Life history and development - A framework for understanding developmental plasticity in lower termites

Abstract

Termites (Isoptera) are the phylogenetically oldest social insects, but in scientific research they have always stood in the shadow of the social Hymenoptera. Both groups of social insects evolved complex societies independently and hence, their different ancestry provided them with different life-history preadaptations for social evolution. Termites, the ‘social cockroaches’, have a hemimetabolous mode of development and both sexes are diploid, while the social Hymenoptera belong to the holometabolous insects and have a haplodiploid mode of sex determination. Despite this apparent disparity it is interesting to ask whether termites and social Hymenoptera share common principles in their individual and social ontogenies and how these are related to the evolution of their respective social life histories. Such a comparison has, however, been much hampered by the developmental complexity of the termite caste system, as well as by an idiosyncratic terminology, which makes it difficult for non-termitologists to access the literature.
Here, we provide a conceptual guide to termite terminology based on the highly flexible caste system of the “lower termites”. We summarise what is known about ultimate causes and underlying proximate mechanisms in the evolution and maintenance of termite sociality, and we try to embed the results and their discussion into general evolutionary theory and developmental biology. Finally, we speculate about fundamental factors that might have facilitated the unique evolution of complex societies in a diploid hemimetabolous insect taxon. This review also aims at a better integration of termites into general discussions on evolutionary and developmental biology, and it shows that the ecology of termites and their astounding phenotypic plasticity have a large yet still little explored potential to provide insights into elementary evo-devo questions.

Full text

Biol. Rev. (2008), 83, pp. 295–313. 295
doi:10.1111/j.1469-185X.2008.00044.x
Life history and development - a framework
for understanding developmental plasticity in
lower termites
Judith Korb
1
*
†
and Klaus Hartfelder
2
1
Biologie I, Universita
¨t Regensburg, D-93040 Regensburg, Germany
2
Departamento de Biologia Celular e Molecular e Bioagentes Patoge
ˆ
nicos, Faculdade de Medicina de Ribeira˜o Preto, Universidade de Sa˜o Paulo,
Ribeira˜o Preto, Brazil (E-mail: klaus@fmrp.usp.br)
(Received 17 September 2007; revised 16 April 2008; accepted 08 May 2008)
ABSTRACT
Termites (Isoptera) are the phylogenetically oldest social insects, but in scientific research they have always stood
in the shadow of the social Hymenoptera. Both groups of social insects evolved complex societies independently
and hence, their different ancestry provided them with different life-history preadaptations for social evolution.
Termites, the ‘social cockroaches’, have a hemimetabolous mode of development and both sexes are diploid,
while the social Hymenoptera belong to the holometabolous insects and have a haplodiploid mode of sex
determination. Despite this apparent disparity it is interesting to ask whether termites and social Hymenoptera
share common principles in their individual and social ontogenies and how these are related to the evolution of
their respective social life histories. Such a comparison has, however, been much hampered by the developmental
complexity of the termite caste system, as well as by an idiosyncratic terminology, which makes it difficult for
non-termitologists to access the literature.
Here, we provide a conceptual guide to termite terminology based on the highly flexible caste system of the
‘‘lower termites’’. We summarise what is known about ultimate causes and underlying proximate mechanisms in
the evolution and maintenance of termite sociality, and we try to embed the results and their discussion into
general evolutionary theory and developmental biology. Finally, we speculate about fundamental factors that
might have facilitated the unique evolution of complex societies in a diploid hemimetabolous insect taxon. This
review also aims at a better integration of termites into general discussions on evolutionary and developmental
biology, and it shows that the ecology of termites and their astounding phenotypic plasticity have a large yet still
little explored potential to provide insights into elementary evo-devo questions.
Key words: social insect, caste, polyphenism, developmental plasticity, pseudergate, neotenic, wing development,
moult, metamorphosis, juvenile hormone.
CONTENTS
I. Introduction ...................................................................................................................................... 296
II. Caste patterns in termites ................................................................................................................. 298
III. Ultimate causes influencing caste development in lower termites .................................................. 301
IV. Proximate mechanisms underlying caste development .................................................................... 302
(1) Social and environmental triggers of polyphenisms in termites and other hemimetabolous insects 302
(2) Endocrine control of caste development in lower termites ....................................................... 302
* Address for correspondence: (Tel: 0049 541 9693496; Fax: 0049 541 9692830; E-mail: judith.korb@biologie.uni-osnabrueck.de).
†
Present address: Verhaltensbiologie, Universita
¨t Osnabru
¨ck, D-49080 Osnabru
¨ck, Germany (E-mail: judith.korb@biologie.
uni-osnabrueck.de).
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

(3) Juvenile hormone in progressive moults of lower termites ....................................................... 303
(4) Molecular underpinnings of caste in lower termites ................................................................. 304
V. Perspectives on hormonal regulation, caste development and termite evolution ........................... 305
(1) Ecdysteroids and insulin signaling .............................................................................................. 305
(2) What is the status of the larval stages? ...................................................................................... 306
(3) Regressive moults in lower termites, a major enigma ............................................................... 307
VI. Conclusions ....................................................................................................................................... 308
VII. Acknowledgments ............................................................................................................................. 308
VIII. References ......................................................................................................................................... 308
I. INTRODUCTION
Caste differentiation in termites is one of the most
conspicuous examples of facultative polyphenism in ani-
mals, in which individuals show various phenotypes despite
the same genetic background. Within a termite colony -
with few exceptions where a genetic component might be
involved (Hayashi et al., 2007) - the offspring of a single king
and queen can develop into workers, soldiers, and two
sexual morphs depending on environmental and social
stimuli. Each caste generally exhibits a particular behav-
ioural repertoire and often caste-specific morphological
characters. As in social Hymenoptera (ants, and some bees
and wasps), this caste system of termites is considered to be
the basis for the evolutionary and ecological success of these
eusocial insects (Oster & Wilson, 1978).
Obviously, such elaborate societies evolved from solitary
ancestors. Consequently, a focus of current research is to
explain how complex phenotypes can evolve from ancestral
solitary forms. Often differences in complex traits among
species are not the result of the presence or absence of
particular genes, but arise from changes in the mechanisms of
gene regulation affecting when and where a gene or an entire
regulatory module is expressed. As noted by (Brakefield,
2006; p. 362): ‘There is a limited genetical tool kit and
much of the morphological diversity evolution is about old
genes performing new tricks. Although existing genetic
pathways can be co-opted and subsequently elaborated
upon to do something different, and specific genes can take
on additional tasks at new times during development and in
different tissues via gene duplication and divergence, de novo
evolution of new pathways appears to be rare’.
Besides the reproductive polyphenisms in social insects and
that of male weaponry in dung beetles, other major types of
polyphenisms in insects are sequential (seasonal) polyphen-
isms in wing length (in its extreme form, winged and wingless
morphs) or coloration; these are also often connected to dif-
ferences in reproductive strategies (for review see Hartfelder &
Emlen,2005).Inbothtermitesandants,thecastesyndrome
is coupled to a wing dimorphism. This clearly represents
convergent evolution since the wingless workers of ants are
adults of a holometabolan clade, whereas the wingless
workers in termites are either immatures (in the wood-nesting
termite families Termopsidae and Kalotermitidae, and
Prorhinotermes) or have conserved an immature phenotype
[the worker caste of the Mastotermitidae, Rhinotermitidae
(except Prorhinotermes), Hodotermitidae, Serritermitidae, and
Termitidae]. Additional wingless morphs in termites are the
neotenic reproductives and soldiers, for which there is no real
equivalent in ants. Termites evolved complex societies more
than 130 million years ago, probably during the upper
Jurassic, (Thorne, Grimaldi & Krishna, 2000) from a
cockroach-like ancestor similar to the wood-roaches (Nalepa
& Bandi, 2000) which form the sister group to present
termites (Eggleton, 2001; Inward, Beccaloni & Eggleton,
2007a).Cockroachesaswellastermitesarecharacterizedby
a highly flexible development (Roth, 1981; Nalepa, 1994).
Even though termites are phylogenetically distant from
ants, and the two groups have been considered either to be
engaged in arms races (Longhurst, Johnson & Wood, 1978;
Deligne, Quennedy & Blum, 1981) or in more peaceful co-
evolutionary interactions (Dejean, Durand & Bolton, 1996),
they share several characters as a result of convergent
evolution. Both ants and termites have clearly evolved
independently from winged solitary or primitively social
ancestors (Wilson, 1971); winglessness is an adaptation to
burrowing activities and a nest structure which functions as
a fortress against predators. The question, thus, becomes
which mechanisms underlie the flexible expression of wing
phenotypes in ants and termites, where wings develop in the
alate reproductives and where wing development is shut
down in wingless reproductives and in the worker and
soldier castes.
Pattern formation in wings, including their transforma-
tion into very different forewing and hindwing types, is best
described in Drosophila melanogaster; not only the individual
signaling pathway elements, but even entire gene regulatory
networks are found to be highly conserved across species,
orders and even phyla (Carroll, Grenier & Weatherbee,
2005). The D. melanogaster wing formation network has been
successfully applied to a study on the loss of wings in
workers of several ant species (Abouheif & Wray, 2002)
which showed that wing disc development in workers is not
brought to a halt at a single unique interruption point in the
network, but rather can stop at different points among ant
genera. Caste polyphenism and its relation to wing bud
development has not yet been studied to this extent in
termites, but recent progress on endocrine regulation and
associated gene expression differences is now gradually
shedding light on regulatory mechanisms underlying caste
development, especially the fascinating (and notoriously
confusing) plasticity of caste fate in the lower termites.
We will summarise here the state of termite research
regarding this developmental question. First, we will explain
termite caste systems and their distribution among taxa.
This is supplemented by summaries of the classification of
Judith Korb and Klaus Hartfelder296
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

termite species (Table 1), developmental terms (Table 2) and
classification of castes (Table3) to provide a comprehensive
introduction to termite terminology, essential for wider
recognition and comprehension of termite research. We will
mainly concentrate on phylogenetically basal termite taxa,
but occasional notes on higher termite taxa are given where
appropriate. Second, we will summarise what is known
about ultimate causes influencing cooperation and altruism
in termite societies. Third, we will describe current views
about proximate mechanisms underlying social complexity
in termites starting with environmental and social triggers
that affect different developmental trajectories, then out-
lining the underlying endocrinology of termite caste de-
velopment, and presenting the current understanding of
differential gene expression during termite caste differenti-
ation. Finally, by embedding our knowledge within a general
framework of insect development, we will argue that termite
studies can add novel facets to our understanding of the
evolution of holometabolous development from hemime-
tabolous ancestors and we speculate about fundamental
facilitators that may help to explain the exceptional position
of termites as the only diploid group within the highly social
insects.
There is a plethora of literature published on termites and
on how best to erradicate them. Undeniably, we have come
a long way since the pioneering descriptions on termite life
cycles by Euge
`ne Nielen Marais (1937), but there remain
enormous gaps in our knowledge on this group of insects; we
can still share his observation that ‘The entomologist who
made the acquaintance of the termite for the first time,
would be justified in thinking it to be an immigrant from
a different planet’ (Marais, 1937, chap. 9, par 1).
Table 1. Classification of termite species
Based on their ecology and particularly nesting and feeding habits, termite species can be grouped into two life types:
One-piece nesting termites Multiple-pieces nesting termites (incl. Abe’s intermediate type; Abe, 1990)
dlive inside single piece of dead wood serving both
as nest and food source
dwell-defined nest separate from foraging grounds
dwith exception of winged sexual’s individuals
never leave nest
dworkers exploit new food resources outside the nest
dcolony life limited by food availability dcolony life not limited by food availability
dbasal life history
dhighly flexible individual development (see false workers)dtrue, morphologically differentiated worker caste
with reduced reproductive potential
dTermopsidae, Kalotermitidae, Prorhinotermes dMastotermitidae, Hodotermitidae, Serritermitidae,
Rhinotermitidae (except Prorhinotermes), Termitidae
An alternative traditional classification of termites is based on their gut symbionts:
Lower termites Higher termites
dbacteria and flagellates in gut dbacteria only
dall termites except Termitidae dTermitidae
Table 2. Developmental terms applied to termites
Lower termites are characterized by a unique flexibility in
development which is generated through three molting types:
dProgressive moult - a moult characterizing the gradual development
from egg via several instars into an adult. Associated with
progressive moults is an increase in body size and morphological
development. This is the default developmental program in all
hemimetabolous and holometabolous insects.
dStationary moult - an intermittent moult that is associated with
a lack of increase in body size and morphological development.
This type of development occurs in several insect species and is
frequently associated with periods of food shortage, when a larva
or nymph is not capable of passing a critical mass threshold in
an instar. In some termites it might also be linked to the wear of
mandibles.
dRegressive moult - a moult that is characterized by a decrease in
body size and/or regression of morphological development,
generally a reduction of wing bud size in nymphal instars. This
type of development is unique to termites.
In contrast to other hemimetabolous insects where postembryonic
development is characterized by a progression of nymphal instars,
termites show two distinct types of instars:
dLarval instar(s) - instar(s) without externally visible wing buds
dNymphal instars - instars with externally visible wing buds; these
instars characterize the gradual, progressive development into
winged sexuals.
Larval instars are sometimes further split into:
dDependent larvae - the first up to the second or third larval instar,
when larvae supposedly still depend on brood care by ‘workers’
(but see ‘false workers’ in Table 3)
dIndependent larvae - all other larval instars, when larvae care for
themselves. They comprise the false workers and the pseudergates
sensu lato (see Table 3).
Developmental plasticity in termites 297
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

II. CASTE PATTERNS IN TERMITES
In termites two reproductive and two non-reproductive
castes can be distinguished: primary and neotenic repro-
ductives on the one hand, and soldiers and true workers on
the other, with the latter two forming the majority of the
individuals of a colony (Noirot, 1990; Roisin, 2000). The
occurrence of the castes differs among families (Roisin,
2000) (Fig.1). While soldiers are present in all families with
exception of a few genera in the Termitidae (higher
Table 3. Classification of termite castes
Reproductives
Individuals that reproduce within a colony, generally one female (queen) and one male (king). These can be:
dPrimary reproductives - reproductives that found a new colony after a nuptial dispersal flight. They develop gradually via several nymphal
instars into winged sexuals (alates) that shed their wings (dealates) after the nuptial flight. They are characterized by stark sclerotization,
the presence of compound eyes and wing marks (remnants of the wings’ articulation after they have been shed).
dAdultoids - alates that shed their wings and reproduce within the natal nest (they are not neotenics).
dNeotenic reproductives - wingless reproductives that develop within the natal colony via a single moult from any instar after the third larval
instar. At this neotenic moult, their gonads grow and they develop some imaginal characters while maintaining an otherwise larval
appearance; some characters, like wing pads, may regress. Neotenic reproductives are characterized by the absence of wings and usually
by the lack of compound eyes. The cuticle is less sclerotized than in primary reproductives. They are subdivided into:
dReplacement reproductives if they develop after the death of the same-sex reproductive of a colony.
dSupplementary reproductives if they develop in addition to other same-sex reproductive(s) already present within a colony.
Depending on the termite life type and the instar from which they develop the neotenics can be further classified into:
dNeotenics (sensu stricto): They can be either apterous neotenics developing from a larval instar or brachyterous neotenics developing from
a nymphal instar. They are found in lower termites. Neotenics developing from nymphs are sometimes also called secondary
reproductives, while those developing from workers are called tertiar y reproductives.
dErgatoids: neotenics developing from workers in higher termites
dNymphoids: neotenics developing from nymphs in higher termites
‘Workers’
The majority of individuals within a colony belong to the so-called ‘worker’ caste, although they do not necessarily have to work (see ‘false
workers’). With a few exceptions among some higher termites, they are not restricted to a specific sex.
A clear separation should be drawn between the ‘workers’ of the one- and multiple-pieces nesting termites (see Table1) as they are not
equivalent in function and development:
dFalse workers - the majority of the individuals within a colony of one-piece nesting termites. They differ from the (true) workers of
multiple-pieces nesting termites as they are totipotent larvae that lack morphological differentiations. Correspondingly, they are less
involved in truly altruistic working tasks, such as foraging, brood care, or building behaviours. Therefore, they may rather be regarded as
large immatures that delay reproductive maturity (‘hopeful reproductives’).
dTrue workers - workers in colonies of the multiple-pieces nesting termites. They can be considered altruistic individuals as they perform
most tasks within a colony (e.g. foraging, brood care, and building behaviour) except for reproduction and specialized defence. Although
they sometimes, especially in lower termites, still have some reproductive options (for instance as neotenic reproductives), their
morphological differentiations (especially their sclerotisation) largely restrict their developmental capability (a notable exception is
Mastotermes darwiniensis). In functional terms, these true workers, often just called workers, are equivalent to the workers of the social
Hymenoptera, even though the latter are imagoes, whereas the true workers here are preimaginal stages.
An alternative technical term can be found that distinguishes workers with a flexible development and options for direct reproduction from
workers with restricted developmental trajectories:
dPseudergates - ‘workers’ of many lower termites (including one- and multiple-pieces nesting species) that have broad developmental
options, generally including progressive, stationary and regressive moults. Current use of this term often lacks the precision of its original
definition (Grasse & Noirot,1947) for individuals that develop regressively from nymphal instars to ‘worker’ instars without wing buds.
We propose here the following definitions which we use consistently throughout our review:
dPseudergates sensu stricto - individuals that develop via regressive moults from instars with wing buds (‘nymphal instars’) to instars without
wing buds (‘larval instars’), as defined by Grasse & Noirot (1947).
dPseudergates sensu lato - ‘workers’ of the lower termites that have the potential to undergo progressive, stationary, and regressive moults.
They include the ‘false workers’ of the one-piece nesting termites and those ‘true workers’ of the multiple-pieces nesting termites that
belong to the lower termites and have a flexible development. They comprise larval and nymphal instars.
Soldiers
This caste is unique among social insects in function and development. Soldiers are ancestral in termites and evolved prior to a true worker
caste. Unlike the soldiers found in other social insects, this caste is monophyletic in termites.
dSoldiers - a clearly altruistic caste that is always sterile and that is morphologically and behaviourally specialized for defence of the colony
against predators and competitors.
dPresoldiers - a single transitional instar during development from ‘worker’ to soldier.
Judith Korb and Klaus Hartfelder298
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

termites), true workers are absent in the families Termop-
sidae (dampwood termites) and Kalotermitidae (drywood
termites), and in the genus Prorhinoter mes within the
Rhinotermitidae. In these taxa the majority of individuals
are large immatures (also called workers, helpers, false
workers or pseudergates) which differ from true workers in
several aspects. As termite terminology can be very
confusing, we recommend the use of ‘false workers’ for
the large immatures of the Termopsidae, Kalotermitidae
and Prorhinotermes: the term ‘pseudergate’ originally had
a more restricted definition (individuals that developed
regressively from nymphs; Grasse & Noirot, 1947; see also
Table 3), and there is accumulating evidence that they are
not engaged in brood care or raising of siblings to the extent
that the term ‘helper’ might imply (Korb, 2007b;
B. Rosengaus personal communication). The workers of
all other termites are ‘true workers’ (see Table 3). True
workers with flexible development including regressive
moults together with false workers have often been called
‘pseudergates’. We propose here to use the terms ‘pseu-
dergate sensu lato’ for workers with flexible development
including false workers, and ‘pseudergate sensu stricto’
exclusively for individuals that develop via regressive
moult(s) from nymphs (i.e. pseudergates as defined by
Grasse & Noirot, 1947; see Table 3).
The false workers of the Termopsidae, Kalotermitidae
and Prorhinotermes lack morphological differentiations and
are not a final caste. They are a mixed group of several
instars which are not arrested in their development, but
rather retain the option to develop into sterile soldiers (a few
individuals within each colony) or one of the two types of
reproductives (Fig. 2A). Thus, in contrast to what was
recently reported for a Reticulitermes species (Hayashi et al.,
2007), caste development in these taxa is not genetically
determined. In the pathway leading to reproductives’ false
workers can either develop into winged sexuals that found
a new nest (primary reproductives) or they can become
neotenic replacement reproductives in the natal nest if the
same-sex reproductive of the colony is unhealthy or dies.
Neotenic reproductives in Termopsidae, Kalotermitidae
and Prorhinotermes are characterized by the absence of wings
and compound eyes and a less sclerotized cuticle than in
alates. They may originate via a single moult from any instar
after the third. At this neotenic moult, their gonads grow
and they develop some imaginal characters while main-
taining an otherwise larval appearance; some characters,
like wing pads, may even regress. This stands in contrast to
the development of winged sexuals which is gradual and
generally occurs via several nymphal instars with increasing
wing pad development.
Associated with the lack of a true worker caste in
Termopsidae, Kalotermitidae and Prorhinotermes is a charac-
teristic life type. These termites live inside a piece of wood
that serves both as food and nest and which they, except
occasionally in a few species, never leave to exploit new
resources (one-piece nesting termites; Abe, 1987, 1990).
Consequently, when the food is depleted, the colony dies. In
contrast to the typical drywood and dampwood termites,
however, it has recently been shown for some Prorhinotermes
species that they can move to a new nest site when food
supplies are low (Roisin & Parmentier, 2006). Interestingly,
neotenic development can be triggered by adding new
potential nest sites (i.e. pieces of wood) at a distance of a few
centimetres, and some data indicate that these neotenics
?
Mastotermitidae (MP): p, n, w, s
Hodotermitidae (MP): p, n, w,s
Termopsidae (OP): p, n, s
Kalotermitidae (OP): p, n, s
Serritermitidae (MP): p, n, w, s ?
Rhinotermitidae (OP): p, n, s
(MP): p, n, w, s
+ Termitidae (MP): p, (n), w, (s)
Lower termites
Higher termites
Phylogenetic tree Termite family (life type): castes Classification
Fig. 1. Phylogenetic tree with life types and the occurrence of different castes in termites. OP- one-piece life type termites which
nest in a piece of wood that serves both as shelter and food; MP- multiple-pieces life type termites where nest and food are
separated. Unresolved positions are shown in grey or marked ?. Traditionally, termites are classified into lower and higher termites
according to the presence or the absence of protozoan gut symbionts. For the monotypic Serritermitidae the caste system is
unknown (Roisin, 2000). Castes: p ¼primary reproductives, n ¼neotenic reproductives, w ¼(true) workers, s ¼soldiers; castes in
parentheses indicate that the caste is not present in all species of this family. In taxa lacking workers, totipotent large immatures
(often also called pseudergates, helpers, or false workers; see Table 3) occur. As they do not represent a developmental endpoint,
they are not equivalent to the other termite castes and therefore are not listed as a caste here.
Developmental plasticity in termites 299
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

become the kings and queens of the new nest in a budding
process (Roisin, 2006). Thus, although these neotenics do
not reproduce in the natal nest, as occurs in the
Termopsidae and Kalotermitidae, immatures of Prorhino-
termes species can become reproductives without a costly
winged dispersal process.
Strikingly, this option to become an unwinged reproduc-
tive that avoids a costly nuptial flight also exists in the most
basal termite family, the monotypic Mastotermitidae;
Mastotermes darwiniensis lacks totipotent large immatures
and has true workers (Fig. 2B). Recent results show that
similar to derived termite taxa, such as the Termitidae,
there is an early separation into two developmental
pathways, the apterous and nymphal lines, in M. darwiniensis
(Watson & Abbey, 1985; Watson & Sewell, 1985; Parment-
ier, 2006). These have been called neuter and alate lines
respectively, but the former terms seem to be more
appropriate (see Roisin, 2000). Individuals can either
develop via nymphal instars into winged sexuals (nymphal
line) or they can become workers (apterous line) (Fig. 2B).
Interestingly, these true workers have the option to develop
into wingless neotenic reproductives (ergatoids; neotenics
derived from workers) which have been shown to head most
of the field colonies that originate primarily via colony
budding (Goodisman & Crozier, 2002; M. Lenz, personal
communication). Although alates do occur, they are
apparently less successful in colony foundation; field
colonies with primary reproductives were not found in
recent studies (Goodisman & Crozier, 2002; M. Lenz,
personal communication). Thus, M. darwiniensis displays
mosaic evolution of ancestral and highly derived traits.
Convergently to derived taxa, like some Rhinotermitidae
and the Termitidae (Fig. 2C), it evolved a separate apterous
worker line, although from a developmental point of view,
this worker line is not equivalent to those in the derived
taxa; both the nymphal and apterous lines in M. darwiniensis
have direct reproductive options, the nymphal line in the
form of winged sexuals and the apterous worker line in the
form of unwinged neotenic reproductives. In M. darwiniensis
the neotenics, on which colony reproduction largely relies,
originate from the workers, whereas in most other termite
species nymph-derived neotenics are found (M. Lenz,
personal communication). Clearly, in all basal termite
clades neotenic reproduction is common (Fig. 1) and
presents an alternative to winged dispersing sexuals.
Neotenic reproduction, which only has been lost
secondarily in some derived clades, thus, can be considered
a synapomorphy that characterizes termites. Its ancestral
origin goes hand in hand with the transition to eusociality
and has long been claimed to be fundamental for the
A– One-piece nesting termites
B – Mastotermes darwiniensis
C – Multiple-pieces nesting termites (except M. darwiniensis)
Neotenic reproductive
Neotenic
reproductive
Nymphoids3Adultoids4
Ergatoids4
Fig. 2. Developmental pathways in termites. Italics indicate winged reproductives; bold type indicates wingless reproductives; /:
progressive moult; ): regressive moult; 4: stationary moults.
1
: workers develop from different instars depending on the species;
they are partly polymorphic: i.e. major and minor workers;
2
: in some species nymphal instars can have regressive moults;
3–5
:
reproductives that stay in the nest to reproduce, present in some species (Myles, 1999);
3
nymphoids: neotenic reproductives
developing from nymphs;
4
adultoids: non-dispersing reproductives developing from alates, mainly in Termitidae;
5
ergatoids:
neotenic reproductives developing from (true) workers.
Judith Korb and Klaus Hartfelder300
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

evolution of eusociality in termites (Myles, 1988; Thorne,
1997; but see also Roisin, 1999). Reproduction as wingless
neotenics might be regarded as an alternative breeding
tactic that avoids the cost of winged dispersal. Early
separation during development into a nymphal and
apterous line in species with true workers [Rhinotermitidae
(except Prorhinotermes), Serritermitidae, Hodotermopsidae,
Termitidae] which forage outside the nest (multiple-pieces
nesting termites, sensu Abe, 1987) (Fig. 2C) might further
imply that the termite caste system, including true workers,
arose from a wing-polyphenism reflecting alternative
breeding tactics (sensu Gross, 1996).
III. ULTIMATE CAUSES INFLUENCING CASTE
DEVELOPMENT IN LOWER TERMITES
Their developmental flexibility combined with a basal
phylogenetic position make one-piece nesting termites ideal
subjects to study the ultimate causes for the development of
immatures into soldiers and two types of reproductives
(wingless neotenics and dispersing winged sexuals). In
solitary insects, wing polyphenisms reflect alternative
breeding options: winged individuals disperse to breed
elsewhere, while wingless adults reproduce at the natal nest
(philopatric breeding) or close to it. The ‘developmental
decision’ whether to stay or leave in these solitary insects
generally depends on density and food availability at the
natal nest, which will both influence reproductive oppor-
tunities and the direct fitness of philopatric breeding (Roth,
1981; Mu
¨ller, Williams & Hardie, 2001; Braendle et al.,
2006). The default option in solitary insects is to become
a winged sexual, as dispersal in all organisms is generally
selected for to avoid competition with relatives and
inbreeding (Hamilton & May, 1977). Abundant food
resources at the nest, opportunities to meet unrelated
mating partners close to the nest, or a lack of incest
avoidance (e.g. through secondary mechanisms that protect
against inbreeding depression) combined with high dis-
persal costs may, however, select for philopatric breeding
and the evolution of wingless morphs (e.g. Alexander, 1974;
Braendle et al., 2006).
In the ancestors of termites, it can be hypothesised that
the first individuals to remain at the nests ‘chose’ philopatric
breeding to avoid costly dispersal, while helping evolved
only secondarily. This is illustrated by an extant drywood
termite species, Cryptotermes secundus. We will concentrate on
this species here because it not only exhibits an ancestral
one-piece nesting life type (Korb, 2007a) but also is one of
the most thoroughly studied. References to other lower
termites will be given where appropriate. In Cryptotermes
secundus, like in all one-piece nesting termites, totipotent
individuals have the option to develop into winged sexuals
or to stay in the nest with a chance of becoming a wingless
neotenic replacement sexual when the same-sex reproduc-
tive of the colony dies. There is no local resource
competition, as the nest constitutes a bonanza food resource
that generally outlasts the lifetime of the founding primary
reproductives (Korb, 2008). If the nest quality declines,
individuals are predicted to show an increased tendency to
develop into winged sexuals. Experiments have confirmed
this, showing that reduced food availability (Korb & Lenz,
2004; Korb & Schmidinger, 2004) or a high parasite load
(Korb & Fuchs, 2006) at the nest together with large group
sizes lead to increased development of dispersing sexuals.
Furthermore, the number of individuals developing into
winged sexuals that leave the nest can be explained by the
relative probability of inheriting the nest versus successfully
founding a new colony (Korb, 2008). As has been suggested
for termites in general (Nutting, 1969), the probability of
successfully founding a nest is extremely low in C. secundus
(<1 %). The chances of inheriting the colony are on the
same order of magnitude and depend on colony size, age of
the present reproductives, and the potential longevity of the
nest (Korb & Schneider, 2007). These three variables
explain the variation in the number of individuals develop-
ing into dispersing sexuals in field colonies. This suggests
that individuals remaining at the nest as neotenics gain
direct fitness benefits as has also been proposed by Myles
(1988): dispersal is risky, while the nest presents a safe haven
(sensu Kokko & Ekman, 2002). Similarly, in Zootermopsis
nevadensis inheritance of the natal breeding position after
intercolonial encounters seems to favour the development of
soldier-like neotenics (i.e. neotenics with soldier-like traits,
misleadingly also called reproductive soldiers) which are
a peculiarity of the Termopsidae (Thorne, Breisch &
Muscedere, 2003). At least in C. secundus, indirect fitness
benefits gained through raising siblings seem to be less
important (Korb, 2007b): the ‘decision’ to develop into
a winged sexual is independent of the number of young
present in the nest. If individuals were staying in order to
raise young, one would predict a negative correlation
between these variables: individuals would be less likely to
leave the colony when there are more offspring to raise. The
lack of a correlation was explained by subsequent obser-
vations of an absence of brood care in this species (Korb,
2007b).
Although C. secundus is the only species for which we have
such detailed results derived from field as well as laboratory
experiments, these results might apply to one-piece nesting
termites in general (Korb, 2008) for several reasons: (i) all
one-piece nesting termites have totipotent false workers that
can develop into winged sexuals and neotenic reproduc-
tives; (ii) reports exist for many one-piece nesting termites
that a reduction in food availability triggers the develop-
ment of winged sexuals (Buchli, 1958; Lenz, 1976, 1994; La
Fage & Nutting, 1978; Korb & Lenz, 2004); (iii) they live
within a bonanza-type food resource, removing the value of
food provisioning for nestmates as all individuals have easy
access to food, and a lack of specialized brood care has been
recorded for at least four other species [Zooter mopsis
nevadensis (Howse, 1968); Zootermopsis angusticollis (Rosengaus
& Traniello, 1993); Cryptotermes domesticus and Cryptotermes
cynocephalus (J. Korb, personal observations)]; (iv) they live in
a wooden nest that is well protected against predators, while
dispersal is relatively risky (Nutting, 1969; Myles, 1988).
Taken together, it appears that breeding opportunities as
neotenic reproductives offer high incentives for staying at
the nest in one-piece nesting termites. Hence, as in solitary
Developmental plasticity in termites 301
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

insects with wing polymorphism, there seem to be two
alternative tactics: stay and breed as wingless sexuals, or
leave as winged sexuals to reproduce elsewhere.
It can therefore be suggested that the safety of the nest,
together with the chance of inheriting the natal breeding
position, led to delayed dispersal and the formation of family
groups. In C. secundus, as well as in other termites (Haverty,
1977, 1979; Haverty & Howard, 1981; Shellman-Reeve,
1997), the first soldier only develops after such a group has
become established. In C. secundus this commonly is at the
end of the first year after colony foundation, when about 20
false workers are present. Thereafter, the number of soldiers
within a colony is adjusted to colony size, so that a rather
constant and species-specific proportion of soldiers will be
present within a nest (Haverty, 1977, 1979; Haverty &
Howard, 1981). These sterile soldiers defend their family
against predators and competitors, hence gaining indirect
fitness benefits. This was shown in a field experiment with
C. secundus in which soldiers were removed and their re-
development hormonally suppressed: soldier-less colonies
had a lower fitness than similar-sized control colonies (Roux
& Korb, 2004).
IV. PROXIMATE MECHANISMS UNDERLYING
CASTE DEVELOPMENT
(1) Social and environmental triggers of
polyphenisms in termites and other
hemimetabolous insects
For most organisms displaying alternative phenotypes,
neither phenotype exhibits higher fitness overall. Rather,
there is a trade-off, with the relative fitness of the different
phenotypes being contingent upon environmental condi-
tions. The evolution and maintenance of polyphenisms,
therefore, requires and is a consequence of variation in the
environment. For the evolution of polyphenisms several
conditions must be met. First, environmental conditions
must influence development to generate different pheno-
types. Second, the resulting phenotypes must exhibit higher
than average fitness in their respective environments. The
factors acting as triggering cues may be the same as the
selective agent, or they may be different. As the develop-
mental environment of a phenotype often precedes the
selective environment for the adult organism, an environ-
mental cue must at least be correlated with future selective
factors (West-Eberhard, 2003). Environmental control of
alternative phenotypes can, therefore, evolve in organisms
living in spatially or temporally variable environments in
which cues can be used to predict reliably the future
selective environment (Moran, 1992).
In termites, the cues triggering wing development are
identical with the selective agents that favour dispersal of
individuals. Reduced food availability or increased parasite
load (both causing reduced fitness benefits for individuals
staying at the nest) immediately induce a change in
behaviour of C. secundus linked to development into winged
sexuals (Korb & Schmidinger, 2004): false workers increase
food acquisition behaviours and spend less time moving.
More strikingly, while proctodeal trophallaxis (anal feeding)
is reduced at the colony level, those individuals which are
the most active feeders of other individuals develop
progressively into nymphal instars. These results support
a long-standing, but so far unproven suggestion that
inhibitory substances are transmitted within the colony via
proctodeal trophallaxis (Lu
¨scher, 1974). Similarily, large
group sizes which decrease an individual’s chance of
inheriting the nest, function as triggers of winged sexual
development.
Assuming that termite societies evolved as a consequence
of immatures (false workers) following conditionally two
alternative reproductive tactics (dispersing or staying at the
nest), and that such alternative tactics (‘‘should I lay or
should I go’’) are common in solitary hemimetabolous
insects and are frequently associated with wing length
polyphenisms, one would expect that the mechanisms
underlying the development of alternative phenotypes are
evolutionarily conserved. Indeed, similar to one-piece
nesting termites, food quality or quantity and population
density (i.e. group sizes) are factors known to affect wing
development and reproductive physiology in a large
number of hemimetabolous insects, including aphids,
locusts and crickets (reviewed by Zera, 2003; Hartfelder &
Emlen, 2005; Braendle et al., 2006). Thus, the plethora of
results on reproductive physiology in cockroaches (for
review see Raikhel, Brown & Belle
´s, 2005) may heuristically
guide future in-depth studies on this major aspect of
sociality in termites.
The second non-reproductive caste in termites, the
soldiers, which are the only individuals in one-piece nesting
termites that lose the ability to reproduce, also result from
an environmentally induced polyphenism (Lu
¨scher, 1958;
Lenz, 1976; Korb, Roux & Lenz, 2003). Their development
is triggered by the presence of reproductives, food
availability and the size of a colony, while the presence of
soldiers inhibits further soldier differentiation (Miller, 1942;
Lu
¨scher, 1969; Springhetti, 1969; Haverty & Howard,
1981; Bordereau & Han, 1986; Liu et al., 2005a). This
results in colonies having a more or less constant proportion
of soldiers in relation to colony size (Haverty, 1977; Noirot
& Darlington, 2000). Whether predation pressure also may
influence soldier numbers is still controversially debated
(Noirot & Darlington, 2000).
(2) Endocrine control of caste development in
lower termites
The coordination of growth and tissue differentiation within
modular systems, such as the segmented body plan of
insects, requires both long-range and short-range signaling
by hormones. Ecdysteroids and juvenile hormones are the
key factors that drive an insect through larval and nymphal
moults and metamorphosis, and their chemistry, haemo-
lymph titres and mode of action in target tissues have been
investigated in great detail in a wide variety of hemi-
metabolous and holometabolous species (for reviews see
Nijhout, 1994; Riddiford, 1996; Hartfelder, 2000; Goodman
Judith Korb and Klaus Hartfelder302
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

& Granger, 2005; Henrich, 2005; Lafont et al., 2005).
Furthermore, models have been developed from hormone
titre analyses and hormone application experiments that
explain the transition from hemimetabolous to holometab-
olous development through the exploitation of the pronym-
phal stage and successive recruitment of epidermal cells into
imaginal discs (Truman & Riddiford, 1999, 2002). This
already highly successful life-history transition has been
further extended by the introduction of alternative
phenotypes (polyphenisms), which represent adaptive
responses to environmental changes without disrupting
successful genotypic combinations.
As detailed above, caste development in the lower
termites is highly plastic (with certain restrictions in
multiple-pieces nesting termites) (Fig. 2), making them
a challenge to any model on endocrine regulation in
hemimetabolous development, because moults are not only
progressive or stationary, but can even be regressive. Below
we review the literature on morphogenetic hormones and
their actions in termite caste development and try to
integrate these still fragmentary findings with the much
better studied cockroaches, within which the termites are
nested (Inward et al., 2007a; see also Lo et al., 2007), and
with general ideas on the evolution of insect metamorpho-
sis. The focus will be on the lower termites because caste
fate in the higher termites (Termitidae) is generally
determined rather early in development and may even
involve embryonic predisposition via maternally deposited
hormones (Lanzrein, Gentinetta & Fehr, 1985b). Further-
more, moults in higher termites are generally progressive
and there are no records of regressive moults in Termitidae.
(3) Juvenile hormone in progressive moults of
lower termites
As illustrated in Fig. 2, progressive moults of particular
importance to caste fate in lower termites are (i) the late
instar larvae to presoldier/soldier transition, (ii) the
transition from late instar larvae to nymph (first nymphal
moult, to the nymphal line), and (iii) from late instar larvae/
nymphs to a neotenic replacement reproductive. Of these,
the presoldier-soldier transition is at present the best
understood.
Although the induction of soldier differentiation by
juvenile hormone (JH) and juvenile hormone analogues
(JHAs) was one of the earliest findings in JH research
(Lu
¨scher, 1969; Howard & Haverty, 1979; for a recent
summary see Hrdy et al., 2006), the regulation of JH titres
and its mode of action is only now becoming clear, one of
the main problems being the role of the social environment.
When monitoring JH and ecdysteroid titres in isolated
larvae of the rhinotermitid Reticuliter mes flavipes, Okot-
Kotber et al. (1993) noted a gradual increase in levels of
ecdysteroids followed by a steeper increase in JH titre, both
peaking at day 9 after isolation, shortly before a presoldier
moult normally initiates in isolated pseudergates sensu lato of
this species. A radiochemical assay for measuring JH
biosynthesis rates was employed to monitor corpora allata
activity in pseudergates sensu lato of Reticulitermes flavipes that
were kept isolated from their nest in groups of 12–50
individuals (Elliott & Stay, 2008). The results showed an
increase in JH synthesis around the time point when some
of the pseudergates sensu lato were expected to develop into
neotenic reproductives or presoldiers. Also, while JH
synthesis rates were generally higher in presoldiers than in
pseudergates sensu lato or soldiers, corpora allata activity was
considerably lower in presoldiers than in pseudergates sensu
lato or neotenics in the pharate stage, that is during the
subsequent moulting phase. The authors interpreted this as
similar to the terminal stage of development in cockroaches.
A caveat in the interpretation of these results is that the
developmental profiles of JH are strongly modulated by
season, food availability and colony composition. In the
rhinotermitid species, Coptoter mes formosanus, meticulous
studies revealed a strong seasonal cycling of JH titres in
soldiers and pseudergates sensu lato (Liu et al., 2005b), similar
to earlier observations on corpora allata (CA) volumes in
the kalotermitid Kalotermes flavicollis (Lu
¨scher, 1972). JH titre
levels in pseudergates sensu lato were either negatively
affected by increasing the percentage of soldiers in
experimental groups (Mao et al., 2005; Park & Raina,
2005), or were positively affected by improved food or
temperature conditions (Liu et al., 2005a). Furthermore, the
time course of JH synthesis rates for Reticulitermes flavipes
pseudergates sensu lato kept isolated from their nest was
shown to be dependent on group size (Elliott & Stay, 2008).
Such social influences on factors controling caste develop-
ment are widespread and not limited to soldier develop-
ment in termites, but rather are a facet of the pleiotropic
functions of JH in insect development and reproduction.
For example, JH both prevents wing shedding and
precocious ovarian activity in immature alates of the
termopsid Zootermopsis angusticollis, yet stimulates oogenesis
in mated queens in this species (Brent, Schal & Vargo,
2005). This pleiotropic role of JH also became apparent in
a study on corpora allata activity in apterous and
brachypterous neotenic Reticulitermes flavipes females, where
an increase in the number of vitellogenic ovarioles was
accompanied by an increase in corpora allata activity
(Elliott & Stay, 2007). The critical questions, therefore, are
(a) how is the hormone titre regulated, and (b) which are the
molecular targets for JH action in termites?
The first inhibitors of JH biosynthesis were discovered in
cockroaches and were termed allatostatins (Woodhead et al.,
1989; Stay et al., 1991). Members of this large family of
neuropeptides have now been identified in all major orders
of insects (for review see Stay & Tobe, 2007). Allatostatin
immunoreactivity has recently been described in the
rhinotermid, Reticulitermes flavipes, showing that lateral and
medial neurosecretory cells in the brain innervate the
corpus allatum (Yagi et al., 2005). Furthermore, in vitro
incubation of termite corpora allata (CA) in the presence of
two cockroach allatostatins significantly inhibited JH pro-
duction (Yagi et al., 2005). Allatostatins, however, do not
only regulate JH synthesis but, due to their widespread
occurrence and origin as gut-brain peptides, can affect
a large suite of body functions, including food intake by
modulating gut contraction (Aguilar et al., 2003; Aguilar,
Maestro & Belles, 2006). The effects of food quality on
Developmental plasticity in termites 303
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

soldier induction (Liu et al., 2005a) may partially be
mediated through these routes. In conjunction with their
allatostatic effect, these peptides would then link reduced
individual food intake due to deteriorating colony con-
ditions to a reduction in the percentage of larvae that enter
the presoldier-soldier pathway. The pleiotropic functions of
such endogenous regulatory peptides could, thus, provide
a switch mechanism that links the social with the internal
environment. Links between the social and the internal
environment are still little understood, not only in termites
but also in the caste development of other social insects.
Even though inhibitory pheromones have long been
thought to be involved in the development of reproductives
and in the adjustment of caste ratios in termites (Lu
¨scher,
1964; Lefeuve & Bordereau, 1984; Noirot, 1991), the neural
substrates for their perception and transmission to the
endocrine system are unknown, except for the architecture
of the allatostatin-expressing neurons (Yagi et al., 2005).
After its release into the haemolymph, the extremely
lipophilic JH molecule must be bound to transport proteins
in order to reach target organs and to remain protected
against degradation by JH esterases and/or JH epoxide
hydrolases (reviewed by Goodman & Granger, 2005). JH
binding proteins have been identified by photoaffinity
labeling in haemolymph of the rhinotermitids Reticulitermes
flavipes, Coptotermes formosanus and the termopsid Zootermopsis
nevadensis (Okot-Kotber & Prestwich, 1991a,b). While these
proteins bind JH with high affinity, two general transport
and storage proteins, hexamerin 1 (Hex-1) and hexamerin 2
(Hex-2) have recently emerged as major candidates for
regulators of soldier development in the rhinotermitid
Reticulitermes flavipes. As in other insects, fat body expression
levels and haemolymph titres of these proteins are strongly
modulated during development and seem to be affected by
JH, especially in the case of Hex-2 (Scharf et al., 2005a,b;
Zhou et al., 2006b; Zhou, Faith & Scharf, 2006a; Zhou,
Traver & Scharf, 2007b). In turn, JH regulation of
hexamerin expression has a marked feedback effect on JH
availability to target tissues. Whereas hexamerins of other
insects are known to bind JH with low affinity (Tawfik et al.,
2006), they apparently do this in a peculiar way in R. flavipes.
Based on RNA interference (RNAi) results and Western
blotting with a JH-specific antiserum, Zhou et al. (2006b)
conclude that Hex-1 might covalently bind JH and that
these two proteins interact and form a sink for JH that, at
the colony level, would allow the fine-tuning of worker to
soldier caste ratios. This termite therefore seems to have
exploited an abundant haemolymph protein with pleiotro-
pic functions and co-opted it into a regulatory network of
social organization (Zhou et al., 2006a). A similar feedback
network involving a hormone of pleiotropic functions (JH)
and a phylogenetically equally old transport/storage protein
(vitellogenin) has recently been demonstrated to regulate
age polyethism in honey bee workers (Amdam et al., 2003,
2006).
While these examples illustrate the co-option of ground
plan components of insect developmental and reproductive
physiology, there are still a number of questions to be
answered. Firstly, the proposed covalent binding of JH is
peculiar and raises the following questions: (a) is Hex-1-
bound JH accessible to degradation by JH esterase or JH
epoxide hydrolase in haemolymph, (b) is it sequestered in
Hex-1-bound form into the fat body, and if so, what
happens there, (c) can JH-sensitive target tissues also
sequester Hex-1-bound JH, and what are the subsequent
effects and, most importantly, (d) is this mechanism of JH
sequestration a general property of lower termites or is it
restricted to Reticulitermes flavipes?
Additional questions have been raised by Hrdy et al.
(2006) who showed that the racemic JH-III used in the
above experiments on hexamerin function is not a particu-
larly strong inducer of soldier development in Reticulitermes
species, when compared to other JHAs. A feasible explana-
tion for the low activity of JH-III would be its well-known
lower metabolic stability in insect haemolymph, where it is
efficiently degraded by insect JH-esterases (Oakeshott et al.,
2005). In addition, the fraction of JH-III that is covalently
bound to Hex-1 in Reticulitermes flavipes haemolymph would
essentially be unavailable for physiological functions. An
important question to answer would, thus, be whether JHAs
modulate hexamerin expression similar to JH, that is, do
they transcriptionally mimic JH-III and do they also bind to
Hex-1?
Regarding intracellular JH clearing, the finding of a JHA-
induced expression of a cytochrome P450 transcript in the
fat body of the termopsid Hodotermopsis sjostedti (Cornette
et al., 2006) is interesting. This enzyme could be a candidate
for intracellular JH degradation, since cytochrome P450s
are not only general detoxifying enzymes, but also have
been specifically implicated in the metabolism of methyl
farnesoate to JH-III (Helvig et al., 2004). Similarly, Zhou
et al. (2007a) showed for the rhinotermitid Reticulitermes
flavipes that several fat-body-related P450s (CYP4) were
differentially expressed after JH treatments, and we found
that a cytochrome P450 enzyme was overexpressed in
female Cryptotermes secundus neotenics compared to false
workers (Weil, Rehli & Korb, 2007; see below).
(4) Molecular underpinnings of caste in lower
termites
Attempts to unravel the molecular basis of caste develop-
ment have also focused on the soldier differentiation
pathway, because of its relative ease of induction by JH
and JHA applications, and also because of the marked
morphological differences in soldiers. The first differential
gene expression screens performed on the termopsid
Hodotermopsis sjostedti (formerly H. japonica) (Miura et al.,
1999; Miura, 2001) led to the identification of a gene with
soldier-specific expression (SOL1) in the mandibular gland
of this dampwood termite. It encodes a putative member of
the lipocalin family that may be a soldier-specific secretory
product of this gland. In the termitid Hospitaliter mes
medioflavus, this gland develops from a disc-like structure
once a presoldier-differentiating moult has been induced by
a high JH titre (Miura & Matsumoto, 2000).
In a follow-up study performed as a differential display
reverse-transcription polymerase chain reaction (DDRT-
PCR) screen on RNA extracted from mandibles of
Judith Korb and Klaus Hartfelder304
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

Hodotermopsis sjostedti, Koshikawa et al. (2005) investigated
tissue-specific differentiation processes induced by the
application of a JHA. They confirmed the expected
overexpression of cuticle proteins in the mandibles of
developing soldiers and also revealed a set of putative
transcription and translation regulators (including a staufen
orthologue), an actin-binding protein possibly involved in
cellular morphogenesis and also a member of the aldehyde
dehydrogenase (Adh) family.
Macroarray screens on whole-body pseudergate sensu lato,
presoldier, soldier and nymphal RNA of the rhinotermitid
Reticulitermes flavipes (Scharf et al., 2003) detected 25
differentially expressed sequence tags (ESTs), 16 of which
represented orthologues (E-values <e
[6
). Pseudergates
sensu lato showed a strong overexpression of endosymbiont
cellulase genes and soldiers overexpressed cytoskeletal
proteins, especially ones related to skeletal muscle, and
a cytochrome oxidase I encoding gene. While these were
not surprising findings considering the functions of these
two castes, the overexpression of vitellogenin in presoldiers
is perplexing. Another remarkable point is that most of the
unknown genes (no BLAST matches) were overexpressed
in pseudergates sensu lato or in soldiers and, thus, may
represent novel genes typical for either of these castes, like
the SOL1 transcript of the termopsid H. sjostedti.
The molecular basis of development from pseudergates
sensu lato to reproductives (especially into neotenic replace-
ment reproductives, since this requires only a single moult)
is now under investigation in the kalotermitid Cryptotermes
secundus (Weil et al., 2007). Using a highly sensitive suppres-
sion subtractive hybridization strategy (Representational
Difference Analysis), this study led to the complete or
partial cloning of five differentially expressed genes, as
validated by quantitative real-time PCR. These genes (a
member of the esterase-lipase family, a putative beta-
glycosidase, a cytochrome P450 gene, vitellogenin, and an
unknown gene) were markedly overexpressed in female
neotenics. Surprising findings for vitellogenin transcripts
were the lack of sex-specificity (highly expressed both in
neotenic females and males) and lack of compartmentali-
zation (expressed in the head, thorax, and abdomen).
Vitellogenin has long been considered a sex-specific protein
exclusively required for oogenesis. This paradigm has lately
undergone considerable change, especially in social insects.
In the honey bee, vitellogenin has been shown to be involved
in the regulation of task performance, via repression of the JH
titre (Guidugli et al.,2005a), and in longevity (Amdam et al.,
2005; Corona et al., 2007), and vitellogenin expression has
also been detected in larval stages (Guidugli et al.,2005b). In
cockroaches, the induction of vitellogenin by JH is also not
restricted to adult females, but has been reported to occur in
preadult stages (Lanzrein, 1974; Cruz et al., 2003) and in
males (Mundall, Tobe & Stay, 1979).
Such differential expression of several metabolism-related
genes in termite caste development as well as the over-
expression of a cytochrome C oxidase subunit III in nymphs
and pseudergates sensu lato of the rhinotermitid Reticulitermes
santonensis (Lienard et al., 2006), may shed light on a largely
overlooked aspect of caste development, the importance of
metabolic regulation. In this respect, termites parallel the
honey bee, where similar observations have emerged from
several expression screens that can now be explored in
depth on the basis of genomic information (Cristino et al.,
2006).
Whereas these studies are informative on metabolic
regulation and on the differential expression of structural
genes they tell us relatively little about patterning
mechanisms involved in the shaping of caste-specific
structures. In particular, the shutting down of wing
development in termite castes (true workers, neotenics and
soldiers) has not yet been investigated at the molecular level,
although studies have been initiated to reveal underlying
cellular mechanisms (Miura et al., 2004).
V. PERSPECTIVES ON HORMONAL
REGULATION, CASTE DEVELOPMENT AND
TERMITE EVOLUTION
(1) Ecdysteroids and insulin signaling
Whereas the role of JH in termite caste development has
received much attention due to its potential in pest control
strategies, ecdysteroids have only rarely been studied,
despite their importance in moulting cycles and meta-
morphosis. JH and ecdysteroid titres rise concomitantly in
isolated pseudergates sensu lato of the rhinotermitid
Reticulitermes flavipes (Okot-Kotber et al., 1993), and the
deposition and sclerotization of cuticle in the mandible of
soldiers is dependent on an interaction between these two
hormones (Okot-Kotber, 1983).
For molecular studies, the ecdysteroid signaling pathway
may be more productive since, in distinction to JH
signaling, the ecdysone receptor(s) (EcRs) and their
dimerization partners, the orphan nuclear receptors of the
rexinoid/ultraspiracle (RXR/USP) family have been iden-
tified in a large number of species (Thummel, 1996;
Henrich, 2005). An EcR-A isoform has recently been
identified in the cockroach Blattella germanica, and RNAi
experiments have revealed that it is involved in adult-
specific developmental processes, including wing develop-
ment (Cruz et al., 2006). Similar functional RNAi analyses
have also been performed for the RXR/USP orthologue of
this cockroach showing that BgRXR knockdown arrests the
nymphal to adult moult (Martin et al., 2006). The
identification of Drosophila EcR and RXR/USP orthologues
in cockroaches and the results of the functional assays
demonstrate that this nuclear receptor pathway is highly
conserved in hemimetabolous and holometabolous insects.
Furthermore, besides being a dimerization partner for EcR,
the RXR/USP protein has also been suggested as a possible
JH receptor in Drosophila melanogaster (Jones & Sharp, 1997;
Xu et al., 2002; Jones et al., 2006) and also in the honey bee
(Barchuk, Maleszka & Simoes, 2004). Alternatively, with
their low-affinity binding characteristics for specific ligands,
the RXR/USP nuclear receptors could function as lipid
sensors, and thus provide a direct link between nutritional
status and moulting or metamorphosis induction (Chawla
et al., 2001).
Developmental plasticity in termites 305
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

This link between the classic morphogenetic hormones,
JH and ecdysteroids, and nutritional status could also
involve another phylogenetically old signaling pathway
which functions as a local growth regulator, the insulin-/-
insulin-like signaling (IIS) and the associated target of
rapamycin (TOR) pathway (Brogiolo et al., 2001; Edgar,
2006). These flexible response systems to the endogenous
nutrient milieu regulate cell growth and cell division, and
thus affect organ and body size. Furthermore, they have
been shown to interact directly with the function of the
prothoracic gland (Colombani et al., 2005; Mirth, Truman
& Riddiford, 2005). Through this gateway, an important
aspect of metamorphosis, the critical size threshold,
becomes directly amenable to natural selection.
In the honey bee, the availability of the complete genome
sequence has facilitated the annotation of insulin/insulin-
like peptides, insulin receptors, TOR, as well as most of
the downstream elements of these signaling pathways (The
Honey Bee Genome Sequencing Consortium, 2006). The
expression of some of the IIS components has been shown
to be strongly affected by the diet provided to honey bee
larvae (Wheeler, Buck & Evans, 2006), in correspondence
with the caste-specific differential expression profiles found
for an insulin receptor (Azevedo & Hartfelder, 2008);
interference with TOR signaling by rapamycin administra-
tion or RNAi has been shown to affect queen/worker
differentiation directly (Patel et al., 2007). The high
conservation of IIS components should make this pathway
an interesting target for studies on termite caste differen-
tiation. This group of social insects is of singular interest
because of its dietary specialization, especially in those
lower termite species which continuously reside in and feed
on a single piece of dry or damp wood and for which food
availability has been shown to trigger different caste
developmental trajectories (see above). The question of
critical size in the context of nutrition and caste divergence
and its consequences for direct and indirect fitness should,
thus, be of singular importance to the highly flexible systems
of caste determination in the lower termites.
(2) What is the status of the larval stages?
A major problem in termite developmental biology is
actually a terminological one. While they are unquestion-
ably hemimetabolous insects, their early postembryonic
instars are nevertheless denominated as larvae and not as
nymphs, the typical terminology for growth stages of
solitary hemimetabolans. In termites, the term nymph is
reserved for later developmental stages that exhibit pro-
gressively growing wing pads and, thus, are gradually
committed to winged dispersal (Noirot, 1990). In some
termites only the early stages (first to third instars) are
termed larvae, but de facto, the false workers of the one-piece
nesting lower termites can also be considered larvae, as long
as they have an unsclerotized cuticle, simple mandibles and
do not show external wing pads or compound eyes. The
traditional terminology of termite developmental stages has
been called into question by the discovery of an early larval
stage with wing buds in the genus Termitogeton (Rhinotermi-
tidae). Their wing buds regress in successive moults until
they reappear in the single nymphal stage (Parmentier &
Roisin, 2003). A second major problem is how to describe
stationary and regressive moults in the lower termites.
From their morphological characters and possibly also
their developmental status, the early postembryonic stages
of termites (Fig. 3) are more similar to the grub-like larvae of
lepidopterans or coleopterans than to the more adult-like
cockroach or locust nymphs. They are holometabolan-like
with respect to the developmental status of their wing and
eye primordia and the cuticle structure, and they are
hemimetabolan-like with respect to the body tagmata (more
pronounced thorax-abdomen division). In terms of evolu-
tionary trajectories, the postembryonic stages of termites
(probably including parts of the false worker/pseudergates
sensu lato stage, which is not a single developmental stage but
a composite resulting from progressive, stationary and
regressive moulting events) could actually be considered as
equivalent to the pronymphal stage of Orthoptera and
Blattodea.
This hypothesis would be consistent with the scenario
proposed by Truman & Riddiford (1999, 2002) on the
transition from hemimetabolous to holometabolous devel-
opment (but see also Heming, 2003; Minelli et al., 2006; for
a wider discussion and alternative viewpoints). In termites
external wing buds only start to grow during the nymphal
moults (but are present in Termitogeton; Parmentier & Roisin,
2003), and larval termites do not have compound eyes but
only primordia (Miura, 2005). In holometabolous insects,
wings develop from imaginal discs and the eye primordia of
termites resemble the eye imaginal discs of Manduca sexta
(Champlin & Truman, 1998; MacWhinnie et al., 2005). The
pseudergates sensu lato, which can comprise several instars,
Fig. 3. Developmental stages of the drywood termite Cryptotermes secundus. (A) ‘Dependent larva’, (B) false workers (larval and
nymphal instars).
Judith Korb and Klaus Hartfelder306
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

could, thus, be conceived as a platform for flexibility in
developmental decisions, leading to soldiers, alates, or
neotenic reproductives.
So far, this hypothesis remains to be tested. Some data
hint at this scenario, such as a change in mandible shape
and an increased degree of sclerotization observed in
rhinotermitids following JHA application (Lenz, 1976; Lelis
& Everaerts, 1993), similar to results obtained in locust
pronymphs (Truman & Riddiford, 1999), and the formation
of soldier-alate intermediates (Lelis & Everaerts, 1993;
Koshikawa, Matsumoto & Miura, 2002; Miura, 2005) in
Hodotermopsis sjostedti (Termopsidae), which can be inter-
preted as a trade-off between wing/eye development and
that of defence structures on the head.
A revision of termite embryogenesis is clearly required.
There are excellent descriptions on the embryonic devel-
opment of Kalotermes flavicollis and Zootermopsis nevadensis
(Striebel, 1960) and Crypotermes brevis (Kawanishi, 1975), but
these do not address the question of whether termites
undergo embryonic moults. In locusts, such moults are
important for the appearance of a pronymphal stage, before
eclosion of the larva from the egg. Identification of
a pronymphal stage would require detailed JH and
ecdysteroid titre measurements to determine whether there
are differences in the titre profiles between the larval and
nymphal/presoldier stages and whether and how these titre
patterns differ from those established for a well-studied
basal hemimetabolan insect, the cockroach Nauphoeta cinerea
(Lanzrein et al., 1985a). An alternative would be to
investigate candidate genes for gene regulatory networks
underlying moulting and metamorphosis, in an approach
similar to that taken by Abouheif & Wray (2002) for
understanding the prevention of wing development in ant
workers. Both approaches should enable us to clarify the
position of the stationary and especially the peculiar
regressive moults in lower termites.
(3) Regressive moults in lower termites,
a major enigma
The regulatory network for the stationary moults of
pseudergates sensu lato may be equivalent to that of
supernumerary moults of solitary hemimetabolous insects,
which are associated with little growth and no differentiation.
The driving force for stationary moults in the lower termite
pseudergates sensu lato may actually be the wear of their
mandibles (Roisin & Lenz, 1999). Renewing this structure in
a moult would allow large immatures (i.e. pseudergates sensu
lato) to remain for an extended period in the nest with
a chance of eventually inheriting the natal breeding position.
Regressive moults are observed in the nymphal-alate
transition, where nymphs that did reach the alate moult in
one year regress to the false worker stage, accompanied by
wing pad reduction (pseudergates sensu stricto)(Korb&
Katrantzis, 2004). So why do these individuals not remain
as nymphs in the nest and develop into alates early in the
next swarming season? This is especially interesting as there
seem to be developmental deadlines for each nymphal
instar; individuals that fail to reach these deadlines cannot
become alates.
There might be two explanations. First, a regressive
moult in the nymphal-alate transition could be a conse-
quence of wing pad mutilation. Mutilated wing buds were
found in several lower termite species and it was
hypothesized that they were the result of manipulations
by siblings or parents (Zimmermann, 1983; Myles, 1988;
Roisin, 1994, 2006; Miura et al., 2004) similar to the
mutilation of gemmae in some queenless ponerine ants
(Peeters & Higashi, 1989; Ramaswamy et al., 2004).
However, as nobody has actually observed the process of
mutilation in nature, its causes are unknown. Results for
the drywood termite Cryptotermes secundus suggestthatsuch
damage is an artefact of handling conditions (Korb, 2005).
From a theoretical point of view there should be no
selective advantage in monogamous colonies to sibling or
parental manipulation (Korb, 2005), especially when false
workers do not provide much help in raising siblings, as
recent data for C. secundus suggest (Korb, 2007b). Why then
should parents or siblings manipulate their nestmates to
stay in the nest, when these nestmates do not provide
costly brood care and when breeding opportunities outside
the nest are not limited by intraspecific competition?
Roisin (2006) suggested that intracolonial competition
might exist for some high-quality resources or brood care.
But why then can individuals, which do not disperse under
abundant food conditions, can become winged sexuals
when food availability is reduced? If individuals are
capable to develop into alates under reduced food
conditions they should be even more capable to become
alates under abundant food conditions. These consider-
ations suggest that individuals are staying voluntarily
under abundant food conditions.
Conflicts among totipotent individuals over dispersal
might occur in fused termite colonies with a within-colony
relatedness below 0.5 and if dispersal is a better option then
staying at the nest. Further research must show whether
these latter conditions are met in any lower termite species
under natural conditions. So far, although scant, the
available field data suggest that staying in the nest with
a chance to inherit the colony is not an inferior reproductive
tactic compared to leaving the colony as winged sexual
(Korb, 2007b, 2008).
An alternative explanation is that regression might be
part of an ‘honest signal’ in a test for alate competence. All
individuals have to start from the same stage as apterous
individuals. From this stage, only those individuals that
reach the developmental deadlines are the most competent
for alate development. Individuals that are less competent
would stay for another year and try to gain enough
resources for the subsequent nuptial flight.
At the proximate level, the general paradigm for
hormonal regulation in metamorphic moults provides
a series of predictions for regressive moults in the
nymphal-alate transition, since the nature of the subsequent
moult is always determined during the preceding inter-
moult period (Riddiford, 1994). A critical factor in the last
preimaginal stage is the JH level at the rise of the moult-
inducing ecdysteroid peak. We predict that only nymphs
with a sufficiently suppressed level of JH will become alates,
while an above-threshold JH level during the ecdysteroid
Developmental plasticity in termites 307
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

peak could block wing bud expansion and reverse the
moulting type in terms of cuticle protein expression.
VI. CONCLUSIONS
(1) ‘‘Pseudergates sensu lato’’ are a fascinating example of
ample developmental plasticity with far-reaching conse-
quences in terms of the ecological and evolutionary success
of termites. We compile here current knowledge on the
ecology and life history of lower termites, and set this into
a framework of developmental biology, especially the
endocrine regulation underlying caste differentiation.
(2) The so-called one-piece nesting termites with their
wood-nesting life style are considered a basal group (Roisin,
2000; Thorne & Traniello, 2003; Korb, 2007a). From
phylogenetic data on termite families it is not possible to
reconstruct the evolution of true workers unambiguously
(Thompson et al., 2000, 2004; Grandcolas & D’Haese,
2004; Inward, Vogler & Eggleton, 2007b). Other results
leave less doubt about the basal position of false workers,
and thus of the one-piece nesting life style (reviewed in
Korb, 2007a): (i) a recently published comprehensive
phylogenetic analysis on Dictyopterans places the termites
within the cockroaches, Blattodea, as a sister group to the
woodroaches, Cryptocercidae (Inward et al., 2007a; see also
Lo et al., 2007) which have a similar life style to the one-
piece nesting termites; (ii) the presence of true workers in
the basal group Mastotermes darwiniensis, which are not
equivalent to the true workers of other multiple-pieces
nesting termites (Parmentier, 2006), suggests at least two
independent origins of true workers, implying a basal
position for one-piece nesting termites.
(3) The one-piece nesting termites show the greatest
flexibility in developmental decisions related to caste.
Developmental plasticity is, thus, a basal character in
termites. Termite caste allometries (as e.g. shown by
Koshikawa et al., 2002) are results of phenotypic plasticity
that does not involve genetic change. The demonstration of
major epigenetic effects on honey bee caste fate through
differential genome methylation (Kucharski et al., 2008)
furthermore shows that social insect polyphenisms can be
interpreted as split developmental reaction norms in terms
of Schlichting & Pigliucci’s (1998) approach to phenotypic
evolution.
(4) We propose that a pronymphal stage similar to that of
the holometabolan clade (Truman & Riddiford, 1999, 2002)
may have been translocated from within the egg to
a prolonged postembryonic stage at the beginning of
termite evolution, in association with the suppression of
wing development due to a burrowing feeding habit in
a woodroach-like ancestor. Under this scenario, the
prolonged postembryonic stage, exemplified by the false
worker/pseudergates sensu lato stage, provides a platform for
developmental trajectories into wingless and winged
reproductives, which present alternative breeding tactics,
and, probably as a subsequent evolutionary step, also
development into soldiers. The extension of the pronym-
phal stage of basal hemimetabolans into a sequence of
postembryonic instars in such a termite ancestor would then
be considered a pre-adaptation for developmental plasticity
contingent on the nutritional and social environment. A
next step could then have been the assimilation of this
contingency into a morphogenetic program that directs the
development of alternative phenotypes via endocrine
control.
(5) The role of JH has long been explored in proposals for
termite pest control strategies and based on such data and
hormone titre measurements we propose a scenario of
endocrine regulation of caste development that incorpo-
rates critical periods and response thresholds (Nijhout,
1994; Hartfelder & Emlen, 2005). Recent results on the
differential expression of hexamerin genes in the genus
Reticulitermes can be interpreted as a co-option of a general
storage protein that belongs to an ancient family of
arthropod proteins (Burmester, 2002).
(6) Even though the focus of the current review is on
lower termites, the castes of higher termites can be easily
integrated into this framework. The restricted plasticity in
the apterous line and their much earlier commitment to
reproductive castes can be conceived as a change in the
timing of hormone-dependent determination steps. The
higher JH levels observed in Macrotermes michaelseni eggs
dedicated to become reproductives (Lanzrein et al., 1985b)
indicate that the lines split at at least two stages, an early
one in the embryonic phase, for the apterous/nymphal
decision, and a later one in the early nymphal stages, for the
worker/soldier decision (Fig. 2).
VII. ACKNOWLEDGMENTS
We would like to thank Michael Lenz, Paul Eggleton and
two anonymous referees for their helpful comments on this
manuscript. We also acknowledge financial support from
the Deutsche Forschungsgemeinschaft (DFG, KO 1895/6)
and from a Brazilian/German (CAPES/DAAD) coopera-
tion program (PROBRAL 261/07).
VIII. REFERENCES
ABE, T. (1987). Evolution of life types in termites. In Evolution and
Coadaptation in biotic communities (ed. S. Kawano, J. H. Connell
and T. Hidaka), pp. 125–148. University of Tokyo Press, Tokyo.
ABE, T. (1990). Evolution of worker caste in termites. In Social
Insects and the Environment (eds. G. K. Veeresh, B. Mallik and
C. A. Viraktamath), pp. 29–30. Oxford & IBH, New Delhi.
ABOUHEIF,E.&WRAY, G. A. (2002). Evolution of the genetic net-
work underlying wing polyphenism in ants. Science 297, 249–252.
AGUILAR, R., MAESTRO,J.L.&BELLES, X. (2006). Effects of
myoinhibitory peptides on food intake in the German
cockroach. Physiological Entomology 31, 257–261.
AGUILAR, R., MAESTRO, J. L., VILAPLAN, L., PASCUAL, N., PIULACHS,
M. D. & BELLES, X. (2003). Allatostatin gene expression in brain
and midgut, and activity of synthetic allatostatins on feeding-
related processes in the cockroach Blattella germanica.Regulatory
Peptides 115, 171–177.
ALEXANDER, R. D. (1974). The evolution of social behavior. Annual
Reviews of Ecology and Systematics 5, 325–383.
Judith Korb and Klaus Hartfelder308
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

AMDAM,G.V.,AASE,A.L.T.O.,SEEHUUS, S. C., FONDRK, M. K.,
NORBERG,K.&HARTFELDER, K. (2005). Social reversion of
immunosenescence in honey bee workers. Experimental Gerontology
40, 939–947.
AMDAM,G.V.,CSONDES, A., FONDRK,M.K.&PAG E, R. E. (2006).
Complex social behaviour derived from maternal reproductive
traits. Nature 439, 76–78.
AMDAM,G.V.,NORBERG, K., HAGEN,A.&OMHOLT, S. W. (2003).
Social exploitation of vitellogenin. Proceedings of the National
Academy of Sciences of the United States of America 100, 1799–1802.
AZEVEDO,S.V.&HARTFELDER, K. (2008). The insulin signaling
pathway in honey bee (Apis mellifera) caste development-
differential expression of insulin-like peptides and insulin
receptors in queen and worker larvae. Journal of Insect Physiology,
(in press; doi 10.1016/j.jinsphys.2008.04.009).
BARCHUK, A. R., MALESZKA,R.&SIMOES, Z. L. P. (2004). Apis
mellifera ultraspiracle: cDNA sequence and rapid up-regulation by
juvenile hormone. Insect Molecular Biology 13, 459–467.
BORDEREAU,C.&HAN, S. H. (1986). Stimulatory influence of the
queen and king on soldier differentiation in the higher termites
Nasutitermes lujae and Cubiter mes fungifaber.Insectes Sociaux 33,
296–305.
BRAENDLE, C., DAVI S , G. K., BRISSON,J.A.&STERN, D. L. (2006).
Wing dimorphism in aphids. Heredity 97, 192–199.
BRAKEFIELD, P. M. (2006). Evo-devo and constraints on selection.
Trends in Ecology and Evolution 21, 362–368.
BRENT, C. S., SCHAL,C.&VARGO, E. L. (2005). Endocrine changes
in maturing primary queens of Zootermopsis angusticollis.Journal of
Insect Physiology 51, 1200–1209.
BROGIOLO, V., STOCKER, H., IKEYA, T., RINTELEN, F., FERNANDEZ,R.
&H
AFEN, E. (2001). An evolutionarily conserved function of the
Drosophila insulin receptor and insulin-like peptides in growth
control Current Biology,11, 213–221.
BUCHLI, H. R. (1958). L’origine des castes et les potentialite
´s
ontoge
´niques des termites europe
´ens du genre Reticulitermes
Holmgren. Annales des Sciences Naturelles Zoologie et Biologie Animale
11, 267–429.
BURMESTER, T. (2002). Origin and evolution of arthropod
hemocyanins and related proteins. Journal of Comparative
Physiology, Series B 172, 95–107.
CARROLL, S. B., GRENIER,J.K.&WEATHERBEE, S. D. (2005). From
DNA to Diversity - Molecular Genetics and the Evolution of Animal
Design, 2 edition. Blackwell, Oxford.
CHAMPLIN,D.T.&TRUMAN, J. W. (1998). Ecdysteroids govern two
phases of eye development during metamorphosis of the moth,
Manduca sexta. Development 125, 2009–2018.
CHAWLA, A., REPA,J.J.,EVANS,R.M.&MANGELSDORF,D.J.
(2001). Nuclear receptors and lipid physiology: opening the
X-files. Science 294, 1866–1870.
COLOMBANI, J., BIANCHINI, L., LAYALLE, S., PONDEVILLE, E.,
DAUPHIN-VILLEMANT,C.,ANTONIEWSKI,C.,CARRE
´,C.,NOSELLI,
S. & LEOPOLD, P. (2005). Antagonistic actions of ecdysone
and insulins determine final size in Drosophila.Science 310, 667–670.
CORNETTE, R., KOSHIKAWA, S., HOJO, M., MATSUMOTO,T.&
MIURA, T. (2006). Caste-specific cytochrome P450 in the damp-
wood termite Hodotermopsis sjostedti (Isoptera, Termopsidae). Insect
Molecular Biology 15, 235–244.
CORONA, M., VELARDE, R. A., REMOLINA, S., MORAN-LAUTER, A.,
WANG, Y., HUGHES,K.A.&ROBINSON, G. E. (2007).
Vitellogenin, juvenile hormone, insulin signaling, and queen
honey bee longevity. Proceedings of the National Academy of Science of
the United States of America 104, 7128–7133.
CRISTINO, A. S., NUNES, F. M., LOBO, C. H., BITONDI,M.M.G.,
SIMO
˜ES, Z. L. P., FONTOURA COSTA, L., LATTORFF,H.M.G.,
MORITZ,R.F.A.,EVA N S,J.D.&HARTFELDER,K.(2006).
Caste development and reproduction: a genome-wide analysis
of hallmarks of insect eusociality. Insect Molecular Biology 15,
703–714.
CRUZ , J., MANE-PADROS, D., BELLES,X.&MARTIN, D. (2006).
Functions of the ecdysone receptor isoform-A in the hemi-
metabolous insect Blattella germanica revealed by systemic RNAi
in vivo.Developmental Biology 297, 158–171.
CRUZ , J., MARTIN, D., PASCUAL, N., MAESTRO, J. L., PIULACHS,M.D.
&B
ELLES, X. (2003). Quantity does matter: Juvenile hormone
and the onset of vitellogenesis in the German cockroach. Insect
Biochemistry and Molecular Biology 33, 1219–1225.
DEJEAN, A., DURAND,J.L.&BOLTON, B. (1996). Ants inhabiting
Cubitermes termitaries in African rain forests. Biotropica 28, 701–713.
DELIGNE, A., QUENNEDY,A.&BLUM, M. S. (1981). The enemies
and defense mechanisms of termites. In Social Insects vol. 2
(ed. H. R. Hermann), pp. 1–76. New York Academic Press,
New York.
EDGAR, B. A. (2006). How flies get their size: genetics meets
physiology. Nature Review Genetics 7, 907–916.
EGGLETON, P. (2001). Termites and trees: a review of recent
advances in termite phylogenetics. Insectes Sociaux 48, 187–193.
ELLIOTT,K.L.&STAY, B. (2007). Juvenile hormone synthesis as
related to egg development in neotenic reproductives of the
termite Reticulitermes flavipes, with observations on urates in fat
body. General and Comparative Endocrinology 152, 102–110.
ELLIOTT,K.L.&STAY, B. (2008). Changes in juvenile hormone
synthesis in the termite Reticulitermes flavipes during development
of soldiers and neotenic reproductives from groups of isolated
workers. Journal of Insect Physiology 54, 492–500.
GOODISMAN,M.A.D.&CROZIER, R. H. (2002). Population and
colony genetic structure of the primitive termite Mastotermes
darwiniensis.Evolution 56, 70–83.
GOODMAN,W.G.&GRANGER, N. A. (2005). The juvenile
hormones. In Comprehensive Insect Molecular Science, vol. 3 (ed. L. I.
Gilbert, K. Iatrou and S. Gill), pp. 319–408. Elsevier, Oxford.
GRANDCOLAS,P.&D’HAESE, C. (2004). The origin of a ’true’
worker caste in termites: phylogenetic evidence is not decisive.
Journal of Evolutionary Biology 15, 885–888.
GRASSE
´,P.P.&NOIROT, C. (1947). Le polymorphisme social du
termite a cou jaune (Kalotermes flavicollis F.) Les faux-ouvriers ou
pseudergates et les mues regressive. Comptes Rendus de Academie des
Sciences 214, 219–221.
GROSS, M. R. (1996). Alternative reproductive strategies and
tactics: diversity within sexes. Trends in Ecology and Evolution 11,
92–98.
GUIDUGLI, K. R., NASCIMENTO, A. M., AMDAM,G.V.,BARCHUK,
A. R., OMHOLT,S.W.,SIMO
˜ES,Z.L.P.&HARTFELDER,K.
(2005a). Vitellogenin regulates hormonal dynamics in the
worker caste of a eusocial insect. FEBS Letters 579, 4961–4965.
GUIDUGLI, K. R., PIULACHS, M. D., BELLES, X., LOURENCO,A.P.&
SIMOES, Z. L. P. (2005b). Vitellogenin expression in queen
ovaries and in larvae of both sexes of Apis mellifera.Archives of
Insect Biochemistry and Physiology 59, 211–218.
HAMILTON,W.D.&MAY, R. M. (1977). Dispersal in stable habitats.
Nature 269, 578–581.
HARTFELDER, K. (2000). Arthropoda - Insecta: Caste differentia-
tion. In Progress in Developmental Endocrinology, vol. X part B.
Reproductive Biology of Invertebrates (ed. A. Dorn), pp. 185–204.
Wiley, Chichester.
Developmental plasticity in termites 309
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

HARTFELDER,K.&EMLEN, D. J. (2005). Endocrine control of
insect polyphenism. In Comprehensive Insect Molecular Science, vol. 3
(ed. L. I. Gilbert, K. Iatrou and S. Gill), pp. 651–703. Elsevier,
Oxford.
HAVERTY, M. I. (1977). The proportion of soldiers in termite colo-
nies: a list and a bibliography (Isoptera). Sociobiology 2, 199–216.
HAVERTY, M. I. (1979). Soldier production and maintenance of
soldier proportions in laboratory experimental groups of
Coptotermes formosanus Shiraki. Insectes Sociaux 26, 69–84.
HAVERTY,M.I.&HOWARD, R. W. (1981). Production of soldiers
and maintenance of soldier proportions by laboratory experi-
mental groups of Reticulitermes flavipes (Kollar) and Reticulitermes
virginicus (Banks) (Isoptera: Rhinotermitidae). Insectes Sociaux 28,
32–39.
HAYASHI, Y., LO, N., MIYATA,H.&KITADE, O. (2007). Sex-linked
genetic influence on caste determination in a termite. Science
318, 985–987.
HELVIG, C., KOENER, J. F., UNNIHATHAN,G.C.&FEYEREISEN,R.
(2004). CYP15A1, the cytochrome P450 that catalyzes epoxi-
dation of methylfarnesoate to juvenile hormone III in cockroach
corpora allata. Proceedings of the National Academy of Sciences of the
United States of America 101, 4024–4029.
HEMING, B. S. (2003). Insect Development and Evolution. Cornell
University Press, Ithaca and London.
HENRICH, V. C. (2005). The ecdysteroid receptor. In Comprehensive
Insect Molecular Science, vol. 3 (ed. L. I. Gilbert, K. Iatrou and
S. Gill), pp. 243–285. Elsevier, Oxford.
HOWARD,R.W.&HAVERTY, M. I. (1979). Termites and juvenile
hormone analogues: a review of methodology and observed
effects. Sociobiology 4, 269–278.
HOWSE, P. E. (1968). On the division of labour in the primitive
termite Zootermopsis nevadensis (Hagen). Insectes Sociaux 15, 45–50.
HRDY, I., KULDOVA, J., HANUS,R.&WIMMER, Z. (2006). Juvenile
hormone III, hydroprene and a juvenogen as soldier caste
differentiation regulators in three Reticulitermes species: potential
of juvenile hormone analogues in termite control. Pest
Management Science 62, 848–854.
INWARD, D., BECCALONI,G.&EGGLETON, P. (2007a). Death of an
order: a comprehensive molecular phylogenetic study confirms
that termites are eusocial cockroaches. Biology Letters 3, 331–335.
INWARD,D.J.G.,VOGLER,A.P.&EGGLETON, P. (2007b). A
comprehensive phylogenetic analysis of termites (Isoptera)
illuminates key aspects of their evolutionary biology. Molecular
Phylogenetics and Evolution 44, 953–967.
JONES, G., JONES, D., TEAL, P., SAPA,A.&WOZNIAK, M. (2006). The
retinoid-X receptor ortholog, ultraspiracle, binds with nano-
molar affinity to an endogenous morphogenetic ligand. FEBS
Journal 273, 4983–4996.
JONES,G.&SHARP, P. A. (1997). Ultraspiracle: An invertebrate
nuclear receptor for juvenile hormones. Proceedings of the National
Academy of Sciences of the United States of America 94, 13499–13503.
KAWANISHI, C. Y. (1975). Embryonic development of the drywood
termite Cryptotermes brevis. Hawaii Agricultural Experimental Station,
College of Agriculture, UNiversity of Hawaii, Technical Bulletin 95,35pp.
KOKKO,H.&EKMAN, J. (2002). Delayed dispersal as a route to
breeding: Territorial inheritance, safe havens, and ecological
constraints. American Naturalist 160, 462–484.
KORB, J. (2005). Regulation of sexual development in termites:
mutilation, pheromonal manipulation or honest signal? Natur-
wissenschaften 92, 45–49.
KORB, J. (2007a). Termites. Current Biology 17, R995–999.
KORB, J. (2007b). Workers of a drywood termite do not work.
Frontiers in Zoology 4, 7 pp.
KORB, J. (2008). The ecology of social evolution in termites. In
Ecology of social evolution (ed. J. Korb and J. Heinze), pp. 151–174.
Springer, Heidelberg.
KORB,J.&FUCHS, A. (2006). Termites and mites - adaptive
behavioural responses to infestation? Behaviour 143, 891–907.
KORB,J.&KATRANTZIS, S. (2004). Influence of environmental
conditions on the expression of the sexual dispersal phenotype
in a lower termite: implications for the evolution of workers in
termites. Evolution & Development 6, 342–352.
KORB,J.&LENZ, M. (2004). Reproductive decision-making in the
termite, Cryptotermes secundus (Kalotermitidae), under variable
food conditions. Behavioral Ecology 15, 390–395.
KORB, J., ROUX,E.A.&LENZ, M. (2003). Proximate factors
influencing soldier development in the basal termite Cryptotermes
secundus (Hill). Insectes Sociaux 50, 299–303.
KORB,J.&SCHMIDINGER, S. (2004). Help or disperse? Cooperation
in termites influenced by food conditions. Behavioral Ecology and
Sociobiology 56, 89–95.
KORB,J.&SCHNEIDER, K. (2007). Does kin structure explain the
occurrence of workers in a lower termite? Evolutionary Ecology 21,
817–828.
KOSHIKAWA, S., CORNETTE, R., HOJO, M., MAEKAWA, K.,
MATSUMOTO,T.&MIURA, T. (2005). Screening of genes
expressed in developing mandibles during soldier differentiation
in the termite Hodotermopsis sjostedti.FEBS Letters 579, 1365–
1370.
KOSHIKAWA, S., MATSUMOTO,T.&MIURA, T. (2002). Morphomet-
ric changes during soldier differentiation of the damp-wood
termite Hodotermopsis japonica (Isoptera, Termopsidae). Insectes
Sociaux 49, 245–250.
KUCHARSKI, R., MALESZKA, J., FORET,S.&MALESZKA, R. (2008).
Nutritional control of reproductive status in honeybees via DNA
methylation. Science 319, 1827–1830.
LAFAGE,J.P.&NUTTING, W. L. (1978). Nutrient dynamics of
termites. In Production ecology of ants and termites (ed. M. V. Brian),
pp. 165–232. Cambridge University Press, Cambridge.
LAFONT, R., DAUPHIN -VILLEMANT, C., WARREN,J.T.&REES,H.
(2005). Ecdysteroid chemistry and biochemistry. In Comprehensive
Insect Molecular Science, vol. 3 (ed. L. I. Gilbert, K. Iatrou and S.
Gill), pp. 125–195. Elsevier, Oxford.
LANZREIN, B. (1974). Influence of a juvenile hormone analog on
vitellogenin synthesis and oogenesis in larvae of Nauphoeta cinerea.
Journal of Insect Physiology 20, 1871–1885.
LANZREIN, B., GENTINETTA, V., ABEGGLEN, H., BAKER,F.C.,
MILLER,C.A.&SCHOOLEY, D. A. (1985a). Titers of ecdysone,
20-hydroxyecdysone and juvenile hormone III throughout the
life cycle of a hemimetabolous insect, the ovoviviparous
cockroach Nauphoeta cinerea.Experientia 41, 913–917.
LANZREIN, B., GENTINETTA,V.&FEHR, R. (1985b). Titres of
juvenile hormone and ecdysteroids in reproduction and eggs of
Macrotermes michaelseni: Relation to caste determination. In Caste
Differentiation in Social Insects (ed. J. A. L. Watson, B. M. Okot-
Kotber and C. Noirot), pp. 307–327. Pergamon Press, Oxford.
LEFEUVE,P.&BORDEREAU, C. (1984). Soldier formation regulated
by a primer pheromone from the soldier frontal gland in
a higher termite, Nasutitermes lujae.Proceedings of the National
Academy of Sciences of the United States of America 81, 7665–7668.
LELIS,A.T.&EVERAERTS, C. (1993). Effects of juvenile-hormone
analogs upon soldier differentiation in the termite Reticulitermes
Judith Korb and Klaus Hartfelder310
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

santonensis (Rhinotermitidae, Heterotermitinae). Journal of Mor-
phology 217, 239–261.
LENZ, M. (1976). The dependence of hormone effects in caste
determination on external factors. In Phase and Caste Deter mination
in Insects. Endocrine Aspects (ed. M. Lu
¨scher), pp. 73–90. Pergamon
Press, Oxford.
LENZ, M. (1994). Food resources, colony growth and caste
development in wood-feeding termites. In Nourishment and
Evolution in Insect Societies (ed. J. Hunt and C. A. Nalepa), pp.
159–210. Westview Press, Boulder.
LIE
´NARD, M. A., LASSANCE, J.-M. X. S., PAULMIER, I., PICIMBON, J.-F.
&L
O
¨FSTEDT, C. (2006). Differential expression of cytochrome
c oxidase subunit III gene in castes of the termite Reticuliter mes
santonensis.Journal of Insect Physiology 52, 551–557.
LIU, Y. X., HENDERSON, G., MAO,L.X.&LAINE, R. A. (2005a).
Effects of temperature and nutrition on juvenile hormone titers
of Coptotermes formosanus (Isoptera: Rhinotermitidae). Annals of the
Entomological Society of America 98, 732–737.
LIU, Y. X., HENDERSON, G., MAO,L.X.&LAINE, R. A. (2005b).
Seasonal variation of juvenile hormone titers of the formosan
subterranean termite, Coptotermes for mosanus (Rhinotermitidae).
Environmental Entomology 34, 557–562.
LO, N., ENGEL, M. S., CAMERON, S., NALEPA, C. A., TOKUDA,G.,
GRIMALDI, D., KITADE, O., KRISHNA, K., KLASS,K.D.,
MAEKAWA, K., MIURA,T.&THOMPSON, G. J. (2007). Save
Isoptera: A comment on Inward et al. Biology Letters 3,562–
563.
LONGHURST, C., JOHNSON,R.A.&WOOD, T. G. (1978). Predation
by Megaponera foetens (Fabr.) (Hymenoptera: Formicidae) on the
termites in the Nigerian Southern Guinea savanna. Oecologia 32,
101–107.
LU
¨SCHER, M. (1958). U
¨ber die Entstehung der Soldaten bei
Termiten. Revue Suisse Zoologie 65, 372–377.
LU
¨SCHER, M. (1964). Die spezifische Wirkung ma
¨nnlicher und
weiblicher Ersatzgeschlechtstiere auf die Entstehung von
Geschlechtstieren bei der Termite Kalotermes flavicollis (Fab.).
Insectes Sociaux 11, 79–90.
LU
¨SCHER, M. (1969). Die Bedeutung des Juvenilhormons fu
¨r die
Differenzierung der Soldaten bei der Termite Kalotermes
flavicollis.InProceedings VI. Congress of IUSSI, pp. 165–170, Bern.
LU
¨SCHER, M. (1972). Environmental control of juvenile hormone
(JH) secretion and caste differentiation in termites. General and
Comparative Endocrinology Suppl. 3, 509–504.
LU
¨SCHER, M. (1974). Kasten und Kastendifferenzierung bei niederen
Te r mi t e n. In Sozialpolymorphismus bei Insekten (ed. G. H. Schmidt),
pp. 694–739. Wissenschaftliche Verlagsgesellschaft, Stuttgart.
MACWHINNIE,S.G.B.,ALLEE, J. P., NELSON,C.A.,RIDDIFORD,L.M.,
TRUM AN ,J.W.&CHAMPLIN, D. T. (2005). The role of nutrition in
creation of the eye imaginal disc and initiation of metamorphosis in
Manduca sexta.Developmental Biology 285, 285–297.
MAO, L. X., HENDERSON, G., LIU,Y.X.&LAINE, R. A. (2005).
Formosan subterranean termite (Isoptera: Rhinoter mitidae)
soldiers regulate juvenile hormone levels and caste differentiation
in workers. Annals of the Entomological Society of America 98, 340–345.
MARAIS, E. N. (1937). The Soul of the White Ant. Methuen, London
MARTIN, D., MAESTRO, J. L., CRUZ , J., MANE-PADROS,D.&BELLES,
X. (2006). RNAi studies reveal a conserved role for RXR in
molting in the cockroach Blattella germanica.Journal of Insect
Physiology 52, 410–416.
MILLER, E. M. (1942). The problem of castes and caste differ-
entiation in Prorhinotermes simplex (Hagen). Bulletin of the University
of Miami 15, 3–27.
MINELLI, A., BRENA, C., DEFLORIAN, G., MARUZZO,D.&FUSCO,G.
(2006). From embryo to adult - beyond the conventional
periodization of arthropod development. Development, Genes and
Evolution 216, 373–383.
MIRTH, C., TRUMAN,J.W.&RIDDIFORD, L. M. (2005). The role of
the prothoracic gland in determining critical weight for
metamorphosis in Drosophila melanogaster.Current Biology 15, 1–12.
MIURA, T. (2001). Morphogenesis and gene expression in the soldier-
caste differentiation of termites. Insectes Sociaux 48, 216–223.
MIURA, T. (2005). Developmental regulation of caste-specific
characters in social-insect polyphenism. Evolution & Development
7, 122–129.
MIURA, T., KAMIKOUCHI, A., SAWATA, M., TAKEUCHI, H., NATORI,
S., KUBO,T.&MATSUMOTO, T. (1999). Soldier caste-specific
gene expression in the mandibular glands of Hodotermopsis
japonica (Isoptera: Termopsidae). Proceedings of the National Academy
of Sciences of the United States of America 96, 13874–13879.
MIURA, T., KOSHIKAWA, S., MACHIDA,M.&MATSUMOTO, T. (2004).
Comparative studies on alate wing formation in two related
species of rotten-wood termites: Hodotermopsis sjostedti and
Zootermopsis nevadensis (Isoptera, Termopsidae). Insectes Sociaux
51, 247–252.
MIURA,T.&MATSUMOTO, T. (2000). Soldier morphogenesis in
a nasute termite: discovery of a disk-like structure forming
a soldier nasus. Proceedings of the Royal Society of London, Series B
267, 1185–1189.
MORAN, N. A. (1992). The evolutionary maintenance of alternative
phenotypes. American Naturalist 139, 971–989.
MU
¨LLER, C. B., WILLIAMS,I.S.&HARDIE, J. (2001). The role of
nutrition, crowding and interspecific interactions in the devel-
opment of winged aphids. Ecological Entomology 26, 330–340.
MUNDALL, E. C., TOBE,S.S.&STAY, B. (1979). Induction of
vitellogenin and growth of implanted oocytes in male cock-
roaches. Nature 282, 97–98.
MYLES, T. G. (1988). Resource inheritance in social evolution from
termite to man. In Ecology of Social Behavior (ed. C. N.
Slobodchikoff), pp. 379–425. Academic Press, New York.
MYLES, T. G. (1999). Review of secondary reproduction in termites
(Insecta: Isoptera) with comments on its role in termite ecology
and social evolution. Sociobiology 33, 1–91.
NALEPA, C. A. (1994). Nourishment and the origin of ter-
mite eusociality. In Nourishment and Evolution in Insect Societies (ed.
J. H. Hunt and C. A. Nalepa), pp. 57–104. Westview Press,
Boulder.
NALEPA,C.A.&BANDI, C. (2000). Characterizing the ancestors:
paedomorphosis and termite evolution. In Termites: Evolution,
Sociality, Symbioses, Ecology (ed. T. Abe, L. D. E. Bignel and M.
Higashi), pp. 53–76. Kluwer Academic Press, Dordrecht.
NIJHOUT, H. F. (1994). Insect Hormones. Princeton University Press,
Princeton.
NOIROT, C. (1990). Sexual castes and reproductive strategies in
termites. In Social Insects - an Evolutionary Approach to Caste and Repro-
duction (ed. W. Engels), pp. 5–35. Springer Verlag, Heidelberg.
NOIROT, C. (1991). Caste differentiation in Isoptera - basic features,
role of pheromones. Ethology, Ecology & Evolution, 3–7.
NOIROT,C.&DARLINGTON, J. P. E. C. (2000). Termite nests:
architecture, regulation and defence. In Ter mites: Evolution,
Sociality, Symbioses, Ecology. (ed. T. Abe, D. E. Bignell and M.
Higashi), pp. 121–140. Kluwer Academic Press, Dordrecht.
NUTTING. (1969). Flight and colony foundation. In Biology of
Termites, vol. 1 (ed. K. Krishna and F. M. Weesner), pp. 233–282.
Academic Press, New York.
Developmental plasticity in termites 311
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

OAKESHOTT, J. G., CLAUDIANOS, C., CAMPBELL, P. M., NEWCOMB,R.
D. & RUSSELL, R. J. (2005). Biochemical genetics and genomics
of insect esterases. In Comprehensive Molecular Insect Science, vol. 5
(ed. L. I. Gilbert, K. Iatrou and S. S. Gill), pp. 309–381.
Elsevier, San Diego.
OKOT-KOTBER, B. M. (1983). Ecdysteroid levels associated with
epidermal events during worker and soldier differentiation in
Macrotermes michaelseni (Isoptera: Macrotermitinae). General and
Comparative Endocrinology 52, 409–417.
OKOT-KOTBER,B.M.&PRESTWICH, G. D. (1991a). Identification of
a juvenile-hormone binding protein in the castes of the termite,
Reticulitermes flavipes, by photoaffinity labeling. Insect Biochemistry
21, 775–784.
OKOT-KOTBER,B.M.&PRESTWICH, G. D. (1991b). Juvenile
hormone binding proteins of termites detected by photoaffinity
labeling: comparison of Zootermopsis newadensis with two
rhinotermitids Coptotermes formosanus and Reticulitermes flavipes.
Archives of Insect Biochemistry and Physiology 17, 119–128.
OKOT-KOTBER, B. M., PRESTWICH,G.D.,STRAMBI,A.&STRAMBI,
C. (1993). Changes on morphogenetic hormone titers in isolated
workers of the termite Reticulitermes flavipes (Kollar). General and
Comparative Endocrinology 90, 290–295.
OSTER,G.F.&WILSON, E. O. (1978). Caste and Ecology of Social
Insects. Princeton University Press, Princeton.
PARK,Y.I.&RAINA,A.K.(2005).Regulationofjuvenile
hormone titers by soldiers in the Formosan subterranean
termite, Coptotermes formosanus.Journal of Insect Physiolog y 51,
385–391.
PARMENTIER, D. (2006). Developmental Flexibility and Evolution of the
Worker Caste in Termites. PhD thesis, Universite
´Libre de
Bruxelles, Bruxelles.
PARMENTIER,D.&ROISIN, Y. (2003). Caste morphology and
development in Termitogeton nr. planus (Insects, Isoptera, Rhino-
termitidae). Journal of Morphology 255, 69–79.
PATEL, A., FONDRK, M. K., KAFTANOGLU, O., EMORE, C., HUNT,G.,
FREDERICK,K.&AMDAM, G. V. (2007). The Making of a queen:
TOR pathway is a key player in diphenic caste development.
PLoS One 2(6), e509.
PEETERS,C.&HIGASHI, S. (1989). Reproductive dominance
controlled by mutilation in the queenless ant Diacamma australe.
Naturwissenschaften 76, 177–180.
RAIKHEL, A. S., BROW N,M.R.&BELLE
´S, X. (2005). Hormonal
control of reproducive processes. In Comprehensive Molecular Insect
Science, vol. 3 (ed. L. I. Gilbert, K. Iatrou and S. S. Gill), pp.
433–491. Elsevier, Oxford.
RAMASWAMY, K., PEETERS, C., YUVANA,S.P.,VARGHESE,T.,
PRADEEP, H. D., DIETEMANN, V., KARPAKAKUNJARAM, V., COBB,
M. & GADAGKAR, R. (2004). Social mutilation in the Ponerine
ant Diacamma: cues originate in the victims. Insectes Sociaux 51,
410–413.
RIDDIFORD, L. M. (1994). Cellular and molecular actions of
juvenile hormone. I. General considerations and premetamor-
phic actions. Advances in Insect Physiology 24, 213–274.
RIDDIFORD, L. M. (1996). Juvenile hormone: the status of its ‘‘status
quo’’action. Archives of Insect Biochemistry and Physiology 32, 271–
286.
ROISIN, Y. (1994). Intragroup conflicts and the evolution of sterile
castes in termites. American Naturalist 143, 751–765.
ROISIN, Y. (1999). Philopatric reproduction, a prime mover in the
evolution of eusociality? Insectes Sociaux 46, 297–305.
ROISIN, Y. (2000). Diversity and evolution of caste patterns. In
Termites: Evolution, Sociality, Symbioses, Ecology (ed. T. Abe, D. E.
Bignell and M. Higashi), pp. 95–120. Kluwer Academic
Publishers, Dordrecht.
ROISIN, Y. (2006). Life history, life types and caste evolution in
termites. In Life Cycles in Social Insects: Behaviour, Ecology and
Evolution (ed. V. E. Kipyatkov), pp. 85–95. St. Petersburg
University Press, St. Peterburg.
ROISIN,Y.&LENZ, M. (1999). Caste developmental pathways in
colonies of Coptotermes lacteus (Froggatt) headed by primary
reproductives (Isoptera, Rhinotermitidae). Insectes Sociaux 46,
273–280.
ROISIN,Y.&PARMENTIER, D. (2006). Foraging by termites without
workers: implications for the evolution of castes and life types. In
XV International Congress of IUSSI, pp. 101, Washington, D.C.
ROSENGAUS,R.B.&TRANIELLO, J. F. A. (1993). Temporal
polyethism in incipient colonies of the primitive termite
Zootermopsis angusticollis: a single multiage caste. Journal of Insect
Behavior 6, 237–252.
ROTH, L. M. (1981). Introduction. In The American Cockroach (ed. W. J.
Bell and K. G. Adiyodi), pp. 1–14. Chapman and Hall, London.
ROUX,E.A.&KORB , J. (2004). Evolution of eusociality and the
soldier caste in termites: a validation of the intrinsic benefit
hypothesis. Journal of Evolutionary Biology 6, 342–352.
SCHARF, M. E., RATLIFF, C. R., WU-SCHARF, D., ZHOU,X.G.,
PITTENDRIGH,B.R.&BENNETT, G. W. (2005a). Effects of juvenile
hormone III on Reticulitermes flavipes: changes in hemolymph
protein composition and gene expression. Insect Biochemistry and
Molecular Biology 35, 207–215.
SCHARF, M. E., WU-SCHARF, D., PITTENDRIGH,B.R.&BENNETT,G.
W. (2003). Caste and development-associated gene expression in
a lower termite. Genome Biology 4, R:62.
SCHARF, M. E., WU-SCHARF, D., ZHOU, X., PITTENDRIGH,B.R.&
BENNETT, G. W. (2005b). Gene expression profiles among
immature and adult reproductive castes of the termite
Reticulitermes flavipes.Insect Molecular Biology 14, 31–44.
SCHLICHTING,C.D.&PIGLIUCCI, M. (1998). Phenotypic Evolution -
a Reaction Norm Perspective. Sinauer Associates, Sunderland,
Massachusetts.
SHELLMAN-REEVE, J. S. (1997). The spectrum of eusociality in
termites. In The Evolution of Social Behaviour in Insects and Arachnids
(ed. J. C. Choe and B. J. Crespi), pp. 52–93. Cambridge
University Press, Cambridge.
SPRINGHETTI, A. (1969). Influenza dei reali sulla differenziazione
dei soldati di Kalotermes flavicollis Fabr. (Isoptera). Proceedings of the
6th Congress of IUSSI, Bern, 267–273.
STAY,B.&TOBE, S. S. (2007). The role of allatostatins in juvenile
hormone synthesis in insects and crustaceans. Annual Review of
Entomology 52, 277–299.
STAY, B., WOODHEAD, A. P., JOSHI,S.&TOBE, S. S. (1991).
Allatostatins, neuropeptide inhibitors of juvenile hormone
biosynthesis in brain and corpora allata of cockroaches,
Diploptera punctata.InInsect Neuropeptides (ed. J. J. Menn, T. J.
Kelly and E. P. Masler), pp. 164–176. American Chemical
Society, Washington, D.C.
STRIEBEL, H. (1960). Zur Embryonalentwicklung der Termiten.
Acta Tropica 17, 193–260
TAWFIK, A. I., KELLNER, R., HOFFMANN,K.H.&LORENZ,M.W.
(2006). Purification, characterisation and titre of the haemo-
lymph juvenile hormone binding proteins from Schistocerca
gregaria and Gryllus bimaculatus.52, 255–268.
THE HONEY BEE GENOME SEQUENCING CONSORTIUM (2006).
Insights into social insects from the genome of the honeybee
Apis mellifera.Nature 443, 931–949.
Judith Korb and Klaus Hartfelder312
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society

THOMPSON, G. J., KITADE, O., LO,N.&CROZIER, R. H. (2000).
Phylogenetic evidence for a single, ancestral origin of a ‘true’
worker caste in termites. Journal of Evolutionary Biolog y 13, 869–881.
THOMPSON,G.J.,KITADE, O., LO,N.&CROZIER, R. H. (2004). On
the origin of termite workers: weighing up the phylogenetic
evidence. Journal of Evolutionary Biology 17, 217–220.
THORNE, B. L. (1997). Evolution of eusociality in termites. Annual
Review of Ecology and Systematics 28, 27–54.
THORNE, B. L., BREISCH,N.L.&MUSCEDERE, M. L. (2003).
Evolution of eusociality and the soldier caste in termites:
Influence of intraspecific competition and accelerated inheri-
tance. Proceedings of the National Academy of Sciences of the United States
of America 100, 12808–12813.
THORNE, B. L., GRIMALDI,D.A.&KRISHNA, K. (2000). Early fossil
history of the termites. In Termites: Evolution, Sociality, Symbioses,
Ecology (ed. T. Abe, D. E. Bignell and M. Higashi), pp. 77–94.
Kluwer Academic Publishers, Dordrecht.
THORNE,B.L.&TRANIELLO, J. F. A. (2003). Comparative social
biology of basal taxa of ants and termites. Annual Review of
Entomology 48, 283–306.
THUMMEL, C. (1996). Flies on steroids - Drosophila metamorphosis
and the mechanisms of steroid hormone action. Trends in Genetics
12, 306–310.
TRUMAN,J.W.&RIDDIFORD, L. M. (1999). The origin of insect
metamorphosis. Nature 401, 447–452.
TRUMAN,J.W.&RIDDIFORD, L. M. (2002). Endocrine insights into
the evolution of metamorphosis in insects. Annual Review of
Entomology 47, 467–500.
WATSON,J.A.L.&ABBEY, H. M. (1985). Development of neotenics in
Mastotermes darwiniensis Froggatt: an alternative strategy. In Caste
Differentiation in Social Insects (ed. J. A. L. Watson, B. M. Okot-
Kotber and C. Noirot), pp. 107–124. Pergamon Press, Oxford.
WATSON,J.A.L.&SEWELL, J. J. (1985). Caste development in
Mastotermes and Kalotermes: which is primitive? In Caste
Differentiation in Social Insects (ed. J. A. L. Watson, B. M. Okot-
Kotber and C. Noirot), pp. 27–40. Pergamon Press, Oxford.
WEIL, T., REHLI,M.&KORB, J. (2007). Molecular basis for the
reproductive division of labour in a lower termite. BMC Genomics
8, e198.
WEST-EBERHARD, M. J. (2003). Developmental Plasticity and Evolution.
Oxford University Press, Oxford, New York.
WHEELER,D.E.,BUCK,N.&EVA NS ,J.S.(2006).Expressionof
insulin pathway genes during the period of caste determination
in the honey bee, Apis mellifera.Insect Molecular Biology 15,
597–602.
WILSON, E. O. (1971). The Insect Societies. The Belknap Press of
Harvard University Press, Cambridge, MA.
WOODHEAD, A. P., STAY, B., SEIDEL, S. L., KHAN,M.A.&TOBE,
S. S. (1989). Primary structure of 4 allatostatins - neuropeptide
inhibitors of juvenile hormone synthesis. Proceedings of the
National Academy of Sciences of the United States of America 86,
5997–6001.
XU, Y., FANG, F., CHU, Y., JONES,D.&JONES, G. (2002). Activation
of transcription through the ligand-binding pocket of the
orphan nuclear receptor ultraspiracle. European Journal of
Biochemistry 269, 6026–6036.
YAGI, K. J., KWOK, R., CHAN, K. K., SETTER, R. R., MYLES,T.G.,
TOBE,S.S.&STAY, B. (2005). Phe-Gly-Leu-amide allatostatin in
the termite Reticulitermes flavipes: Content in brain and corpus
allatum and effect on juvenile hormone synthesis. Journal of Insect
Physiology 51, 357–365.
ZERA, A. J. (2003). The endocrine regulation of wing poly-
morphism in insects: State of the art, recent surprises, and future
directions. Integrative and Comparative Biology 43, 607–616.
ZHOU, X., FAITH,M.O.&SCHARF, M. E. (2006a). Social
exploitation of hexamerin: RNAi reveals a major caste-
regulatory factor in termites. Proceedings of the National Academy
of Sciences of the United States of America 103, 4499–4504.
ZHOU, X., SONG, C., GRZYMALA, T. L., OI,F.M.&SCHARF,M.E.
(2007a). Juvenile hormone and colony conditions differentially
influence cytochrome P450 gene expression in the termite
Reticulitermes flavipes.Insect Molecular Biology 15, 749–761.
ZHOU, X., TARVER, M. R., BENNETT,G.W.,OI,F.M.&SCHARF,M.
E. (2006b). Two hexamerin genes from the termite Reticulitermes
flavipes: Sequence, expression, and proposed functions in caste
regulation. Gene 376, 47–58.
ZHOU, X., TARVER,M.R.&SCHARF, M. E. (2007b). Hexamerin-
based regulation of juvenile hormone-dependent gene expres-
sion underlies phenotypic plasticity in a social insect. Development
134, 601–610.
ZIMMERMANN, R. B. (1983). Sibling manipulation and indirect
fitness in termites. Behavioral Ecology and Sociobiology 12, 143–145.
Developmental plasticity in termites 313
Biological Reviews 83 (2008) 295–313 Ó2008 The Authors Journal compilation Ó2008 Cambridge Philosophical Society