Audible, but not ultrasonic, calls reflect surface-dwelling or subterranean specialization in pup and adult Brandt’s and mandarin voles

Abstract

For human-audible vocalizations (below 20 kHz) of rodents, subterranean lifestyle results in low-frequency calls coupled with low-frequency hearing. For ultrasonic vocalizations (above 20 kHz), the effect of subterranean lifestyle on the acoustics is unknown. This study fills this gap of knowledge, by comparing vocalizations of two closely related species, the surface-dwelling Brandt’s vole Lasiopodomys brandtii (17 pups, 19 adults) and the subterranean mandarin vole L. mandarinus (15 pups, 16 adults). As predicted, the audible calls (AUDs) were substantially higher-frequency in L. brandtii than in L. mandarinus, in either pups or adults. In contrast to AUDs, the ultrasonic calls (USVs) did not differ in frequency variables between species, in either pups or adults. Interspecies differences were found in duration: AUDs were shorter in adult L. brandtii than in adult L. mandarinus, USVs were longer in pup L. brandtii than in pup L. mandarinus, and the low-frequency USVs of adult L. brandtii were longer than low-frequency USVs of adult L. mandarinus. We advance a hypothesis that shift towards higher-frequency AUDs in L. brandtii compared to L. mandarinus was triggered by the evolutionary emergence of the high-frequency audible alarm call, which is only present in L. brandtii but absent in L. mandarinus. We discuss that USVs may be resistant to these selection pressures as close-distant social signals.

Significance statement
Relationship between ecological specialization, such as subterranean or surface-dwelling lifestyle, and the acoustic traits of communicative signals in rodents evoke interest for over than 30 years. So far, the relationship between vocalization and subterranean life (low-frequency calls and low-frequency hearing) was only reported for calls produced by rodents in human-audible range of frequencies. No data was available for ecological adaptations of ultrasonic calls; moreover, even the existence of ultrasonic calls in subterranean rodents was unknown to recent time. Comparative studies of closely related subterranean and surface-dwelling rodent species might highlight the evolution of acoustic traits related to these ecological specializations.

Full text

Vol.:(0123456789)
1 3
Behavioral Ecology and Sociobiology (2022) 76:106
https://doi.org/10.1007/s00265-022-03213-6
ORIGINAL ARTICLE
Audible, butnotultrasonic, calls reflect surface‑dwelling
orsubterranean specialization inpup andadult Brandt’s andmandarin
voles
MargaritaM.Dymskaya1· IlyaA.Volodin2,3 · AntoninaV.Smorkatcheva1 · NinaA.Vasilieva4 ·
ElenaV.Volodina3
Received: 7 December 2021 / Revised: 8 July 2022 / Accepted: 12 July 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022
Abstract
For human-audible vocalizations (below 20kHz) of rodents, subterranean lifestyle results in low-frequency calls coupled
with low-frequency hearing. For ultrasonic vocalizations (above 20kHz), the effect of subterranean lifestyle on the acoustics
is unknown. This study fills this gap of knowledge, by comparing vocalizations of two closely related species, the surface-
dwelling Brandt’s vole Lasiopodomys brandtii (17 pups, 19 adults) and the subterranean mandarin vole L. mandarinus
(15 pups, 16 adults). As predicted, the audible calls (AUDs) were substantially higher-frequency in L. brandtii than in L.
mandarinus, in either pups or adults. In contrast to AUDs, the ultrasonic calls (USVs) did not differ in frequency variables
between species, in either pups or adults. Interspecies differences were found in duration: AUDs were shorter in adult L.
brandtii than in adult L. mandarinus, USVs were longer in pup L. brandtii than in pup L. mandarinus, and the low-frequency
USVs of adult L. brandtii were longer than low-frequency USVs of adult L. mandarinus. We advance a hypothesis that shift
towards higher-frequency AUDs in L. brandtii compared to L. mandarinus was triggered by the evolutionary emergence
of the high-frequency audible alarm call, which is only present in L. brandtii but absent in L. mandarinus. We discuss that
USVs may be resistant to these selection pressures as close-distant social signals.
Signicance statement
Relationship between ecological specialization, such as subterranean or surface-dwelling lifestyle, and the acoustic traits
of communicative signals in rodents evoke interest for over than 30years. So far, the relationship between vocalization and
subterranean life (low-frequency calls and low-frequency hearing) was only reported for calls produced by rodents in human-
audible range of frequencies. No data was available for ecological adaptations of ultrasonic calls; moreover, even the existence
of ultrasonic calls in subterranean rodents was unknown to recent time. Comparative studies of closely related subterranean
and surface-dwelling rodent species might highlight the evolution of acoustic traits related to these ecological specializations.
Keywords Acoustic communication· Arvicolinae species· Audible and ultrasonic vocalization· Subterranean rodents
Introduction
Subterranean lifestyle is known in more than 250 rodent
species (Nevo 1999; Lacey etal. 2000; Begall etal. 2007a).
Life underground under poor ventilation and high humidity
leads to many morphological, physiological, and behavioral
adaptations (Begall etal. 2007a; Park etal. 2017; Vejmělka
etal. 2021) governed by the respective genes (e.g., Jiao etal.
2019; Bondareva etal. 2021; Sahm etal. 2021).
Deprivation from most sensory stimuli due to under-
ground life also affects rodent acoustic communication,
which is especially important in the conditions of dark
burrow tunnels, where visual communication is obstructed
(Begall etal. 2007b; Burda etal. 2007). Rodent human-audi-
ble calls (AUDs, below 20kHz) are more variable in social
than in solitary subterranean species (Dvořáková etal. 2016;
Communicated by E. Korpimäki
* Ilya A. Volodin
volodinsvoc@gmail.com
Extended author information available on the last page of the article

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
106 Page 2 of 17
Schleich and Francescoli 2018) and display a convergent
evolution of sound-producing and auditory systems (Begall
etal. 2007b). This convergent evolution involves the using of
low-frequency AUDs for communication (Nevo etal. 1987;
Credner etal. 1997; Begall etal. 2007b; Knotková etal.
2009; Pepper etal. 2017; Okanoya etal. 2018; Schleich and
Francescoli 2018) and the maximum hearing sensitivity at
low frequencies (Müller and Burda 1989; Heffner and Hef-
fner 1992; Kӧssl etal. 1996; Brückmann and Burda 1997;
Dent etal. 2018; Okanoya and Screven 2018). Communica-
tion with low-frequency AUDs is effective in burrows, as
these calls propagate to a distance of a few meters with-
out weakening and even become more intense due to bur-
row stethoscope effect (Heth etal. 1986; Lange etal. 2007;
Schleich and Antenucci 2009; Okanoya and Screven 2018).
Regarding the underground ultrasonic communication in
rodents, data are scarce. The concept of convergent evolution
of low-frequency hearing and low-frequency AUDs suggests
that ultrasonic calls (USVs, above 20kHz) might not be
used by subterranean rodents (Begall etal. 2007b). However,
recent studies show that some subterranean Arvicolinae spe-
cies, as adult northern mole voles Ellobius talpinus (Volodin
etal. 2022) and pup mandarin voles Lasiopodomys man-
darinus (Yu etal. 2011), are capable of producing USVs.
Furthermore, one recent study reports the audition in the
ultrasonic range of frequencies in the subterranean rodent,
the coruro Spalacopus cyanus (Caspar etal. 2021). So, a
convergent evolution of ultrasonic vocalization and high-
frequency hearing is potentially expectable in subterranean
rodents.
Comparative studies of AUDs and USVs of subterranean
and surface-dwelling Arvicolinae species might highlight
the relationship between lifestyle and the acoustic traits of
these calls (Rutovskaya 2018). Sister species, the Brandt’s
vole L. brandtii and mandarin vole L. mandarinus, represent
a promising comparative model for revealing the acoustic
adaptations related to surface-dwelling or subterranean life-
style. Whereas L. mandarinus is adapted to subterranean
life, L. brandtii displays surface-dwelling lifestyle (Tai and
Wang 2001; Dong etal. 2018; Sun etal. 2019; Cui etal.
2020). While L. brandtii forage on grass aboveground (Cui
etal. 2020), L. mandarinus primarily forage in tunnels of
up to 95m long with multiple (up to 70) exits and on sur-
face in immediate vicinity to burrow entrance (Dmitriev
etal. 1980; Smorkatcheva etal. 1990). Phylogenetic anal-
yses suggest that L. brandtii and L. mandarinus were the
last two species diverged from the common trunk of the
genus Lasiopodomys approximately 0.5–0.95 million years
ago (Abramson etal. 2009; Li etal. 2017; Shi etal. 2021).
Lifestyle (surface-dwelling or subterranean) of the com-
mon ancestor of L. brandtii and L. mandarinus is unknown.
Both species are steppe-dwellers in China, Mongolia, and
Transbaikalia (Russia), with partly overlapping distribution
areas between species (Smorkatcheva etal. 1990; Smith and
Xie 2008; Alexeeva etal. 2015; Lebedev etal. 2016). Both
species live in extended family-based groups (L. brandtii:
Dmitriev etal. 1980; L. mandarinus: Smorkatcheva 1999;
Tai and Wang 2001). Both species become mature early: in
L. brandtii, sexual maturity for males and females is reached
at about 35days of age (Zorenko and Jakobsone 1986). In L.
mandarinus, males and females are sexually mature at 55–60
and 38–45days of age, respectively (Zorenko etal. 1994;
Smorkatcheva 1999).
As in Arvicolinae rodents, pup age and body size influ-
ence the acoustic parameters of both AUDs and USVs (Ter-
leph 2011; Yurlova etal. 2020; Volodin etal. 2021; Warren
etal. 2022); correct interspecies comparison of the acoustics
is only possible between matched age classes and between
species matched in body size. So, careful control of animal
body size and age would be necessary for the comparative
study of vocalizations in vole species.
Adults of both vole species are active throughout 24-h
cycle (L. brandtii: Khruscelevsky and Kopylova 1957; L.
mandarinus: Smorkatcheva etal. 1990). Aboveground activ-
ity of adult L. brandtii is primarily diurnal (Khruscelevsky
and Kopylova 1957; Wan etal. 2006; Cui etal. 2020), but
adult L. mandarinus emerge to ground surface in dark time
(Dmitriev etal. 1980; Smorkatcheva etal. 1990). Newborns
of both species are raised at similar conditions of burrow (L.
brandtii: Khruscelevsky and Kopylova 1957; L. mandarinus:
Smorkatcheva etal. 1990).
Acoustic structure of adult AUDs differs between L.
brandtii and L. mandarinus. Audible sharp squeaks, occur-
ring in all types of interactions between animals from
friendly to aggressive, are twice higher in fundamental fre-
quency (f0) in L. brandtii (4.1–7.5kHz) than in L. man-
darinus (1.5–1.8kHz) (Rutovskaya 2011, 2012, 2018).
Male courtship songs are substantially higher-frequency
in L. brandtii (13.7kHz) than in L. mandarinus (1.2kHz)
(Rutovskaya 2018). Only L. brandtii produce audible high-
frequency (10.2–10.7kHz) alarm calls, which potentially
evolved in this species for defense against avian predators
(Rutovskaya 2012, 2018). Pup isolation USVs are only
described in 2-–14-day-old L. mandarinus (Yu etal. 2011),
whereas pup isolation AUDs or adult USVs have yet to be
studied in either L. brandtii or L. mandarinus. Whereas a
usual procedure of short-term isolation of pups from the
nest is sufficient for eliciting AUDs and USVs in Arvicoli-
nae pup voles (Yu etal. 2011; Yurlova etal. 2020), for adult
voles, more elaborated call-eliciting procedures are applied,
e.g., touch with a cotton bud, handling, and body measure-
ments (Yurlova etal. 2020; Klenova etal. 2021; Volodin
etal. 2021).
The aim of this study was to compare between captive
L. brandtii and L. mandarinus the acoustics of AUDs and
USVs, emitted by pups and adults of both species during

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
Page 3 of 17 106
short-term isolation and handling. For AUDs, we predicted
that pup and adult f0s might be higher-frequency in sur-
face-dwelling L. brandtii than in subterranean L. mandari-
nus. This prediction was based on published data reporting
acoustic differences of AUDs between adult L. brandtii
and L. mandarinus (Rutovskaya 2018) and on published
data reporting the low-frequency hearing in subterranean
rodents (Heffner and Heffner 1992; Gerhardt etal. 2017;
Okanoya etal. 2018). For USVs, we had not any special
prediction, in the lack of comparative data.
Methods
Study site, animals anddates
Calls (AUDs and USVs) of pup and adult L. brandtii
and L. mandarinus of 1–4 generation in captivity were
recorded from the beginning of March 2019 to mid-Octo-
ber 2020 in captive colonies of the Biological Institute of
Saint Petersburg University (Russia). To decrease observer
bias, blinded methods were mostly used when behavio-
ral data were recorded and/or analyzed: audio recording
trials conducted by one researcher (IAV) were primarily
analyzed by another researcher (MMD) and vice versa.
Colony founders were 7 L. brandtii obtained in 2017 from
the Chita region, Transbaikalia, Russia, and 20 L. man-
darinus (7 individuals obtained in 2017 from the Djida
region, Buryatia, Russia, and 13 individuals obtained in
2019 from the Selenga region, Buryatia, Russia).
Subject L. brandtii were 17 2–5-day-old unsexed pups
from 14 litters delivered by 13 parental pairs, 1–2 pups per
litter, and 19 adults (10 males, 9 females) aged from 72 to
391days old. Subject L. mandarinus were 15 2–5-day-old
unsexed pups from 14 litters delivered by 10 parental pairs,
1–2 pups per litter, and 16 adults (9 males, 7 females) aged
from 65 to 867days old. Day of pup birth was considered
zero day of pup life (Supplementary TableS1).
Housing
Animals were kept in pairs with one or a few subsequent
litters in glass terraria (25 × 50 × 30cm or 30 × 60 × 40cm
depending on group size) with wire-mesh roofs, with saw-
dust layer of 15–20cm, toilet paper as nest material, and
one or two wooden hides. The animals were fed each sec-
ond day with rabbit chow, oat (grain and sprouts), and
willow branches. Carrot, beet, and apples were provided
adlibitum as a source of both food and water.
Experimental procedure
Call-eliciting trials were conducted in a separate room
where only the focal animal was present. The experimental
procedure (following Zaytseva etal. 2019) was the same
for pups and adults and for both species. The focal ani-
mal was tested singly in only one trial; therefore, all calls
could be identified as belonging to the focal individual.
Trials were conducted in daytime at room temperature
20–25°C and natural lighting from the window. All elec-
tric equipment (lamps, fridges, computers) were turned off
for reducing the audible and ultrasonic background noise.
The elicited calls were related to moderate discomfort,
experienced by pups due to the cooling out of the nest,
and experienced by adults due to short-term social isola-
tion and handling. These calls were not distress-related
for pups and for adults, as pup cooling was short term and
moderate; whereas for the adults, the short-term isolation
from mates and human handling was reminiscent of rou-
tine procedure during regular cage cleanings occurring
every 5–7days, to which the animals were habituated.
A focal animal was transferred in a small clean plastic
container from a home cage to the experimental room on
the same floor within 60s and subjected to the 4-stage
480-s experimental procedure provoking vocalization. Test
trial included four stages: (1) isolation for 120s in a plas-
tic container 190 × 130 × 70mm (for pups) or in a plastic
cylinder without bottom with diameter 320mm, height
400mm (for adults); (2) touch with a cotton bud for 120s
approximately 2 times per second; (3) handling by fixing
in human hand and keeping with belly up for 120s; and
(4) body measurements for about 120s. The start of each
trial stage was indicated with voice mark of experimenter
(MMD or IAV), and the end of measurements was the end
of a trial. After the trial, the focal animal was weighted on
the electronic scales G&G TS-100 (G&G GmbH, Neuss,
Germany, accurate to 0.01g), in the same container that
served for the animal transfer.
The measurements included successive measuring of
body length (from tip of muzzle to anus) and head length
(from tip of muzzle to occiput), with electronic cali-
pers (Kraft Tool Co., Lenexa, Kansas, USA) accurate to
0.01mm. This cycle of measurements was repeated thrice
and the average values were calculated. Weighting and
measurement data were used for estimating the potential
differences in body size between study species.
Focal animal was returned to home cage immediately
after the end of a trial and weighting. Before the next test
trial, the experimental setup was washed with soapy water
and rubbed with cotton with alcohol, to avoid potential
odor effects on vocalization.

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
106 Page 4 of 17
Call recording
During each 4-stage test trial, a continuous recording of
AUDs and USVs of the focal individual was conducted. For
recording AUDs at sampling rate 48kHz and resolution 16
bit, we used a solid-state recorder Marantz PMD 660 (D&M
Professional, Kanagawa, Japan) with cardioid microphone
Sennheiser K6-ME64 (Sennheiser Electronic, Wedemark,
Germany).
For recording USVs at sampling rate 256kHz and reso-
lution 16 bit, we used a Pettersson D1000X recorder with
built-in microphone (Pettersson Electronik AB, Uppsala,
Sweden) and an Echo Meter Touch 2 PRO recorder (Wildlife
Acoustics, Inc., Maynard, MA USA), which also served for
tracking the real-time spectrogram of USVs on compatible
smartphone. As parallel recordings with Pettersson and Echo
Meter displayed similarly high quality and did not affect
the measured acoustic variables, we could use for acoustic
analyses calls recorded with either system.
The sonic and ultrasonic microphones were set at 25cm
above the focal animal, providing a good signal-to noise
ratio during recording. Acoustic recordings for each test trial
were stored as two wave files, one for AUD and one for USV
recording.
Call samples
For acoustic analysis, we selected calls (AUDs and USVs) of
best quality (not superimposed with strikes or other noises,
with a good signal-to noise ratio). To minimize potential
pseudoreplication, we avoided taking the calls following
each other, because successive calls can be more similar
to each other than calls separated with other calls. Calls for
analysis were evenly taken from different parts of the trial
stages. In addition, contour shape or presence of nonlin-
ear phenomena was not taken into account during selection
of calls for analysis. We also limited the number of calls
included in analysis per individual.
For analysis of acoustic variables of AUDs, we selected
AUDs from 10 individual L. brandtii and 10 individual L.
mandarinus pups (20 AUDs per pup) and selected AUDs
from 12 individual L. brandtii and 10 individual L. mandari-
nus adults (7–21 AUDs per adult). Pup AUDs were evenly
taken from different parts of the 1st (isolation) trial stage.
Adult AUDs were primarily taken from the 3rd (handling)
and 4th (body measurements) trial stages, because adult
voles did not emit AUDs at 1st (isolation) trial stage and
only 14 AUDs of L. brandtii could be taken from 2nd (touch)
stage. In total, we included in analysis 802 AUDs: 200 pup
AUDs per species and 201 adult AUDs per species (Sup-
plementary TableS2).
For analysis of acoustic variables of USVs, we selected
USVs from 11 individual L. brandtii and 11 individual L.
mandarinus pups (20 USVs per pup, but one individual only
provided 14 calls). Pup USVs were primarily taken from
the 1st (isolation) trial stage and, in addition, some calls (30
USVs of L. brandtii pups and 36 USVs of L. mandarinus
pups) were taken from the 2nd stage, because the number
of calls from the 1st stage was limited.
Adult USVs of each species were split in two non-over-
lapping categories, the low-frequency USVs (LF USVs) and
the high-frequency USVs (HF USVs) (see the “Results”
section). Thus, for investigating variation of the full set of
acoustic parameters between the two categories of adult
USVs and comparing them with pup USVs, we classified
all USVs to three categories: (1) pup USVs; (2) adult LF
USVs, and (3) adult HF USVs; and called the corresponding
nominal variable as “USV category.”
Adults rarely produced USVs, so we included in analysis
all USVs produced by 15 individual adult L. brandtii and
11 individual adult L. mandarinus, 6–23 USVs of each cat-
egory, the low-frequency (LF USVs) and the high-frequency
(HF USVs). Adult USVs were taken from all the four trial
stages. In total, we included in analysis 1072 USVs: 220 pup
USVs of L. brandtii, 214 pup USVs of L. mandarinus, 211
LF USVs and 139 HF USVs of adult L. brandtii, and 105
LF USVs and 183 HF USVs of adult L. mandarinus (Sup-
plementary TableS3).
Call analysis
Acoustic variables of AUDs and USVs were measured using
Avisoft SASLab Pro (Avisoft Bioacoustics, Berlin, Ger-
many); data of measurements were automatically exported
to Microsoft Excel (Microsoft Corp., Redmond, WA, USA).
Before measurements, we high-pass filtered all wav-files at
0.2kHz (for AUDs) or at 10kHz (for USVs), to remove
background noise.
Spectrograms for measurements were created at sampling
rate 48kHz (for AUDs) or 256kHz (for USVs), Hamming
window, fast Fourier transform (FFT) 1024 points, frame
50%, and overlap 93.75% for AUDs and 87.5% for USVs.
For each AUD or USV, we manually measured, in the spec-
trogram window of Avisoft, the duration with the standard
marker cursor, and the maximum fundamental frequency
(f0max), the minimum fundamental frequency (f0min), the
fundamental frequency at the beginning of a call (f0beg),
and the fundamental frequency at the end of a call (f0end)
with the reticule cursor. For each AUD or USV, we measured
the peak frequency (fpeak) in the power spectrum window
of Avisoft (Fig.1).
AUD andUSV contours
By visual inspection of call in the spectrogram window
of Avisoft, we classified AUDs and USVs to one of five

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
Page 5 of 17 106
contours (Fig.2): flat, chevron, upward, downward, and
complex (following Yurlova etal. 2020; Kozhevnikova etal.
2021). Flat contour was determined if the difference between
f0min and f0max was less than 0.6kHz (in AUDs) or less
than 6kHz (in USVs). In cases where the difference between
f0min and f0max was equal or larger than 0.6kHz or 6kHz,
respectively, a call contour could be classified as chevron (up
and down), upward (ascending from start to end), downward
(descending from start to end), or complex (up and down a
few times or U-shaped) (Fig.2).
Nonlinear phenomena andnote composition
inAUDs andUSVs
Each AUD and USV was checked for presence of nonlinear
phenomena (Fig.3): biphonations, subharmonics, deter-
ministic chaos, and frequency jumps (Wilden etal. 1998;
Yurlova etal. 2020; Kozhevnikova etal. 2021). Biphonation
was noted when two independent fundamental frequencies,
the low (f0) and the high (g0), as well as their combina-
tory frequency bands (g0 minus f0; g0 minus 2f0; etc.) were
present in call spectrum. Subharmonics were noted when
frequency bands of 1/2 or 1/3 of f0 were present in call spec-
trum (Fig.3). Deterministic chaos was noted when a chaotic
segment (sometimes with residual fundamental frequency)
was present in call spectrum (Fig.3). We only noted the
presence of deterministic chaos and/or subharmonics, if the
duration of call fragments containing these nonlinear phe-
nomena comprised at least 10% of the entire call duration
(Yurlova etal. 2020; Kozhevnikova etal. 2021).
We noted a presence of frequency jumps, when the f0
increased jump-like up or down for ≥ 1kHz (for AUDs) or
for ≥ 10kHz (for USVs) (Fig.3). As frequency jumps break
the f0 contour to separate notes, we considered the calls
without frequency jumps as one-note calls, the calls with
one frequency jump as two-note calls, and calls with two
or more frequency jumps as multi-note calls (Fig.3). For
determining the type of f0 contour in the calls containing
frequency jumps, we virtually joined the parts of the broken
contour of f0, following (Yurlova etal. 2020; Kozhevnikova
etal. 2021).
In addition, we classified USVs accordingly to the three
possible note compositions (1-note, 2-note, multi-note)
based on presence of up or/and down frequency jumps over
10kHz (Fig.3). The 1-note USVs lacked frequency jumps;
the 2-note USVs had one frequency jump (up or down); and
the multi-note USVs had two or more frequency jumps (see,
e.g., Zaytseva etal. 2019).
Statistical analyses
Statistical analyses were made with STATISTICA, v. 8.0
(StatSoft, Tulsa, OK, USA) and R 4.1.0 (R Development
Core Team 2021). Means were presented as mean ± SD, and
all tests were two-tailed and differences were considered
significant whenever p < 0.05. We used one-way ANOVA for
estimating the effect of factors species (separately for pups
and for adults) and sex (only for adults, separately for each
species) on the morphometric (body size-related) variables.
To analyze the acoustics of AUDs and USVs, we per-
formed linear mixed effect models (LMM) using package
nlme (Pinheiro etal. 2021) implemented in R. For AUDs,
age (pup vs. adult), species, and their interaction were fitted
as fixed terms. For USVs, LMMs included the USV category
(pup USVs, adult LF USVs and adult HF USVs), species,
and their interaction as fixed predictors. Individual identity
was fitted as a random term in all models. Post hoc compari-
sons were performed with Tukey HSD test using emmeans
package in R (Lenth 2021).
Fig. 1 Measured acoustic variables in: a ultrasonic (USV) call of pup
L. mandarinus; b audible (AUD) call of adult L. brandtii. Spectro-
gram (right) and mean power spectrum of the entire call (left). Desig-
nations: duration – call duration; f0beg – the fundamental frequency
at the onset of a call; f0max – the maximum fundamental frequency;
f0end – the fundamental frequency at the end of a call; f0min – the
minimum fundamental frequency; fpeak – the frequency of maxi-
mum amplitude. Spectrogram was created using sampling frequency
256kHz (for USVs) or 48kHz (for AUDs), Hamming window, fast
Fourier transform (FFT) 1024 points, frame 50%, overlap 87.5% (for
USVs) or 93.75% (for AUDs)

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
106 Page 6 of 17
We used Fisher’s exact test to compare percentages.
We used discriminant function analysis (DFA) standard
procedure to estimate the differences in values of acous-
tic variables between two categories of adult USVs (LF
USVs and HF USVs). Variables with most contribution to
discrimination were determined based on Wilk’s lambda
values.
Results
Age andbody size variables
ANOVA did not reveal any interspecies differences regard-
ing the age, body weight, body length, and head length,
either for pups or for adults (Table1). The comparison of
body weight, body length, and head length between adult
male and female L. brandtii did not reveal any significant
differences (F1,17 = 2.02; p = 0.17; F1,17 = 0.11; p = 0.75;
F1,17 = 1.29; p = 0.27, respectively). In adult L. mandari-
nus, body weight (F1,14 = 0.42; p = 0.53), and body length
(F1,14 = 1.84; p = 0.20) did not differ between sexes as well,
whereas head length was larger in males (F1,14 = 5.50;
p = 0.03; 32.05 ± 1.38 mm and 30.50 ± 1.22 mm,
respectively).
AUD contours andnonlinear phenomena
Pup and adult AUDs of L. brandtii and L. mandarinus dis-
played all the five possible contour shapes (Fig.4). Chev-
ron contour prevailed, being equally frequent in pups of
both species (p = 0.91, Fisher’s exact test). Upward contour
was more frequent in pup L. mandarinus than in pup L.
brandtii (p < 0.001). Complex contour was more frequent
in pup L. brandtii than in pup L. mandarinus (p < 0.001).
The remaining two (downward and flat) contours were
both rare in pups of either species, without significant dif-
ferences in the occurrence between them (Fig.4).
As in pups, chevron contour prevailed in adults of either
species; however, it was more frequent in adult L. brandtii
than in adult L. mandarinus (p < 0.001, Fisher’s exact test)
(Fig.4). In adult L. brandtii, flat contour practically lacked
and upward contour was rare, whereas in adult L. man-
darinus, flat and upward contours were similarly frequent
(p < 0.001 in both cases). Complex contour was more
frequent in adult L. brandtii than in adult L. mandarinus
(p < 0.001). Downward contour was infrequent compared
to other contours in adults of either species (p = 0.06)
(Fig.4).
Overall, chevron contour prevailed in pup and adult
AUDs of either species, whereas all other contours were
substantially less frequent. Complex contour was a few
Fig. 2 Five contour shapes in audible (AUDs) and ultrasonic (USVs)
vole calls. AUDs: a – flat in AUD of adult female L. mandarinus; b –
chevron in AUD of 5-day-old pup L. mandarinus; c – upward in AUD
of 4-day-old pup L. brandtii; d – downward in AUD of adult male
L. mandarinus; e – complex in AUD of 5-day-old pup L. brandtii.
USVs: a – flat in USV of 5-day-old pup L. brandtii; b – chevron
in USV of 4-day-old pup L. mandarinus; c – upward in USV of
5-day-old pup L. brandtii; d – downward in USV of 2-day-old pup
L. brandtii; e – complex in USV of 4-day-old pup L. brandtii. Spec-
trogram was created using sampling frequency 256kHz (for USVs)
or 44.1 kHz (for AUDs), Hamming window, fast Fourier transform
(FFT) 1024 points, frame 50%, overlap 87.5% (for USVs) or 96.87%
(for AUDs)

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
Page 7 of 17 106
times more frequent in L. brandtii, whereas the upward
and flat contours were a few times more frequent in L.
mandarinus.
In both pup and adult L. brandtii and L. mandarinus,
AUDs could contain three kinds of nonlinear phenomena:
frequency jumps, subharmonics and deterministic chaos
(Fig.5). Nonlinear phenomena were found in 68.0% of
AUDs in pup L. brandtii, but in only 35.5% of AUDs in
pup L. mandarinus (p < 0.001, Fisher’s exact test). In pup
L. brandtii, most frequent nonlinear phenomena were sub-
harmonics and deterministic chaos. Frequency jump was
the least frequent in pup L. brandtii and lacked entirely in
pup L. mandarinus. In pup L. brandtii, percentage of AUDs
with deterministic chaos was comparable with those in pup
L. mandarinus (p = 0.59), whereas subharmonics were less
frequent in pup AUDs of L. mandarinus (p < 0.001) (Fig.5).
In adult AUDs of both species, percentage of calls with
nonlinear phenomena was small: 7.0% in adult L. brandtii
and 17.9% in adult L. mandarinus (differences are signifi-
cant, p = 0.001) (Fig.5). In adult AUDs of L. brandtii, sub-
harmonics practically lacked, whereas in adult AUDs of L.
mandarinus, subharmonics were most widespread among
nonlinear phenomena (differences are significant, p < 0.001).
Deterministic chaos was present at the same level in adults
of either species (p = 0.39). Frequency jumps occurred rarely
and only in adults (Fig.5).
Fig. 3 Nonlinear phenomena in audible (AUDs) and ultrasonic
(USVs) vole calls. AUDs: a – subharmonics in AUD of 5-day-old
pup L. mandarinus; b – deterministic chaos in AUD of 5-day-old pup
L. mandarinus; c – frequency jump up in AUD of 2-day-old pup L.
brandtii. USVs: a – biphonation in USV of 4-day-old pup L. man-
darinus; b – frequency jump down-up in USV of 2-day-old pup L.
mandarinus; c – subharmonics in USV of an adult male L. brandtii.
Spectrogram was created using sampling frequency 256 kHz (for
USVs) or 44.1 kHz (for AUDs), Hamming window, fast Fourier
transform (FFT) 1024 points, frame 50%, overlap 87.5% (for USVs)
or 96.87% (for AUDs)
Table 1 Values (mean ± SD) for age and body size variables of pup
and adult L. brandtii and L. mandarinus, and the results of interspe-
cies comparison. n – number of individuals
Variable L. brandtii L. mandarinus ANOVA
Pups n = 17 n = 15
Age (days) 3.11 ± 1.18 3.53 ± 1.13 F1,30 = 0.99;
p = 0.33
Body weight (g) 3.67 ± 0.83 4.13 ± 1.12 F1,30 = 1.75;
p = 0.20
Body length (mm) 33.80 ± 3.05 34.12 ± 4.60 F1,30 = 0.05;
p = 0.82
Head length (mm) 16.53 ± 1.18 17.39 ± 1.33 F1,30 = 3.71;
p = 0.06
Adults n = 19 n = 16
Age (days) 201.0 ± 87.2 302.9 ± 229.9 F1,31 = 3.04;
p = 0.09
Body weight (g) 35.80 ± 7.51 35.92 ± 5.02 F1,33 = 1.71;
p = 0.20
Body length (mm) 93.9 ± 7.56 94.10 ± 5.56 F1,33 = 0.01;
p = 0.93
Head length (mm) 31.60 ± 1.61 31.38 ± 1.50 F1,33 = 0.18;
p = 0.67
Fig. 4 Percentages of five different contour shapes in pup and adult
audible calls (AUDs) of L. brandtii and L. mandarinus. Contour
names are provided on the Fig. n – number of calls

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
106 Page 8 of 17
AUD acoustics
For AUDs, LMM showed that the effect of factor species
was significant for the duration and all f0 variables, but not
for fpeak (Table2). Factor age affected significantly the
duration, f0end, f0min, and fpeak. Interaction of factors
species and age was significant only for duration and f0end
(Table2), reflecting the differences in age-related dynamics
between species for these acoustic variables. Specifically,
the duration markedly decreased with age in L. brandtii and
increased in L. mandarinus.
Between species, pup AUDs did not differ by duration
and fpeak (Table2). All f0 variables of AUDs were higher in
pup L. brandtii than in pup L. mandarinus. In adults, AUDs
were shorter in L. brandtii than in L. mandarinus. As in
pups, in adults, the values of all f0 variables of AUDs were
higher in L. brandtii than in L. mandarinus, whereas fpeak
did not differ between species (Table2).
Comparison between ages within species showed, that in
L. brandtii, adult AUDs were shorter and lower in f0end and
fpeak than pup AUDs, whereas f0beg, f0max, and f0min did
not differ between pups and adults (Table2). In L. mandari-
nus, adult AUDs did not differ from pup AUDs in duration
and fpeak, whereas the values of all f0 variables were lower
in adults than in pups (Table2).
Two categories (LF andHF) ofadult USVs
While pups only produced one type of USVs, distribu-
tions of fpeak and f0max values of adult USVs were two-
humped, thus indicating a presence of two non-overlapping
call categories, the low-frequency USVs (LF USVs) and
the high-frequency USVs (HF USVs) (Fig.6). So, based
on values of fpeak and f0max, each USV was assigned to
one of these call categories (Fig.6). For L. brandtii, LF
Fig. 5 Percentages of three kinds of nonlinear phenomena in pup and
adult AUDs of L. brandtii and L. mandarinus. Nonlinear phenomena
names are provided on the Fig. n – number of calls
Table 2 Values (mean ± SD) of acoustic variables of audible calls
(AUDs) of pup and adult L. brandtii and L. mandarinus, and LMM
results for the effects of species and age on the acoustics. The B ± SE
correspond to parameter estimates and standard errors in LMM. Indi-
vidual identity was introduced as a random term in all LMMs. dura-
tion – call duration, f0beg – the fundamental frequency at the begin-
ning of a call, f0end – the fundamental frequency at the end of a call,
f0max – the maximum fundamental frequency, f0min – the minimum
fundamental frequency, fpeak – the peak frequency, n – number of
calls. The same superscripts indicate the values, which are non-sig-
nificantly different from other values by the given acoustic parameter
(post hoc Tukey HSD test, p < 0.05)
Variable Pups Adults LMM
L. brandtii,
n = 200 L. mandarinus,
n = 200 L. brandtii,
n = 201 L. mandarinus,
n = 201
Species Age Species × Age
interaction
Duration (s) 0.103 ± 0.027a0.101 ± 0.026a0.071 ± 0.040b0.128 ± 0.047aB = 1.4 ± 0.3,
p < 0.001 B = 1.0 ± 0.3,
p < 0.001 B = − 1.5 ± 0.4,
p < 0.001
f0beg (kHz) 2.63 ± 1.42a1.34 ± 0.24b2.14 ± 1.20a0.96 ± 0.20cB = − 1.5 ± 0.3,
p < 0.001 B = 0.4 ± 0.3,
p = 0.16 B = 0.3 ± 0.4,
p = 0.34
f0max (kHz) 7.30 ± 1.35a2.43 ± 0.29b6.49 ± 2.69a1.67 ± 0.34cB = − 2.0 ± 0.1,
p < 0.001 B = 0.3 ± 0.1,
p = 0.07 B = 0.3 ± 0.2,
p = 0.13
f0end (kHz) 4.88 ± 1.42a1.62 ± 0.43b2.46 ± 1.10c1.10 ± 0.28dB = − 1.2 ± 0.2,
p < 0.001 B = 1.1 ± 0.2,
p < 0.001 B = − 0.5 ± 0.2,
p = 0.04
f0min (kHz) 2.57 ± 1.37a1.28 ± 0.23b1.88 ± 0.88a0.92 ± 0.19cB = − 1.4 ± 0.3,
p < 0.001 B = 0.5 ± 0.3,
p = 0.04 B = 0.2 ± 0.4,
p = 0.65
fpeak (kHz) 11.62 ± 3.44a8.90 ± 4.53a,b 8.83 ± 2.20b7.48 ± 1.51bB = − 0.3 ± 0.3,
p = 0.24 B = 0.9 ± 0.3,
p = 0.005 B = − 0.4 ± 0.4,
p = 0.29

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
Page 9 of 17 106
USVs had f0max values ≤ 48kHz, whereas HF USVs had
f0max values ≥ 50kHz. For L. mandarinus, LF USVs had
f0max values ≤ 53kHz, whereas HF USVs had f0max val-
ues ≥ 53kHz, although one LF USV had a value of 56kHz
and one HF USV had a value of 50kHz (Fig.6). For L.
brandtii, LF USVs had fpeak values ≤ 44kHz, whereas HF
USVs had fpeak values ≥ 44kHz. For L. mandarinus, LF
USVs had fpeak values ≤ 46kHz, whereas HF USVs had
fpeak values ≥ 47kHz (Fig.6).
In adults of either species, LF USVs were more frequent
at stages 1 (isolation) and 2 (touch) of test trials: 210 of
211 (99.5%) LF USVs in L. brandtii and 67 of 105 (64%)
LF USVs in L. mandarinus. In adults of either species,
HF USVs occurred nearly exclusively at stages 3 (han-
dling) and 4 (body measurements) of test trials: 139 of 139
(100%) HF USVs in L. brandtii and 172 of 183 (94%) HF
USVs in L. mandarinus.
DFA based on six acoustic variables (duration, f0beg,
f0max, f0end, f0min, fpeak) confirmed subdivision of
adult USVs to the two categories, LF USVs and HF USVs
(Table3). Parameters most contributing to discrimination,
in the order of decreasing importance, were the duration,
f0max and fpeak in L. brandtii and fpeak, f0max and f0beg
in L. mandarinus.
USV contours, note compositions, andnonlinear
phenomena
Pups and adults of both species produced USVs with all the
five contour shapes (Fig.7). In pup L. brandtii, chevron con-
tour was more frequent than in pup L. mandarinus, whereas
pup L. mandarinus produced more frequently USVs with
upward contour (p < 0.001 in both cases, Fisher’s exact test).
Fig. 6 Distribution of ultrasonic
calls (USVs) of adult L. brandtii
and L. mandarinus according
to the maximum fundamental
frequency (f0max) and peak fre-
quency (fpeak), n = 350 USVs
for L. brandtii and n = 288
USVs for L. mandarinus. The
averaged over 11 points smooth-
ing lines are shown
Table 3 Percent of correct classifying of ultrasonic calls of adult
L. brandtii and L. mandarinus to the two USV categories, low-fre-
quency USVs (LF USVs) and high-frequency USVs (HF USVs),
based on discriminant function analysis (DFA) standard procedure
USV category Classifying to a predicted
category
Total Correct
classifying
(%)
LF USV HF USV
L. brandtii
LF USV 211 0 211 100.0
HF USV 5 134 139 96.4
Total 216 134 350 98.6
L. mandarinus
LF USV 105 0 105 100.0
HF USV 2 181 183 98.9
Total 107 181 288 99.3

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
106 Page 10 of 17
Flat, downward, and complex contours occurred at the same
levels in pups of either species (Fig.7).
In adults, we did not find any significant interspecies dif-
ference in the occurrence of different contours of LF USVs
(Fig.7). In either species, flat contour was most widespread
(present in 74% LF USVs of L. brandtii and in 80% LF
USVs of L. mandarinus); the upward and downward con-
tours were less frequent. In either species, chevron was the
rarest contour, occurring in only 0.95% LF USVs, and com-
plex contour was not found at all (Fig.7).
In HF USVs of adults of both species, the most wide-
spread contour was upward, which was more frequent in L.
brandtii than in L. mandarinus (p = 0.024) (Fig.7). Flat con-
tour, second in order by the occurrence, was more frequent
in adult L. mandarinus than in adult L. brandtii (p = 0.024);
remaining three contours were equally present in both spe-
cies (Fig.7).
Nonlinear phenomena detected in pup and adult USVs
of both species were frequency jumps, subharmonics, and
biphonations (Fig.8). Deterministic chaos, which was usual
nonlinear phenomenon in AUDs, lacked in USVs. In pups,
nonlinear phenomena were present in about half of USVs (in
43.2% USVs of pup L. brandtii and in 53.8% USVs of pup L.
mandarinus, differences are marginally significant, Fisher’s
exact test, p = 0.055) (Fig.8). In pup USVs, the most wide-
spread nonlinear phenomenon was frequency jump (42.3%
USVs of L. brandtii and 50.9% USVs of L. mandarinus) and
biphonations were rare (in 5.0% and 7.0% USVs, respec-
tively); subharmonics were the rarest nonlinear phenomena
(0.5% USVs, only in pup L. brandtii, all differences between
species were non-significant (Fig.8).
In LF USVs and HF USVs of adults of both species,
nonlinear phenomena (from 13 to 20% USVs) occurred
rarer than in pup USVs (Fig.8). Frequency jumps were
most frequent, subharmonics occurred rarer, and biphona-
tions lacked (Fig.8). All differences between species were
non-significant.
Overall, pup USVs contained more nonlinear phenomena
than adult USVs. Most widespread nonlinear phenomenon
in both pups and adults and in both species was frequency
jump. Pup USVs often contained biphonations, lacking in
adults, however, adult USVs often contained subharmonics,
practically lacking in pups.
In pups of both species, 1-note USVs prevailed (interspe-
cies differences are non-significant, p = 0.08, Fisher’s exact
test) (Fig.9). Two-note pup USVs were rarer than 1-note
USVs, but more often in L. mandarinus than in L. brandtii
(p = 0.016, Fisher’s exact test). Multi-note pup USVs were
rare in either species (interspecies differences are non-sig-
nificant, p = 0.51, Fisher’s exact test) (Fig.9).
In adults of both species, most LF USVs and HF USVs
were 1-note calls; 2-note calls occurred rarer and multi-note
USVs were scarce (Fig.9). All interspecies differences were
non-significant.
USV acoustics
We observed a significant variation between species on
duration but not on fpeak or f0 variables for USVs (LMM,
Table4). The USV category significantly affected all meas-
ured acoustic variables of USVs. Significant interaction of
Fig. 7 Percentages of five
different USV contours in pup
and adult L. brandtii and L.
mandarinus. Contour names
are provided on the figure. LF
USV – low-frequency USV, HF
USV – high-frequency USV, n –
number of calls

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
Page 11 of 17 106
species and USV category was observed only in a model
with the duration of USVs as a response (Table4).
Between species, pup USVs only differed by duration:
L. brandtii pup USVs were longer than L. mandarinus
pup USVs (Tukey post hoc test, p < 0.05; Table4). Values
of fpeak and all f0 variables of pup USVs did not differ
between species. In LF USVs of adults, the duration of
AUDs was significantly longer in L. brandtii than in L.
mandarinus (Tukey post hoc test, p < 0.05). As in pups of
these species, all other variables of adult LF USVs did not
differ between species (Table4). In HF USVs of adults, the
duration, fpeak, and all f0 variables did not differ between
species (Table4).
In L. brandtii, pup USVs were longer than adult LF
USVs, whereas all f0 variables and fpeak did not differ
from adult LF USVs and were significantly and substantially
(twice) lower compared to adult HF USVs (Tukey post hoc
test, p < 0.05; Table4).
Fig. 8 Percentages of differ-
ent nonlinear phenomena in
different USV categories: pup
USVs, adult LF USVs and
adult HF USVs for L. brandtii
and L. mandarinus. Nonlinear
phenomena names are provided
on the figure LF USV – low-fre-
quency USV, HF USV – high-
frequency USV, n – number of
calls
Fig. 9 Percentages of three pos-
sible note compositions (1-note,
2-note, multi-note) in pup and
adult USVs of L. brandtii and L.
mandarinus. Note compositions
are provided on the figure. LF
USV – low-frequency USV, HF
USV – high-frequency USV, n –
number of calls

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
106 Page 12 of 17
In L. mandarinus, pup USVs were longer than adult LF
USVs, and their f0beg and f0max were higher than in adult
LF USVs (Tukey post hoc test, p < 0.05). Values of f0end
and f0min and of fpeak did not differ between USVs of pups
and LF USVs of adults. Compared to HF USVs of adults,
pup L. mandarinus USVs were longer and had twice lower
values of all f0 variables and of fpeak (Table4).
Discussion
The most remarkable novelty of the results of this study
was that, in contrast to AUDs, USVs were surprisingly simi-
lar between species in frequency parameters: the values of
f0 and fpeak did not differ between species in either pups
or adults. So, our data suggest that ultrasonic vocalization
is unaffected with way of living, subterranean or surface-
dwelling. These results indicate that selection pressure for
frequency parameters of USVs is evidently lacking at any
lifestyle. At the same time, the audible vocalization was
strongly affected with lifestyle in both adults and in pups
(in spite of pup living in the same acoustic environment in
both species).
This study confirmed the results by Rutovskaya (2018)
that AUDs are much higher in f0 in adult surface-dwelling
L. brandtii (mean f0max 6.5kHz) than in adult subterra-
nean L. mandarinus (mean f0max 1.7kHz). In addition, our
study revealed that this trend of interspecies differences can
be expanded on AUDs of pups (mean f0max 7.3kHz in L.
brandtii vs 2.4kHz in L. mandarinus). We can therefore
conclude that, at least in the two Lasiopodomys vole species,
frequency parameters of USVs remain resistant to the shifts
from surface-dwelling to subterranean lifestyle or vice versa.
We can also conclude that lifestyle of these species is only
reflected in frequency parameters of AUDs, but not USVs.
Consistently, selection on ultrasonic call rate in neonatal
laboratory rats Rattus norvegicus affects low-frequency,
but not ultrasonic, vocalizations in adult rats (Lesch etal.
2020). Thus, our study provides additional evidence that
rodent AUDs may be more reactive to selection pressures
for behavior than rodent USVs.
Our results on the lower-frequency AUDs in the subter-
ranean species, L. mandarinus, compared to the surface-
dwelling L. brandtii, and on similar-frequency USVs in both
species, can be explained by acoustic adaptation hypothesis
(Ey and Fisher 2009). Better sound transmission in burrows
is only applicable to low-frequency AUDs, as these calls
Table 4 Values (mean ± SD) of USV acoustic variables of pup and
adult L. brandtii and L. mandarinus, and LMM results for the effects
of species and USV category (pup USV/Adult LF USV/Adult HF
USV) on the acoustics. The B ± SE correspond to parameter estimates
and standard errors in LMM. Individual identity was introduced as a
random term in all LMMs. duration – call duration, f0beg– the fun-
damental frequency at the beginning of a call, f0end – the fundamen-
tal frequency at the end of a call, f0max – the maximum fundamental
frequency, f0min – the minimum fundamental frequency, fpeak – the
peak frequency, n – number of calls. The same superscripts indicate
the values, which are non-significantly different from other values by
the given acoustic parameter (post hoc Tukey HSD test, p < 0.05)
Variable Pup USV Adult LF USV Adult HF USV LMM
L. brandtii,
n = 220 L. mandari-
nus, n = 214 L. brandtii,
n = 211 L. mandari-
nus, n = 105 L. brandtii,
n = 139 L. mandari-
nus, n = 183
Species USV category Species × USV
category interac-
tion
Duration (s) 0.082 ± 0.039a0.055 ± 0.037b0.043 ± 0.020b0.019 ± 0.019c0.015 ± 0.012c0.017 ± 0.016cB = 0.4 ± 0.2,
p = 0.05 B = 1.5 ± 0.1,
p < 0.001
B = 2.2 ± 0.2,
p < 0.001
B = − 1.4 ± 0.1,
p < 0.001
B = 1.0 ± 0.3,
p = 0.001
f0beg (kHz) 31.66 ± 7.39a,b 36.37 ± 7.79b30.43 ± 7.33a30.11 ± 7.62a71.89 ± 13.75c68.33 ± 12.34cB = − 0.1 ± 0.2,
p = 0.7 B = − 1.9 ± 0.1,
p < 0.001
B = − 1.9 ± 0.1,
p < 0.001
B = − 0.1 ± 0.1,
p = 0.34
B = 0.4 ± 0.2,
p = 0.06
f0max (kHz) 39.66 ± 9.80a,b 44.77 ± 9.11b33.26 ± 7.48a33.40 ± 7.90a81.25 ± 13.83c78.50 ± 15.19cB = − 0.1 ± 0.1,
p = 0.6 B = − 2.0 ± 0.1,
p < 0.001
B = − 1.8 ± 0.1,
p < 0.001
B = − 0.1 ± 0.1,
p = 0.33
B = 0.3 ± 0.2,
p = 0.10
f0end (kHz) 29.63 ± 7.09a33.41 ± 7.98a30.14 ± 7.27a29.86 ± 7.66a76.03 ± 13.04b73.44 ± 15.60bB = − 0.02 ± 0.1,
p = 0.9 B = − 1.9 ± 0.1,
p < 0.001
B = − 1.9 ± 0.1,
p < 0.001
B = − 0.04 ± 0.1,
p = 0.69
B = 0.3 ± 0.2,
p = 0.14
f0min (kHz) 25.24 ± 5.25a28.80 ± 7.00a27.35 ± 6.43a27.74 ± 7.21a67.23 ± 11.71b64.42 ± 11.51bB = − 0.05 ± 0.1,
p = 0.7 B = − 1.9 ± 0.1,
p < 0.001
B = − 2.1 ± 0.1,
p < 0.001
B = − 0.1 ± 0.1,
p = 0.39
B = 0.3 ± 0.2,
p = 0.09
fpeak (kHz) 30.33 ± 5.30a34.11 ± 8.22a29.58 ± 6.73a30.56 ± 6.80a72.99 ± 12.25b71.46 ± 12.59bB = − 0.01 ± 0.1,
p = 1.0 B = − 2.0 ± 0.1,
p < 0.001
B = − 1.9 ± 0.1,
p < 0.001
B = − 0.01 ± 0.1,
p = 0.98
B = 0.2 ± 0.2,
p = 0.15

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
Page 13 of 17 106
propagate to a distance of a few meters (Heth etal. 1986)
being even increased by burrow stethoscope effect (Lange
etal. 2007; Schleich and Antenucci 2009; Okanoya and
Screven 2018. In contrast, USVs rapidly attenuate, greater
scatter, have poorer localizability compared to the low-fre-
quency AUDs (Musolf and Penn 2012) and propagate to a
shorter distance irrespectively on surface or in burrow, thus
being indifferent to selection pressure for propagation abil-
ity. Compared to the low-frequency calls, the high-frequency
calls can only be heard at close range and can therefore be
only used for communication in the immediate vicinity with
conspecifics, as suggested for other rodents (Wilson and
Hare 2006) and for canids (Sibiryakova etal. 2021).
A question remains, whether the transit to subterranean
lifestyle provoked the shift towards lower-frequency AUDs
in L. mandarinus or, alternately, the transit to surface-
dwelling lifestyle provoked the shift towards higher-fre-
quency AUDs in L. brandtii. For other vole species, related
studies indicate that f0max of their audible sharp squeaks
commonly does not exceed 2–3kHz (Rutovskaya 2018,
2019a, b, c). Exclusions are the Harting’s vole (Microtus
hartingi) (10.2–17.6kHz: Pandourski 2011; Rutovskaya
2019a) and the narrow-headed vole (Lasiopodomys grega-
lis) (3.6–5.6kHz: Rutovskaya 2018). In surface-dwelling L.
brandtii and L. gregalis, AUDs are high-frequency, whereas
in subterranean L. mandarinus they are low-frequency. At
the same time, f0max of AUDs (sharp squeaks) of L. man-
darinus (up to 2kHz) overlaps with those of other vole spe-
cies, whereas AUDs of L. brandtii are distinctively high-
frequency (f0 up to 6kHz) compared to AUDs of other vole
species (Rutovskaya 2018). Thus, Lasiopodomys species dis-
play a relatively large range of f0max values. We can there-
fore advance a hypothesis that shift of AUDs towards higher
frequencies in L. brandtii outcomes from surface-dwelling
lifestyle and the respective emergence of the high-frequency
(10.2–10.7kHz, Rutovskaya 2012) audible alarm call in this
species. This hypothesis is alternative to the hypothesis sug-
gesting that shift of AUDs towards lower frequencies in L.
mandarinus outcomes from the subterranean lifestyle of this
species.
This study provides new results that f0 of AUDs dis-
plays an ontogenetic shift towards lower frequencies with
maturation in both species. The ontogenetic lowering of f0
in AUDs is typical for mammals, because of age-related
increase of the vocal fold length and mass, resulting in the
lower vibration rate at phonation (Fitch and Hauser 2002).
The ontogenetic lowering of f0 in AUDs was reported, e.g.,
in giant panda (Ailuropoda melanoleuca) (Charlton etal.
2009), steppe marmot (Marmota bobak), and great gerbil
(Rhombomys opimus) (Nikolskii 2007). Exclusions from this
common rule (i.e., mammals with non-lowering f0 with age)
are found, e.g., in a few species of ground squirrels (Matros-
ova etal. 2007, 2011; Swan and Hare 2008; Schneiderová
etal. 2015) and in two shrew species (Schneiderová 2014;
Volodin etal. 2015).
We found, for the first time for Arvicolinae species, that,
with maturation, f0 values of USVs split to two categories:
the low-frequency USVs (LF USVs of 27–33kHz), indis-
tinguishable in f0 from pup USVs, and the high-frequency
USVs (HF USVs of 65–81kHz) lacking in pups. In pups, we
only analyzed USVs emitted at the 1st (isolation) and 2nd
(touch) trial stages, but in adults of both L. brandtii and L.
mandarinus, LF USVs were more frequent at 1st and 2nd
(isolation and touch) trial stages, whereas HF USVs were
more frequent at 3rd and 4th (handling and body measure-
ments) trial stages. In adult rodents, the isolation and touch
procedures are potentially related to a weaker negative emo-
tional arousal than handling and body measurements, dur-
ing which the animals try to escape or bite a human hand
(Klenova etal. 2021). We can propose that, in Lasiopodomys
voles, emission of LF USVs and HF USVs may be related
to different levels of caller’ negative emotional arousal.
This is reminiscent of the situation in laboratory rat, the
species in which initially broad range of pup USV frequen-
cies (30–65kHz) split in ontogeny to two call categories, of
22-kHz and 50-kHz USVs, related in adults to negative and
positive emotional arousal, respectively (Brudzynski etal.
1999; Brudzynski 2005; Riede 2011; Riede etal. 2015).
Thus, the ontogenetic split of pup USVs to the two different
categories with maturation is not to be unique for labora-
tory rat.
Interspecies differences in USVs were only found in dura-
tion. The USVs were longer in pup L. brandtii than in pup L.
mandarinus, and LF USVs of adult L. brandtii were longer
than LF USVs of adult L. mandarinus. Age-related changes
of USV duration from pups to adults may represent a com-
mon trend for rodents. Aside L. brandtii and L. mandarinus
(this study), the shortening of USVs from pups to adults
was reported for yellow steppe lemming (Yurlova etal.
2020) and for five Gerbillinae species (Zaytseva etal. 2019;
Kozhevnikova 2021). At the same time, the age-related
changes of duration from pups to adults in AUDs seem to
be not a common trend in mammals. For example, in speck-
led ground squirrels Spermophilus suslicus, the alarm call
duration increases with maturation (Volodina etal. 2010);
whereas in some other species of ground squirrels, it remains
unchanged (Swan and Hare 2008; Volodina etal. 2010).
Duration of audible squeaks decreases with maturation in
fat-tailed gerbil Pachyuromys duprasi (Zaytseva etal. 2020)
and in yellow steppe lemming Eolagurus luteus (Volodin
etal. 2021).
In our study, body size and age of pups and adults did
not differ between species, so these factors could not be
responsible for the detected acoustic differences. This is an
additional argument that the two closely related species, L.
brandtii and L. mandarinus, represent a most convenient

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
106 Page 14 of 17
model for comparative studies of the effects of surface-
dwelling and subterranean lifestyle on the acoustics (this
study) as well as on other biological aspects, as physiol-
ogy, genetics and behavior (e.g., Dong etal. 2018; Sun
etal. 2020).
Predation could be one of the key factors affecting the
transit to underground or surface-dwelling lifestyle in L.
brandtii and L. mandarinus. Potentially, predation of rap-
tors (e.g., Zhong etal. 2022) could promote the emergence
of the high-frequency audible alarm call (10.2–10.7kHz,
Rutovskaya 2012, 2018) in L. brandtii. The alarm call of
L. brandtii might not be well audible for raptors, who hear
well the audible frequencies below 6–8kHz (Yamazaki
etal. 2004; McGee etal. 2019).
Distinctive to raptors, which primarily rely on their
vision for capturing small mammalian prey (Potier etal.
2020), the nocturnal avian predators (owls) may primarily
rely on their hearing (de Koning etal. 2020). Although the
potential effect of owl predation on transit of L. brandtii to
diurnal lifestyle was not yet considered by any study, this
hypothesis seems to be reasonable for habitats with a high
press of owl hunting on the voles.
Pup isolation calls and adult discomfort-related calls
analyzed in both vole species in this study are not directly
related to predation, being either addressed to parents (pup
calls) or expressing the internal state of discomfort of a
caller (adult calls). However, potentially, terrestrial preda-
tors (foxes and mustelids) can rely on hearing these calls
(AUDs and USVs) when hunting by digging out the ani-
mals from burrows, as foxes, or coming into the burrow,
as mustelids. Mustelids and foxes can hear all the range
of both audible and ultrasonic calls of voles, up to 51kHz
in the least weasel Mustela nivalis (Heffner and Heffner
1985) and up to 48–51kHz in red fox Vulpes vulpes (Mal-
kemper etal. 2015). Red foxes hear substantially better
the audible calls around 2kHz than ultrasonic frequencies
(Peterson etal. 1969; Malkemper etal. 2015) and there-
fore can rely on rustling sounds of voles rather than their
audible or ultrasonic calls (Frey etal. 2016). In addition,
pup voles can be potentially predated in burrows by infan-
ticidal conspecifics or other species of rodents, e.g., by
Daurian ground squirrels Spermophilus dauricus, which
can be captured in burrows at the same colonies as voles
(own observations by the authors). Previously, the poten-
tial effect of infanticide on the acoustics of alarm calls was
considered for ground squirrels (Matrosova etal. 2007).
Supplementary Information The online version contains supplemen-
tary material available at https:// doi. org/ 10. 1007/ s00265- 022- 03213-6.
Acknowledgements We thank the staff of the Biological Institute of
Saint Petersburg University for help and support. We thank two anony-
mous reviewers for their valuable and constructive comments.
Author contribution IAV, AVS, and EVV designed the study. MMD,
IAV, AVS, and EVV collected the data. MMD and IAV performed
acoustic analyses. NAV and IAV performed statistical analyses. All
authors wrote the first draft of the manuscript, commented on and
approved the final manuscript before submission.
Funding This study was supported by the Russian Science Founda-
tion grant number 19–14-00037, for the audio recording and analysis
(to IAV and EVV) and by the Russian Foundation for Basic Research,
grant number 19–04-00538a, for behavioral experiments with animals
(to AVS).
Data availability The datasets used in this study are available from the
supplementary information files.
Declarations
Ethics approval The authors adhered to the “Guidelines for the treat-
ment of animals in behavioural research and teaching” (Anim Behav
(2020) 159:I-XI) and the legal requirements of Russia pertaining to
the protection of animal welfare. The experimental procedure was
approved by the Committee of Bio-ethics of Lomonosov Moscow State
University, research protocol # 2011–36.
Consent for publication All authors approved the final manuscript
before submission.
Conflict of interest The authors declare no competing interests.
References
Abramson NI, Lebedev VS, Tesakov AS, Bannikova AA (2009)
Supraspecies relationships in the subfamily Arvicolinae (Roden-
tia, Cricetidae): an unexpected result of nuclear gene analysis.
Mol Biol 43:834. https:// doi. org/ 10. 1134/ S0026 89330 90501 48
Alexeeva N, Erbajeva M, Khenzykhenova F (2015) Lasiopodomys
brandti in Pleistocene of Transbaiklia and adjacent territories:
distribution area, evolutionary development in context of global
and regional events. Quatern Int 355:11–17. https:// doi. org/ 10.
1016/j. quaint. 2014. 08. 017
Begall S, Burda H, Schleich CE (eds) (2007a) Subterranean rodents.
News from Underground. Springer, Berlin
Begall S, Lange S, Schleich C, Burda H (2007b) Acoustics, audition
and auditory system. In: Begall S, Burda H, Schleich CE (eds)
Subterranean rodents: News from underground. Springer, Berlin,
pp 97–111
Bondareva OV, Potapova NA, Konovalov KA, Petrova TV, Abram-
son NI (2021) Searching for signatures of positive selection in
cytochrome b gene associated with subterranean lifestyle in fast-
evolving arvicolines (Arvicolinae, Cricetidae, Rodentia). BMC
Ecol Evol 21:92. https:// doi. org/ 10. 1186/ s12862- 021- 01819-4
Brückmann G, Burda H (1997) Hearing in blind subterranean Zambian
mole-rats (Cryptomys sp.): collective behavioural audiogram in
a highly social rodent. J Comp Physiol A 181:83–88. https:// doi.
org/ 10. 1007/ s0035 90050 095
Brudzynski SM (2005) Principles of rat communication: quantitative
parameters of ultrasonic calls in rats. Behav Genet 35:85–92.
https:// doi. org/ 10. 1007/ s10519- 004- 0858-3
Brudzynski SM, Kehoe P, Callahan M (1999) Sonographic structure
of isolation-induced ultrasonic calls of rat pups. Devel Psychobiol
34:195–204. https:// doi. org/ 10. 1002/ (sici) 1098- 2302(199904)
34:3% 3c195:: aid- dev4% 3e3.0. co;2-s

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
Page 15 of 17 106
Burda H, Šumbera R, Begall S (2007) Microclimate in burrows of
subterranean rodents - revisited. In: Begall S, Burda H, Schleich
CE (eds) Subterranean rodents: News from underground.
Springer, Berlin, pp 21–33
Caspar KR, Heinrich A, Mellinghaus L, Gerhardt P, Begall S (2021)
Evoked auditory potentials from African mole-rats and coruros
reveal disparity in subterranean rodent hearing. J Exp Biol
224:243371. https:// doi. org/ 10. 1242/ jeb. 243371
Charlton BD, Keating JL, Rengui L, Huang Y, Swaisgood RR (2009)
Female giant panda (Ailuropoda melanoleuca) chirps advertise
the caller’s fertile phase. Proc R Soc Lond B 277:1101–1106.
https:// doi. org/ 10. 1098/ rspb. 2009. 1431
Credner S, Burda H, Ludescher F (1997) Acoustic communication
underground: vocalization characteristics in subterranean social
mole-rats (Cryptomys sp., Bathyergidae). J Comp Physiol A
180:245–255. https:// doi. org/ 10. 1007/ s0035 90050 045
Cui C, Xie Y, Hua Y, Yang S, Yin B, Wei W (2020) Brandt’s vole
(Lasiopodomys brandtii) affects its habitat quality by altering
plant community composition. Biologia 75:1097–1104. https://
doi. org/ 10. 2478/ s11756- 020- 00469-z
de Koning M, Beatini JR, Proudfoot GA, Gall MD (2020) Hearing in
3D: directional auditory sensitivity of northern saw-whet owls
(Aegolius acadicus). Integr Comp Biol 60:1058–1067. https://
doi. org/ 10. 1093/ icb/ icaa0 24
Dent ML, Screven LA, Kobrina A (2018) Hearing in rodents. In:
Dent ML, Fay RR, Popper AN (eds) Rodent Bioacoustics.
Springer, Cham, pp 71–105
Dmitriev PP, Tamir Z, Davaa N (1980) Characteristics of habitats
of Microtus brandtii in the East Khangai. Zool Zh 59:274–282
[in Russian]
Dong Q, Shi L, Li Y etal (2018) Differential responses of Lasio-
podomys mandarinus and Lasiopodomys brandtii to chronic
hypoxia: a cross-species brain transcriptome analysis. BMC
Genomics 19:901. https:// doi. org/ 10. 1186/ s12864- 018- 5318-1
Dvořáková V, Hrouzková E, Šumbera R (2016) Vocal repertoire
of the social Mashona mole-rat (Fukomys darlingi) and how
it compares with other mole-rats. Bioacoustics 25:253–266.
https:// doi. org/ 10. 1080/ 09524 622. 2016. 11411 17
Ey E, Fischer J (2009) The ‘acoustic adaptation hypothesis’ – a
review of the evidence from birds, anurans and mammals. Bio-
acoustics 19:21–48. https:// doi. org/ 10. 1080/ 09524 622. 2009.
97536 13
Fitch WT, Hauser MD (2002) Unpacking “Honesty”: Vertebrate vocal
production and the evolution of acoustic signals. In: Simmons A,
Fay RR, Popper AN (eds) Acoustic Communication. Springer,
New York, pp 65–137
Frey R, Volodin IA, Fritsch G, Volodina EV (2016) Potential sources
of high frequency and biphonic vocalization in the dhole (Cuon
alpinus). PLoS ONE 11:e0146330. https:// doi. org/ 10. 1371/ journ
al. pone. 01463 30
Gerhardt P, Henning Y, Begall S, Malkemper EP (2017) Audiograms
of three subterranean rodent species (genus Fukomys) determined
by auditory brainstem responses reveal extremely poor high-fre-
quency hearing. J Exp Biol 220:4377–4382. https:// doi. org/ 10.
1242/ jeb. 164426
Heffner RS, Heffner HE (1985) Hearing in mammals: the least weasel.
J Mammal 66:745–755. https:// doi. org/ 10. 2307/ 13808 01
Heffner RS, Heffner HE (1992) Hearing and sound localization in blind
mole rats (Spalax ehrenbergi). Hearing Res 62:206–216. https://
doi. org/ 10. 1016/ 0378- 5955(92) 90188-s
Heth G, Frankenberg E, Nevo E (1986) Adaptive optimal sound for
vocal communication in tunnels of a subterranean mammal
(Spalax ehrenbergi). Experientia 42:1287–1289. https:// doi. org/
10. 1007/ BF019 46426
Hrouzková E, Bernasová E, Šklíba J (2020) Eavesdropping on a
heterospecific alarm call in the giant root-rat (Tachyorytes
macrocephalus), an important prey of the Ethiopian wolf
(Canis simensis). J Ethol 38:121–124. https:// doi. org/ 10. 1007/
s10164- 019- 00618-1
Jiao H, Hong W, Nevo E, Li K, Zhao H (2019) Convergent reduction
of V1R genes in subterranean rodents. BMC Evol Biol 19:176.
https:// doi. org/ 10. 1186/ s12862- 019- 1502-4
Khruscelevsky VP, Kopylova OA (1957) Materials on ecology of
Brandt’s vole. Report 5. Peculiarities of seasonal and daily activ-
ity. Trans Antiplague Inst Siberia Far East 16:69–76 [in Russian]
Klenova AV, Volodin IA, Ilchenko OG, Volodina EV (2021) Dis-
comfort-related changes of call rate and acoustic variables
of ultrasonic vocalizations in adult yellow steppe lemmings
Eolagurus luteus. Sci Rep 11:14969. https:// doi. org/ 10. 1038/
s41598- 021- 94489-7
Knotková E, Veitl S, Šumbera R, Sedláček F, Burda H (2009) Vocalisa-
tions of the silvery mole-rat: comparison of vocal repertoires in
subterranean rodents with different social systems. Bioacoustics
18:241–257. https:// doi. org/ 10. 1080/ 09524 622. 2009. 97536 04
Kozhevnikova JD (2021) Comparative analysis of parameters of ultra-
sonic calls in adults and pups of four Gerbillinae species. MSc
thesis, Moscow State University [in Russian]
Kozhevnikova JD, Volodin IA, Zaytseva AS, Ilchenko OG, Volodina
EV (2021) Pup ultrasonic isolation calls of six gerbil species and
the relationship between acoustic traits and body size. R Soc Open
Sci 8:201558. https:// doi. org/ 10. 1098/ rsos. 201558
Lacey EA, Patton JL, Cameron GN (eds) (2000) Life underground:
The biology of subterranean rodents. The University of Chicago
Press, Chicago
Lange S, Burda H, Wegner RE, Dammann P, Begall S, Kawalika M
(2007) Living in a “stethoscope”: burrow-acoustics promote audi-
tory specializations in subterranean rodents. Naturwissenschaften
94:134–138. https:// doi. org/ 10. 1007/ s00114- 006- 0168-0
Lebedev VS, Bannikova AA, Ya A, Shar S, Surov AV (2016) A revised
checklist of Mongolian mammal species. Explor Biol Resour
Mongolia 13:349–360
Lenth RV (2021) Emmeans: estimated marginal means, aka least-
squares means. R package version 1.6.1. https:// cran.r- proje ct.
org/ web/ packa ges/ emmea ns/ index. html
Lesch R, Orozco A, Shilling M, Zimmerberg B, Fitch WT (2020)
Selection on ultrasonic call rate in neonatal rats affects low
frequency, but not ultrasonic, vocalizations in adults. Ethology
126:1007–1018. https:// doi. org/ 10. 1111/ eth. 13075
Li K, Kohn MH, Zhang S, Wan X, Shi D, Wang D (2017) The colo-
nization and divergence patterns of Brandt’s vole (Lasiopodomys
brandtii) populations reveal evidence of genetic surfing. BMC
Evol Biol 17:145. https:// doi. org/ 10. 1186/ s12862- 017- 0995-y
Kӧssl M, Frank G, Burda H, Müller M (1996) Acoustic distortion prod-
ucts from the cochlea of the blind African mole rat, Cryptomys
spec. J Comp Physiol A 178:427–434. https:// doi. org/ 10. 1007/
BF001 93979
Malkemper EP, Topinka V, Burda H (2015) A behavioral audiogram
of the red fox (Vulpes vulpes). Hear Res 320:30–37. https:// doi.
org/ 10. 1016/j. heares. 2014. 12. 001
Matrosova VA, Volodin IA, Volodina EV, Babitsky AF (2007) Pups
crying bass: Vocal adaptation for avoidance of age-dependent pre-
dation risk in ground squirrels? Behav Ecol Sociobiol 62:181–
191. https:// doi. org/ 10. 1007/ s00265- 007- 0452-9
Matrosova VA, Blumstein DT, Volodin IA, Volodina EV (2011)
The potential to encode sex, age, and individual identity in the
alarm calls of three species of Marmotinae. Naturwissenschaften
98:181–192. https:// doi. org/ 10. 1007/ s00114- 010- 0757-9
McGee JA, Nelson PB, Ponder JB, Marr J, Redig P, Walsh EJ (2019)
Auditory performance in bald eagles and red-tailed hawks: a com-
parative study of hearing in diurnal raptors. J Comp Physiol A
205:793–811. https:// doi. org/ 10. 1007/ s00359- 019- 01367-9

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
106 Page 16 of 17
Müller M, Burda H (1989) Restricted hearing range in a subter-
ranean rodent, Cryptomys hottentotus. Naturwissenschaften
76:134–135. https:// doi. org/ 10. 1007/ BF003 66611
Musolf K, Penn DJ (2012) Ultrasonic vocalizations in house mice:
a cryptic mode of acoustic communication. In: Macholan M,
Baird SJE, Munclinger P, Pialek J (eds) Evolution of the house
mouse. Cambridge University Press, Cambridge, pp 253–277
Nevo E (1999) Mosaic evolution of subterranean mammals: regres-
sion, progression and global convergence. Oxford University
Press, Oxford
Nevo E, Heth G, Beiles A, Frankenberg E (1987) Geographic dia-
lects in blind mole rats: role of vocal communication in active
speciation. P Natl Acad Sci USA 84:3312–3315. https:// doi. org/
10. 1073/ pnas. 84. 10. 3312
Nikolskii AA (2007) Comparative analyses of the alarm call fre-
quency in different age rodent groups. Zool Zh 86:499–504 [in
Russian]
Okanoya K, Screven LA (2018) Rodent vocalizations: adaptations to
physical, social, and sexual factors. In: Dent M, Fay R, Popper A
(eds) Rodent Bioacoustics. Springer, Cham, pp 13–41
Okanoya K, Yosida S, Barone CM, Applegate DT, Brittan-Powell EF,
Dooling RJ, Park TJ (2018) Auditory-vocal coupling in the naked
mole-rat, a mammal with poor auditory thresholds. J Comp Phys-
iol A 204:905–914. https:// doi. org/ 10. 1007/ s00359- 018- 1287-8
Pandourski I (2011) Case of alarm vocalization in a colony of Microtus
guentheri (Danford & Alston, 1880) (Mammalia, Rodentia, Arvi-
colidae) from Southern Bulgaria. ZooNotes 20:1–3
Park TJ, Reznick J, Peterson BL etal (2017) Fructose-driven glyco-
lysis supports anoxia resistance in the naked mole-rat. Science
356:307–311. https:// doi. org/ 10. 1126/ scien ce. aab38 96
Pepper JW, Stanton HB, Lacey EA, Sherman PW (2017) Vocalization
of the naked mole-rat. In: Sherman PW, Jarvis JUM, Alexander
RD (eds) The biology of the naked mole-rat. Princeton University
Press, Princeton, NJ, pp 243–274
Peterson EA, Heaton WC, Wruble SD (1969) Levels of auditory
response in fissiped carnivores. J Mammal 50:566–578. https://
doi. org/ 10. 2307/ 13787 84
Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team (2021) nlme:
linear and nonlinear mixed effects models. R package version
3.1–153, https:// CRAN.R- proje ct. org/ packa ge= nlme
Potier S, Mitkus M, Kelber A (2020) Visual adaptations of diurnal and
nocturnal raptors. Semin Cell Dev Biol 106:116–126. https:// doi.
org/ 10. 1016/j. semcdb. 2020. 05. 004
R Development Core Team (2021) R: a language and environment
for statistical computing. R Foundation for Statistical Computing,
Vienna, Austria, http:// www.R- proje ct. org
Riede T (2011) Subglottal pressure, tracheal airflow, and intrinsic
laryngeal muscle activity during rat ultrasound vocalization. J
Neurophysiol 106:2580–2592. https:// doi. org/ 10. 1152/ jn. 00478.
2011
Riede T, Zhao Y, LeDoux MS (2015) Vocal development in dystonic
rats. Physiol Rep 3:12350. https:// doi. org/ 10. 14814/ phy2. 12350
Rutovskaya MV (2011) Acoustic communication in the Mandarin vole
(Lasiopodomys mandarinus, Rodentia). Biol Bull 38:911–918.
https:// doi. org/ 10. 1134/ S1062 35901 10900 68
Rutovskaya MV (2012) The sound signals of Brandt’s vole (Lasiopod-
omys brandti). Sens Systems 26:31–38 [In Russian]
Rutovskaya MV (2018) Audible acoustic communication in voles of
Arvicolinae subfamily. KMK Press, Moscow [In Russian]
Rutovskaya MV (2019a) Sound communication in social voles (sub-
genus Sumeriomys). Bioacoustics 28:503–521. https:// doi. org/ 10.
1080/ 09524 622. 2018. 14827 87
Rutovskaya MV (2019b) Vocal communication in Microtus (Terricola)
schelkovnikovi and M. (T.) daghestanicus in the audible range
of frequencies. J Ethol 37:187–195. https:// doi. org/ 10. 1007/
s10164- 019- 00587-5
Rutovskaya MV (2019c) Inheritance of the acoustic signal parameters
in interspecific hybrids of the bank (Myodes glareolus) and the
Tien Shan (M. centralis) voles. BMC Evol Biol 19:44. https:// doi.
org/ 10. 1186/ s12862- 019- 1374-7
Sahm A, Platzer M, Koch P etal (2021) Increased longevity due to
sexual activity in mole-rats is associated with transcriptional
changes in the HPA stress axis. eLife 10:57843. https:// doi. org/
10. 7554/ eLife. 57843
Schleich CE, Antenucci DC (2009) Sound transmission and burrow
characteristics of the subterranean rodent Ctenomys talarum
(Rodentia: Ctenomyidae). Acta Theriol 54:165–170. https:// doi.
org/ 10. 1007/ BF031 93172
Schleich CE, Francescoli G (2018) Three decades of subterranean
acoustic communication studies. In: Dent M, Fay R, Popper A
(eds) Rodent Bioacoustics. Springer, Cham, pp 43–69
Schneiderová I (2014) Vocal repertoire ontogeny of the captive Asian
house shrew Suncus murinus suggests that the male courtship call
develops from the caravanning call of the young. Acta Theriol
59:149–164. https:// doi. org/ 10. 1007/ s13364- 013- 0141-1
Schneiderová I, Schnitzerova P, Uhlıkova J, Brandl P, Zouhar J, Mateju
J (2015) Differences in alarm calls of juvenile and adult European
ground squirrels (Spermophilus citellus): findings on permanently
marked animals from a semi-natural enclosure. Zoo Biol 34:503–
512. https:// doi. org/ 10. 1002/ zoo. 21233
Shi L, Liu L, Li X, Wu Y, Tian X, Shi Y, Wang Z (2021) Phylogeny
and evolution of Lasiopodomys in subfamily Arvivolinae based
on mitochondrial genomics. Peer J 9:e10850. https:// doi. org/ 10.
7717/ peerj. 10850
Sibiryakova OV, Volodin IA, Volodina EV (2021) Polyphony of
domestic dog whines and vocal cues to body size. Curr Zool
67:165–176. https:// doi. org/ 10. 1093/ cz/ zoaa0 42
Smith AT, Xie Y (2008) A Guide to the Mammals of China. Princeton
University Press, Princeton, NJ
Smorkatcheva AV (1999) The social organization of the mandarin
vole, Lasiopodomys mandarinus, during the reproductive period.
Z Saugetierkd 64:344–355
Smorkatcheva AV, Aksenova TG, Zorenko TA (1990) The ecology of
Lasiopodomys mandarinus (Rodentia, Cricetidae) in the Transbai-
cal area. Zool Zh 69:115–124 [in Russian]
Sun H, Zhang Y, Wang B, LiY XuW, Mao R, Wang Z (2019) Investiga-
tion on oxygen and carbon dioxide fluctuations in Lasiopodomys
mandarinus burrows. Pakistan J Zool 51:1519–1526. https:// doi.
org/ 10. 17582/ journ al. pjz/ 2019. 51.4. 1519. 1526
Sun H, Ye K, Liu D, Pan D, Gu S, Wang Z (2020) Evolution of hemo-
globin genes in a subterranean rodent species (Lasiopodomys
mandarinus). Biology 9:106. https:// doi. org/ 10. 3390/ biolo gy905
0106
Swan DC, Hare JF (2008) Signaler and receiver ages do not affect
responses to Richardson’s ground squirrel alarm calls. J Mammal
89:889–894. https:// doi. org/ 10. 1644/ 07- MAMM-A- 228.1
Tai F, Wang T (2001) Social organization of mandarin voles in burrow
system. Acta Theriol Sin 21:50–56
Terleph T (2011) A comparison of prairie vole audible and ultrasonic
pup calls and attraction to them by adults of each sex. Behaviour
148:1277–1296. https:// doi. org/ 10. 1163/ 00057 9511X 600727
Vejmělka F, Okrouhlík J, Lövy M, Šafa G, Nevo E, Bennett NC,
Šumbera R (2021) Heat dissipation in subterranean rodents: the
role of body region and social organisation. Sci Rep 11:2029.
https:// doi. org/ 10. 1038/ s41598- 021- 81404-3
Volodin IA, Zaytseva AS, Ilchenko OG, Volodina EV (2015) Small
mammals ignore common rules: a comparison of vocal reper-
toires and the acoustics between pup and adult piebald shrews
Diplomesodon pulchellum. Ethology 121:103–115. https:// doi.
org/ 10. 1111/ eth. 12321
Volodin IA, Yurlova DD, Ilchenko OG, Volodina EV (2021) Ontogeny
of audible squeaks in yellow steppe lemming Eolagurus luteus:

Behavioral Ecology and Sociobiology (2022) 76:106
1 3
Page 17 of 17 106
trend towards shorter and low-frequency calls is reminiscent of
those in ultrasonic vocalization. BMC Zool 6:27. https:// doi. org/
10. 1186/ s40850- 021- 00092-8
Volodin IA, Dymskaya MM, Smorkatcheva AV, Volodina EV (2022)
Ultrasound from underground: cryptic communication in subter-
ranean wild-living and captive northern mole voles (Ellobius tal-
pinus). Bioacoustics 31:414–434. https:// doi. org/ 10. 1080/ 09524
622. 2021. 19601 91
Volodina EV, Matrosova VA, Volodin IA (2010) An unusual effect of
maturation on the alarm call fundamental frequency in two spe-
cies of ground squirrels. Bioacoustics 20:87–98. https:// doi. org/
10. 1080/ 09524 622. 2011. 97536 34
Wan X, Liu W, Wang G, Wang M, Zhong W (2006) Seasonal changes
of the activity patterns of Brandt’s vole (Lasiopodomys brandtii)
in the typical steppe in Inner Mongolia. Acta Theriol Sin
26:226–234
Warren MR, Campbell D, Borie AM, Ford CL, Dharani AM, Young
LJ, Liu RC (2022) Maturation of social-vocal communication in
prairie vole (Microtus ochrogaster) pups. Front Behav Neurosci
15:814200. https:// doi. org/ 10. 3389/ fnbeh. 2021. 814200
Wilden I, Tembrock G, Herzel H, Peters G (1998) Subharmonics,
biphonation, and deterministic chaos in mammal vocalization.
Bioacoustics 9:171–196. https:// doi. org/ 10. 1080/ 09524 622. 1998.
97533 94
Wilson DR, Hare JF (2006) The adaptive utility of Richardson’s ground
squirrel Spermophilus richardsonii short-range ultrasonic alarm
signals. Can J Zool 84:1322–1330. https:// doi. org/ 10. 1139/
z06- 120
Yamazaki Y, Yamada H, Murofushi M, Momose H, Okanoya K (2004)
Estimation of hearing range in raptors using unconditioned
responses. Ornithol Sci 3:85–92. https:// doi. org/ 10. 2326/ osj.3. 85
Yu P, Wang J, Tai F, Broders H, An S, Zhang X, He F, An X, Wu
R (2011) The effects of repeated early deprivation on ultrasonic
vocalizations and ontogenetic development in mandarin vole pups.
Behav Process 88:162–167. https:// doi. org/ 10. 1016/j. beproc. 2011.
09. 001
Yurlova DD, Volodin IA, Ilchenko OG, Volodina EV (2020) Rapid
development of mature vocal patterns of ultrasonic calls in a fast-
growing rodent, the yellow steppe lemming (Eolagurus luteus).
PLoS ONE 15:e0228892. https:// doi. org/ 10. 1371/ journ al. pone.
02288 92
Zaytseva AS, Volodin IA, Ilchenko OG, Volodina EV (2019) Ultra-
sonic vocalization of pup and adult fat-tailed gerbils (Pachyuro-
mys duprasi). PLoS ONE 14:e0219749. https:// doi. org/ 10. 1371/
journ al. pone. 02197 49
Zaytseva AS, Volodin IA, Ilchenko OG, Volodina EV (2020) Audible
calls and their ontogenetic relationship with ultrasonic vocali-
zation in a rodent with a wide vocal range, the fat-tailed gerbil
(Pachyuromys duprasi). Behav Process 180:104241. https:// doi.
org/ 10. 1016/j. beproc. 2020. 104241
Zhong Z, Li G, Sanders D, Wang D, Holt RD, Zhang Z (2022) A
rodent herbivore reduces its predation risk through ecosystem
engineering. Curr Biol 32:1869–1874. https:// doi. org/ 10. 1016/j.
cub. 2022. 02. 074
Zorenko TA, Jakobsone GK (1986) Peculiarities of the postnatal devel-
opment of the Brandt’s vole, Lasiopodomys brandtii Radde. In:
Zorenko TA (ed) Conservation, ecology and ethology of animals.
University of Latvia, Riga, pp 25–44 [in Russian]
Zorenko TA, Smorkatcheva AV, Aksyonova TG (1994) Reproduction
and postnatal ontogenesis of the mandarin vole Lasiopodomys
mandarinus Milne-Edwards, and a comparison with Brandt’s vole
L. brandti Radde (Rodentia, Cricetidae). Laboratornye Zhyvotnye
4:5–16
Publisher's note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Authors and Aliations
MargaritaM.Dymskaya1· IlyaA.Volodin2,3 · AntoninaV.Smorkatcheva1 · NinaA.Vasilieva4 ·
ElenaV.Volodina3
Margarita M. Dymskaya
rita.dym@yandex.ru
Antonina V. Smorkatcheva
tonyas1965@mail.ru
Nina A. Vasilieva
ninavasilieva@gmail.com
Elena V. Volodina
volodinsvoc@mail.ru
1 Department ofVertebrate Zoology, St. Petersburg State
University, St.Petersburg199034, Russia
2 Department ofVertebrate Zoology, Faculty ofBiology,
Lomonosov Moscow State University, Vorobievy Gory, 1/12,
Moscow119234, Russia
3 Department ofBehaviour andBehavioural Ecology,
Severtsov Institute ofEcology andEvolution,
Russian Academy ofSciences, Leninsky prospect 33,
Moscow119071, Russia
4 Department ofPopulation Ecology, Severtsov Institute
ofEcology andEvolution, Russian Academy ofSciences,
Leninsky prospect, 33, Moscow119071, Russia