Armed reproductives: Evolution of the frontal gland in imagoes of
, Thomas Bourguignon
, David Sillam-Dussès
, Robert Hanus
Research Team of Infochemicals, Institute of Organic Chemistry and Biochemistry, Flemingovo nám. 2, 166 10 Prague, Czech Republic
Department of Zoology, Faculty of Science, Charles University in Prague, Vini
cná 7, 128 44 Prague, Czech Republic
Evolutionary Biology and Ecology, Université Libre de Bruxelles, Belgium
Department of Biological Science, National University of Singapore, 117543 Singapore, Singapore
Institut de Recherche pour le Développement, Unité Mixte de Recherche 211 Biogéochimie et Ecologie des Milieux Continentaux,
Interactions Biologiques dans les Sols, 32 avenue Henri Varagnat, 93143 Bondy, France
Laboratoire d’Ethologie Expérimentale et Comparée, Equipe d’accueil 4443, Université Paris 13, Sorbonne Paris Cité, 99 avenue Jean-Baptiste Clément,
93430 Villetaneuse, France
Faculty of Forestry and Wood Sciences, Czech University of Life Sciences, Kamýcká 129, 165 21 Prague, Czech Republic
Received 6 February 2013
Accepted 2 April 2013
The frontal gland of termites is a structure without any equivalent among other animals. Although this
gland is well known in soldiers, it received almost no attention in other castes. Recently, we described it
in imagoes of Rhinotermitidae and Serritermitidae. In order to provide a complete picture of the evo-
lution of this gland in termite imagoes, we studied it in additional 34 species of Termitidae, representing
7 of the 8 subfamilies. The frontal gland of these species is formed by class 1 secretory cells only, and
occurs in two basic shapes: epithelial with reservoir in Foraminitermitinae and Macrotermitinae, and
epithelial without reservoir in all other subfamilies. The size variability of the gland is high, not only
among Termitidae subfamilies, but also within subfamilies. Our data suggest that the ancestral form of
the frontal gland is epithelial with reservoir, as found in Rhinotermitidae, Serritermitidae, and basal
Termitidae. The reduction of the reservoir occurred at least two times and the gland was lost two times
independently: in Protermes sp. and in Microtermes toumodiensis (both Macrotermitinae).
Ó2013 Elsevier Ltd. All rights reserved.
Termites are ecologically dominant in a variety of tropical and
subtropical habitats, where they are extremely abundant, repre-
senting a rich food source for a wide variety of predators (Deligne
et al., 1981). The established colonies are well defended by pas-
sive adaptations such as a hidden wayof life, or the nest and gallery
architecture (Perna et al., 2008), and by active adaptations con-
sisting in the presence of a rich set of morphological and behav-
ioural defensive features in a specialized caste of defenders, the
soldiers (Deligne et al., 1981;
Sobotník et al., 2010a). Consequently,
individuals residing inside the nest are well protected, but they are
subjected to predation as soon as they leave the nest. Predation is
known to be important, both on workers during foraging activities
and on alate imagoes during dispersal ﬂights and early phases of
colony establishment (Dial and Vaughan, 1987;Lepage, 1991;Korb
and Salewski, 2000).
During the combat, termites generally try to bite the opponent,
and often combine the mandibular action with the release of a
defensive secretion from specialized glands such as the labial
glands (Moore, 1968;Plasman et al., 1999) or the labral glands
Sobotník et al., 2010b). Several taxa also
developed their own innovative defensive strategies as exempliﬁed
by the dehiscent glands in Ruptitermes (Costa-Leonardo, 2004)or
the dorsal apparatus in Neocapritermes taracua (
Sobotník et al.,
2012). However, the most prominent defensive organ of termites
is undoubtedly the frontal gland, occurring in a clade comprising
the families Rhinotermitidae, Stylotermitidae, Serritermitidae and
Termitidae (Emerson, 1971).
The frontal gland is a structure without any equivalent among
other animals (Noirot, 1969). In the soldier caste, it is an unpaired
organ, epithelial gland with reservoir (according to classiﬁcation
*Corresponding author. Faculty of Forestry and Wood Sciences, Czech University
of Life Sciences, Kamýcká 129, 165 21 Prague, Czech Republic. Tel.: þ420 23438
3776; fax: þ420 23438 3739.
E-mail addresses: sobotnik@ﬂd.czu.cz,firstname.lastname@example.org (J.
Contents lists available at SciVerse ScienceDirect
Arthropod Structure & Development
journal homepage: www.elsevier.com/locate/asd
1467-8039/$ esee front matter Ó2013 Elsevier Ltd. All rights reserved.
Arthropod Structure & Development 42 (2013) 339e348
Billen, 2011), opening at the top of the head through the fontanelle
(Noirot, 1969;Prestwich and Collins, 1982;Quennedey, 1984;
Sobotník et al., 2010d). It can either be restricted only to the head,
as in many Termitidae (Noirot, 1969), or reach deep into the
abdomen as in most Rhinotermitidae and Serritermitidae (
et al., 2004,2010a). The frontal gland structure and secretion is well
known in soldiers, but received almost no attention in othercastes,
even though it has been reported in presoldiers of some species
(Prestwich, 1984a;Bordereau et al.,1997;
Sobotník et al., 2004), in
some workers (
Sobotník et al., 2010c) and in all imagoes of Rhi-
notermitidae and Serritermitidae (
Sobotník et al., 2010d). Contrary
to the soldiers, the frontal gland is never epithelial with reservoir in
workers, but occurs as an epithelial thickening only (
Sobotník et al.,
2010c). In imagoes, both forms are observed but up to now, only
superﬁcial descriptions have been published of Termitidae imagoes
(Holmgren, 1909;Bugnion, 1913;Noirot, 1969).
The frontal gland of soldiers produces manyclasses of chemicals
(hydrocarbons, alcohols, aldehydes, ketons, macrolactones, mono-,
sesqui- and diterpenes, aromatic compounds, nitro-compounds,
ceramids and others), which act as contact poisons, irritants, re-
pellents, immobilizing agents, antihealants, or alarm pheromones
(Howse, 1984;Prestwich, 1984b;
Sobotník et al., 2010a). The
composition of the frontal gland secretion of imagoes is known
only in Prorhinotermes spp. (Rhinotermitidae), where it contains
toxic and irritant chemical compounds, like the soldiers’secretion
(Piskorski et al., 2007,2009;
Sobotník et al., 2010a).
To complete the story of the evolution of the frontal gland in
termite imagoes, we investigated the anatomy of this gland in a
representative set of Termitidae species. Termitidae is the most
abundant family of termites, with the largest ecological and
behavioural diversity, and contains over 85% of genera and 70% of
species (Engel et al., 2009). In the present work, we studied the
frontal gland in imagoes of 34 species, representing seven of the
eight currently recognized Termitidae subfamilies (Sphaero-
termitinae, Macrotermitinae, Foraminitermitinae, Apicotermitinae,
Termitinae, Syntermitinae, and Nasutitermitinae sensu Engel et al.,
2. Materials and methods
Specimens used for this study are listed in Table 1.They were
either preserved in 80% alcohol or ﬁxed for electron microscopy (for
Sobotník et al., 2010d). Semithin sections (0.5 or 1
thick) were cut with an Ultracut Reichert-Jung and stained with
Azure II or toluidin blue solutions. Sections were studied with a Carl
Zeiss Amplival microscope combined with a Canon EOS 500 D
digital camera. All ﬁgures represent sagittal or parasagittal sections
of heads of imagoes, with mouth parts to the left.
List of the species studied and their origin.
Conservation method Place Date Number of specimens
Ancistrotermes cavithorax (Sjoestedt, 1899) Alcohol Lamto, Côte d’Ivoire 21.3.1968 1_,1\
Macrotermes bellicosus (Smeathman, 1781) Dubosq-Brasil, alcohol Abidjan, Côte d’Ivoire 3.1970 1_,1\
Microtermes toumodiensis Grassé, 1937 Alcohol Lamto, Côte d’Ivoire 2.2.1968 2_,2\
Odontotermes pauperans (Silvestri, 1912) Alcohol Lamto, Côte d’Ivoire 1.4.1968 2_,2\
Protermes sp. Holmgren, 1910 Alcohol Ouesso, Congo 27.12.1956 2_
Pseudacanthotermes militaris (Hagen, 1858) Alcohol Lamto, Côte d’Ivoire 1968 2_,2\
Pseudacanthotermes spiniger (Sjoestedt, 1900) Dubosq-Brasil, alcohol Gabon 5.1989 1_,1\
Sphaerotermes sphaerothorax (Sjoestedt, 1911) Alcohol Ebogo II (near Mbalmayo),
Foraminitermes coatoni Krishna, 1963 Alcohol Leopoldville (Kalina), Congo 6.4.1948 2_,1\
Anoplotermes-group sp. MFAA Nouragues, French Guiana 31.3.2006 1\
Anoplotermes janus Alcohol Nouragues, French Guiana 24.3.2006 1_
Aparatermes sp. Fontes, 1986 Glutaraldehyde Nouragues, French Guiana 21.4.2008 2_,2\
Astalotermes quietus (Silvestri, 1914) Alcohol Lamto, Côte d’Ivoire 21.3.1968 1_,1\
Longustitermes manni (Snyder, 1922) Alcohol Nouragues, French Guiana 28.3.2006 2_,1\
Ruptitermes sp. Mathews, 1977 Alcohol Camp Patawa, French Guiana 10.2.2007 1_,1\
Amitermes beaumonti Banks, 1918 Alcohol Gamboa, Panama 17.5.1991 2_,2\
Crepititermes verruculosus Emerson, 1925 Alcohol Nouragues, French Guiana 31.3.2006 1_,2
Dentispicotermes brevicarinatus (Emerson, 1950) Alcohol Nouragues, French Guiana 17.4.2008 2\
Ephelotermes argutus (Hill, 1929) Alcohol Wipim, Papua New Guinea 30.3.1989 1_,1\
Neocapritermes araguaia Krishna and Araujo, 1968 Glutaraldehyde Petit Saut, French Guiana 8.1.2010 2_,2\
Pericapritermes papuanus Bourguignon and Roisin, 2008 Alcohol Wipim, Papua New Guinea 31.3.1989 1_,1\
Protocapritermes odontomachus (Desneux, 1905) Alcohol Manus Island, Papua New Guinea 6.6.1984 2_,2\
Termes fatalis Linnaeus, 1758 Glutaraldehyde Petit Saut, French Guiana 9.1.2010 2_,2\
Termes sp. BGlutaraldehyde Petit Saut, French Guiana 2.2.2008 2_,2\
Silvestritermes holmgreni (Snyder, 1926) Alcohol Nouragues, French Guiana 29.3.2006 1_,1\
Embiratermes neotenicus (Holmgren, 1906) Glutaraldehyde Petit Saut, French Guiana 8.1.2010 2_,2\
Rhynchotermes perarmatus (Snyder, 1925) Alcohol Gamboa, Panama 26.4.1991 2_
Syntermes molestus (Burmeister, 1839) Alcohol Nouragues, French Guiana 24.3.2006 1\
Diwaitermes kanehirae (Oshima, 1914) Alcohol Topo, Irian Jaya, Indonesia 28.11.1995 1_,1\
Grallatotermes grallator (Desneux, 1905) Alcohol Yapsiei, Papua New Guinea 11.3.1994 1\
Hospitalitermes papuanus Ahmad, 1947 Alcohol Yapsiei, Papua New Guinea 10.3.1994 2_,1\
Nasutitermes princeps (Desneux, 1905) Alcohol Kaimana, Irian Jaya 21.11.1995 1_,1\
Subulitermes sp. Holmgren, 1910 Glutaraldehyde Petit Saut, French Guiana 8.1.2010 2_,1\
K. Kutalová et al. / Arthropod Structure & Development 42 (2013) 339e348340
A few specimens, namely one male and one female of Apara-
termes sp. and one female of Termes sp. B were ﬁxed for ultra-
structure with glutaraldehyde; one female of Anoplotermes-group
sp. M and one male of Pericapritermes papuanus from 80% ethanol,
were also studied using TEM (see Table 1, see
Sobotník et al., 2010c
for preparation protocol).
Height (H) and length (L) of the frontal gland were measured on
sagittal sections with the ImageJ software. The width (W)was
estimated from the number of parasagittal sections on which the
frontal gland was present. These parameters were used for frontal
gland volume calculation. The frontal gland shape was estimated as
an ellipsoid if the gland was epithelial with reservoir, and the
(H/2) (L/2) Wwas then used. If the
frontal gland was present in the form of an epithelial gland without
reservoir, it was considered conical in shape and the volume was
estimated by the equation of V¼1/3
(W/2) (L/2) H. The
calculation of relative frontal gland size (Fg) was performed by
comparing frontal gland volume and head length (X¼distance
between clypeo-frons boundary and posterior margin of head) of
each specimen using the formula Fg ¼V/X
. The values were then
converted to multiples of the smallest value.
To investigate the evolution of the frontal gland, we recon-
structed its ancestral state with the software Mesquite (Maddison
and Maddison, 2010). Three possible states were considered: fron-
tal gland with reservoir, frontal gland without reservoir and frontal
gland missing. We used the likelihood method analysis using the
Mk1 model, as well as parsimony analysis. Several phylogenetic
trees were used to compute these analyses: the tree provided by
Inward et al. (2007) with branch length set to one, a tree we
computed using COII, 12S and 28S GeneBank deposited sequences
and reconstructed with the maximum likelihood method, and a set
of 101 trees computed using COII, 12S and 28S GeneBank deposited
sequences and Bayesian analysis. Sequences were aligned with the
Mega_5.0 package (Tamura et al., 2011), using the ClustalW algo-
rithm and then concatenated with Mesquite (Maddison and
Maddison, 2010). Sequence alignments were visually checked and
ambiguous alignments were ignored to produce a ﬁnal data matrix
with 1602 bp. The maximum likelihood method was implemented
in DNAml in the Phylip-3.69 suite of programs for Mac OS X to
generate a phylogeny, under default setting (Felsenstein, 2005). The
Bayesian trees were computed using Markov chain Monte Carlo
algorithm, following a GTR model with gamma-distributed rate
variation across sites and implemented in Mr Bayes version 3.1.1
(Ronquist and Huelsenbeck, 2003). The run length was one million
generations, from which we sampled one tree every 5000 genera-
tions started from the 500,000th generation. All trees were con-
structed using 267 species and were pruned to leave only the species
of interest. Finally, we investigated whether there is a trend towards
a sexual dimorphism in the frontal gland size of termites, using a
Wilcoxon test computed on the data available for 31 species.
Because our Wilcoxon test did not correct the phylogenetic auto-
correlation between species, we treated these results as explorative.
3.1. Common features of imaginal frontal glands
The frontal gland is always located posteriorly to the brain,
behind the posterior attachment of the fronto-labral muscle. Two
basic types of frontal gland were observed, either epithelial with or
without the reservoir (the gland being present only as an epidermal
thickening of lenticular or conical shape); glands of the latter type
are generally smaller (Table 2). Because the shape of both types of
frontal gland differs, we used two different equations to estimate
the relative size of the frontal gland (Table 2).
In frontal glands with reservoir, the cuticle overlaying the
secretory epithelium is always highly modiﬁed. The gland opening
(fontanelle) occurs only in this type of gland as a simple narrow
pore located above the posterior part of the brain. A pair of
tentorial-fontanellar muscles is always stretched between the
gland epithelium (below the fontanelle or at the base of the
secretory cells of the largest height) and the tentorium. No sex
differences were observed in studied samples concerning the
structure of the gland. When the frontal gland is developed as
epithelial without reservoir, the overlaying cuticle shows only
slight modiﬁcations, namely multiple perforations of the epicuticle,
enlargement of pore canals, and reduction of endocuticle thickness
due to lack of the most basal layers developed elsewhere (Fig. 1A).
3.2. Systematic survey
The frontal gland with reservoir is present in Foramini-
termitinae and in the majority of Macrotermitinae, except for Pro-
termes sp. and Microtermes toumodiensis (both Macrotermitinae), in
which the frontal gland is absent, even though the tentorial-
fontanellar muscles are preserved. The relative size of the gland is
larger in Pseudacanthotermes spp. (Macrotermitinae) and Fora-
minitermes coatoni (Foraminitermitinae) compared to other Mac-
rotermitinae: Ancistrotermes cavithorax,Macrotermes bellicosus and
Odontotermes pauperans (see Table 2). The fontanelle is placed on a
more or less marked protuberance (see Fig. 2). The glandular
epithelium is made of large cuboidal or collumnar cells.
In the only representative of Sphaerotermitinae, Sphaerotermes
sphaerothorax, imagoes possess the frontal gland as a very small
epidermal thickening, made only of a few secretory cells.
All studied species of Apicotermitinae present the frontal gland
without reservoir, which is always made of narrow columnar cells.
There is a marked protuberance in Aparatermes sp. where the gland
is located (see Fig. 2F). The gland size varies considerably among
studied species. It is quite large in Aparatermes sp. and Astalotermes
quietus, while fairly small in e.g. Ruptitermes sp.
The frontal gland occurs as an epithelial gland without reservoir
in all representatives of Termitinae (Fig. 3). Overall frontal gland
size is rather small, the largest one occurring in Dentispicotermes
brevicarinatus, while the two Termes species and Ephelotermes have
the smallest gland (for details see Table 2). The glandular cuticle is
highly modiﬁed in Amitermes beaumonti, and the basal layers of
endocuticle extend into the head at the place of the tentorial-
fontanellar muscles attachment (Fig. 3A); the frontal gland itself
is quite small and made only of a few cuboidal cells.
All studied species of Syntermitinae present the frontal gland
without reservoir. The secretorycells of the frontal gland are always
columnar. The largest gland is present in Rhynchotermes per-
armatus, in which the gland is placed on protruded cuticle. The
glandular cuticle is considerably thinner in Embiratermes
In Nasutitermitinae, the frontal gland also lacks the reservoir.
The frontal gland is made only of a few secretory cells in Grallato-
termes grallator, which has the smallest gland of all studied Nasu-
titermitinae, while the largest gland occurs in Hospitalitermes
papuanus (Fig. 4F).
3.3. Ultrastructural features of imaginal frontal glands
The frontal gland of Aparatermes sp. imagoes is very similar to
the one of workers (compare Fig. 1 with
Sobotník et al., 2010c) and
only differs in the following features: the gland is approximately
twice as large, reveals smaller amounts of thread-like biocrystals
which more often turn into myelin ﬁgures, and shows higher
secretory activity through small lucent vesicles abundantly
K. Kutalová et al. / Arthropod Structure & Development 42 (2013) 339e348 341
released from the secretory cells (Fig. 1B). The frontal gland of
Termes sp. B female is clearly a secretory organ, as evidenced by
modiﬁcations of the cuticle overlaying the gland (enlargement of
pore canals, epicuticle perforation), occurrence of microvilli at the
cell apices, presence of numerous lucent secretory vesicles. Such
vesicles, about 1
m in diameter, are predominantly found at the
secretory cell bases and are often surrounded by mitochondria;
smaller vesicles, about 100 nm in diameter, predominantly occur at
the cell apex. Cell ultrastructure was destroyed in samples of Ano-
plotermes-group sp. M and P. papuanus (both stored in ethanol), but
the cuticle reveals the same modiﬁcations as in Termes sp. B and
Aparatermes sp., highlighting a secretory function of the frontal
gland of imagoes.
3.4. Evolution of the frontal gland
All our analyses clearly point out that the ancestral form of the
frontal gland is epithelial with reservoir,not only in Rhinotermitidae
and Serritermitidae, but also in the common ancestor of Termitidae
(Fig. 5). Ancestral state reconstruction on the tree provided by
Inward et al. (2007) suggests that the reduction of the reservoir
occurred at least two times, once in Psammotermes and once in the
ancestor of a group including all Termitidae but Macrotermitinae,
Sphaerotermitinae and Foraminitermitinae. An additional ambigu-
ity occurs on the Foraminitermes eSphaerotermes branch, on which
the reservoir was either lost once in Sphaerotermes or reacquired
once in Foraminitermes. The ancestral state reconstruction supports
that the frontal gland was completely lost two times independently,
in the genera Protermes and Microtermes. On the 101 Bayesian trees,
one suggests a scenario similar to Inward’s tree, while the other 100
trees all suggest that the frontal gland with reservoir was reduced
into an epithelial gland without reservoir three times independently
and that the frontal gland was completely lost two times indepen-
dently in the above mentioned lineages. Finally, the maximum
likelihood tree suggests two independent losses of the reservoir,
Sphaerotermes being retrieved as the sister group of all Termitidae
but Macrotermitinae and Foraminitermitinae. The treealso suggests
two independent losses of the frontal gland (Figs. 5 and S1). Even
though we did not apply correction for the phylogenetic autocor-
relation, the Wilcoxon test did not point out any signiﬁcant differ-
ence between sexes (n¼31, V¼191, p¼0.272).
The frontal gland is a defensive organ of prime importance in
soldiers of advanced termites, occurring also in presoldiers, ima-
goes and workers (Quennedey, 1984;
Sobotník et al., 2010a,c,d).
While its structure and function is well-understood in soldiers, only
scarce data on other castes have been gathered. The original form of
the frontal gland was epithelial with reservoir, as evidenced by the
reconstruction of its ancestral state in the common ancestor of all
studied species. The frontal gland reservoir was later subjected to
Dimensions and relative size of the frontal gland in the studied species.
Type of frontal gland Length of head
Recounted to the
Ancistrotermes cavithorax Epithelial with reservoir 1.005e1.045 47.3899e51.6070 50.8406 111.20e136.13
Macrotermes bellicosus Epithelial with reservoir 1.832e1.865 198.5736e228.9475 30.6115e37.2357 81.97e99.70
Microtermes toumodiensis Absent 0.680e0.729 0.0000 0.0000 0.00
Odontotermes pauperans Epithelial with reservoir 1.587e1.912 61.7651e122.9417 8.8365e29.7354 23.66e79.62
Protermes sp. Absent 0.736e0.761 0.0000 0.0000 0.00
Pseudacanthotermes militaris Epithelial with reservoir 1.158e1.304 130.2193e212.0038 78.6036e105.4789 210.47e282.43
Pseudacanthotermes spiniger Epithelial with reservoir 1.229e1.279 859.9981e1021.9833 463.2782e488.4636 1240.49e1307.93
Sphaerotermes sphaerothorax Epithelial without reservoir 0.660e0.710 0.4400e0.5750 1.3108e2.0001 3.5098e5.3555
Foraminitermes coatoni Epithelial with reservoir 0.548e0.579 55.5591e66.0347 286.9756e396.9096 768.42e1062.78
Anoplotermes-group sp. M Epithelial without reservoir 0.628 2.2663 9.1506 24.50
Anoplotermes janus Epithelial without reservoir 0.444 0.8529 9.7440 26.09
Aparatermes sp. Epithelial without reservoir 0.667e0.747 5.92058e15.7461 15.0949e52.3538 40.42e140.18
Astalotermes quietus Epithelial without reservoir 0.557e0.56 2.7674e4.1879 16.0142e23.8468 42.88e63.85
Longustitermes manni Epithelial without reservoir 0.384e0.403 0.1306e0.1461 1.9958e2.5799 5.34e6.91
Ruptitermes sp. Epithelial without reservoir 1.242e1.349 1.1968e1.5024 0.4875e0.7842 1.31e2.10
Amitermes beaumonti Epithelial without reservoir 0.707e0.762 0.5383e1.2818 1.2167e2.9784 3.26e7.98
Crepititermes verruculosus Epithelial without reservoir 0.501e0.509 0.3796e0.9928 3.019e7.7085 8.08e20.64
Dentispicotermes brevicarinatus Epithelial without reservoir 0.825e0.829 9.0581e10.6631 15.8992e19.0245 42.57e50.94
Ephelotermes argutus Epithelial without reservoir 0.593e0.610 0.1524e0.2285 0.7308e1.0068 1.96e2.70
Neocapritermes araguaia Epithelial without reservoir 0.717e0.814 1.2466e1.8249 3.1396e3.4555 6.09e9.25
Pericapritermes papuanus Epithelial without reservoir 0.898e0.904 5.1010e7.1214 7.0565e9.6396 18.89e25.81
Protocapritermes odontomachus Epithelial without reservoir 0.711e0.736 0.9191e1.3823 2.3529e3.4670 6.30e9.28
Termes fatalis Epithelial without reservoir 0.629e0.657 0.1429e0.2810 0.5744e1.1079 1.54e2.97
Termes sp. BEpithelial without reservoir 0.654e0.681 0.3883e0.4697 1.2413e1.5203 3.32e4.07
Silvestritermes holmgreni Epithelial without reservoir 0.818e0.842 0.9472e1.3164 1.7305e2.2053 4.63e5.90
Embiratermes neotenicus Epithelial without reservoir 1.149e1.265 2.6263e4.5867 1.4252e3.0237 3.82e8.10
Rhynchotermes perarmatus Epithelial without reservoir 1.025e1.038 6.2135e7.5722 5.7699e6.7707 15.45e18.13
Syntermes molestus Epithelial without reservoir 1.908 16.6710 2.4020 6.43
Diwaitermes kanehirae Epithelial without reservoir 0.769e0.790 0.8870e0.8887 1.8025e1.9505 4.83e5.22
Grallatotermes grallator Epithelial without reservoir 1.465 1.1743 0.3735 1.00
Hospitalitermes papuanus Epithelial without reservoir 1.252e1.367 25.4794e30.4495 11.0908e13.1195 29.70e35.13
Nasutitermes princeps Epithelial without reservoir 1.359e1.489 6.9405e15.7329 2.1024e6.2683 5.63e16.78
Subulitermes sp. Epithelial without reservoir 0.496e0.506 0.0578e0.1203 0.4737e0.9284 1.27e2.49
K. Kutalová et al. / Arthropod Structure & Development 42 (2013) 339e348342
Fig. 1. The structure of the frontal gland in Aparatermes sp. male. A: Structure of cuticle overlaying the frontal gland. B: Detail of cytoplasm in the mid part of secretory cell. Scale
bars represent 2
m. Abbreviations: m, mitochondria; mv, microvilli; s, secretion; sv, secretory vesicles; asterisks mark enlarged pore canals in the cuticle.
several independent reductions in (i) Psammotermes (Rhinotermi-
Sobotník et al., 2010d), (ii) in a clade comprising Apico-
termitinae, Termitinae, Syntermitinae, Nasutitermitinae and
Cubitermitinae (all Termitidae) and (iii) possibly in Sphaerotermes
(Termitidae). Future phylogenetic reconstructions will likely help to
resolve this uncertainty, as Sphaerotermes might either bethe sister
group of Foraminitermes,ﬁguring an independent reduction of the
frontal gland; or the sister group of all Termitidae but Macro-
termitinae and Foraminitermes, in which case the frontal gland was
reduced once in the Termitidae (Fig. 5). Thus, Termitidae sub-
families differ in the shape of their frontal gland: epithelial
with reservoir in Foraminitermitinae and in Macrotermitinae,
reservoir reduced in all other subfamilies. The frontal gland
completely disappeared two times in smaller species, belonging to
The frontal gland in Termitidae imagoes is relatively small and
always conﬁned to the head, while it may reach the thorax in
Heterotermes paradoxus or even the abdomen in Rhinotermitinae
Sobotník et al., 2010c). Nearly all Termitidae representatives
possess a functional frontal gland either with a reservoir, as in some
Macrotermitinae (as already stated by Bugnion,1913;Noirot,1969)
and Foraminitermitinae, or without it as in other subfamilies. The
frontal gland is often reduced in size, and surprisingly completely
absent in the two species of Protermes and Microtermes we exam-
ined (both Macrotermitinae). The variability in size of frontal glands
is quite high not only among Termitidae representatives (although
Fig. 2. The development of the frontal gland in Termitidae: Sphaerotermitinae, Macrotermitinae, Foraminitermitinae and Apicotermitinae. A: Sphaerotermes sphaerothorax, male
head. Bar represents 0.3 mm. B: Ancistrotermes cavithorax, female head. Bar represents 0.3 mm. C: Macrotermes bellicosus, female head. Bar represents 0.1 mm. D: Foraminitermes
coatoni, male head. Bar represents 0.3 mm. E: Longustitermes manni, female head. Bar represents 0.1 mm. F: Aparatermes sp. female head. Bar represents 0.3 mm. Arrows mark
tentorial-fontanellar muscle. Asterisks mark frontal gland. Abbreviations: b, brain (supraoesophageal ganglion); c, clypeus; fb, fat body; h, hypopharynx; l, labrum; lb, labium; mm,
mandibular muscles; p, pharynx.
K. Kutalová et al. / Arthropod Structure & Development 42 (2013) 339e348344
considerably lower than in Rhinotermitidae), but also within
particular subfamilies. Although the gland is represented as only an
epithelial thickening in all advanced Termitidae, there is a sign of
reversal in Amitermes beaumonti (Termitinae), in which the glan-
dular cuticle is deeply indented below the level of the surrounding
cuticle, making the fontanelle in fact re-appearing (see Fig. 3A).
Therefore the re-acquisition of the frontal gland in Foramini-
termitinae, as suggested by the evolutionary analysis on phyloge-
netic trees, appears plausible.
The frontal gland is in general formed exclusively by class 1
secretory cells (after Noirot and Quennedey, 1974), with exception
of the Rhinotermitidae Coptotermes and Heterotermes (Quennedey,
Sobotník et al., 2010d). On the other hand, class 3 secretory
cells frequently occur by the fontanelle or in the canaliculus (the
evacuating channel of the frontal gland) in Nasutitermitinae
soldiers (Grassé, 1982; pers. observ.), and such cells were also
observed in some alates (see Fig. 4B). We expect that they anyway
do not contribute to the frontal gland function, since their structure
is identical to common class 3 secretory cells scattered all over the
termite body (
Sobotník et al., 2010a, pers. observ.).
Surprisingly, the tentorial-fontanellar muscles occur in all castes
and species studied so far, irrespectively of the frontal gland
development (Noirot, 1969;
Sobotník et al., 2010d). While their
function seems clear for glands with reservoir, where they keepthe
gland opening free for secretion ﬂow outwards (
Sobotník et al.,
2010d), there is no clear explanation of their function for epithe-
lial glands without reservoir; the muscles may only convey a spe-
ciﬁc function e.g. during moulting.
According to our observations, the frontal gland is always a
functional secretory organ. A defensive role has been convincingly
Fig. 3. The development of the frontal gland in Termitidae: Termitinae. A: Amitermes beaumonti, female head. Bar represents 0.2 mm. Inset: Parasagittal section of the frontal gland
showing the frontal gland cuticle modiﬁcation. Bar represents 0.1 mm. B: Dentispicotermes brevicarinatus, female head. Bar represents 0.2 mm. C: Crepititermes verruculosus, female
head. Bar represents 0.2 mm. D: Protocapritermes odontomachus, male head. Bar represents 0.2 mm. E: Pericapritermes papuanus, male head. Bar represents 0.2 mm. F: Neo-
capritermes araguaia, female head. Bar represents 0.2 mm. Arrows mark tentorial-fontanellar muscle. Asterisks mark frontal gland. Abbreviations: b, brain (supraoesophageal
ganglion); fb, fat body; mm, mandibular muscles; p, pharynx.
K. Kutalová et al. / Arthropod Structure & Development 42 (2013) 339e348 345
proven in several species possessing a reservoir (see Piskorski et al.,
2009;Krasulová et al., 2010;
Sobotník et al., 2010d). In fact, we have
repeatedly failed in detecting compounds produced by frontal
glands without reservoir, but existing evidence clearly shows that
some secretion is produced, but its function may differ from
epithelial glands with reservoir. Frontal glands lacking the reservoir
may produce proteinaceous secretions (e.g. like in Aparatermes
alates and workers;
Sobotník et al., 2010b), so that its function
might shift to antibacterial or antifungal, which seems reasonable
for young de-alate couples starting to reproduce by their own.
Generally, defensive glands are equipped with a reservoir
(Chapman, 1998), because the defensive efﬁcacy is related to the
amount of the available defensive agent ready to be used. There-
fore, it seems logical that smaller species would rather be subject to
gland reduction (Ephelotermes argutus,Longustitermes manni)or
disappearance (Protermes sp., Microtermes toumodiensis) compared
to larger ones. Our data support this view (frontal gland losses
occurred in the smallest representatives of Macrotermitinae),
although not without exceptions (the relatively large frontal gland
of A. quietus).
Possible factors favouring the frontal gland reduction are the
development of new predator avoidance strategies, such as speciﬁc
ﬂight patterns (seasonal, daily, height of ﬂight etc.) or predator
saturation by synchronization of dispersal ﬂights over a large scale
(Nutting, 1969;Bourguignon et al., 2009). Moreover, ﬂights of
species of Termitidae may be diurnal, nocturnal or crepuscular
(Weesner, 1960;Nutting, 1969). The scarce knowledge about the
timing of the termites’ﬂights does not allow an extensive
Fig. 4. The development of the frontal gland in Termitidae: Syntermitinae and Nasutitermitinae. A: Silvestritermes holmgreni, female head. Bar represents 0.3 mm. B: Embiratermes
neotenicus, male head. Bar represents 0.2 mm. Arrowhead marks class 3 secretory cell. C: Rhynchotermes perarmatus, male head. Bar represents 0.3 mm. D: Diwaitermes kanehirae,
male head. Bar represents 0.3 mm. E: Grallatotermes grallator, female head. Bar represents 0.3 mm. F: Hospitalitermes papuanus, male head. Bar represents 0.3 mm. Arrows mark
tentorial-fontanellar muscle. Asterisks mark frontal gland. Abbreviations: b, brain (supraoesophageal ganglion); c, clypeus; fb, fat body; h, hypopharynx; l, labrum; mm, mandibular
muscles; p, pharynx.
K. Kutalová et al. / Arthropod Structure & Development 42 (2013) 339e348346
comparison of frontal gland development and ﬂight timing, but
available data do not suggest any correlation. For example, swarms
occur at dusk or at night in species without a frontal gland reservoir
like Hospitalitermes spp. (Kalshoven, 1958) and Microcerotermes
heimi (Nasir, 2010), but also in species with a frontal gland with
reservoir like M. bellicosus (Ruelle, 1964). Moreover, ﬂights appear
to be characteristic for a given species and can be very different
even between species of the same genus, as in Microtermes,in
which particular species swarm either in the mid-afternoon, early
in the evening, or during the night (Wood, 1981). Another example
is given by Pseudacanthotermes militaris and Pseudacanthotermes
spiniger which ﬂy during the afternoon and at dusk, respectively
(Connétable et al., 2012), while their frontal glands are comparable
in shape and size. The chronological difference between dispersal
ﬂights in related species can thus be explained rather by the
reproductive isolation than by the development of the frontal
gland, especially in Pseudacanthotermes whose sex pheromones are
identical (Bordereau and Pasteels, 2011).
While the effect of dispersal patterns can hardly be studied, the
function of soldier frontal gland secretion is clearly defensive
(Richardson and Levitan, 1994;Lubin and Montgomery, 1981), and
the alate frontal gland secretion is similar to that of soldiers
(Piskorski et al., 2007,2009;Krasulová et al., 2010). The composi-
tion of defensive secretions is species-speciﬁc, and the defensive
compounds are de novo synthesized (Prestwich, 1984b;
et al., 2010a), thus the defensive secretion synthesis may repre-
sent a substantial part of the defence costs. Species probably face a
trade-off between the production of fewer well-defended de-
scendants (as in e.g. Rhinotermitinae, see
Sobotník et al., 2010d)
and the production of larger numbers of poorly defended de-
scendants as in species with a reduced frontal gland; speciﬁc life
strategies are too complex to be understood only from the view-
point of frontal gland development.
We are grateful to Christian Bordereau and Kumar Krishna for
providing part of material for this study. We are grateful to Mirek
s for the help with TEM. This research was funded by the Czech
Science Foundation (project No. P506/10/1570), by the Institute of
Organic Chemistry and Biochemistry, Academy of Sciences of the
Czech Republic, Prague (project RVO: 61388963). J
S thanks to
project 20124364 of Internal Grant Agency of Faculty of Forestry
and Wood Sciences (Speciﬁc research of the Czech University of Life
Appendix A. Supplementary material
Supplementary data related to this article can be found at http://
Billen, J., 2011. Exocrine glands and theirkey function in the communication system
of social insects. Formosan Entomologist 31, 75e84.
Bordereau, C., Pasteels, J.M., 2011. Pheromones and chemical ecology of dispersal
and foraging in termites. In: Bignell, D.E., Roisin, Y., Lo, N. (Eds.), Biology of
Termites, a Modern Synthesis. Springer, Dordrecht, pp. 279e320.
Bordereau, C., Robert, A., Van Tuyen, V., Peppuy, A., 1997. Suicidal defensive
behavior by frontal gland dehiscence in Globitermes sulphureus Haviland sol-
diers (Isoptera). Insectes Sociaux 44, 289e297.
Bourguignon, T., Leponce, M., Roisin, Y., 2009. Insights into the termite assemblage
of a neotropical rainforest from the spatio-temporal distribution of ﬂying alates.
Insect Conservation and Diversity 2, 153e162.
Bugnion, E., 1913. Le Termes horni Wasm. de Ceylan. Revue Suisse de Zoologie 21,
Chapman, R.F., 1998. The Insects: Structure and Function, fourth ed. Cambridge
Univ. Press, Cambridge.
Connétable, S., Robert, A., Bordereau, C., 2012. Dispersal ﬂight and colony devel-
opment in the fungus-growing termites Pseudacanthotermes spiniger and
P. militaris. Insectes Sociaux 59, 269e277.
Costa-Leonardo, A.M., 2004. A new interpretation of the defense glands of
neotropical Ruptitermes (Isoptera, Termitidae, Apicotermitinae). Sociobiology
Deligne, J., Quennedey, A., Blum, M.S., 1981. The enemies and defense mechanisms
of termites. In: Hermann, H.R. (Ed.), Social Insects, vol. II. Academic Press, New
York, pp. 1e76.
Dial, K.P., Vaughan, T.A., 1987. Opportunistic predation on alate termites in Kenya.
Biotropica 19, 185e187.
Emerson, A.E., 1971. Tertiary fossil species of the Rhinotermitidae (Isoptera), phy-
logeny of genera, and reciprocal phylogeny of associated Flagellata ( Protozoa)
and Staphylinidae (Coleoptera). Bulletin of the American Museum of Natural
History 146, 143e303.
Engel, M.S., Grimaldi, D., Krishna, K., 2009. Termites (Isoptera): their phylogeny,
classiﬁcation, and rise to ecological dominance. American Museum Novitates
Felsenstein, J., 2005. PHYLIP (Phylogeny Inference Package) Version 3.6. Distributed
by the Author. Department of Genome Sciences, University of Washington,
Grassé, P.P., 1982. Termitologia, Tome I. Masson, Paris.
Fig. 5. Mapping the evolution of the frontal gland in termite imagoes on phylogenetic trees. A: Tree by Inward et al. (2007). B: Maximum likelihood tree. C: Bayesian tree after one
million generations (Species names missing when different species were used for DNA sequencing and our structural study). For further details see Fig. S1.
K. Kutalová et al. / Arthropod Structure & Development 42 (2013) 339e348 347
Holmgren, N., 1909. Termitenstudien. I. Anatomische Untersuchungen. In: Kungliga
Svenska Vetenskapsakademiens Handlingar 44, pp. 1e215.
Howse, P.E., 1984. Sociochemicals of termites. In: Bell, W.J., Cardé, W.J. (Eds.),
Chemical Ecology of Insects. Chapman & Hall, London, pp. 475e519.
Inward, D.J.G., Vogler, A.P., Eggleton, P., 2007. A comprehensive phylogenetic
analysis of termites (Isoptera) illuminates key aspects of their evolutionary
biology. Molecular Phylogenetics and Evolution 44, 953e967.
Kalshoven, L.G.E., 1958. Observations on the black termites, Hospitalitermes spp., of
Java and Sumatra. Insectes Sociaux 5, 9e30.
Korb, J., Salewski, V., 2000. Predation on swarming termites by birds. African
Journal of Ecology 38, 173e174.
Krasulová, J., Hanus, R.,
Sobotník, J., Bourguignon, T., Cva
cka, J., Valterová, I., 2010.
Chemistry and anatomy of the frontal gland in termite imagoes. In: 26th Annual
Meeting of the International Society of Chemical Ecology, Tours, France, July
Lepage, M.G., 1991. Predation on the termite Macrotermes michaelseni reproductives
and post settlement survival in the ﬁeld (Isoptera: Macrotermitinae). Socio-
biology 18, 153e166.
Lubin, Y.D., Montgomery, G.G., 1981. Defenses of Nasutitermes termites (Isoptera:
Termitidae) against tamandua anteaters (Edentata: Myrmecophagidae). Bio-
tropica 13, 66e76.
Maddison, W.P., Maddison, D.R., 2010. Mesquite: a Modular System for Evolutionary
Analysis. Version 2.73. http://mesquiteproject.org.
Moore, B.P.,1968. Studies on the chemical composition and function of the cephalic
gland secretion in Australian termites. Journal of Insect Physiology 14, 33e39.
Nasir, S., 2010. Diversity and Swarming Patterns of Termites of District Mianwali.
Ph.D. dissertation, p. 261.
Noirot, C., 1969. Glands and secretions. In: Krishna, K., Weesner, F.M. (Eds.), Biology
of Termites, vol. 1. Academic Press, New York, pp. 89e123.
Noirot, C., Quennedey, A., 1974. Fine structure of insect epidermal glands. Annual
Review of Entomology 19, 61e80.
Nutting, W.L., 1969. Flight and colony foundation. In: Krishna, K., Weesner, F.M.
(Eds.), Biology of Termites, vol. 1. Academic Press, New York, pp. 233e282.
Perna, A., Jost, C., Couturier, E., Valverde, S., Douady, S., Theraulaz, G., 2008. The
structure of gallery networks in the nests of termite Cubitermes spp. revealed by
X-ray tomography. Naturwissenschaften 95, 877e884.
Piskorski, R., Hanus, R., Va
cková, S., Cva
Sobotník, J., Svato
s, A., Valterová, I.,
2007. Nitroalkenes and sesquiterpene hydrocarbons from the frontal gland
secretion of three Prorhinotermes termite species. Journal of Chemical Ecology
Piskorski, R., Hanus, R., Kalinová, B., Valterová, I., K
cek, J., Bourguignon, T.,
Sobotník, J., 2009. Temporal and geographic variations in the
morphology and chemical composition of the frontal gland in imagoes of Pro-
rhinotermes species (Isoptera: Rhinotermitidae). Biological Journal of the
Linnean Society 98, 384e392.
Plasman, V., Daloze, D., Braekman, J.C., Connétable, S., Robert, A., Bordereau, C., 1999.
New macrolactones from the defensive salivary secretion of soldiers of the
African termite Pseudacanthotermes spiniger.Tetrahedron Letters 40, 9229e9232.
Prestwich, G.D., 1984a. Interspeciﬁc variation of deterpene composition of
Cubitermes soldier defense secretions. Journal of Chemical Ecology 10, 1219e
Prestwich, G.D., 1984b. Defense mechanisms of termites. Annual Review of Ento-
mology 29, 201e232.
Prestwich, G.D., Collins, M.S., 1982. Chemical defense secretions of the termite
soldiers of Acorhinotermes and Rhinotermes (Isoptera, Rhinotermitinae): ke-
tones, vinyl ketones and
-ketoaldehydes derived from fatty acids. Journal of
Chemical Ecology 8, 147e161.
Quennedey, A., 1984. Morphology and ultrastructure of termite defense glands. In:
Hermann, H.R. (Ed.), Defensive Mechanisms in Social Insects. Praeger, New
York, pp. 151e200.
Richardson, P.R.K., Levitan, C.D.,1994. Tolerance of aardwolves todefense secretions
of Trinervitermes trinervoides. Journal of Mammalogy 75, 84e91.
Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes version 3.0: Bayesian phylogenetic
inference under mixed models. Bioinformatics 19, 1572e1574.
Ruelle, J.E., 1964. L’essaimage de Macrotermes natalensis Haviland dans la région de
Leopoldville (Isoptera, Macrotermitinae). In: Bouillon, A. (Ed.), Études sur les
termites africains. Masson, Paris, pp. 231e245.
Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5:
molecular evolutionary genetics analysis using maximum likelihood, evolu-
tionary distance, and maximum parsimony methods. Molecular Biology and
Evolution 28, 2731e2739.
Weesner, F.M., 1960. Evolution and biology of the termites. Annual Review of
Entomology 5, 153e170.
Wood, T.G.,1981. Reproductive isolating mechanisms among speciesof Microtermes
(Isoptera, Termitidae) in the Southern Guinea savanna near Mokwa, Nigeria. In:
Howse, P.E., Clément, J.L. (Eds.), Biosystematics of Social Insects. Academic Press,
London, pp. 309e332.
Sobotník, J., Weyda, F., Hanus, R., Kyjaková, P., Doubský, J., 2004. Ultrastructure of
the frontal gland in Prorhinotermes simplex (Isoptera: Rhinotermitidae) and
quantity of the defensive substance. European Journal of Entomology 101,
Sobotník, J., Jiro
sová, A., Hanus, R., 2010a. Chemical warfare in termites. Journal of
Insect Physiology 56, 1012e1021.
Sobotník, J., Bourguignon, T., Hanus, R., Weyda, F., Roisin, Y., 2010b. Structure and
function of defensive glands in soldiers of Glossotermes oculatus (Isoptera:
Serritermitidae). Biological Journal of the Linnean Society 99, 839e848.
Sobotník, J., Sillam-Dussès, D., Weyda, F., Dejean, A., Roisin, Y., Hanus, R.,
Bourguignon, T., 2010c. The frontal gland in workers of neotropical soldierless
termites. Naturwissenschaften 97, 495e503.
Sobotník, J., Bourguignon, T., Hanus, R., Sillam-Dussès, D., Pﬂegerová, J., Weyda, F.,
Kutalová, K., Vytisková, B., Roisin, Y., 2010d. Not only soldiers have weapons:
evolution of the frontal gland in imagoes of the termite families Rhinotermi-
tidae and Serritermitidae. PLoS ONE 5, e15761.
Sobotník, J., Bourguignon, T., Hanus, R., Demianová, Z., Pytelková, J., Mare
Foltýnová, P., Preisler, J., Cva
cka, J., Krasulová, J., Roisin, Y., 2012. Explosive
backpacks in old termite workers. Science 337, 436.
K. Kutalová et al. / Arthropod Structure & Development 42 (2013) 339e348348