© 2021 Nordic Society Oikos. Published by John Wiley & Sons Ltd
Subject Editor: Jerome Spitz
Editor-in-Chief: Dries Bonte
Accepted 31 August 2021
00: 1–12, 2021
Consumers feeding at the aquatic–terrestrial ecosystem interface may obtain a mixture
of aquatic and terrestrial diet resources that vary in nutritional composition. However,
in lake riparian spiders, the relative signiﬁcance of aquatic versus terrestrial diet sources
remains to be explored. We investigated the trophic transfer of lipids and polyunsatu-
rated fatty acids (PUFA) from emergent aquatic and terrestrial insects to spiders at
varying distances from the shoreline of a subalpine lake in Austria, using diﬀerences
in fatty acid proﬁles and compound-speciﬁc stable carbon (δ13C) and hydrogen (δ2H)
isotopes. e omega-3 PUFA content of emergent aquatic insects was higher than that
of terrestrial insects. Emergent aquatic insects contained on average 6.6 times more
eicosapentaenoic acid (EPA) and 1.2 times more α-linolenic acid (ALA) than terres-
trial insects, whereas terrestrial insects contained on average 2.6 times more linoleic
acid (LIN) than emergent aquatic insects. Spiders sampled directly on the lake and
in upland habitats had similar EPA contents, but this EPA was derived from diﬀerent
diet sources, depending on the habitat. e δ13CEPA and δ2HEPA values of ‘lake spiders’
revealed an aquatic diet pathway (i.e. EPA of aquatic origin). In contrast, EPA of spi-
ders collected in terrestrial habitats was depleted in both 13C and 2H compared to any
potential food sources, and their ALA isotopic values, suggesting that EPA was partly
bioconverted from its dietary precursor ALA (i.e. internal pathway). e δ2H values of
fatty acids clearly indicated that diet sources diﬀered depending on the spider’s habitat,
which was less evident from the δ13C values of the fatty acids. Our data highlight that
spiders can use two distinct pathways (trophic versus metabolic) to satisfy their physi-
ological EPA demand, depending on habitat use and dietary availability.
Keywords: bioconversion, carbon isotopes of fatty acids, eicosapentaenoic acid,
emerging aquatic insects, hydrogen isotopes of fatty acids, riparian consumers
Dietary availability determines metabolic conversion of long-
chain polyunsaturated fatty acids in spiders: a dual compound-
specific stable isotope approach
Margaux Mathieu-Resuge, Matthias Pilecky, Cornelia W. Twining, Dominik Martin-Creuzburg,
Tarn Preet Parmar, Simon Vitecek and Martin J. Kainz
M. Mathieu-Resuge (https://orcid.org/0000-0001-6927-3400) ✉ (email@example.com), M. Pilecky (https://orcid.org/0000-0002-3404-5923),
S. Vitecek (https://orcid.org/0000-0002-7637-563X) and M. J. Kainz (https://orcid.org/0000-0002-2388-1504), WasserCluster Lunz – Biologische Station
GmbH, Lunz am See, Austria. SV also at: Univ. of Natural Resources and Life Sciences, Vienna, Inst. of Hydrobiology and Aquatic Ecosystem Management,
Vienna, Austria. MJK also at: Faculty of Medicine and Health, Danube Univ. Krems, Krems, Austria. – C. W. Twining (https://orcid.org/0000-0002-4346-
8856), Max Planck Inst. of Animal Behavior, Radolfzell, Germany. – CWT, D. Martin-Creuzburg (https://orcid.org/0000-0002-4248-0730) and T. P.
Parmar (https://orcid.org/0000-0002-1321-6893), Limnological Inst., Univ. of Konstanz, Konstanz, Germany.
Nutrient and energy transfer across ecosystem boundaries is
a key process ensuring ecosystem functioning and food web
stability (Barnes et al. 2018). Insects that spend their juvenile
life stages in aquatic ecosystems before emerging (i.e. emergent
aquatic insects) into terrestrial ecosystems as adults are impor-
tant vectors of dietary nutrients, including omega-3 long-chain
polyunsaturated fatty acids (n-3 LC-PUFA; ≥ 20C). In aquatic
ecosystems, n-3 LC-PUFA are mostly synthesized by aquatic
primary producers and subsequently transferred across trophic
levels within the aquatic food web, while these dietary nutri-
ents are generally lacking in terrestrial food webs (Twining et al.
2016a, 2019). During their larval stages, emergent aquatic
insects acquire n-3 LC-PUFA, such as eicosapentaenoic acid
(EPA, 20:5n-3), from aquatic resources (Torres-Ruiz et al.
2007, Martin-Creuzburg et al. 2017, Scharnweber et al. 2019,
Twining et al. 2019). e n-3 LC-PUFA are considered to be
essential for animals because they are required to maintain vital
physiological processes but cannot be synthesized de novo by
many taxa (Cook and McMaster 2002, Twining et al. 2016a,
Fritz et al. 2017). However, many consumers are able to synthe-
size n-3 LC-PUFA from dietary C18 PUFA precursors, albeit
in quantities that are often insuﬃcient to cover their needs
(Twining et al. 2018). Even when bioconversion from dietary
precursors is possible, it is likely more energetically demand-
ing than dietary acquisition of speciﬁc n-3 LC-PUFA (Parrish
2009). erefore, many consumers can beneﬁt from obtaining
n-3 LC-PUFA through their diet to meet their physiological
demands. e diﬀerences in fatty acid proﬁles between aquatic
and terrestrial insects can help reveal aquatic–terrestrial linkages,
e.g. by comparing fatty acid proﬁles of consumers, like spiders,
with their potential prey (Fritz et al. 2017, Chari et al. 2020).
Spiders and other insectivorous riparian consumers prey
upon a wide range of insects and thus obtain a mixture of
resources diﬀering in nutritional quality (Paetzold et al. 2005,
Fritz et al. 2017, Twining et al. 2019, Chari et al. 2020). Web-
constructing spiders are relatively stationary predators and
mostly insectivorous (Nyﬀeler 1999); they occur along broad
ecological gradients and can disperse by means of ballooning
(Bonte 2013). In spiders, random colonisation by ballooning
can be elicited by kin competition (Berger-Tal et al. 2016)
or poor resource availability (Mestre and Bonte 2012). is
strategy is risky and energetically costly (Bonte 2013), but
can provide access to high quality resources (e.g. emergent
aquatic insects rich in n-3 LC-PUFA). In contrast to spiders
that drift within terrestrial habitats, spiders that inciden-
tally drift to ﬂoating structures (e.g. docks, boats, emergence
traps) on aquatic habitats likely encounter food sources that
are richer in n-3 LC-PUFA, i.e. aquatic insects. e avail-
ability of n-3 LC-PUFA from aquatic resources can aﬀect the
performance of riparian insectivores (Twining et al. 2016b,
2018, 2019), including spiders (Fritz et al. 2017). However,
how spiders process diets from aquatic or terrestrial origin has
not been studied yet.
Bulk carbon and nitrogen stable isotopes as well as fatty
acid proﬁles have been successfully used to reveal food sources
and the degree to which aquatic or terrestrial resources
are used by diﬀerent consumers (Vander Zanden et al.
1999, Iverson et al. 2004, Perga et al. 2006, Iverson 2009,
Galloway et al. 2015), even though their application is con-
strained by the fact that several resources may share similar
stable isotope/fatty acid compositions (Cloern et al. 2002,
Guo et al. 2018, Twining et al. 2020, Ebm et al. 2021).
Compound-speciﬁc stable isotope analysis (CSIA) oﬀers an
innovative and promising approach to these limitations. e
stable isotope values of fatty acids can indicate their dietary
or metabolic origin (Bec et al. 2011, Burian et al. 2020,
Kühmayer et al. 2020, Pilecky et al. 2021). Twining et al.
(2020) recently proposed that stable hydrogen isotopes of
fatty acids can be used at much ﬁner-scale resolutions to
assess the origin of dietary resources and trophic interactions
than it would be possible from stable carbon isotopes, espe-
cially in systems in which the carbon stable isotope values of
potential resource fatty acids overlap (Pilecky et al. 2021).
Here, we applied these methods to determine whether spi-
ders obtain physiologically important n-3 LC-PUFA, such as
EPA, from aquatic or terrestrial sources.
Few analyses of spiders’ responses to variations in diet
composition and quality at small spatial scales have been
conducted so far, and such studies have largely focused on
river ecosystems (Paetzold et al. 2005, Chari et al. 2020,
Siebers et al. 2021). In addition, in spite of studies suggest-
ing that aquatic-derived n-3 LC-PUFA can be beneﬁcial for
riparian spiders (Fritz et al. 2017), previous studies have not
resolved whether lake riparian spiders acquire n-3 LC-PUFA
from aquatic or terrestrial sources. In this context, the aim of
our study was to characterize the food resources and origin
of n-3 LC-PUFA in lake riparian consumers as well as their
metabolic fate. We used carbon and hydrogen stable isotopes
of fatty acids to investigate the origin of fatty acids and dem-
onstrate their potential as dietary tracers in ﬁeld studies. To
understand how habitat changes aﬀect resource use by spi-
ders, we assessed PUFA sources of (long-jawed) orb-weaving
and sheet-weaving spiders (Tetragnathidae, Araneidae and
Linyphiidae) within the riparian zone of a subalpine lake. We
collected potential insect prey emerging from the lake and
the surrounding terrestrial habitat as well as Tetragnathid,
Araneid and Linyphiid spiders. Based on total lipid content,
fatty acid and stable isotope proﬁles, as well as CSIA of fatty
acids, we tested the following hypotheses: 1) total lipid and
n-3 LC-PUFA contents of emerging aquatic insects are higher
than those of terrestrial insects; 2) spider fatty acid and bulk
stable isotope proﬁles reﬂect those of their insect prey; 3) the
origin of spider fatty acids diﬀers between spiders collected
directly on the lake and those collected around the lake.
Material and methods
Study sites and sampling
Insects and spiders were collected twice a week, from mid-
June to the end of September 2019, at the subalpine Lake
Lunz, Austria and in its adjacent terrestrial habitat (47°85′N,
15°05′E). Six ﬂoating emergence traps were deployed on
Lake Lunz (Fig. 1A–B), to collect emerging aquatic insects.
Araneidae, Tetragnathidae and Linyphiidae spiders coloniz-
ing emergence traps (‘lake spiders’ hereafter) were collected
by hand. Nine window/malaise hybrid traps were placed
along three diﬀerent terrestrial transects, each along a distance
gradient from the lake shore (1, 70 and 150 m; Fig. 1A, C).
Two transects of three window traps each were deployed in
an orchard (roughly corresponding to CLC 2.4.2 Complex
Cultivation Patterns), while the third transect was installed in a
coniferous forest (CLC 3.1.2), all along the southern margin of
Lake Lunz; again, Araneidae, Tetragnathidae and Linyphiidae
spiders colonizing window traps were collected by hand.
Traps collected a broad range of taxa that were grouped as
‘aquatic insects’ and ‘terrestrial insects’. ‘Aquatic insects’ com-
prised Chironomidae and Trichoptera (Phryganeidae and
Limnephilidae) from emergence traps on Lake Lunz, while
Cicadidae and a set of terrestrial Diptera (mostly Muscidae,
Phoridae, Psychodidae, Sciaridae and Dolichopodidae)
from window traps were considered ‘terrestrial insects’. Each
month, Araneidae, Tetragnathidae and Linyphiidae spiders
(total n = 5 per habitat) from the outside of traps were col-
lected in each habitat. Correspondingly, spiders collected on
the lake traps are considered ‘lake spiders’, while spiders col-
lected on terrestrial traps are considered ‘terrestrial spiders’.
Each emergence trap placed on Lake Lunz consisted of
four ﬂoatable tubes covering a surface area of 0.36 m2 form-
ing a pyramid-shaped construction covered with extra ﬁne
mosquito net (mesh size ~500 μm) (Martin-Creuzburg et al.
2017, Fig. 1B). Window traps (0.36 m2 area) were cov-
ered with same nets as the lake traps. Terrestrial traps were
suspended from trees and equipped with collecting vials
All collected insects and spiders were transported to the
laboratory within 1 hour, frozen at −80°C, then freeze-dried
for 24 h and identiﬁed to order or family level. Both spiders
and insects were counted and put in pre-weighed tin cups,
weighed and stored at −20°C until further analyses.
Fatty acid analysis
After freeze-drying, a minimum of 2 mg of insect (n = 10 per
order and per habitat) and spider (n = 5 per habitat) samples
were homogenized and lipids were extracted according to the
method described by Guo et al. (2016). Extracted lipids were
transmethylated to obtain fatty acid methyl esters (FAME)
that were subsequently analyzed on a gas chromatograph
(Trace GC; ermo Scientiﬁc; FID 250°C, carrier gas: He:
1 ml min−1, detector gases: H2: 35 ml min−1, make-up gas
ﬂow 30 ml min−1, air: 350 ml min−1, temperature ramp of
the oven: 140°C at 20°C min−1 for 5 min, to 170°C at 4°C
min−1 and to 240°C at 2°C min−1 for 8 min), equipped with
a ﬂame-ionization detector (FID, set at 250°C). FAME were
separated by a Supelco SP-2560 column (100 m, 0.25 mm
i.d., 0.2 mm ﬁlm thickness), identiﬁed by comparison of their
retention times with standards (37-component FAME Mix,
Supelco 47885-U; Bacterial Acid Methyl Ester Mix, Supelco
47080-U) and quantiﬁed with reference to seven-point cali-
bration curves based on known standard dilution raw con-
centrations. All fatty acids were measured and reported as
FAME, and their contents are expressed in mass fractions (i.e.
mg FAME g dw−1), and in percentages (%) of total fatty acids.
Bulk and compound-speciﬁc stable isotope analyses
Freeze-dried and homogenized samples (ca 0.3 mg) of insects
(n = 10 per order and per habitat) and spiders (n = 5 per
habitat) were put into tin capsules. eir bulk stable isotope
Figure 1. Location of the sampling habitats around Lake Lunz (A; 47°85′N, 15°05′E; Austria). Lake traps are represented by dark blue
squares, and terrestrial traps by red squares. Floating traps on the lake (B) and window trap on terrestrial habitat (C).
(δ13C and δ15N) values were quantiﬁed using an A ﬂash HT
Plus CNSOH elemental analyzer interfaced with a Conﬂo
IV device to a continuous ﬂow stable isotope ratio mass spec-
trometer (Delta V Advantage IRMS). Values were normalized
against reference gas injections of N2 and CO2 and standard-
ized using international standards using international stan-
dards IAEA-N-1, and IAEA-N-2 for nitrogen, and USGS24,
and IAEA-CH-7 for carbon.
Compound-speciﬁc stable isotope analyses (CSIA) were
performed to assess the isotopic composition of selected FA.
We selected insect samples (n = 5 per order, i.e. n = 10 per
habitat) collected in both habitats in August, and spiders
(n = 5 per habitat) of each habitat from June to September.
FAME were separated using a gas chromatograph linked to
the Delta V Advantage IRMS via Isolink 2 and Conﬂo IV.
A Split/Splitless Liner with Single Taper (4 × 6.3 × 78.5
mm, vat. no. 453A1355) was used, the injector temperature
was kept at 250°C and all samples were injected in splitless
mode. For δ13C, FAME were separated on a VF-WAXms 60
m/0.25 mm i.d./0.25 µm ﬁlm thickness column (Agilent
Technologies) at a ﬂow rate of 1.2 ml min−1, followed by oxi-
dation to CO2 in a combustion reactor, ﬁlled with Ni, Pt
and Cu wires, at a temperature of 1000°C. For δ2H, FAME
were separated on a VF-WAXms 30 m/0.32 mm i.d./1 µm
ﬁlm thickness column) at a ﬂow rate of 1.0 ml min−1, fol-
lowed by reduction to H2 by passing through a high thermal
conversion reactor (empty ceramic tube) kept at 1420°C.
e temperature gradient for δ13C analysis started at 80°C,
which was kept for 2 min, after which the temperature was
raised by 30°C min−1 to 175°C, by 5°C min−1 to 200°C and
ﬁnally by 2.4°C min−1 to 250°C, which was maintained for
30 min. e temperature gradient for δ2H analysis started at
80°C, which was kept for 2 min, after which the temperature
was raised by 30°C min−1 to 175°C, and then by 5°C min−1
to 240°C, and held for 35 min. FAME were identiﬁed as
for GC-FID using 37-component FAME Mix. Results are
expressed in delta (δ) units with respect to international stan-
dards (Vienna Standard Mean Ocean Water for δ2H, Vienna
Pee Dee Belemnite for δ13C and atmospheric nitrogen for
δ15N), following the equation: δ2H, δ13C or δ15N = [(Rsample/
Rstandard) − 1] × 103 (expressed in ‰), where R is 2H/1H,
13C/12C or 15N/14N. e 16:1n-7 and 16:1n-9, as 18:1n-7
and 18:1n-9 coelute on the column, therefore they are simul-
taneously analyzed and reported as Σ16:1 and Σ18:1.
Normality distribution and homoscedasticity of data were
analyzed using Shapiro–Wilk’s and Bartlett’s tests; both pre-
requisites were not met and thus data were analyzed using
non-parametric tests. First, we tested for diﬀerences in spi-
ders collected along a distance gradient (i.e. aquatic, ripar-
ian and more distant terrestrial). As there were no signiﬁcant
diﬀerences in δ13C and δ2H values of fatty acids among these
three sites (due to low number of replicates in diﬀerent sites),
we combined riparian and distant terrestrial spiders and com-
pared ‘lake’ versus terrestrial spiders.
To assess the diﬀerence of nutritional values of insects
depending on their habitats of origin, Kruskal–Wallis (KW)
tests followed by Conover–Iman multiple comparisons with
Bonferroni adjustment method (post hoc tests) allowed us
to compare the mean seasonal lipid content; as well as LIN,
ALA, EPA, total PUFA contents and for each other single
fatty acid found in insects of each habitat (in %, Table 1).
e diet of the spiders was characterized according to
their habitats. For that, we ﬁrst compared the diets’ lipid and
PUFA contents using KW tests. To represent lipid contents,
we combined the ‘violin’ and ‘box’ plots to show the density
distribution of our data together with the median, and quar-
tiles of it, where dots represent outliers. Violin plots show
the probability density of the total lipid contents (Hintze
and Nelson 1998; Fig. 2). Particular fatty acid contents (mg
FAME g−1) of interest are depicted in histograms (Fig. 3).
e KW tests also allowed to test for diﬀerences in the bulk
stable isotope values (carbon and nitrogen) between habitats
of spiders and between their potential preys, for each month
independently. Bulk stable isotope values (δ13C and δ15N)
of insects and spiders (i.e. including only Tetragnathidae) in
August were then used to assess diﬀerences in dietary sources
and consumers between both habitats (95% conﬁdent ellipses
for spiders were estimated using the stat_ellipse function of
ggplot2. Fig. 4).
Finally, the mean δ13C and δ2H values of ‘terrestrial’ and
‘aquatic’ spider fatty acids were compared using the same
method of KW tests (Supporting information), yielding dif-
ferences of PUFA origins between habitats. en, to trace the
origin of EPA in spiders, δ13C and δ2H values of EPA and
ALA were compared between spiders (i.e. consumers includ-
ing only Tetragnathidae) and insects (i.e. prey) from August
in each habitat, and represented by boxplots (ﬁg. 5).
All statistical analyses were performed and visualized with
R ver. 3.6.1 (<www.r-project.org>), using vegan, stat and
PMCMR, lm4 packages.
Differences in the nutritional value of emergent aquatic
versus terrestrial insects
e total lipid content of emergent aquatic insects was on
average 1.2 times higher than in terrestrial insects (KW test,
H25 = 6.00, p < 0.05, Fig. 2). e total PUFA content of
insects did not diﬀer signiﬁcantly between aquatic and ter-
restrial habitats (KW test, H25 = 3.40, p = 0.06; Fig. 3D).
On average, terrestrial insects contained 2.6 times more
LIN than emergent aquatic insects (KW test, H25 = 21.06,
p < 0.001; Fig. 3A), whereas emergent aquatic insects con-
tained 1.2 times more ALA than terrestrial insects (KW
test, H25 = 9.10, p < 0.01; Fig. 3B). In particular, the EPA
content of emergent aquatic insect was on average 6.6 times
higher than that of terrestrial insects (KW test, H25 = 63.40,
p < 0.001; Fig. 3C). However, emergent aquatic insects
had higher n-3 PUFA, but lower n-6 PUFA than terrestrial
insects, indeed the n-3/n-6 ratios of emergent aquatic insects
were on average 5.4 times higher than those of terrestrial
insects (Table 1).
Diet of spiders after dispersal in different habitats
Unlike insects, spiders collected on the lake (‘lake spiders’)
had on average 1.5 times lower lipid contents than spi-
ders from the terrestrial habitat (KW test, H25 = 10.40, p
< 0.01, Fig. 2). Total PUFA (KW test, H25 = 13.10, p <
0.001), LIN (KW test, H25 = 14.30, p < 0.001) and ALA
(KW test, H25 = 11.40, p < 0.001) contents were on aver-
age 1.7, 2.2 and 2.3 times higher in terrestrial than in ‘lake
spiders’, respectively (Fig. 3). e EPA contents of spiders
did not diﬀer signiﬁcantly between the two habitats (KW
test, H25 = 0.03, p = 0.87; Fig. 3C). e n-3/n-6 ratios of
spiders from both habitats were < 1, with ‘lake spiders’ show-
ing a signiﬁcantly higher ratio (0.8) compared to terrestrial
spiders (0.5) (KW test, H25 = 7.90, p < 0.001). e EPA
content of emergent insects did not diﬀer compared to ‘lake
spiders’ (KW test, H25 = 1.86, p = 0.17; Table 1), while
the terrestrial spiders have higher EPA contents than ter-
restrial insects (KW test, H25 = 21.86, p < 0.001; Table 1).
Terrestrial spiders accumulated higher contents of n-3 fatty
acids (included n-3 HUFA) compared to terrestrial insects
(KW test, H25 = 5.77, p < 0.05; Table 1). In comparison,
emergent insects had higher n-3 HUFA than ‘lake spiders’
(KW test, H25 = 10.22, p < 0.01; Table 1).
Emergent aquatic insects from Lake Lunz were 13C depleted
compared to insects from terrestrial habitats throughout the
study period (KW tests, p < 0.05; Table 2), and compared to
spiders from both habitats (KW tests, p < 0.05). Despite the
diﬀerences between insects, bulk carbon stable isotope values
did not diﬀer signiﬁcantly between ‘lake spiders’ and spiders
collected from terrestrial traps (Table 2) with the exception of
July, when ‘lake spiders’ were more 13C depleted than riparian
spiders (KW tests, H25 = 5.80, p < 0.05; Table 2). e bulk
nitrogen stable isotope values of spiders from both habitats
were also similar over the season, as well as when comparing
insects with spiders (Table 2, Fig. 4). For example, in August,
spiders from both habitats did not diﬀer in their bulk stable
isotope values (Fig. 4), but inter-individual variation within
‘lake spiders’ was larger compared to inter-individual varia-
tion within terrestrial spiders.
Stable carbon and hydrogen values of PUFA in
e δ13CALA values in emergent aquatic insects were lower
compared to those in terrestrial insects (Supporting infor-
mation), but the δ13CEPA values did not signiﬁcantly diﬀer
between habitats (Supporting information). e δ2HEPA val-
ues were lower in emergent aquatic insects than in terrestrial
insects (Δδ2H = 121.9‰; Supporting information). e
δ13CEPA values in ‘lake spiders’ were higher compared to those
of terrestrial spiders (Δδ13C = 3.8‰; KW test, H25 = 4.56,
p < 0.05; Supporting information), while the δ13C values of
other PUFA were not signiﬁcantly diﬀerent between habi-
tats (Supporting information). e δ2HALA values were higher
in ‘lake spiders’ than in terrestrial spiders (Δδ2H = 60.6‰;
Supporting information) and the δ2HEPA values not sig-
niﬁcantly diﬀerent between habitats, but ‘lake spiders’ were
isotopically slightly depleted compared to terrestrial spiders
To assess the origin of EPA in spiders, the δ13C and δ2H
values of ALA and EPA were compared between insects and
spiders in each habitat (Fig. 5). e δ13CALA values of ‘lake
spiders’ and emergent aquatic insects did not diﬀer signiﬁ-
cantly (KW test, H25 = 5.40, p = 0.07; Fig. 5A). e δ2HALA
values of ‘lake spiders’ were signiﬁcantly higher than those of
Trichoptera and similar to those of Chironomidae (KW test,
H25 = 7.30, p < 0.05; Fig. 5B). e δ13CEPA values of ‘lake
spiders’ were similar to those of emergent Chironomidae,
and signiﬁcantly higher than those of Trichoptera (KW
test, H25 = 9.12, p < 0.05; Fig. 5C), while the δ2HEPA val-
ues of ‘lake spiders’ were signiﬁcantly higher than those of
Trichoptera and similar to those of Chironomidae (KW
test, H25 =5.90, p < 0.05; Fig. 5D). Terrestrial spiders had
Table 1. Total lipids (TL mean ± SE; mg g−1) and fatty acids contents
(mean ± SE; mass percentage of total FA, %) of insects and spiders
from both aquatic and terrestrial habitats. Only the fatty acids
accounting for > 1% in at least one sample are shown. Different
letters indicate signiﬁcant difference between insect and spider (KW
tests and Conover–Iman multiple comparisons, signiﬁcant level p <
0.05), for each environment.
Lake Lunz Terrestrial
Insects Spiders Insects Spiders
ΣBFA 6.6 ± 0.6 8.1 ± 0.8 2.1 ± 0.3a4.3 ± 0.7b
14:0 2.5 ± 0.3a1.8 ± 0.4b1.9 ± 0.4a2.6 ± 0.4b
16:0 17.2 ± 0.5a13.4 ± 0.8b15.8 ± 0.7 15.1 ± 0.6
18:0 6.5 ± 0.2a12.3 ± 0.6b6.6 ± 0.6a8.1 ± 0.2b
ΣSFA 28.2 ± 0.7 30.0 ± 1.0 25.8 ± 1.1 27.2 ± 0.9
16:1n-7 9.8 ± 0.8a6.8 ± 0.8b9.3 ± 1.4 7.8 ± 0.7
18:1n-9 13.3 ± 1.1a16.2 ± 1.5b23.4 ± 1.3 24.3 ± 1.3
18:1n-7 3.8 ± 0.4a6.1 ± 0.8b0.8 ± 0.2a3.5 ± 0.6b
ΣMUFA 28.1 ± 1.1 30.0 ± 2.0 34.7 ± 1.9 36.7 ± 1.5
18:2n-6 (LIN) 10.2 ± 0.7a18.0 ± 1.1b25.8 ± 2.1 21.2 ± 1.3
18:3n-3 (ALA) 9.7 ± 1.1a3.9 ± 0.5b7.3 ± 1.5a5.1 ± 0.6b
18:4n-3 1.2 ± 0.2a0.2 ± 0.0b0.0 ± 0.0a0.2 ± 0.1b
20:4n-6 2.5 ± 0.2 3.8 ± 0.6 2.1 ± 0.4 2.3 ± 0.2
20:5n-3 (EPA) 14.9 ± 1.4 12.2 ± 1.3 2.2 ± 0.5a6.5 ± 0.8b
ΣPUFA 41.9 ± 1.2 39.6 ± 2.6 38.6 ± 1.8 35.9 ± 1.5
Σn-3 26.1 ± 1.7a17.1 ± 1.6b9.7 ± 1.4a11.9 ± 0.9b
n-3 HUFA 15.3 ± 1.4 13.0 ± 1.5 2.3 ± 0.5a6.6 ± 0.9 b
Σn-6 13.6 ± 0.8a23.0 ± 1.6b28.6 ± 1.9 24 ± 1.2
n-3/n-6 2.7a0.8b0.5 0.5
TL 182.7 ± 7.8 165.8 ± 13.1 155.49 ± 8.7a 246.6 ± 18.2b
SFA: saturated fatty acids; MUFA: monosaturated fatty acids; PUFA:
polyunsaturated fatty acids; BFA: Bacterial fatty acids (sum of 15:0,
iso15:0, anteiso15:0, iso16:0, 17:0, iso17:0 and anteiso17:0, 18:1n-
7, 18:1n-6); Σn-3: sum of n-3 fatty acids; Σn-6: sum of n-6 fatty
acids; n-3 HUFA: n-3 highly unsaturated fatty acids (sum of 20:3n-3,
20:4n-3, 20:5n-3, 22:3n-3, 22:5n-3, 22:6n-3); n-3/n-6: sum of n-3
fatty acids/sum of n-6 fatty acids.
signiﬁcantly lower δ13CALA & EPA, δ2HALA & EPA values than ter-
restrial Diptera (KW tests, H25 = 7.30, 6.20 and 4.2, p <
0.05, respectively; Fig. 5).
Emerging aquatic insects were richer in n-3 LC-PUFA, espe-
cially EPA, than terrestrial insects that contained more of
the n-6 PUFA LIN, indicating the higher nutritional value
(i.e. n-3 LC-PUFA) of emergent aquatic insects. Our data
revealed that riparian spiders used dietary sources from both
ecosystems and obtained their n-3 LC-PUFA via two distinct
pathways. ‘Lake spiders’ collected from Lake Lunz acquired
their EPA directly via the consumption of emergent aquatic
insects, most likely from Chironomidae, while terrestrial spi-
ders biosynthesized their EPA from dietary precursors, i.e.
ALA and/or stearidonic acid (18:4n-3). e combined use
of stable hydrogen and carbon isotopes of fatty acids empha-
sized the signiﬁcance of these two distinct pathways for spi-
ders in the acquisition of EPA, which was not possible to
reveal from fatty acid proﬁles or bulk stable isotopes.
PUFA of terrestrial and emergent aquatic insects as
e higher n-3 LC-PUFA contents, in particular EPA, in
emergent aquatic insects compared to terrestrial insects, are in
line with previous ﬁndings (Twining et al. 2019). ese diﬀer-
ences in the nutritional quality of insects appear to be a result
of ecosystem-based diﬀerences in the feeding history of insect
larvae (Guo et al. 2018, Scharnweber et al. 2019). Emergent
aquatic insects likely obtain their ALA and EPA from dietary
algae rich in PUFA, especially EPA (Hixson et al. 2015,
Twining et al. 2016a). In contrast, terrestrial insects mainly
derive their LIN and ALA from the base of the terrestrial food
web (Gladyshev et al. 2013, Twining et al. 2019), which pro-
vides mainly ALA, but only traces of EPA (Gladyshev et al.
2009, Hixson et al. 2015, Taipale et al. 2015). Because n-3
LC-PUFA are important for development, somatic growth
and reproduction of animals (Twining et al. 2016b, 2019),
their higher contents in emergent aquatic insects compared
to terrestrial insects results in diﬀerences in dietary quality
supplied to riparian insectivores. erefore, emergent aquatic
insects may generally be of higher nutritional value, in terms
of n-3 PUFA content, than terrestrial insects for insectivo-
rous riparian consumers, such as spiders.
Interestingly, the high ALA contents in both emerging
aquatic and terrestrial insects indicate ALA to be a poor diet
source biomarker in this study. is fatty acid can be syn-
thetized by both aquatic and terrestrial primary producers
(Twining et al. 2016a). ALA is a physiologically important
fatty acid (e.g. as cell membrane component), and serves as
a potential precursor for n-3 LC-PUFA (Brenna et al. 2009,
Hixson et al. 2015). e capacity to convert C18 to C20
PUFA (e.g. ALA to EPA) diﬀers among consumers (Bell
and Tocher 2009). In addition, this conversion is metaboli-
cally cost-intensive (Parrish 2009). Hence, riparian spiders
that have direct access to dietary EPA may have a metabolic
Figure 2. Total lipid contents (mg g dw−1) of (A) insects and (B) spiders from aquatic and terrestrial habitat. Signiﬁcant diﬀerences (KW
tests and Conover–Iman multiple comparisons) between origin of insects (i.e. aquatic versus terrestrial) are indicated with lowercase letters,
while diﬀerences between origin of spiders (i.e. aquatic versus terrestrial) are indicated with capital letters. e violin plots present the prob-
ability density distribution of the total lipids (therefore the ‘distribution’ of the data). In the boxplots, the median is represented by the thick
horizontal line; the box limits are the 25% (lower part) and the 75% (upper part) quartiles of the dataset; the vertical bars represent 1.5
times the interquartile range (IQR (i.e. the diﬀerence between the ﬁrst and third quartile) above the upper quartile and below the lower
quartile; and dots represent outliers which are therefore the observations that are above q0.75 + 1.5 × IQR or below q0.25 − 1.5 × IQR.
advantage over spiders that need to biosynthesize EPA from
Diet of spiders after random colonization of
Dietary sources of spiders were diﬃcult to discern based on
bulk carbon and nitrogen stable isotope values. Bulk δ13C
and δ15N values of ‘lake spiders’ and terrestrial spiders did not
diﬀer and thus no dietary distinction was possible between
spiders from aquatic and terrestrial habitats based on bulk
stable isotope data. e turnover of the stable isotopes in
spiders ranged between one and three weeks (Belivanov and
Hambäck 2015). erefore, it is most likely that ‘lake spi-
ders’ have recently dispersed to the aquatic environment, as
reﬂected by the stable isotope values closer from the terres-
trial habitat, where they have been feeding before dispersal.
In a recent study, bulk stable carbon isotope values of stream
riparian spiders did not match the ones of aquatic source,
suggesting that they did not obtain carbon from dietary
Figure 3. Contents in (A) LIN, (B) ALA, (C) EPA and (D) total PUFA of insects and spiders from aquatic and terrestrial habitats (mean ±
SE; mg FAME g−1). Signiﬁcant diﬀerences (KW tests and Conover–Iman multiple comparisons) between origin (i.e. aquatic versus terres-
trial) of insects are indicated with lowercase letters, while diﬀerences between origin (i.e. aquatic versus terrestrial) of spiders are indicated
with capital letters.
aquatic resources (Siebers et al. 2021). Chitin and proteins
represent up to 90% of the material found in spider cuticles
(Sewell 1955, Nentwig 2012), therefore their nitrogen iso-
topic composition should reﬂect the one of the habitat in
which they were hunting before moulting. However, these
limitations can be overcome by combining elemental analysis
(e.g. bulk δ13C) with molecular (e.g. fatty acids) biomarkers
(Perga et al. 2006, Jardine et al. 2015), and/or compound-
speciﬁc stable isotope analyses to trace dietary sources and
speciﬁc compounds within food chains (Kohl et al. 2015,
Taipale et al. 2015, Twining et al. 2020).
Spiders collected on the lake traps had lower total lipid,
PUFA, LIN and ALA contents than terrestrial spiders, but the
equal EPA contents suggest preferential retention of this fatty
acid. e particular retention of EPA (%) and n-3 HUFA by
terrestrial spiders compared to terrestrial insects highlights
the importance of long-chain PUFA in riparian consumers.
However, the PUFA composition of terrestrial spiders was
characterized by higher C18 PUFA contents, i.e. LIN and
ALA, compared to ‘lake spiders’. ese diﬀerences in PUFA
contents suggest that terrestrial spiders are part of a distinct
terrestrial food chain rich in LIN and ALA, but poor in EPA
(Budge and Parrish 1998, Twining et al. 2016a). e lower
lipid content observed in ‘lake spiders’ may have resulted
from a lower food availability on the ﬂoating lake traps, and/
or before their dispersal. In contrast, terrestrial spiders may
have been able to maintain higher lipid contents because of
a higher prey density in the riparian zone and the absence of
dispersal costs. It is also possible that ‘lake spiders’ may have
used C18 PUFA preferentially for gaining energy, but retained
EPA in their cell membranes, suggesting a hierarchy in the use
of biochemical components when facing changing resources.
Moreover, the high EPA content in emergent aquatic insects
supports the hypothesis that spiders that drifted to aquatic
habitats via ballooning likely have higher dietary access to this
n-3 LC-PUFA. Our results corroborate that EPA is an excel-
lent biomarker of aquatic-derived subsidies in riparian con-
sumers (Chari et al. 2020), and highlight the importance of
n-3 LC-PUFA for both riparian spiders . In particular, these
data reveal two diﬀerent trophic trajectories through which
spiders, or riparian insectivores in general, obtain PUFA.
Origin of PUFA in spiders
e CSIA data suggest that spiders acquired their EPA from dif-
ferent sources. ‘Lake spiders’ likely obtained their EPA directly
Table 2. Bulk carbon and nitrogen isotope composition (mean ± SD; δ13C and δ15N, ‰) of insects and spiders from aquatic (lake) and ter-
restrial habitats. Different letters indicate signiﬁcant differences (KW tests and Conover–Iman multiple comparisons) between habitats;
Δδ13C = difference in δ13C values (‰) between aquatic and terrestrial habitats.
δ13C (‰) Δδ13C
δ15N (‰) δ13C (‰) Δδ13C
Lake Lunz Terrestrial Lake Lunz Terrestrial Lake Lunz Terrestrial Lake Lunz Terrestrial
June −34.2 ± 1.6b−27.7 ± 2.0a6.5 3.0 ± 1.4 2.4 ± 2.3 −28.7 ± 1.9 −28.5 ± 1.3 0.2 2.0 ± 1.4 1.7 ± 0.8
July −31.0 ± 3.1b−27.9 ± 1.6a3.1 1.8 ± 1.8 2.0 ± 2.5 −29.3 ± 1.1b−27.3 ± 0.7a2.0 2.4 ± 1.0 1.5 ± 1.6
August −30.5 ± 1.8b−26.8 ± 1.9a3.7 1.3 ± 2.2 1.8 ± 3.0 −27.9 ± 2 −29.0 ± 0.8 1.1 1.7 ± 1.6 1.3 ± 1.4
September −30.2 ± 2.0b−27.5 ± 2.2a2.7 −0.6 ± 3.6b2.1 ± 2.2a−27.7 ± 1.4 −28.5 ± 0.6 0.8 1.4 ± 2.5 3.2 ± 0.9
Figure 4. Bulk stable isotopes carbon and nitrogen ratios (δ13C and δ15N, ‰) of insects and spiders from aquatic and terrestrial habitats in
August. e 95% normal ellipse from aquatic spiders is represented in blue, and from terrestrial spiders in dark green.
from their aquatic diet, while terrestrial spiders likely obtained
EPA from both diet and bioconversion (i.e. they converted
dietary precursors to EPA intrinsically). Based on δ13CEPA val-
ues, ‘lake spiders’ obtained their EPA from Chironomidae,
one of the most abundant emergent aquatic insect taxa found
in lakes (Armitage 1995, Martin-Creuzburg et al. 2017,
Selene et al. 2020, Mathieu-Resuge et al. 2021b), mak-
ing them a likely food source for ‘lake spiders’. Because the
δ13CEPA values of emergent aquatic insects overlapped with
those from terrestrial insects, it was not possible to trace the
origin of spider EPA using δ13CEPA values alone. However, the
δ2HEPA values of ‘lake spiders’ were close to those of emer-
gent aquatic insects, which were clearly distinct from those of
terrestrial insects (Δδ2H = 121.9‰). us, the δ2HEPA values
provided isotopic evidence for the aquatic origin of EPA in
‘lake spiders’, highlighting the potential of using dual com-
pound-speciﬁc stable isotope approaches in tracing the origin
of fatty acids in natural ecosystems.
Spiders from terrestrial habitats appear more likely to have
obtained their EPA partially from diet sources and partially
Figure 5. Stable carbon and hydrogen isotope values (δ13C and δ2H, ‰) of the n-3 polyunsaturated fatty acids ALA (A and B, respectively)
and EPA (C and D, respectively) of insects and spiders collected from aquatic (Lake Lunz) and terrestrial habitats in August. e letters
indicate signiﬁcant diﬀerences (KW tests and Conover–Iman multiple comparisons) between stable isotope values of n-3 PUFA between
taxa from aquatic and terrestrial habitats.
through bioconversion from dietary ALA. As it was the case
for ‘lake spiders’, δ13CEPA & ALA values of aquatic and terres-
trial insects overlapped, precluding us from drawing strong
inferences about fatty acid origin for terrestrial habitat spi-
ders based upon δ13CEPA & ALA. Terrestrial spiders had δ2HEPA
& ALA values between aquatic and terrestrial insects, suggesting
that EPA can come from both aquatic (i.e. direct assimila-
tion and accumulation) and terrestrial (i.e. bioconversion
from precursors) sources. However, in contrast to the δ2HEPA
values of ‘lake spiders’, the δ2HEPA values of terrestrial spi-
ders were substantially higher than those of emergent aquatic
insects (Δδ2H = 79‰), making it highly unlikely that they
only obtained their EPA directly from feeding on emergent
aquatic insects. As terrestrial insects contained very little
EPA, compared to terrestrial spiders and because the δ2HEPA
values of terrestrial spiders were lower than those of terres-
trial insects (Δδ2H = −43‰) it is also unlikely that spiders
obtained their EPA directly through diet from terrestrial
insects. Instead, terrestrial spiders had isotopically lighter val-
ues of δ13CEPA & ALA and δ2HE PA & ALA compared to terrestrial
insects, pointing towards innate PUFA conversion, as sug-
gested for other taxa (Bec et al. 2011, Burian et al. 2020,
Twining et al. 2020). Moreover, the δ2HE PA values were
lower than the δ2HALA (Δδ2H = 13‰), supporting the idea
of internal bioconversion from ALA to EPA. Indeed ALA
(%) was high in insects from both environments, but was
also speciﬁcally accumulated by terrestrial spiders. Because
isotopically lighter ALA should be faster converted to EPA
(Bec et al. 2011, Twining et al. 2020), the remaining pool
of ALA remains isotopically more enriched compare to the
EPA bioconverted. Lower dietary access to emerging aquatic
resources may select for increased bioconversion in spider
populations in terrestrial habitats, or increased bioconversion
may be the result of plasticity in response to the relative avail-
ability of EPA and its precursors. Future laboratory feeding
studies will be necessary to distinguish between these pos-
sibilities, by characterising fatty acid fractionations depend-
ing on dietary and/or metabolic acquisition. erefore, these
results imply that terrestrial spiders do not necessarily rely on
dietary EPA from aquatic sources, but given that bioconver-
sion is costly this suggests that riparian spiders preferentially
use dietary EPA whenever it is available.
Compound-speciﬁc stable isotope analysis on fatty acids
is a powerful tool to better understand trophic transfer of
dietary energy in and across ecosystems (Pilecky et al. 2021).
e simultaneous use of stable hydrogen and carbon isotopes
of fatty acids allows to better discriminate the origin of fatty
acids. Even if both hydrogen and carbon stable isotopes allow
discrimination among food sources at the compound-speciﬁc
level (Burian et al. 2020, Twining et al. 2020), the isotopic
discrimination between source and consumer is greater for
hydrogen than carbon stable isotopes (Pilecky et al. 2021). In
our study, the distinction between terrestrial and ‘lake spiders’
and their food sources were much more evident for δ2HEPA
(Δδ2H = 42.9‰ and 121.9‰, respectively) than for δ13CEPA
values (Δδ13C = 3.80‰ and 2.50‰, respectively). However,
this study lacks information on how H stable isotopes are
processed during lipid metabolism (i.e. elongation, desatu-
ration and retro-conversion) and, more precisely, what are
the speciﬁc values of hydrogen isotopic fractionations during
incorporation and internal biosynthesis of fatty acids. One
way to solve these uncertainties and fully characterise the
fractionation processes is by conducting laboratory feeding
experiments (e.g. knowing the food source fatty acids’ stable
isotope composition). Such controlled experiments can eﬀec-
tively characterise precisely the changes in isotopic fraction-
ations of fatty acids between dietary and metabolic pathways.
Our study therefore illustrates the added value that can be
gained from this new method in trophic ecology, allowing
us to explore how the fatty acid requirements of consumers
change with habitat and how bioconversion is triggered by
is ﬁeld study underlines the importance of aquatic
resources in terrestrial consumer diets (e.g. n-3 PUFA) and
thus the interconnectivity between aquatic and terrestrial
habitats. Our results highlight the intrinsic hierarchy of bio-
chemical compounds in spiders, with a preferential retention
of EPA, compared to LIN and ALA. e dual use of stable
hydrogen and carbon isotopes of fatty acids is a novel and
promising approach for trophic ecology to characterize vari-
ous diet sources and pathways to consumers. Results of this
ﬁeld study suggest that spiders acquire EPA from both ter-
restrial and aquatic resources using two pathways: 1) direct
dietary acquisition (trophic pathway), and/or, 2) bioconver-
sion (intrinsic pathway). Speciﬁcally, spiders collected from
Lake Lunz acquired their EPA directly from emergent aquatic
insects – most likely from Chironomidae – while terrestrial
spiders seemed to have obtained their EPA partially from
diet sources (i.e. from aquatic insects) and partially through
bioconversion from dietary precursors (e.g. ALA and/or stea-
ridonic acid). us, spiders appear to be capable of biosynthe-
sizing their n-3 LC-PUFA from dietary precursors depending
on dietary PUFA availability. Finally, the combined use of
stable hydrogen and carbon isotopes of fatty acids provides
better understanding of dietary energy transfer in aquatic and
terrestrial food webs than it would be possible from using
only fatty acid proﬁles or bulk stable isotope values, but it
will be necessary to better calibrate these methods with labo-
ratory feeding experiments to better understand fractionation
along dietary and metabolic pathways.
Acknowledgements – We thank H. H. Hager, K. Winter, S.-K.
Kämmer, E. Wassenaar, L. Perez and S. Damodaran for their ﬁeld
assistance, and lipid analyses.
Funding – is study was funded by the Austrian Science Fund
(FWF; I 3855-B25) and the German Research Foundation (DFG;
MA 5005/8-1) within the framework of the DACH collaboration
Conﬂict of interest – e authors declare no conﬂict of interest.
Margaux Mathieu-Resuge: Conceptualization (equal);
Formal analysis (equal); Investigation (lead); Methodology
(equal); Validation (equal); Visualization (lead); Writing
– original draft (lead). Matthias Pilecky: Formal analy-
sis (equal); Methodology (equal); Validation (equal);
Visualization (equal); Writing – review and editing (equal).
Cornelia W. Twining: Investigation (equal); Methodology
(equal); Validation (equal); Visualization (equal); Writing –
review and editing (equal). Dominik Martin-Creuzburg:
Conceptualization (equal); Funding acquisition (lead);
Investigation (equal); Methodology (equal); Project admin-
istration (lead); Supervision (equal); Validation (equal);
Visualization (equal); Writing – review and editing (equal).
Tarn-Preet Parmar: Methodology (equal); Validation (equal);
Visualization (equal); Writing – review and editing (equal).
Simon Vitecek: Methodology (equal); Validation (equal);
Visualization (equal); Writing – review and editing (equal).
Martin J. Kainz: Conceptualization (equal); Funding acqui-
sition (lead); Investigation (lead); Methodology (equal);
Project administration (lead); Resources (lead); Supervision
(lead); Validation (equal); Visualization (equal); Writing –
original draft (equal).
Data availability statement
Data are available from the Dryad Digital Repository:
Resuge et al. 2021a).
Armitage, P. D. 1995. Chironomidae as food. – In: Armitage P. D.
et al. (eds), e Chironomidae: biology and ecology of non-
biting midges. Chapman & Hall, pp. 423–435.
Barnes, A. D. et al. 2018. Energy ﬂux: the link between multi-
trophic biodiversity and ecosystem functioning. – Trends Ecol.
Evol. 33: 186–197.
Bec, A. et al. 2011. Assessing the reliability of fatty acid–speciﬁc
stable isotope analysis for trophic studies. – Methods Ecol. Evol.
Belivanov, Y. K. and Hambäck, P. A. 2015. e time scale of isotope
signals in spiders: molting the remains of a previous diet. –
Entomol. Exp. Appl. 156: 271–278.
Bell, M. V. and Tocher, D. R. 2009. Biosynthesis of polyunsatu-
rated fatty acids in aquatic ecosystems: general pathways and
new directions. – In: Lipids in aquatic ecosystems. Springer, pp.
Berger-Tal, R. et al. 2016. Good reasons to leave home: proximate
dispersal cues in a social spider. – J. Anim. Ecol. 85:
Bonte, D. 2013. Cost–beneﬁt balance of dispersal and the evolu-
tion of conditional dispersal strategies in spiders. – In: Spider
ecophysiology. Springer, pp. 67–78.
Brenna, J. T. et al. 2009. α-Linolenic acid supplementation and con-
version to n-3 long-chain polyunsaturated fatty acids in humans.
– Prostaglandins Leukot. Essent. Fatty Acids 80: 85–91.
Budge, S. M. and Parrish, C. C. 1998. Lipid biogeochemistry of
plankton, settling matter and sediments in Trinity Bay, New-
foundland. II. Fatty acids. – Org. Geochem. 29: 1547–1559.
Burian, A. et al. 2020. e potential of fatty acid isotopes to trace
trophic transfer in aquatic food-webs. – Phil. Trans. R. Soc. B
Chari, L. D. et al. 2020. Dietary fatty acids of spiders reveal spatial
and temporal variations in aquatic–terrestrial linkages. – Food
Webs 24: e00152.
Cloern, J. E. et al. 2002. Stable carbon and nitrogen isotope com-
position of aquatic and terrestrial plants of the San Francisco
Bay estuarine system. – Limnol. Oceanogr. 47: 713–729.
Cook, H. W. and McMaster, C. R. 2002. Fatty acid desaturation
and chain elongation in eukaryotes. – In: New comprehensive
biochemistry. Elsevier, pp. 181–204.
Ebm, N. et al. 2021. Polyunsaturated fatty acids in ﬁsh tissues more
closely resemble algal than terrestrial diet sources. – Hydrobio-
logia 848: 371–383.
Fritz, K. A. et al. 2017. Subsidies of essential nutrients from aquatic
environments correlate with immune function in terrestrial
consumers. – Freshwater Sci. 36: 893–900.
Galloway, A. W. et al. 2015. A fatty acid based Bayesian approach
for inferring diet in aquatic consumers. – PLoS One 10:
Gladyshev, M. I. et al. 2009. Preliminary estimates of the export of
omega-3 highly unsaturated fatty acids (EPA+ DHA) from
aquatic to terrestrial ecosystems. – In: Lipids in aquatic ecosys-
tems. Springer, pp. 179–210.
Gladyshev, M. I. et al. 2013. Production of EPA and DHA in
aquatic ecosystems and their transfer to the land. – Prostaglan-
dins Other Lipid Mediat. 107: 117–126.
Guo, F. et al. 2016. High-quality algae attached to leaf litter boost
invertebrate shredder growth. – Freshwater Sci. 35: 1213–1221.
Guo, F. et al. 2018. Feeding strategies for the acquisition of high-
quality food sources in stream macroinvertebrates: collecting,
integrating and mixed feeding. – Limnol. Oceanogr. 65:
Hintze, J. L. and Nelson, R. D. 1998. Violin plots: a box plot-
density trace synergism. – Am. Stat. 52: 181–184.
Hixson, S. M. et al. 2015. Production, distribution and abundance
of long-chain omega-3 polyunsaturated fatty acids: a funda-
mental dichotomy between freshwater and terrestrial ecosys-
tems. – Environ. Rev. 23: 414–424.
Iverson, S. J. 2009. Tracing aquatic food webs using fatty acids:
from qualitative indicators to quantitative determination. – In:
Lipids in aquatic ecosystems. Springer, pp. 281–308.
Iverson, S. J. et al. 2004. Quantitative fatty acid signature analysis:
a new method of estimating predator diets. – Ecol. Monogr.
Jardine, T. D. et al. 2015. Reconciling the role of organic matter
pathways in aquatic food webs by measuring multiple tracers
in individuals. – Ecology 96: 3257–3269.
Kohl, L. et al. 2015. Distinct fungal and bacterial δ13C signatures
as potential drivers of increasing δ13C of soil organic matter
with depth. – Biogeochemistry 124: 13–26.
Kühmayer, T. et al. 2020. Preferential retention of algal carbon in
benthic invertebrates: stable isotope and fatty acid evidence
from an outdoor ﬂume experiment. – Freshwater Biol. 65:
Martin-Creuzburg, D. et al. 2017. Cross-ecosystem ﬂuxes: export
of polyunsaturated fatty acids from aquatic to terrestrial ecosys-
tems via emerging insects. – Sci. Total Environ. 577: 174–182.
Mathieu-Resuge, M. et al. 2021a. Data from: Dietary availability
determines metabolic conversion of long-chain polyunsaturated
fatty acids in spiders: a dual compound-speciﬁc stable isotope
approach. – Dryad Digital Repository, <http://dx.doi.
Mathieu-Resuge, M. et al. 2021b. Taxonomic composition and lake
bathymetry inﬂuence fatty acid export via lake emerging insects.
– Freshwater Biol. in press. https://doi.org/10.1111/fwb.13819
Mestre, L. and Bonte, D. 2012. Food stress during juvenile and
maternal development shapes natal and breeding dispersal in a
spider. – Behav. Ecol. 23: 759–764.
Nentwig, W. 2012. Ecophysiology of spiders. – Springer Science &
Nyﬀeler, M. 1999. Prey selection of spiders in the ﬁeld. – J. Arach-
nol. 27: 317–324.
Paetzold, A. et al. 2005. Aquatic terrestrial linkages along a braided-
river: riparian arthropods feeding on aquatic insects. – Ecosys-
tems 8: 748–759.
Parrish, C. C. 2009. Essential fatty acids in aquatic food webs. – In:
Lipids in aquatic ecosystems. Springer, pp. 309–326.
Perga, M.-E. et al. 2006. Carbon pathways to zooplankton: insights
from the combined use of stable isotope and fatty acid biomark-
ers. – Freshwater Biol. 51: 2041–2051.
Pilecky, M. et al. 2021. Compound-speciﬁc stable hydrogen (δ2H)
isotope analyses of fatty acids: a new method and perspectives
for trophic and movement ecology. – Rapid Commun. Mass
Spectrom. 35: e9135.
Scharnweber, K. et al. 2019. e emergence of fatty acids – aquatic
insects as vectors along a productivity gradient. – Freshwater
Biol. 65: 565–578.
Selene, P. et al. 2020. Changes in midge assemblages (Diptera Chi-
ronomidae) in an alpine lake from the Italian western Alps: the
role and importance of ﬁsh introduction. – Hydrobiologia 847:
Sewell, M. T. 1955. e histology and histochemistry of the cuticle
of a spider, Tegenaria domestica (L.). – Ann. Entomol. Soc. Am.
Siebers, A. R. et al. 2021. Riparian hunting spiders do not rely on
aquatic subsidies from intermittent alpine streams. – Aquat. Sci.
Taipale, S. J. et al. 2015. Inferring phytoplankton, terrestrial plant
and bacteria bulk δ13C values from compound speciﬁc analyses
of lipids and fatty acids. – PLoS One 10: e0133974.
Torres-Ruiz, M. et al. 2007. Trophic relations in a stream food web:
importance of fatty acids for macroinvertebrate consumers. – J.
North Am. Benthol. Soc. 26: 509–522.
Twining, C. W. et al. 2016a. Highly unsaturated fatty acids in
nature: what we know and what we need to learn. – Oikos 125:
Twining, C. W. et al. 2016b. Omega-3 long-chain polyunsatu-
rated fatty acids support aerial insectivore performance more
than food quantity. – Proc. Natl Acad. Sci. USA 113:
Twining, C. W. et al. 2018. Conversion eﬃciency of α-linolenic
acid to omega-3 highly unsaturated fatty acids in aerial insecti-
vore chicks. – J. Exp. Biol. 221(3): jeb165373.
Twining, C. W. et al. 2019. Aquatic and terrestrial resources are not
nutritionally reciprocal for consumers. – Funct. Ecol. 33:
Twining, C. W. et al. 2020. Stable isotopes of fatty acids: current
and future perspectives for advancing trophic ecology. – Phil.
Trans. R. Soc. B 375: 20190641.
Vander Zanden, M. J. et al. 1999. Patterns of food chain length in
lakes: a stable isotope study. – Am. Nat. 154: 406–416.