyalellidae) in Humid Pampas: molecular diversity and a provisional new species

Abstract

Hyalella is a genus of epigean freshwater amphipods endemic to the Americas. The study of morphological characters alone has traditionally dominated the description of new species. Recently, molecular systematics tools have contributed to identifying many cryptic species and a high level of convergent evolution in species complexes from North America and the South American highlands. In this study, we evaluate for the first time the molecular diversity in Hyalella spp. in Uruguay, a country located in the humid pampa ecoregion, based on four molecular markers. Thus, we investigate the systematic position of H. curvispina in the context of the available phylogenetic hypothesis for the genus. Phylogenetic and morphological analyses confirm that there is a “curvispina complex”. This complex includes H. curvispina and several similar morphological forms but is paraphyletic in relation to some altiplano species. In addition, we found one provisional new species. The results obtained are contrasted with previous studies to help understand the mechanisms of genetic differentiation and speciation of the genus, which seems to have a strong tendency towards morphological convergence.

Full text

261
Genus Hyalella (Amphipoda: Hyalellidae) in Humid
Pampas: molecular diversity and a provisional new
species
Analisa Waller1, Exequiel R. González2, Ana Verdi1, Ivanna H. Tomasco3
1 Sección Entomología, Facultad de Ciencias, Universidad de la República, Iguá 4225, CP 11400, Montevideo, Uruguay
2 Universidad de Santo Tomás, Avenida Ejército 146, CP 8370003 Santiago. Chile
3 Departamento de Ecología y Evolución, Facultad de Ciencias, Universidad de la República, Iguá 4225, CP 11400, Montevideo, Uruguay
http://zoobank.org/E7CF648B-5235-4412-8C65-14C2A828B954
Corresponding author: Analisa Waller (anawaller@gmail.com)
Received 17 December 2021
Accepted 29 March 2022
Published 28 June 2022
Academic Editors Martin Schwentner, Klaus-Dieter Klass
Citation: Waller A, González ER, Verdi A, Tomasco IH (2022) Genus Hyalella in Humid Pampas: molecular diversity and provisional new species.
Arthropod Syematics & Phylogeny 80: 261–278. https://doi.org/10.3897/asp.80.e79498
Abstract
Hyalella is a genus of epigean freshwater amphipods endemic to the Americas. The study of morphological characters alone has
traditionally dominated the description of new species. Recently, molecular systematics tools have contributed to identifying many
cryptic species and a high level of convergent evolution in species complexes from North America and the South American high-
lands. In this study, we evaluate for the rst time the molecular diversity in Hyalella spp. in Uruguay, a country located in the humid
pampa ecoregion, based on four molecular markers. Thus, we investigate the systematic position of H. curvispina in the context of
the available phylogenetic hypothesis for the genus. Phylogenetic and morphological analyses conrm that there is a “curvispina
complex”. This complex includes H. curvispina and several similar morphological forms but is paraphyletic in relation to some al-
tiplano species. In addition, we found one provisional new species. The results obtained are contrasted with previous studies to help
understand the mechanisms of genetic dierentiation and speciation of the genus, which seems to have a strong tendency towards
morphological convergence.
Keywords
curvispina complex, Uruguay, COI, 12S, 28S, H3, molecular species delimitation, phylogeny
1. Introduction
Hyalella is a genus of epigean freshwater amphipods of
America (Baldinger 2004) and is the only one present in
South America (Reis et al. 2020). Members of this ge-
nus are found in various freshwater environments such
as lakes, ponds, and streams, clinging to vegetation and
burrowing in bottom sediments (da Silva Castiglioni and
Bond-Buckup 2008). They are mainly omnivores, and
due to their feeding habits, they play an essential role in
the food webs facilitating the energy ow in aquatic eco-
systems (da Silva Castiglioni and Bond-Buckup 2008;
Giorgi and Tiraboschi 1999). Dierent species of Hyalel-
la have been used as bioindicators of environmental con-
Arthropod Syematics & Phylogeny 80, 2022, 261–278 | DOI 10.3897/asp.80.e79498
Copyright Analisa Waller et al.: This is an open access article diributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrericted use, dirie-
bution, and reproduction in any medium, provided the original author and source are credited.

Waller et al.: Molecular diversity of Hyalella
262
ditions and pollution; in North America, Hyalella azteca
is a standard organism in bioassays (Casset et al. 2001).
There are 84 described species of Hyalella (Tuparai
Talhaferro et al. 2021; Limberger et al. 2021), but tax-
onomic knowledge of the genus is incomplete. Highly
complex cryptic species with very subtle interspecic
and interpopulation morphological variations (Worsham
et al. 2017) make identifying and dierentiating species
challenging. Traditionally, the taxonomic description and
identication of species have been based exclusively on
morphological characters. Bibliographic research shows
that 70% of the descriptions of the Hyalella species be-
long only to that category. However, speciation is not
always accompanied by morphological change. The ac-
tual number of biological species is likely to be greater
than the current tally of nominal species, most of which
are delineated on purely morphological grounds (Bick-
ford et al. 2007). Recent studies incorporating molecu-
lar data show a high rate of morphological convergences
and cryptic species (Adamowicz et al. 2018; Zapelloni et
al. 2021), which shows the need to integrate genetic and
morphological data for the delimitation of the species of
this genus.
In the last decades, the genus has begun to be studied
by applying molecular systematics tools. In particular, in
the Hyalella genus, the mitochondrial gene for subunit I
of Cytochrome Oxidase C (COI) has been used almost
exclusively, both partial (Adamowicz et al. 2018; Major
et al. 2013; Vergilino et al. 2012; Wellborn and Broughton
2008; Witt et al. 2003; Witt and Hebert 2000; Worsham et
al. 2017) and complete (Major et al. 2013). This gene has
also been used for DNA barcoding (Babin-Fenske et al.
2012; Dionne et al. 2011; Jurado-Rivera et al. 2020; Witt
et al. 2006). Recently, the 13 protein-coding mitochondri-
al genes has been used to resolve phylogenetic relation-
ships among a set of species (Juan et al. 2016; Zapelloni
et al. 2021). A minority party, other independent markers
have been used, such as the 28S nuclear gene (Adamowicz
et al. 2018; Witt et al. 2006; Zapelloni et al. 2021), the
H3 histone gene (Jurado-Rivera et al. 2020), allozymes
(Duan et al. 2000; Witt and Hebert 2000), and dozens of
single-copy nuclear orthologous genes sequences (Zapel-
loni et al. 2021). Several species have relatively restricted
distributions within the Americas, suggesting “groups of
species”. In North America, we found the “azteca com-
plex” with North and Central American species, including
Hyalella azteca (Major et al. 2013; Vergilino et al. 2012;
Witt et al. 2006; Witt and Hebert 2000; Worsham et al.
2017). In South America, several groups have been sug-
gested. One group is endemic of the deep lake Titicaca and
is highly diverse from both molecular and morphological
perspectives (Adamowicz et al. 2018; Jurado-Rivera et al.
2020; Zapelloni et al. 2021). Other ‘groups’ include spe-
cies from the Amazonian basin and its area of inuence,
species from high altitudes in the Andes and low regions
west of the Andes, and species from east of the Andes and
its area of inuence, with anities to H. curvispina; and
the “patagonic complex” distributed in the southern ex-
treme of South America, with some species along the An-
des (González and Watling 2001). Dierent criteria have
been proposed to delimit species due to the presence of
cryptic species and adaptive convergence in these poly-
phyletic complexes. For example, Witt et al. (2006) has
proposed the Species Screening Threshold (SST) method
with COI, which employs a 3.75% maximum within-spe-
cies divergence for delineating relationships among H. az-
teca using K2P distance (Kimura 1980), which posteriors
studies applied (Dionne et al. 2011; Vergilino et al. 2012).
Adamowicz et al. (2018) used Barcode Index Numbers
(BINs). This method implements a species threshold val-
ue of 2% to detect 48 BINs in the South American Hyalel-
la data set, twelve of them occurring at the Titicaca area,
including six uniquely sampled in this lake (Jurado-Rive-
ra et al. 2020). Finally, Jurado-Rivera et al. (2020) used
genetic distances using the K2P and Multi-rate Poisson
Tree Processes method (mPTP), GMYC and ABGD to
delimit species.
The description of Hyalella curvispina had been much
debated until a few years ago. H. curvispina was initial-
ly described by Shoemaker in 1942 as a type locality in
Montevideo. In 1953 de Oliveira described H. curvispina
form H. cangallensis (Schelloenberg) due to the presence
of only one curved setae in the inner ramus of uropod
1. After, Stock and Platvoet (1991) considered that the
description of de Oliveira (1953) was not attributable to
H. curvispina due to the thickness of the palps of maxilli-
peds and because the presence of one or two curved setae
in uropod one is frequently observed in the same popu-
lation without other distinctive features. More recently,
Grosso and Peralta (1999) redescribed the species based
on Chilean material, but a few years after, González and
Watling (2003b) considered that this taxon corresponds to
H. simplex due to the presence of sternal gills in ventral
sternites 3 to 7 while in H. curvispina they are present
in segments 2 to 7. Morphological data (González and
Watling 2001) suggest a species complex to the East of
the Andes characterized by H. curvispina (Peralta and
Grosso 2009). Still, it has not been included in the mo-
lecular phylogenies proposed to date. The morphological
diagnostic features of the “curvispina complex” include
a smooth body surface, the presence of curved setae in
the inner ramus of uropod one, and sternal gills present
in segments 2 to 7. The species that share these charac-
teristics with H. curvispina and inhabit the East of the
Andes are H. pampeana Cavalieri, 1968; H. falklandensis
Bouseld,1996; H. rionegrinae Grosso and Peralta, 1999
and H. bonariensis Dos Santos et al., 2008. Our team
has revealed the presence of dierent morphs of Hyalel-
la curvispina with subtle dierences at the telson level,
internal part of gnathopod 1, and setaes in uropods that
call the status of new species into question. This species
complex inhabits a much more recent and unstable geo-
graphic area than the complexes studied. This region was
formed in the Holocene, between 5000–7000 years ago,
when the Uruguayan coastal lagoons and the rise of the
continental block that includes Uruguay began (del Puer-
to et al. 2011). At the same time, the area presents shallow
lagoons and temporal ponds with periodic drying (Laufer
et al. 2009) that could generate population bottlenecks,
with measurable evolutionary consequences.

Arthropod Syematics & Phylogeny 80, 2022, 261–278 263
In this study, we evaluate the molecular diversity of
Hyalella spp. in Uruguay for the rst time, using four
markers, and we investigate its systematic position in the
framework of available phylogenetic hypothesis for the
genus. Specically, we i) assess the number of Hyalella
cryptic species in Uruguay and ii) infer the phylogeny of
Hyalella comparing Hyalella curvispina with similar and
dierent morphs and place the Uruguayan Hyalella with-
in the clade identied by Zapelloni et al. (2021).
2. Methods
2.1. Field collecting
A previous survey in several populations of Uruguay re-
ported the presence of nine dierent morphs similar to
Hyalella curvispina. In autumn 2020, we visited nine
country points where these dierent morphs came from
and collected 8 of them. Sampling localities were: San
José (H8), Colonia (H1), Durazno (H3), Colonia Rosell
y Rius (H2), Lavalleja (H4), Paso de los Toros (H5), Ba-
toví (H7), Achar (H6), and two localities of Montevideo:
Facultad de Ciencias (FC) and Montevideo type locality
for Hyalella curvispina (MVD) (Fig. 1 and Table S1). We
collected the specimens with a water net with a diameter
of 20 cm and a mesh width of 250 µm. All samples were
stored in liquid N2 at collection sites and preserved there
until the molecular procedure. Two adult males from each
site were dissected and morphologically evaluated under
a stereoscope instead of glass to conrm the presence of
the characteristics of each morph according to the collec-
tion site.
2.2. Molecular procedures
We obtained total genomic DNA extractions from ve
animals for each of the nine morphs previously collected,
following standard protocol for precipitation of proteins
with salts and DNA with ethanol (modied from Miller
et al. 1988). We used only the pereion and pleon of the
animal to avoid contamination of possible microorgan-
isms adhering to the mouthparts, the concentration of
DNA extractions were estimated in the nanodrop spec-
trophotometer (NanoDrop, Thermo Scientic, USA). We
analyzed the genes COI, 12S, 28S and H3 and amplied
them by PCR using the primers LCO and 5587F (Stutz et
al. 2010), 12Samphi-f and 12Samphi-r (Rodrigues 2016),
Rnest and Fnest (Stutz et al. 2010) and H3af and H3ar
(Colgan et al. 1998) respectively. The amplication were
carried out in a PX0.2 Thermal Cycler (Thermo Electron
Corporation) in a total volume of 22 μl containing 10µl
of kit GoTaq® Hot Start Green Master Mix (Promega), 2
µl of each primer (10 mM) and 8µl DNA dilution (1/100).
We included negative controls in all cases, replacing the
DNA dilution with water. Sequences of all morphs were
obtained, except for COI. In particular, we got an average
of 2 COI sequences for ve morphs and H. curvispina,
an average of 3 sequences of 12S, and one sequence of
H3 and 28S for all morphs and H. curvispina of the two
Montevideo localities (FC and MVD).
We amplied a 369 base pair (bp) fragment of the mi-
tochondrial cytochrome c oxidase subunit I (COI) gene
was using the primers LCO and 5587F (Stutz et al. 2010).
The PCR program consisted of initial denaturation for
2 min at 95 ° C followed by 36 cycles of 30 s at 95ºC,
30 s at 45ºC, 1 min at 72ºC, and a 7 min extension step
at 72ºC (modied from Stutz et al. 2010). We amplied
the 12S mitochondrial ribosomal gene using the primers
12Samphi-f and 12Samphi-r (Rodrigues 2016) to obtain
a partial sequence of 452 bp. The PCR cycle was an ini-
tial denaturation for 5 min at 96°C, followed by ten cy-
cles of 30 s at 96°C, 60 s at 55°C (decreasing annealing
temperature by 1°C/cycle during seven cycles and then
decreasing to 47°C in the last cycle), and 60 s at 72°C,
followed again by 30 cycles of 30 s at 96°C, 60 s at 45°C
and 60 s at 72°C, with a nal extension of 5 min at 72°C
(modied from Rodrigues 2016). We amplied the ribo-
somal nuclear gene 28S with primers Rnest and Fnest
(Stutz et al. 2010) to obtain a partial sequence of 605 bp.
The PCR program consisted of a 1 min denaturation step
at 94ºC, 39 cycles of 1 min at 94ºC, 1 min at 51°C, 1
min at 72°C, and a 5 min extension step at 72°C (Stutz
et al. 2010). Finally, we amplied a 332 bp of the H3
protein-coding gene H3af and H3ar (Colgan et al. 1998).
The PCR program consisted of initial denaturation for 5
min at 96°C, followed by 30 cycles of 30 sec at 96°C,
FC
and
MVD
H7
H6
H5
H2
H3
H8
H1
H4
Uruguay
Figure 1. Map of collection sites of the dierent Hyalella
morphs in Uruguay. H1: Colonia (34°26.0501′S 57°49.4717′W);
H2: Colonia Rossell (33°10.9503′S 55°44.363′W); H3: Duraz-
no (33°24.0584′S 56°31.18′W); H4: Lavalleja (34°30.4393′S
55°22.3598′W); H5: Paso de los Toros (32°45.4339′S
56°31.8643′W); H6: Achar (32°23.8234′S 56°9.4333′W); H7:
Batoví (31°52.9283′S 56°0.7117′W); H8: San José (34°18.65′S
56°52.7′W); FC: Montevideo (34°52.8334′S 56°7.05′W); MVD
(34°50.3666′S 56°16.05′W): Montevideo (Hyalella curvispina).

Waller et al.: Molecular diversity of Hyalella
264
45 s at 50°C, and 60 s at 72°C, with a nal extension
of 5 min at 72°C (Rodrigues 2016). PCR products were
checked at electrophoresis with agarose gel (0.8% in TBE
1X), stained with Ethidium Bromide, and visualized with
an ultraviolet light source. PCR products of the expected
size without secondary bands were puried and automat-
ically sequenced (Sanger method, Macrogen Inc., http://
www.macrogen.com) from both ends.
We edited obtained sequences manually using the
PROSEQ 3.2 program (Filatov 2002), considering that
each change corresponded to well-dened peaks in the
chromatogram. In the case of protein-coding genes (COI
and H3), we checked the absence of stop codons or in-
dels that could modify the reading frame to ensure no
pseudogene was present. In nuclear genes, we annotated
the polymorphisms with an IUPAC ambiguity code.
2.3. Molecular analyses
We compared the obtained sequences with sequences
from dierent species of the same genus reported in Gen-
bank (see accession numbers and other details in Supple-
mentary Table S2). All genes were aligned independently
using the MUSCLE algorithm with the MEGA X pro-
gram (Kumar et al. 2018). In all cases, a sequences from
Platorchestia sp. (Table S2), a genus closely related to
Hyalella, was used as the outgroup taxon.
For each gene, nucleotide frequencies, variable sites,
and parsimony-informative (Pi) sites were estimated
using MEGA X software. We calculated global nucle-
otide distances and pairwise distances between morphs
and species with the K2P substitution model (Witt et al.
2006). The most suitable nucleotide substitution model
for each gene was assessed with JMODELTEST v2.2.10
(Darriba et al. 2012), with Akaike Information Criterion
(AIC), and with Bayesian Information Criterion (BIC)
(Posada and Crandall 1998). Additionally, we estimated
phylogenetic reconstruction with MEGAX by apply-
ing dierent criteria with 1000 bootstrap pseudorep-
licates and pairwise deletion options (i.e., eliminate all
positions with less than 95% site coverage). We carried
out the reconstruction by maximum likelihood with the
substitution model suggested by JMODELTEST. The
initial tree(s) for the heuristic search were automatical-
ly obtained by applying Neighbor-Joining and BioNJ
algorithms to a matrix of pairwise distances estimated
using the Maximum Composite Likelihood distance. For
maximum parsimony, we used the Subtree-Pruning-Re-
grafting algorithm with search level 1, in which the initial
trees were obtained by the random addition of sequences
(10 replicates). For neighbor-joining algorithms, evolu-
tionary distances were computed using the Maximum
Composite Likelihood method and are in the units of the
number of base substitutions per site.
Bayesian phylogenetic methods were also performed
in BEAST v.2.6.3 (Suchard et al. 2018) and carried
the option *Beast, to estimate the species tree by con-
sidering the information of all markers. For each gene,
we used the best-t nucleotide substitution model esti-
mated by JMODELTEST v2.2.10 (Darriba et al. 2012).
We explored the nucleotide substitution saturation for
each molecular marker using the Xia test in DAMBE v.
7.2.144 (Xia 2018). For COI, we separated the analysis
and eliminated the third position of the codon. We par-
titioned the COI gene, analyzing the rst and second co-
don positions separated from the third due to saturation of
this marker (see discussion below). We linked the mito-
chondrial genes, the two partitions of COI and 12S, kept
the other two genes independent, and selected the Yule
process. It used the MCMC length of 100.000.000 gen-
erations, sampling every 10.000 with a burn-in of 20%
of the trees sampled. Using TRACER v. 1.6 (Rambaut
2018), we assessed the resulting log les and corrobo-
rated that the Eective Sampling Size (ESS) values were
higher than 200 to ensure adequate sampling and conver-
gence. We created a maximum clade credibility tree in
TREE ANNOTATOR v2.6.3 (Suchard et al. 2018), and
this tree was visualized and edited in FIGTREE v. 1.4.4.
(Rambaut 2018). Due to the diculties in obtaining the
COI gene for all the Uruguayan samples, and because the
evolutionary history of mitochondrial genome is also in
12S (by linkage, as the mitochondrial genome does not
recombine), we estimate two species trees, with and with-
out COI, using the vouchers sequenced for four and three
genes respectively. Finally, we applied a maximum like-
lihood approximation implemented in IQ-TREE v1.6.12
(Nguyen et al. 2015), with a partition by gene and unliked
edges, and an automatic selection of the best substitution
model for each gene, and 1000 pseudoreplicates of ultra-
fast bootstrap (Minh et al. 2013).
We applied three species delimitation methods for
each gene independently and for all of them concatenat-
ed (with and without COI). We used the ABGD method
developed by Puillandre et al. (2012) on the webserver
http://wwwabi.snv.jussieu.fr/public/abgd/abgdweb.html.
After sequence alignment, we computed a matrix of pair-
wise distances using the K2P model (Kimura 1980). We
used Pmin = 0.001 and Pmax = 0.1 and X = 0.5. We also
applied a recently developed method, still based on pair-
wise genetic distances, but whose implementation pro-
vides a score for each dened partition and overcomes
the challenge of a priori dening p, the ASAP (Puillandre
et al. 2021). ASAP was run in webserver https://bioinfo.
mnhn.fr/abi/public/asap/asapweb.html, with K2P model
and default options. Additionally, we used PTP and bPTP
on the webserver https://species.h-its.org/ptp. PTP is a
model for delimiting species on a rooted phylogenetic
tree developed by Zhang et al. (2013). It models specia-
tion or branching events in terms of the number of sub-
stitutions, so it only requires a phylogenetic input tree.
And bPTP is an updated version of the original maximum
likelihood PTP that adds Bayesian support values to de-
limited species on the input tree. Higher BS value on a
node indicates all descendants from this node are more
likely to be from one species.

Arthropod Syematics & Phylogeny 80, 2022, 261–278 265
3. Results
3.1. Sequence data
Because we faced diculties in amplifying and sequenc-
ing COI sequences, we recovered 11 COI sequences that
include only Hyalella curvispina from FC locality in
Montevideo, and ve morphs. We could amplify and se-
quence 30 sequences of the 12S gene, ten 28S and H3 of
H. curvispina, and all morphs (Table 1).
We summarized the sequence length obtained from
each marker and the number of parsimony-informative
sites in Table 2. Although 28S was the longest sequence
(605 bp), the proportion of informative sites was low
(approx. 18%). We analyzed shorter sequences from two
mitochondrial markers, 369 and 452 bp, COI and 12S,
respectively. They presented a high proportion of infor-
mative sites from the point of view of parsimony (39 and
43%, respectively). Histone H3 was the smallest fragment
(332bp) and was the one that presented the highest pro-
portion of parsimoniously informative sites, although, in
absolute terms, there were only a few (22 sites) (Table 2).
Most of the morphs have dierent alleles/haplotypes,
but there were shared haplotypes in all cases, generally
for geographically close localities. In all cases: i) the H.
curvispina samples (topotypes) are identical to those of
FC locality close to 20 km (Montevideo), except for the
H3 gene, ii) localities H1 and H8 with a distance of 120
km between them (Colonia and San José, respectively)
are identical except for the COI, iii) localities H5 and
H6 with a distance of 67 km between both (Paso de los
Toros and Achar) except for COI. For the COI, one of
the samples from the locality H1(1) (Colonia) shares a
haplotype with H. curvispina (type locality and another).
However, for mitochondrial 12S, the H1 and H8 morphs
(from Colonia and San José, respectively) are identical
and dierent from those of H. curvispina (both localities).
In contrast, the morphs H5 and H6 of the localities (Paso
de los Toros and Achar) are identical.
The 28S marker has the lowest variation, and the sam-
ples from populations H5, H6 (Paso de los Toros and
Achar), and H. curvispina (two locations) on the one
hand, as well as those from H1, H8, and H3 (Colonia,
San José, and Durazno), are identical.
For histone H3 the samples from H1, H3, H4, H5, H6,
H7 and H8 (Colonia, Durazno, Lavalleja, Paso de los
Toros, Achar, Batoví and San José) are identical, while
Hyalella curvispina from the two localities of Montevi-
deo do not share haplotype. We kept identical variants in
the analyzes to estimate the species tree with all of them
(Table 3).
Table 1. Number of specimens by morph sequenced and the number of haplotypes for each molecular marker (MM). N: total
number of sequences obtained for each marker. N°ht: number of haplotypes for each marker. Localites: FC, MVD (type locality of
Hyalella curvispina), H1, H2, H3, H4, H5, H6, H7, and H8 (see Fig. 1).
MM N N° hapl. Hyalella curvispina
MVD
Topotype
Hyalella curvispina
FC
H1 H2 H3 H4 H5 H6 H7 H8
COI 11 9 — 2 2 1 2 3 1 — — —
12S 30 7 1 3 4 4 3 3 2 2 4 4
28S 10 61 1 11111111
H3 10 3 1 1 1 1 1 1 1 1 1 1
Table 2. Used molecular markers (MM) with fragment length in base pairs (bp), number of conserved sites (CS), number of variable
sites (VS), and number of parsimony-informative sites (PIS).
Molecular marker (MM) Fragment length (bp) Number of conserved
sites (C)
Number of variable sites (V) Parsimony (Pi)
COI 369 165 203 145
12S 452 182 259 193
28S 605 390 203 110
H3 332 256 74 22
Table 3. Number of variants (Nv, haplotypes or alleles for COI and 12S, or for H3 and 12S, respectively) found per molecular
marker (MM), and shared haplotypes among samples assessed.
MM Nv Samples that shared haplotype/alleles
COI 9 H. curvispina (topotype MVD and FC) and H1(1)
12S 7 H. curvispina (topotype MVD and FC) H1 and H8 H2 and H3 H5 and H6
28S 6H. curvispina (topotype MVD and FC) H1, H8 and H3 H6 and H5 H.curvispina (MVD) and H7
H3 3 H. curvispina (topotype MVD) and H2 H1, H3, H4, H5, H6, H7 and H8

Waller et al.: Molecular diversity of Hyalella
266
The substitution models selected by the AIC criteri-
on were HKY+G for the COI and 12S, the TVM+G for
the 28S, and K80+G for the H3. The substitution models
selected by BIC were TIM3+G for COI and 12S and TM-
V+I+G for 28S and TPM1uf+G for H3.
3.2. Genetic distances
Pairwise genetic distances are summarized in Table 4 and
Supplementary Table S3, S4, S5, S6. A low genetic dis-
tance is observed for all markers (average 10, 4, 1 and
1%, for COI, 12S, 28S, and H3, respectively) among the
Uruguayan morphs, with specimens H4(1) and H4(2v)
(Table S3) being from Lavalleja, the much more diver-
gent morph concerning the rest of the Uruguayan samples
(19, 8, 2 and 1% for COI, 12S, 28S, and H3, respectively).
As expected, for all markers except COI, the average ge-
netic distances between Hyalella spp. and Platorchestia
sp. used as outgroup taxon are much greater than the dis-
tances within the genus Hyalella (Table 4). For the COI,
the average genetic distance between all the species eval-
uated and H. azteca (the most basal species of all those
analyzed according to the study of Zapelloni et al. (2021)
(at the genomic scale) is practically the same as the dis-
tance to the outgroup taxon (24 and 25%, respectively).
The average distances among Uruguayan samples are
moderate (0.10, 0.04, 0.01, 0.01 for COI, 12S, 28S, and
H3, respectively). We found the highest genetic distance
between the H4 (specimens H4(1) and H4(2v)) and the
rest (0.19, 0.08, 0.02 and 0.01, for COI, 12S, 28S, and
H3, respectively), similar to the interspecic distances
between the Uruguayan samples (excluding specimens
H4(1) and H4(2v)) and other species such as H. kochi
2319B.
3.3. Saturation analyses
For all markers, the sequences showed little substitution
saturation. For each marker Iss is signicantly lower
than Iss.c for both symmetric and asymmetric topolo-
gies (these were always lower than the rsts). For COI
sequences, for the three codon positions taken together
(Iss: 0.22 < Iss.c: 0.76 (symmetric topology) Iss.c: 0.52
(asymmetric topology), degree of freedom (DF): 114, p <
0.0001), or for the 1st and 2nd position of the codon tak-
en together (Iss: 0.05 < Iss.c: 0.90 (symmetric topology)
Iss.c: 0.76 (asymmetric topology), DF: 75, p < 0.0001).
For the 12S mitochondrial gene sequences (Iss: 0.13 <
Iss.c: 0.68 (symmetric topology) Iss.c: 0.36 (asymmetric
topology), DF: 256, p < 0.0001). For 28S sequences (Iss:
0.07 < Iss.c: 0.72 (symmetric topology) Iss.c: 0.41 (asym-
metric topology), DF: 521, p < 0.0001) and for histone H3
sequences (Iss: 0.04 < Iss.c: 0.53 (symmetric topology)
Iss.c: 0.41 (asymmetric topology), DF: 95, p < 0.0001).
3.4. Phylogenetic analysis
No gene analyzed independently showed solvency in the
resolution of the most nodes (see phylogenetic recon-
structions (Figs 2–5, and in Supplementary Figs S1–S8).
On the contrary, most of the nodes have low to moderate
support values. Only some were high and varied with the
reconstruction method used. However, except for the H3
gene, the COI and 12S mitochondrial genes and the 28S
nuclear gene show that the Uruguayan samples are para-
phyletic. There is a monophyletic group formed by Uru-
guay, H. montforti 2015 2D, and H. kochi (4747, 2319B,
3TK27), which includes two groups: one wider group
that includes H. curvispina and has a higher anity with
H. montforti 2015 2D and H. kochi 4747, 2319B, 3TK27,
and the other with fewer samples, sister to the previous
one. In these tree topologies, we observed that the sam-
ples annotated as H. kochi are polyphyletic. The H3 gene
presents a very low variation (showed in other studies)
and could not resolve any node with bootstrap values
from moderate to high. We do not recover the monophyly
of the Uruguayan samples, nor the close relationship be-
tween these and H. montforti 2015 2D or some samples
of H. kochi 4747, 2319B, 3TK27.
The Bayesian species tree considering all markers,
each with its most appropriate substitution model, re-
solves most nodes with higher supports of the posterior
probability. Among them, we can highlight 1) the mono-
phyly of the Uruguayan samples together with two sam-
ples of H. kochi 4747, 2319B, and H. montforti 2015 2D,
the basal position of H4(1) concerning this clade, and the
reciprocal monophyly of the Uruguayan samples without
H4(1) on one side and of H. montforti 2015 2D with two
samples of H. kochi (2319B and 4747) on the other, 2) the
basal position of H. azteca and H. armata 26-2A within
the species of the genus analyzed, 3) the monophyly of
two other groups of species H. nefrens 2310E + H. hirsu-
ta 30-5C + H. neveulemairei 30-5D + H. kochi AP18, and
Table 4. Average genetic distances using K2P distance. *Locality H4 represented by specimens H4(1), H4(2v), and H4(3) (the last
one only for 12S).
Mitocondrial Nuclear
Average COI 12S 28S H3
Hyalella sp. vs. H. azteca 0.24 0.22 0.08 0.08
Hyalella sp. vs outgroup 0.25 0.30 0.25 0.12
Uruguay vs. H. kochi 0.18 0.07 0.01 0.01
Uruguay (non-H4 vs H4)* 0.19 0.08 0.02 0.01
Intra Uruguay 0.10 0.04 0.01 0.01

Arthropod Syematics & Phylogeny 80, 2022, 261–278 267
H. cajasi EC3-1 + H. tiwanaku 2304 (Fig. 6), although
with very low statistical support (0.39).
The species tree obtained from the two nuclear genes
and only one mitochondrial (only 12S and excluding
COI) has the same general topology and higher posterior
probability values. The information from the Uruguayan
sample H7(1) (not recovered with COI) is incorporated
and show higher anity with two samples of H. kochi
(4747, 2319B) and H. montforti (2015 2D) than with the
majority group of Uruguayan samples (Fig. 7).
Maximum likelihood reconstructions yielded similar
results (Fig. S9 and Fig. S10) for four or three genes,
respectively). The major dierence in both cases is the
absence of reciprocal monophyly between groups H. ne-
frens 2310E + H. hirsuta 30-5C + H. neveulemairei 30-
5D + H. kochi AP18, and H. cajasi EC3-1 + H. tiwanaku
2304.
3.5. Molecular species delimitation
The results of molecular species delimitation are shown
in the Supplementary Fig. S11. The results for the dier-
ent genes analyzed independently are similar, quite dier-
ent, and COI was the gene that yielded the most dierent
results between methods. Among the four methods tested,
ASAP found more groups in all genes, and diered most
from the other methods.
In general, H. armata 26-2A, H. azteca, H. cajasi EC3-
1, H. tiwanaku 2304, and H4(1), a sample of the “curvispi-
H. neveulemairei 2316D
H. nefrens 2310E
H. neveulemairei 30-5D
H. longipalma 31-10C
H. nefrens 4798A
H. longipalma 1377B
H. cuprea
H. kochi 4822
H. longipalma 1356B
H. kochi AP18
H. kochi 3TK17B
H. hirsuta 30-5C
H. kochi 16 2B
H. simplex
H. fransiscae CHL-1
H. armata 26-2A
H. sp.1 H4(1)
H. sp.1 H4(2v)
H. kochi 2319-B
H. kochi 3TK27
H. montforti 2015 2D
H. kochi 4747
H. curvispina H2(1)
H. curvispina H3(2)
H. curvispina H3(1)
H. curvispina H5(1)
H. curvispina H1(2)
H. curvispina H4(2)
H. curvispina H1(1)
H. curvispina FC1
H. curvispina FC2
H. cajasi EC3-1
H. cajasi EC6-1
H. cajasi ecuador02
H. tiwanaku 4743
H. lucifugax
H. tiwanaku 4743
H. solida 2319-A
H. azteca
P. japonica IZCAS:IA1700i
Figure 2. Phylogeny of Hyalella reconstructed by Maximum Likelihood using all Uruguayan specimens sequenced for COI (369
bp), 28 Hyalella sequences from North and South America, and an outgroup taxon (Platorchestia japonica). Uruguayan samples,
all in “curvispina complex,” are denoted in grey color. Bootstrap values are next to the nodes.

Waller et al.: Molecular diversity of Hyalella
268
na complex”, stand out as well-dierentiated species with
high statistical support. For all genes and methods, sample
H4(1) was always clearly dierent from the rest of the Uru-
guayan samples, with the exception of the ABGD method
applied to COI and concatenated with COI. Besides, a
group of dierent species is considered by these methods
as a single species: H. kochi AP18, H. neveulemairei 30-
5D, H. nefrens 2310E, and H. hirsuta 30-5C. In this group
H. kochi AP18 is dierentiated analyzing H3 and 28S (by
ABGD) and H. hirsuta 30-5C analyzing COI (by PTP and
bPTP). ASAP excludes from this group H. kochi AP18 and
H. hirsuta 30-5C analyzing COI and 12S. The Uruguayan
samples, except H4(1), generally cluster as a single spe-
cies in COI, 12S, and 28S with ABGD, PTP and bPTP, and
the concatenate without COI (by PTP and bPTP).
4. Discussion
This study evaluated genetic diversity based on four loci in
Hyalella curvispina previously identied similar morphs
in Uruguay. In this way, we were able to: i) propose the
H.
neveulemairei
30
-
5D
H.
nefrens
4798A
H.
cuprea
H. azteca
H.
fransiscae
CHL
-
1
H.
armata
26
-
2A
P.
parapacifica
NIBRIV00000000
H. curvispina H8(2)
H. curvispina H8(4)
H. curvispina H8(3)
H. curvispina H8(1)
H. curvispina H1(5)
H. curvispina H1(3)
H. curvispina H1(4)
H. curvispina H1(1)
H. curvispina MVD
H. curvispina FC1
H. curvispina H4(2)
H. curvispina FC2
H. curvispina FC3
H. curvispina H7(1)
H. curvispina H7(3)
H. curvispina H7(4)
H. curvispina H7(2)
H. curvispina H6(2)
H. curvispina H5(1)
H. curvispina H6(4)
H. curvispina H2(4)
H. curvispina H2(2)
H. curvispina H2(3)
H. curvispina H2(1)
H. curvispina H3(1)
H. curvispina H3(2)
H. curvispina H3(3)
H.
kochi
4747
H.
kochi
3TK27
H.
kochi
2319B
H.
montforti
2015 2D
H.
montforti
1410C
H. sp.1 H4(1)
H. sp.1 H4(2v)
H. sp.1 H4(3)
H.
tiwanaku
4743
H.
tiwanaku
2304
H.
lucifugax
H. hirsuta 30
-
5C
H.
kochi
1356B
H.
nefrens
4798A
H.
longipalma
1356B
H.
cajasi
EC3
-
1
H.
cajasi
ecuador02
H.
kochi
3TK10
Figure 3. Phylogeny of Hyalella reconstructed by Maximum Likelihood using all Uruguayan specimens sequenced for 12S (452
bp), 21 Hyalella sequences from North and South America, and an outgroup taxon (Platorchestia parapacica). Uruguayan sam-
ples, all in “curvispina complex,” are denoted in grey color. Bootstrap values are next to the nodes.

Arthropod Syematics & Phylogeny 80, 2022, 261–278 269
paraphyletic status of the “curvispina complex”, ii) esti-
mate its phylogenetic position in the framework of previ-
ously proposed hypotheses of the genus, ii) suggest the
presence at least one provisional species new to Uruguay
that is probably cryptic species as showed in the phyloge-
netic trees reported by Jurado-Rivera et al. (2020).
4.1. Genus Hyalella in Uruguay
The phylogeny obtained incorporates representatives of
the genus Hyalella from Uruguay, including H. curvispi-
na, into the general phylogeny previously proposed for
the genus. As expected, the general topology concerning
the other species coincides with that previously reported
by Zapelloni et al. (2021), which used identical Hyalella
sequences. However, we used only those vouchers se-
quenced for all genes. We recovered all proposed clades
by Adamowicz et al. (2018) (except B, which we did
not include) as monophyletic with high support. In addi-
tion to previous studies, H. azteca and H. armata 26-2A
(Clade C) are more basal than the southern clades. The
closer relationship between these species presents a pos-
terior probability value of less than 0.5 so we cannot trust
the phylogenetic relationships inferred at these nodes, as
Zapelloni et al. (2021) show. The South American species
are clustered together and include dierent clades and
Uruguayan samples. However, the relationships between
groups A, E, D, and F could not be resolved with signi-
cant statistical support (except for basal group C).
The “curvispina complex” forms a monophyletic group
with H. montforti 2015 2D and H. kochi 4747, 2319B,
3TK27, both corresponding to clade E from the northern
Altiplano (Zapelloni et al. 2021). However, H. kochi is
H. montforti 1410C
H. kochi 2319B
H. montforti 2015 2D
H. kochi 3TK27
H. kochi 4747
H. curvispina H7(1)
H. curvispina MVD
H. curvispina FC1
H. curvispina H6(1)
H. curvispina H5(1)
H. curvispina H2(1)
H. curvispina H3(1)
H. curvispina H8(1)
H. curvispina H1(1)
H. sp.1 H4(1)
H. kochi 16 2B
H. kochi 4822
H. kochi 3TK17B
H. kochi AP18
H. tiwanaku 4816B
H. longipalma 31-10C
H. nefrens 2310E
H. longipalma 1377B
H. longipalma 1356B
H. nefrens 4798A
H. neveulemairei 30-5D
H. hirsuta 30-5C
H. tiwanaku 2304
H. cajasi EC6-1
H. neveulemairei 2316D
H. cajasi ecuador02
H. fransiscae CHL-1
H. kochi 3TK10
H. armata 26-2A
H. muerta
H. azteca
H. cajasi EC3-1
P. japonica YP8
Figure 4. Phylogeny of Hyalella reconstructed by Maximum Likelihood using all Uruguayan specimens sequenced for 28S (605
bp), 27 Hyalella sequences from North and South America, and an outgroup taxon (Platorchestia japonica). Uruguayan samples,
all in “curvispina complex,” are denoted in grey color. Bootstrap values are next to the nodes.

Waller et al.: Molecular diversity of Hyalella
270
also elsewhere in the phylogenetic tree, so some of these
vouchers considered as H. kochi are probably cryptic spe-
cies. Then, the close association between H. curvispina
and H. kochi (sensu stricto) should be taken with cau-
tion. In any case, all the samples associated with Uru-
guayan Hyalella come from Peru, which is very distant
geographically. This study is the rst one that includes
Hyalella species to the east of the Andes, i.e., “curvispina
complex”. The close link between the samples from Peru
and Uruguay may reect the scarce knowledge about the
genus. It would be necessary to include many more South
American representatives (in addition to those already in-
cluded from Peru, Bolivia, and Chile). In the future, this
phylogenetic information could shed light on the histori-
cal biogeography of Hyalella and the palaeoclimatic his-
tory of the continent.
On the other hand, we found high genetic variations
between H. curvispina and associated morphs and at least
one provisional new species. We consider that specimens
collected at the type locality of H. curvispina, which also
have the expected morphology, are indeed topotypes of
this species. Other samples, collected in distant lakes
(except H7(1) and H4(1) + H4(2v)), show higher an-
ity with that species, with high statistical support in the
species phylogeny that includes the COI (96%). Indeed,
many of these morphs together with H. curvispina are ge-
netically identical in some of their markers. The genetic
dierentiation found among most of them was moderate,
in the range expected for intraspecic dierentiation, and
consistent among most genes (Supplementary Tables S3,
S4, S5, S6). Thus, we propose that all these morphs are
part of the H. curvispina variation generated in the re-
gion. All the Hyalella specimens found in Uruguay have
morphological characteristics that dene the “curvispi-
na complex”, they are: smooth body surface, presence
of curved setae on inner side of inner ramus of uropod I
and sternal gills present in segments 2 to 7. On the other
hand, we found one morph, H4 specimens H4(1), H4(2v)
and H4(3) (only for 12S gene), which show greater dif-
ferentiation than the rest of the morphs (Fig. 2, 3, 4, 5
H. neveulemairei 30-5D
H. nefrens 2310E
H. longipalma 31-10C
H. hirsuta 30-5C
H. tiwanaku 4816B
H. kochi AP18
H. kochi 4822
H. neveulemairei 2316D
H. curvispina MVD
H. curvispina H2(1)
H. cajasi EC6-1
H. cajasi EC3-1
H. sp.1 H4(1)
H. curvispina H1(1)
H. curvispina H5(1)
H. curvispina H3(1)
H. curvispina H8(1)
H. curvispina H6(1)
H. curvispina H7(1)
H. kochi 4747
H. montforti 2015 2D
H. kochi 2319-B
H. curvispina FC1
H. tiwanaku 4743
H. tiwanaku 2304
H. solida 2319A
H. azteca
H. armata 26-2A
P. pacifica DF1
Figure 5. Phylogeny of Hyalella reconstructed by Maximum Likelihood using all Uruguayan specimens sequenced for H3 (332 bp),
18 Hyalella sequences from North and South America, and an outgroup taxon (Platorchestia pacica). Uruguayan samples, all in
“curvispina complex,” are denoted in grey color. Bootstrap values are next to the nodes.

Arthropod Syematics & Phylogeny 80, 2022, 261–278 271
and Supplementary Tables S3, S4, S5, S6). Although the
COI showed signicant dierentiation between the dif-
ferent morphs, this is not consistent with the information
provided by another mitochondrial marker, 12S, nor by
nuclear markers (see below). We propose maintaining
the complex name Hyalella curvispina for most of these
forms, and suggest one new provisional species corre-
sponding to specimens H4(1), H4(2v) and H4(3)) (La-
0.03
Platorchestia sp.
H. azteca
H. armata 26-2A
H. montforti 2015 2D
H. kochi 2319B
H. kochi 4747
H. curvispina FC1
H. curvispina H1(1)
H. curvispina H2(1)
H. curvispina H3(1)
H. curvispina H5(1)
H. sp.1 H4(1)
H. tiwanaku 2304
H. cajasi EC3-1
H. nefrens 2310E
H. neveulemairei 30-5D
H. hirsuta 30-5C
H. kochi AP18
0.46
0.61
1
0.87
0.52
0.96
0.87
0.48
0.97
0.85
0.99
0.91
0.35
1
0.39
0.96
1
C
E
D
F
A
G
Figure 6. Phylogeny of Hyalella reconstructed by Bayesian analysis. Samples of H. curvispina FC1, H1(1), H2(1), H3(1), H5(1)
and H. sp.1 H4(1) are Uruguayan samples. Eleven Hyalella sequences from North and South America and outgroup taxon (sequenc-
es of dierent specimens of the genus Platorchestia sp.) were included. The consensus tree is based on 1758 bp from COI, 12S, 28S,
and H3 concatenated datasets. Posterior probabilities are noted next the nodes. Clades from A to F were dened in Zapelloni et al.
(2021); Clade G is proposed in the present study.
H. montforti 2015_2D
H. kochi 2319B
H. kochi 4747
H. curvispina FC1
H. curvispina MVD
H. curvispina H7(1)
H. curvispina H1(1)
H. curvispina H8(1)
H. curvispina H2(1)
H. curvispina H3(1)
H. curvispina H5(1)
H. curvispina H6(1)
H. sp.1 H4(1)
H. tiwanaku 2304
H. cajasi EC3-1
H. nefrens 2310E
H. neveulemairei 30-5D
H. hirsuta 30-5C
H. kochi AP18
H. azteca
H. armata 26-2A
Platorchestia sp.
C
E
G
D
F
A
0.03
0.53
0.77
0.66
0.96 0.82
0.36 0.94
0.45 0.9
0.93
0.33
0.88
0.99
0.75
0.52
0.47
0.56
0.65
1
Figure 7. Phylogeny of Hyalella reconstructed by Bayesian analysis. Samples H. curvispina FC1, MVD, H1(1), H2(1), H3(1),
H5(1), H6(1), H7(1), H8(1) and H. sp.1 H4(1) are Uruguayan samples. Eleven Hyalella sequences from North and South America
and outgroup taxon (sequences of dierent specimens of the genus Platorchestia sp.) were included. Shown is the consensus tree
based on 1389 bp from 12S, 28S and H3 concatenated datasets. Posterior probabilities are noted next to the nodes. Clades from A to
F were dened in Zapelloni et al. (2021); Clade G is proposed in the present study.

Waller et al.: Molecular diversity of Hyalella
272
valleja locality) named H. sp.1 in phylogenetic trees. The
morphological characteristics that dierentiate H. sp.1
from H. curvispina are subtle, namely, telson with three
strong setae distributed in two groups in the apical margin
and inner median face of propodus in gnathopod 1 with
an oblique row of 9–10 short pectinate setae.
In addition, all Uruguayan samples are more closely
related to clade E (97%), so they could be considered part
of clade E. Genetic distances between clades measured
as K2P for 28S are in the range of 1.1% to 6.4% (Ada-
mowicz et al. 2018). The H7(1) sample belongs to clade
E, with a genetic distance to it of 0.4%. The remaining
samples from Uruguay, except for H. sp.1, belong to this
clade with a distance of 1.1%. In addition, we suggest a
new clade, clade G, formed by H. sp.1; the genetic dis-
tance between this new clade and clade E is 1.8%.
The information provided by each of the markers in-
dependently, and the markers as a whole, suggests a geo-
graphical dierentiation within the H. curvispina clade
(part of group E). We now considered “H. curvispina
clade”, the clade including the type locality and related
localities, excluding H. sp.1 and the sample H7(1), with
the closest populations being more genetically associat-
ed. However, the relationship between these “geographic
groups” is still uncertain because of low statistical sup-
port. And shallow supports would reect a recent dier-
entiation. A multilocus approach, including thousands of
markers (e.g., obtained from RADseq and NGS approx-
imations), will probably be helpful to resolve the critical
ne-scale aspect of phylogeography needed to ascertain
further details of this dierentiation.
In the locality of Lavalleja, we collected three sam-
ples, two of them with a higher genetic divergence (in
mitochondrial and nuclear markers) regarding the varia-
tion of most Uruguayan variants/morphs and preliminary
morphological evaluation shows dierences between this
sample and H. curvispina (Waller, data not shown). We
suggest that this sample, i.e. specimens H4(1), H4(2v)
and H4(3), would be considered a dierent species H.
sp.1. The other one H4(2) presents low genetic diver-
gence and morphologically corresponds to H. curvispi-
na. Molecular species delimitation analyses strongly
support this conclusion. Dierent methods identify the
sample H4(1) as an independent species for all genes
except COI and their concatenation. Since this sample
belongs to a new species, it conrms sympatric species
living together in the same pool. Several authors have
been observed sympatric distributions in the two species
complexes evaluated at the molecular level. In particu-
lar, the cryptic species of H. azteca from North Ameri-
ca (Wellborn and Cothran 2004; Witt and Hebert 2000)
and in Hyalella species from Brazil (da Silva Castiglioni
and Bond-Buckup 2008a, 2008b, 2009; González et al.
2006). On the other hand, sample H7 shows greater af-
nity to two samples of H. kochi (4747, 2319B) and to
H. montforti 2015 2D than to the rest of the Uruguayan
samples. Although H7 was sequenced for mitochondrial
12S and nuclear markers and not for COI due to techni-
cal problems, we propose this sample be a new species.
However, molecular species dierentiation analyses do
not discriminate it as a dierent species, but it is integrat-
ed into the E clade. It would be necessary to review the
systematic of group E as a whole and establish clearer
species boundaries or suggest synonymy to H. montforti
(by the principle of priority of the zoological nomencla-
tural code). Overall, these results highlight the relevance
of including molecular systematics studies in determin-
ing the genus Hyalella.
4.2. Species identification based on
genetic divergence estimates
The Species Screening Threshold criterion (SST) (Witt
et al. 2006) has been used in the molecular systematics
of the genus for the last two decades (Dionne et al. 2011;
Vergilino et al. 2012) and considers that Hyalella species
diverge from each other by 3.75% for COI sequences
using the K2P distance (Kimura 1980). Following this
criterion, the presence of cryptic species could be consid-
ered for the H. curvispina species complex, particularly
six species of the morphs (excluding H7 not sequenced
for COI). However, our analysis performed with a 369
bp of COI does not provide solid phylogenetic support
for this delimitation. Strikingly several recent Hyalella
COI barcoding studies have used even shorter fragments,
sometimes only 300 to 400 bp in length, for specimen
identication (Dionne et al. 2011; Major et al. 2013; Stutz
et al. 2010).
On the other hand, comparing the information pro-
vided by COI with that from various markers, both mi-
tochondrial and nuclear, reveals the specic diculties
associated with this marker. Firstly, the genetic distanc-
es in the COI marker are practically the same between
dierent levels of variation. In particular, values in the
order of 4 to 29% are found between poorly dierenti-
ated species within the “curvispina complex”, as well as
between highly dierentiated species (some species of
the “curvispina complex” with other Hyalella species),
or between Hyalella species with the outgroup taxon This
condition may be due to saturation (i.e., multiple substi-
tutions at the same site in a sequence leads to underes-
timation of actually occurring mutations) and leads to
homoplasy and an underestimation of divergence times
between haplotypes observed, particularly for older phy-
logenetic events (e.g., Wilke et al. 2009). However, when
analyzing the degree of saturation with the DAMBE pro-
gram, low saturation levels were observed for all mark-
ers, including COI without the third codon position. The
absence of signicant-high saturation for COI could be
due to the sequence size (369 bp), as saturation in COI
for Hyalella has been recorded at sequence sizes larger
than 500 bp (Major et al. 2013; Worsham et al. 2017; Za-
pelloni et al. 2021).
In turn, the 12S gene gives us dierent information to
the COI, although both are mitochondrial genes and are
physically linked because this genome does not recom-
bine (Meyer 1993; Saldamando and Marquez 2012). Al-
though both markers share the evolutionary history, the
12S gene is the most conserved within the mitochondri-

Arthropod Syematics & Phylogeny 80, 2022, 261–278 273
al genome (Arif and Khan 2009). Thus, unlike the COI,
12S might not be saturated (Major et al. 2013; Witt and
Hebert 2000; Worsham et al. 2017) and oers more accu-
rate information at this level of comparison (Arbogast et
al. 2002; Major et al. 2013). On the other hand, pairwise
genetic distances with all markers except COI (with nu-
clear genes, 12S, or even the concatenated construct of all
genes) show dierent sharpie levels of variation within
the genus (i.e., intra- and interspecic) and among genera
(between Hyalella and outgroup taxon). Pairwise genet-
ic distances calculated with COI (Table 4) do not show
this pattern, and variations within and among genera are
of similar magnitude. Thus, we believe that for Hyalella,
12S is more reliable than COI and that the SST criterion
should be applied to another marker.
4.3. General considerations
We found little dierentiation in the markers assessed in
this study, both nuclear and mitochondrial 12S. However,
mitochondrial dierentiation at the COI level is high (in
fact saturated), which has also been reported in studies
conducted for other complexes (Major et al. 2013; Witt
and Hebert 2000; Worsham et al. 2017). These results
may reect the current dierentiation processes of the
genus across regions, in principle dominated by coloni-
zation and extinction events (Zapelloni et al. 2021). Simi-
larly (Duan et al. 2000), and despite the dierence in geo-
graphical scale of the studies, unique variants and high
levels of variation are also found between populations
50–200 km apart. This dierentiation is associated with
the geographic variation. Closer populations are more
phylogenetically linked, suggesting some connectivity
between populations and diversication in the presence
of gene ow within H. curvispina (sensu latu).
Also, in agreement with previous studies about other
species complexes in the genus, we found that the “cur-
vispina complex” is paraphyletic respect to the species
H. kochi 4747, 2319B, 3TK27 (sensu latu) and H. mont-
forti 2015 2D. These results suggest that adaptive and
morphological convergence in this group is high (Ada-
mowicz et al. 2018), and the “curvispina complex” is no
exception. In particular, morphological variation does not
match genetic dierentiation, which may be related to the
recurrent selection of similar morphologies in the face of
the same ecological dierent challenges. In contrast to
the genetic phylogeny, all samples from Uruguay, includ-
ing H. sp.1 and sample H7(1), are more similar morpho-
logically to each other and H. curvispina than to H. kochi
(2319B, 4747 and 3TK27) and H. montforti 2015 2D.
Habitat specialization and trophic regimes could explain
the convergence of morphotypes observed in Hyalella
specimens from Uruguay. Hyalella curvispina has varied
feeding habits: shredder, predator, scrapper, and collec-
tor-collector (Cummins et al. 2005; Giorgi and Tirabos-
chi 1999; Saigo et al. 2009; Wantzen and Wagner 2006).
They are also food for other macroinvertebrates, sh, am-
phibians, and birds (Colla and César 2019). As a defense
mechanism against predation pressure, Hyalella species
occupy dierent habitats (Wellborn 1995). They inhabit a
variety of freshwater environments such as lakes, ponds,
and streams, clinging to vegetation and burrowing in
bottom sediments, where they are important members of
the benthic fauna (da Silva Castiglioni and Bond Buckup
2008B; Grosso and Peralta 1999; Wellborn 1995).
On the other side, the two phylogenetically most
closely related species to the H. curvispina complex (i.e.,
H. kochi and H. montforti) share some morphological
characteristics with it. Regarding H. kochi both species
have a smooth body surface, the inner face of propodus
of gnathopod 1 with seven setae, the presence of curved
setae in the inner ramus of uropod 1. However, the main
characteristic that distinguishes these two species is the
presence of six pairs (from segment 2 to 7) of sternal gills
in H. curvispina, while H. kochi has only ve pairs (from
segment 3 to 7), and the consistency of this character
makes it relevant in the evolutionary relationships within
the genus (González and Watling 2001). Compared with
H. montforti, both species have six pairs of sternal gills
and curved setae in the inner ramus of uropod 1. Still, the
main characteristic that distinguishes these two species is
the body with dorso-posterior anges on pereon segment
7, pleonite 1, 2, and 3 in H. montforti. Since these forms
have a common ancestor and live in dierent environ-
ments, it is reasonable to think that the common morpho-
logical characteristics reect phylogenetic inertia, while
others would be local adaptations.
5. Acknowledgments
This research was supported by Agencia Nacional de Investigación e
Innovación under the code POS NAC 2019 1 157755, and Comisión
Sectorial de Investigación Cientíca (CSIC) de la Universidad de la
República.
6. References
Adamowicz SJ, Marinone MC, Menu-Marque S, Martin JW, Allen
DC, Pyle MN, De los Ríos P, Sobel CN, Ibañez C, Pinto J, Witt
JDS (2018) The Hyalella (Crustacea: Amphipoda) species cloud of
the ancient Lake Titicaca originated from multiple colonizations.
Molecular Phylogenetics and Evolution 125: 232–242. https://doi.
org/10.1016/j.ympev.2018.03.004
Arbogast BS, Edwards SV, Wakeley J, Beerli P, Slowinski JB (2002)
Estimating divergence times from molecular data on phylogenetic
and population genetic timescales. Annual Review of Ecology and
Systematics 33: 707–740. https://doi.org/10.1146/annurev. ecol sys.
3 3.010802.150500
Arif IA, Khan HA (2009) Molecular markers for biodiversity analysis
of wildlife animals: A brief review. Animal Biodiversity and Con-
servation, 32(1): 9–17.
Babin-Fenske JJ, Merritt TJS, Gunn JM, Walsh T, Lesbarrères D (2012)
Phylogenetic analysis of Hyalella colonization in lakes recovering
from acidication and metal contamination. Canadian Journal of
Zoology 90(5): 624–629. https://doi.org/10.1139/z2012-028
Baldinger AJ (2004) A new species of Hyalella (Crustacea: Amphipoda:
Hyalellidae) from Ash Springs, Lincoln County, Nevada, USA, with

Waller et al.: Molecular diversity of Hyalella
274
a key to the species of the genus in North America and the Caribbe-
an region. Journal of Natural History 38(9): 1087–1096. https://doi.
org/10.1080/0022293031000075367
Bickford D, Lohman DJ, Sodhi NS, Ng PKL, Meier R, Winker K, In-
gram KK, Das I (2007) Cryptic species as a window on diversity
and conservation. Trends in Ecology and Evolution 22(3): 148–155.
https://doi.org/10.1016/j.tree.2006.11.004
Bouseld E (1996) A contribution to the reclassication of Neotropi-
cal freshwater Hyalellid amphipods (Crustacea: Gammaridea, Tal-
itroidea). Bollettino del Museo Civico di Storia Naturale di Verona
20: 175–224.
Casset MA, Momo FR, Giorgi ADN (2001) Dinámica poblacional de
dos especies de anfípodos y su relación con la vegetación acuática
en un microambiente de la cuenca del río Luján (Argentina). Ecolo-
gia Austral 11(2): 79–85.
Cavalieri F (1968) Hyalella pampeana sp. n., una nueva especie de
anpodo de agua dulce. (Gammaridea: Hyalellidae). Neotropica
14(59): 107–117.
Colgan DJ, McLauchlan A, Wilson GDF, Livingston SP, Edgecombe
GD, Macaranas J, Cassis G, Gray MR (1998) Histone H3 and U2
snRNA DNA sequences and arthropod molecular evolution. Aus-
tralian Journal of Zoology 46(5): 419–437. https://doi.org/10.1071/
ZO98048
Colla MF, César II (2019) Ecological aspects of natural populations of
Hyalella pampeana (Crustacea, amphipoda, Hyalellidae) from the
natural reserve island of Martín García (Río de La Plata, Argentina).
Anais Da Academia Brasileira de Ciencias 91(1): 1–21. https://doi.
org/10.1590/0001-3765201920170928
Cummins KW, Merritt RW, Andrade PCN (2005) The use of inver-
tebrate functional groups to characterize ecosystem attributes in
selected streams and rivers in south Brazil. Studies on Neotro-
pical Fauna and Environment 40(1): 69–89. https://doi.org/ 10.1-
0 80/01650520400025720
da Silva Castiglioni D, Bond-Buckup G a(2008) Pairing and reproduc-
tive success in two sympatric species of Hyalella (Crustacea, Am-
phipoda, Dogielinotidae) from southern Brazil. Acta Oecologica
33(1): 49–55. https://doi.org/10.1016/j.actao.2007.09.008
da Silva Castiglioni D, Bond Buckup G b(2008) Ecological traits of
two sympatric species of Hyalella Smith, 1874 (Crustacea, Amphi-
poda, Dogielinotidae) from southern Brazil. Acta Oecologica 33(1):
36–48. https://doi.org/10.1016/j.actao.2007.09.007
da Silva Castiglioni D, Bond-Buckup G (2009) Egg production of
two sympatric species of Hyalella Smith, 1874 (Crustacea, Am-
phipoda, Dogielinotidae) in aquaculture ponds in southern Brazil.
Journal of Natural History 43 (21–22): 1273–1289. https://doi.org/
10.1080/00222930902903756
Darriba D, Taboada GL, Doallo R, Posada D (2012) JModelTest 2:
More models, new heuristics and parallel computing. Nature Meth-
ods 9(8): 772. https://doi.org/10.1038/nmeth.2109
de Oliveira PL (1953) Crustacea Amphipoda do Rio de Janeiro. Memo-
rias Do Instituto Osvaldo Cruz 51(1): 288–376.
del Puerto L, García Rodriguez F, Bracco R, Castiñeira C, Blasi A, Inda
H, Mazzeo N, Rodríguez A (2011) Evolución climática holocénica
para el sudeste de Uruguay Análisis multi-proxy en testigos de lagu-
nas costeras. In: García Rodriguez F (Ed.) El Holoceno en la zona
costera de Uruguay 117–147.
Dionne K, Vergilino R, Dufresne F, Charles F, Nozais C (2011) No
evidence for temporal variation in a Cryptic species community of
freshwater amphipods of the Hyalella azteca species complex. Di-
versity 3(3): 390–404. https://doi.org/10.3390/d3030390
Dos Santos ALF, Araujo PB, Bond-Buckup G (2008) New species and
new reports of Hyalella (Crustacea, Amphipoda, Dogielinotidae)
from Argentina. Zootaxa 36(1760): 24–36. https://doi.org/10.11646/
zootaxa.1760.1.2
Duan Y, Guttman SI, Oris JT, Bailer AJ (2000) Genetic structure and
relationships among populations of Hyalella azteca and H. montezu-
ma (Crustacea:Amphipoda). Journal of the North American Bentho-
logical Society 19(2): 308–320. https://doi.org/10.2307/1468073
Filatov DA (2002) PROSEQ: A software for preparation and evoluction-
ary analysis of DNA sequence data sets. Molecular Ecology Notes 2:
621–624. https://doi.org/10.1046/j.1471-8286.2002.00313.x
Giorgi A, Tiraboschi B (1999) Evaluacion experimental del efecto de
dos grupos de macroinvertebrados (anpodos y gasteropodos) sobre
algas epitas. Ecologia Austral 9(1–2): 35–44.
González ER, Bond-Buckup G, Araujo PB (2006) Two new species of
Hyalella from Southern Brazil (Amphipoda: Hyalellidae) with a tax-
onomic key. Journal of Crustacean Biology 26(3): 355–365. https://
doi.org/10.1651/C-2599.1
González ER, Watling L (2001) Three new species of Hyalella from
Chile (Crustacea: Amphipoda: Hyalellidae). Hydrobiologia 464:
175–199. https://doi.org/10.1023/A:1013961904370
González ER, Watling L (2003) Two new species of Hyalella from
Lake Titicaca, and redescriptions of four others in the genus (Crus-
tacea: Amphipoda). Hydrobiologia 497: 181–204. https://doi.org/
10.1023/A:1025451813972
Grosso LE, Peralta M (1999) Anfípodos de agua dulce sudamericanos.
Revisión del género Hyalella Smith. I. Acta Zoologica Lilloana,
45(1): 79–98.
Juan C, Jurado-Rivera JA, Moreno E, Wol C, Jaume D, Pons J (2016)
The mitogenome of the amphipod Hyalella lucifugax (Crustacea)
and its phylogenetic placement. Mitochondrial DNA Part B 1(1):
755–756. https://doi.org/10.1080/23802359.2016.1214554
Jurado-Rivera JA, Zapelloni F, Pons J, Juan C, Jaume D (2020) Mor-
phological and molecular species boundaries in the Hyalella
species ock of Lake Titicaca (Crustacea: Amphipoda). Contri-
butions to Zoology 89(4): 353–372. https://doi.org/10.6084/m9.
gshare.12023304
Kimura M (1980) A simple method for estimating evolutionary rates
of base substitutions through comparative studies of nucleotide se-
quences. Journal of Molecular Evolution 16(2): 111–120. https://
doi.org/10.1007/BF01731581
Kumar S, Stecher G, Li M, Knyaz C, Tamura K (2018) MEGA X: Mo-
lecular evolutionary genetics analysis across computing platforms.
Molecular Biology and Evolution 35(6): 1547–1549. https://doi.
org/10.1093/molbev/msy096
Laufer G, Arim M, Loureiro M, Piñeiro-Guerra JM, Clavijo-Baquet S,
Fagúndez C (2009) Diet of four annual killishes: An intra and inter-
specic comparison. Neotropical Ichthyology 7(1): 77–86. https://
doi.org/10.1590/S1679-62252009000100010
Limberger M, da Silva Castiglioni D, Graichen D, Angelo S (2021) A
new species of freshwater amphipod (Crustacea, Amphipoda, Hy-
alellidae) from Southern Brazil. Zootaxa. 5026 (2): 182–200. https://
doi.org/10.11646/zootaxa.5026.2.2
Major K, Soucek DJ, Giordano R, Wetzel MJ, Soto-Adames F (2013)
The common ecotoxicology laboratory strain of Hyalella azteca is
genetically distinct from most wild strains sampled in Eastern North
America. Environmental Toxicology and Chemistry 32(11): 2637–
2647. https://doi.org/10.1002/etc.2355

Arthropod Syematics & Phylogeny 80, 2022, 261–278 275
Minh BQ, Nguyen MAT, von Haeseler A (2013) Ultrafast approxima-
tion for phylogenetic bootstrap. Molecular Biololgy and Evolution
30:1188-1195. (http://dx.doi.org/10.1093/molbev/mst024)
Meyer A (1993) Evolution of mitochondrial DNA in shes. In: Ho-
chachka P, Mommsen T (Eds.). Biochemistry and molecular biology
of shes. Elsevier. Vol. 1, Issue 7 38 pp.
Miller SA, Dykes D, Polesky H (1988) A simple salting out for ex-
tracting DNA from human nucleated cells. Nucleic Acids Research
16(3): 883–893.
Nguyen L-T, Schmidt HA, von Haeseler A, Minh BQ (2015) IQ-TREE:
A fast and eective stochastic algorithm for estimating maximum
likelihood phylogenies. Molecular Biology and Evolution 32: 268-
274. http://dx.doi.org/10.1093/molbev/msu300
Peralta M, Grosso LE (2009) Crustacea, Syncarida, Amphipoda y De-
capoda. In: Domínguez E, Fernández HR (Eds.). Macroinvertebra-
dos bentónicos sudamericanos. Fundación 469–496.
Posada D, Crandall KA (1998) Modeltest: testing the model of DNA
substitution. In: Bioinformatics Applications Note 9 (14): 817–818.
https://doi.org/10.1093/bioinformatics/14.9.817
Puillandre N, Brouillet S, Achaz G (2021) ASAP: assemble species by
automatic partitioning Molecular Ecology Resources. 21(2): 609–
620. https://doi.org/10.1111/1755-0998.13281
Puillandre N, Lambert A, Brouillet S, Achaz G (2012) Automatic
Barcode Gap Discovery for primary species delimitation. Mo-
lecular Ecology 21(8): 1864–77. https://doi.org/10.1111/j.1365-
294X.2011.05239.x
Rambaut A (2018) Fig tree Version 1.4.4. http://tree.bio.ed.ac.uk/soft-
ware/gtree
Reis GO, Penoni LR, Bueno AAP (2020) First record of the genus Hy-
alella (Amphipoda: Hyalellidae) from Santa Catarina State, Brazil,
with description of two new species. Biota Neotropica 20(2): 2–12.
https://doi.org/10.1590/1676-0611-BN-2019-0879
Rodrigues SG (2016) Filogenia Molecular, Biogeograa e Estrutura
Populacional de Anfípodos de Água Doce (Crustacea, Hyalellidae)
da América do Sul. PhD Thesis Universidade Federal de Lavras.
Brasil.
Saigo M, Marchese M, Montalto L (2009) Hábitos alimentarios de Hy-
alella curvispina Shoemaker, 1942 (Amphipoda : Gammaridea) en
ambientes lénticos de la llanura aluvial del río Paraná medio. Natura
Neotropicalis (40): 43–59.
Saldamando CI, Marquez EJ (2012) Aproximación a la logenia de
Spodoptera (Lepidoptera: Noctuidae) con el uso de un fragmento del
gen de la citocromo oxidasa I (COI). Revista de Biologia Tropical
60(3): 1237–1248. https://doi.org/10.15517/RBT.V60I3.1775
Shoemaker R (1942) A new species of Amphipoda from Uruguay and
Brazil. Journal of the Washington Academy of Sciences 32(1): 80–
83.
Stock JH, Platvoet D (1991) The freshwater amphipoda of the falkland
islands. Journal of Natural History 25(6): 1469–1491. https://doi.
org/10.1080/00222939100770941
Stutz HL, Shiozawa DK, Evans RP (2010) Inferring dispersal of aquat-
ic invertebrates from genetic variation: A comparative study of an
amphipod and mayy in Great Basin springs. Journal of the North
American Benthological Society 29(3): 1132–1147. https://doi.org/
10.1899/09-157.1
Suchard MA, Lemey P, Baele G, Ayres DL, Drummond AJ, Rambaut
A (2018) Bayesian phylogenetic and phylodynamic data integra-
tion using BEAST 1.10. Virus Evolution 4(1): 1–5. https://doi.
org/10.1093/ve/vey016
Talhaferro JT, Bueno AAP, Pires MM, Stenert C, Maltchick L, Kotzian
CB (2021) Three new species of Hyalella (Crustacea, Amphipoda,
Hyalellidae) from the Southern Brazilian Coastal Plain. Zootaxa
4970 (2): 257–292. https://doi.org/10.11646/zootaxa.4970.2.2
Vergilino R, Dionne K, Nozais C, Dufresne F, Belzile C (2012) Ge-
nome size dierences in Hyalella cryptic species. Genome 55(2):
134–139. https://doi.org/10.1139/g11-085
Wantzen KM, Wagner R (2006) Detritus processing by invertebrate
shredders: A neotropical-temperate comparison. Journal of the
North American Benthological Society 25(1): 216–232. https://doi.
org/10.1899/0887-3593(2006)25[216:DPBISA]2.0.CO;2
Wellborn GA (1995) Determinants of reproductive success in freshwa-
ter amphipod species that experience dierent mortality regimes.
Animal Behaviour 50(2): 353–363.
Wellborn GA, Broughton RE (2008) Diversication on an ecologically
constrained adaptive landscape. Molecular Ecology 17(12): 2927–
2936. https://doi.org/10.1111/j.1365-294X.2008.03805.x
Wellborn GA, Cothran RD (2004) Phenotypic similarity and dier-
entiation among sympatric cryptic species in a freshwater amphi-
pod species complex. Freshwater Biology 49(1): 1–13. https://doi.
org/10.1046/j.1365-2426.2003.01160.x
Witt JDS, Blinn DW, Hebert PDN (2003) The recent evolutionary origin
of the phenotypically novel amphipod Hyalella montezuma oers an
ecological explanation for morphological stasis in a closely allied
species complex. Molecular Ecology 12(2): 405–413. https://doi.
org/10.1046/j.1365-294X.2003.01728.x
Witt JDS, Hebert PDN (2000) Cryptic species diversity and evolution in
the amphipod genus Hyalella within central glaciated North Amer-
ica: A molecular phylogenetic approach. Canadian Journal of Fish-
eries and Aquatic Sciences 57(4): 687–698. https://doi.org/10.1139/
f99-285
Witt JDS, Threlo DL, Hebert PDN (2006) DNA barcoding reveals ex-
traordinary cryptic diversity in an amphipod genus: Implications for
desert spring conservation. Molecular Ecology 15(10): 3073–3082.
https://doi.org/10.1111/j.1365-294X.2006.02999.x
Worsham MLD, Julius EP, Nice CC, Diaz PH, Human DG (2017)
Geographic isolation facilitates the evolution of reproductive iso-
lation and morphological divergence. Ecology and Evolution 7(23):
10278–10288. https://doi.org/10.1002/ece3.3474
Xia X (2018) DAMBE7: New and improved tools for data analysis in
molecular biology and evolution. Molecular Biology and Evolution
35(6): 1550–1552. https://doi.org/10.1093/molbev/msy073
Zapelloni F, Pons J, Jurado-Rivera JA, Jaume D, Juan C (2021) Phy-
logenomics of the Hyalella amphipod species-ock of the Andean
Altiplano. Scientic Reports 11(1): 1–14. https://doi.org/10.1038/
s41598-020-79620-4
Zhang J, Kapli P, Pavlidis P, Stamatakis A (2013) A General Species
Delimitation Method with Applications to Phylogenetic Placements.
Bioinformatics 29 (22): 2869–2876.

Waller et al.: Molecular diversity of Hyalella
276
Supplementary material 1
Table S1
Authors: Waller A, González ER, Verdi A, Tomasco IH (2022)
Data type: .xls
Explanation note: Sampling localities of Hyalella in Uruguay and geographic coordinates.
Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/
licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely
share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source
and author(s) are credited.
Link: https://doi.org/10.3897/asp.80.e79498.suppl1
Supplementary material 2
Table S2
Authors: Waller A, González ER, Verdi A, Tomasco IH (2022)
Data type: .xlsx
Explanation note: For each Hyalella sample, species and procedence, voucher number, Molecular Operational Taxo-
nomic Units, and GenBank accession code.
Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/
licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely
share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source
and author(s) are credited.
Link: https://doi.org/10.3897/asp.80.e79498.suppl2
Supplementary material 3
Table S3
Authors: Waller A, González ER, Verdi A, Tomasco IH (2022)
Data type: .xls
Explanation note: Pairwise genetic distance of COI between sequences of Hyalella.
Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/
licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely
share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source
and author(s) are credited.
Link: https://doi.org/10.3897/asp.80.e79498.suppl3

Arthropod Syematics & Phylogeny 80, 2022, 261–278 277
Supplementary material 4
Table S4
Authors: Waller A, González ER, Verdi A, Tomasco IH (2022)
Data type: .xls
Explanation note: Pairwise genetic distance of 12S between sequences of Hyalella.
Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/
licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely
share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source
and author(s) are credited.
Link: https://doi.org/10.3897/asp.80.e79498.suppl4
Supplementary material 5
Table S5
Authors: Waller A, González ER, Verdi A, Tomasco IH (2022)
Data type: .xls
Explanation note: Pairwise genetic distance of 28S between sequences of Hyalella.
Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/
licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely
share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source
and author(s) are credited.
Link: https://doi.org/10.3897/asp.80.e79498.suppl5
Supplementary material 6
Table S6
Authors: Waller A, González ER, Verdi A, Tomasco IH (2022)
Data type: .xls
Explanation note: Pairwise genetic distance of H3 between sequences of Hyalella.
Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/
licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely
share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source
and author(s) are credited.
Link: https://doi.org/10.3897/asp.80.e79498.suppl16

Waller et al.: Molecular diversity of Hyalella
278
Supplementary material 7
Figures S1–S11
Authors: Waller A, González ER, Verdi A, Tomasco IH (2022)
Data type: .pdf
Explanation note: Figure S1: Maximum parsimony phylogeny of a partial COI sequence of Hyalella from North and
South America. — Figure S2: Maximum parsimony phylogeny of a partial 12S sequence of Hyalella from North
and South America. — Figure S3: Maximum parsimony phylogeny of a partial 28S sequence of Hyalella from
North and South America. — Figure S4: Maximum parsimony phylogeny of a partial H3 sequence of Hyalella
from North and South America. — Figure S5: Neighbor-joining phylogeny of a partial COI sequence of Hyalella
from North and South America. — Figure S6: Neighbor-joining phylogeny of a partial 12S sequence of Hyalella
from North and South America. — Figure S7: Neighbor-joining phylogeny of a partial 28S sequence of Hyalella
from North and South America. — Figure S8: Neighbor-joining phylogeny of a partial H3 sequence of Hyalella
from North and South America. — Figure S9: Phylogeny of Hyalella reconstructed by maximum likelihood imple-
mented by IQ-tree based on 1758 bp from COI, 12S, 28S, and H3 concatenated datasets. — Figure S10: Phylogeny
of Hyalella reconstructed by maximum likelihood implemented by IQ-tree based on 1389 bp from 12S, 28S, and
H3 concatenated datasets. — Figure S11: Molecular species delimitation methods (bPTP, PTP, ABGD, and ASAP)
were applied to genes COI, 12S, 28S, and H3 individually and concatenated in the Hyalella genus.
Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/
licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely
share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source
and author(s) are credited.
Link: https://doi.org/10.3897/asp.80.e79498.suppl16