11

INTRAGENERATIONAL  DEVELOPMENTAL  PLASTICITY

The central nervous system can integrate information about the animal’s internal and external environment and use this information to regulate the secretion of hormones.

                                                                                                         H.F. Nijhout

In the course of evolution metazoans have evolved a high potential for adapting within their lifetime to changes and unpredictabilities of the environment, which is known as phenotypic plasticity. They express this potential in the form of a continuum of only quantitatively differing phenotypes, that is the reaction norm, or in the form of sudden discontinuous, discrete qualitative changes in phenotypic characters, that is intragenerational developmental plasticity. This latter form of phenotypic plasticity may be triggered by specific environmental cues (acquired developmental plasticity) or it may appear in the offspring independently of the environmental cues (innate developmental plasticity). As a rule, the intragenerational developmental plasticity is adaptive and involves changes or switches in specific developmental pathways activated by signals originating in the central nervous system. Intragenerational developmental plasticity involves no changes in genes, i.e. it is epigenetically determined and is not transmitted to the offspring.

Developmental Plasticity: Beyond the Reaction Norm

Metazoans are dynamic systems that can change during the lifetime, both in response to external or internal stimuli and as a result of external influences. This is the phenotypic plasticity in the broadest meaning of the term (figure 11.1). Phenotypic plasticity may occur in the form of the norm of reaction (Woltereck, 1909), consisting of a continuum of incrementally varying phenotypes involving no qualitative changes, or in the form of developmental plasticity, which implies appearance of discrete alternative phenotypes.

In describing the phenomenon of the appearance of alternative discrete phenotypes in this work I will use the term developmental plasticity to emphasize the fact that their appearance is determined by switches in developmental pathways.

Both the norm of reaction and developmental plasticity involve no changes in genes or genetic information in general. However, developmental plasticity and reaction norm are two qualitatively different phenomena as far as the nature of the change they bring about and the underlying mechanisms are concerned.

Figure 11.1.  Main forms of phenotypic plasticity in metazoans under natural conditions.

First, the norm of reaction, as the term itself indicates, is a reaction to changes in the environment, whereas developmental plasticity does not always/necessarily depend on environmental stimuli (developmental polymorphisms, e.g.). In cases when the developmental plasticity is related to/depends on environmental stimuli it is an adaptive response rather than a reaction implying that it is not determined by the nature of the stimulus but by the adaptive needs of the organism, which switches to a new developmental pathway for  achieving a discrete phenotypic change.

Second, developmental plasticity is related to switches in developmental pathways and mechanisms, while the reaction norm is not.

Third, and as mentioned earlier, the norm of reaction implies existence of a continuum of phenotypes displaying only quantitative differences between them, whereas developmental plasticity usually implies qualitative changes of the phenotype.

……………………………………………………

……………………………………………………

Developmental Plasticity and Possible Evolutionary Implications

The fact that the developmental plasticity produces discrete traits that are not only quantitatively but qualitatively as well, different from the original phenotype raises the issue of the possible evolutionary implications of the developmental plasticity.

Here I will briefly discuss the possible involvement in the evolutionary process of the  intragenerational developmental plasticity alone. The transgenerational developmental plasticity virtually represents a special form of evolutionary change and therefore it will be considered as a separate issue in the next chapter.

Discussing the role of developmental plasticity in evolution, West-Eberhard has shown that new phenotypic traits, arising in response to the changed environment or as a result of mutational events, may become evolutionarily relevant via processes of phenotypic and genetic accommodation. Phenotypic accommodation implies adjustments that are necessary for integrating the novel phenotypic trait to the general phenotype, in order “to reduce the amount of functional disruption occasioned by a developmental novelty”, whereas genetic accommodation implies the presence of variation in alleles whose frequency will increase under new selection regime determined by the phenotypic novelty (West-Eberhard, 2003d). Other authors also believe that novel phenotypic traits arise by environmental rather than genetic factors and they become evolutionarily relevant via the process of genetic assimilation (Pigliucci and Murren, 2003; Pigliucci et al., 2006). Both hypotheses hold that genetic variation is necessary for phenotypic/developmental plasticity to be evolutionarily relevant. The difference is that the first hypothesis implies that the genetic variation is present in natural populations and selection enables the evolutionary fixation of the phenotypic novelty, whereas the latter holds that phenotypic plasticity increases the chances of survival under changed conditions and mutations that may occur thereafter (genetic assimilation) enable the evolutionary fixation of the novelty.

However, it is difficult not only to prove but even to argue that gene mutations are involved in  the evolution of developmental plasticity. Cases of developmental plasticity described so far, generally, show no signs of “functional disruption” that would require any mutations in genes (Pigliucci and Murren, 2003).

Theoretically, it might be argued that metazoans are capable of producing with the same genes not only small evolutionary changes but even radically different Baupläne (recall amphibian metamorphosis: during its lifetime, i.e. with the same genes, a frog sculptures such radically different Baupläne as that of the class of fish (as a tadpole) and amphibians (as an adult organism), and flies, during their lifetime, sequentially develop both worm and insect Baupläne. The fact that in many cases the plasticity is not adaptive, indicates that accumulation of gradual mutational changes under the action of natural selection is not necessary for the evolution of the developmental plasticity. One should always bear in mind that developmental plasticity involves no changes in genes and hence is essentially a non-genetic, epigenetic phenomenon.

The fact that the developmental plasticity commonly is adaptive to the changed environment in response to which it arises, suggests that it arises not spontaneously or randomly. The plastic response is somehow computed in the meaning that a relation is established between two unrelated elements, the environmental stimulus and the appearance of developmental plasticity. The establishment of this relationship implies as a sine qua non use of new information. The origin and nature of that information is essential for understanding the origin and nature of the epigenetic plasticity, the mechanism by which animals translate environmental stimuli into information for the phenotypic change. It is generally admitted that it is a neuroendocrine mechanism:

Whether there are any cases of adaptive phenotypic plasticity in animals in which development of the relevant trait is directly sensitive to the environmental variable or whether all cases are mediated by evolved integrated systemic processes, as in the case of polyphenisms….Perhaps the most interesting thing about having a hormonal regulation of development is that development comes under the control of the central nervous system. This is because the developmental hormones are directly regulated by neurosecretory factors or are themselves neurosecretory hormones. The central nervous system can integrate information about  the animal’s internal and external environment and use this information to regulate the secretion of hormones (my emphasis - N.C). In this way, development can become responsive to a wide variety of environmental signals, without the need to have developmental processes themselves be sensitive to the environment. (Nijhout, 2003)

All the discrete morphological changes to be presented in this chapter start with, and essentially involve, reception and processing in neural circuits of visual, olfactory, tactile, or interoceptive input, perception or some other sensing by animals of environmental stimuli, including their predators. Since perception, or sensing, in any case takes place in the CNS it is logical to assume that the brain is the place where the causal chain leading from the stimulus to the changed morphology or life history starts (see chapter 2, The Origin and Nature of Epigenetic Information for Metazoan Morphology).

Intragenerational Developmental Plasticity

Adaptive Intragenerational Developmental Plasticity

 Camouflage (Adaptive coloration, Cryptic Coloration, Crypsis)

Camouflage is generally defined as morphological adaptation “intended” to make an individual less visible in its background. It serves mostly preys to hide from their predators, but predators too my extract some hunting profits by being less visible. The adaptive character of the phenomenon is obvious and its basic mechanisms are known to a considerable extent. The chain of events from the visual perception of the environmental background to generation of a body pattern resembling the background often is very complex but already known in some details.

Let’s consider some paradigmatic examples of adaptive coloration in animals and use current knowledge for understanding its mechanism.

Under strong disturbance or provocation, the cuttlefish Sepia officinalis undergoes a series of colour-changes as conspicuous and complete as they are rapid. The first of the successive patterns is one in which two large black spots appear on the dorsal surface of the mantle. Then a rapid and complete paling of the rest of the animal follows, while the black spots themselves become still more saturated and intense... At the same time a black crescent forms beneath each eye, the pupil dilates and the edges of the fins become strongly outlined in black - the rest of the body remaining white.

Alternatively, the two-spot pallid phase may be followed by the animal shooting away, the rapid departure being accompanied by simultaneous darkening of the whole body. Further irritation may lead to total paling and the superimposition of four longitudinal black stripes on the upper surface. These lines flicker vividly over the pallid back, and then suddenly disappear, to be followed by a reappearance of the zebra pattern. All this time, the animal darts about rapidly, as if to avoid the irritation, and its final action when can not do so is to eject a cloud of ink. Then it becomes motionless and hides below the black cloud which it has produced. (Cott, 1966a)

The body pattern of a cephalopod, its gross appearance, is determined by skin chromatophores (from Gr. χρωμα (hroma) - color and φορος (foros) - bear, carry),. The primary function of chromatophores is camouflage. Each chromatophore contains pigment granules and is connected with myocytes and neurons.

Neural signals from the brain determine patterns of muscle contraction and, consequently, the dispersion of pigment granules within the chromatophore, so that in their totality the skin chromatophores produce a pattern that matches the background, achieves general resemblance to the substrate, or breaks up the body’s outline. Neural control of chromatophores enables a cephalopod to change its appearance almost instantaneously, which is a key feature in some escape behaviours and during agonistic signaling. This rapid body pattern polyphenism is adaptive because it may hinder search-image formation by predators.

The change of color or patterning of the body in cephalopods is accomplished by millions of chromatophores, multicellular organs consisting of a central pigment-containing cell attached to a set of 6-20 radially arrayed chromatophore muscles. These muscles are innervated by motor neurons whose cell bodies in the European cuttlefish, Sepia officinalis, are located in the PSEM (posterior subesophageal mass) of the brain (Dubas et al., 1986; Gaston and Tublitz, 2006) (figure 11.2).

Figure 11.2. Diagrammatic representation of lobes in the cephalopod central nervous system that are thought to control chromatophore patterning. Only one side of the brain is represented and the anterior chromatophore lobe is omitted for the sake of clarity. The main pathway is considered to be: optic lobes to lateral basal lobes to chromatophore lobes to chromatophores (From Dubas et al., 1986).

The chromatophores are controlled by a set of lobes in the brain organized hierarchically. At the highest level, the optic lobes, acting largely on visual information, select specific motor programmes (i.e. body patterns); at the lowest level, motor neurons in the chromatophore lobes execute the programmes, their activity or inactivity producing the patterning seen in the skin. Messenger, 2001)

Chromatophore lobes in Octopus vulgaris contain over half a million neurons. The fact that those animals change their body pattern to match their background within a very short time of about a second in cuttlefish (Hanlon and Messenger, 1988) and 2-8 sec in fish seems to exclude any inducer molecules (hormones or secreted proteins) as a possible regulator of body patterning:

Thus, a neural rather than humoral, mechanism must be involved. (Ramachandran et al., 1996)

Now we know that specific small molecule- and peptide neurotransmitters released at the cephalopod neuromuscular junction induce chromatophore functions. The first neurotransmitter to be recognized to have a role in the functioning of chromatophores has been the glutamate. Recently, it has been shown that endogenous neuropeptides of FMRFamide family are involved in the body patterning of Sepia officinalis (Loi et al., 1996; Hanlon and Messenger, 1988).

And since this form of camouflage results from disruptive coloration, i. e. from combination of a varied number of chromatic units, it has been proposed that a physiological unit in the brain must be responsible for each chromatic-morphologicalunit in the skin (Packard, 1982).

Thus, Sepia officinalis first draws the camouflaging image of its body in the brain. The size, contrast, and the number of white squares in the black background are cues the cuttlefish (order Sepiidae) use to switch from uniformly stippled skin patterns (general resemblance) to disruptive skin patterns (Chuan-Chin and Hanlon, 2001).

The pathway of patterning signals from the brain physiological units to the skin chromatophores in cuttlefish, within the broader schema of information flow along the body patterning circuitry, has been schematized as follows:

visual input eyes --> brain optical lobes--> brain lateral basal lobes--> brain chromatophore lobes--> skin (Chuan-Chin and Hanlon, 2001).

Intraspecific signaling and communication is another function of chromatophores that is well documented in several inshore species, and interspecific signaling, using ancient, highly conserved patterns, is also widespread. Neurally controlled chromatophores lend themselves supremely well to communication, allowing rapid, finely graded and bilateral signaling.

Many crustaceans are also able to rapidly modify their color so that it matches the background. Blood-borne factors involved in their phenotypic adaptive coloration belong to two main groups. There is a group of neuropeptides from the X organ-sinus gland complex, which acts directly on chromatophores, causing either dispersion or concentration of the pigment granules. The second group consists partly of neurotransmitters acting within the nervous system by triggering the release of chromatophorotropic neuropeptides. The rest of them, primarily amines, are released in haemolymph from pericardial organs (Gorbman and Davey, 1991). Two main neurohormones involved in adaptive coloration of crustaceans are DRPH (light-adapting distal retinal pigment hormone), with light-adapting function, and its antagonist neurohormone, RPCH (red pigment-concentrating hormone), with the latter being responsible for adaptive coloration of all pigment cells (melanophores, erythrophores, leukophores, and xanthophores) (Josefsson, 1983).

It is demonstrated that electrical stimulation of eyestalks in Crustacea results in the release of peptides with activity on specific types of chromatophores (Gorbman and Davey, 1991d). The neuropeptide, RPCH also controls the color changes in shrimps (Strand, 1999j) and modulates the swimmeret activity rhythms in the crayfish.

Neurohormones regulate color changes and movement of pigments during light-and-dark adaptation of eyes in other arthropods (Pearse et al., 1987).

The central nervous system also determines skin color changes in many fish and amphibians. Fish scales and amphibian skin also contain specialized pigment cells, melanophores. The neurohormonal stimulation of these cells, under conditions of stress/alertness, stimulates melanophores to transport (along microtubules) membrane-enclosed pigment granules toward the periphery (producing darkened cell color) or center (producing pale cell color) (Ramachandran et al., 1996; Lodish et al., 1998). In response to artificial changes in the background, fish (Ramachandran et al., 1996), like cuttlefish (Chuan-Chin and Hanlon, 2001), repattern their body to match the background if the backgrounds display “appreciable differences” (Marshall and Messenger, 1996). This clearly implies visual input and perception of the background and since that perception take place in the brain, the color change and repatterning of the skin is clearly under control of the CNS:

The fish must have independent visual control of each set (or subset) of markings, a possibility that requires verification. There may be ‘feature detectors’ in the fish visual centres that are specialized for detecting different spatial frequencies of textures in the environment and these might exert direct control over the corresponding set of marks on the skin surface. Indeed, there might be a map of effector neurons in (say) the tectum, so that focal electrical stimulation might produce selective contrast enhancement of specific spatial frequency components on the skin. (Ramachandran et al.,1996)

The paradise whiptail, Pentapodus paradiseus, is a fish inhabiting the coastal waters of Queensland, Australia. The fish has colored stripes on its body and is able to change its color from blue to red within less than one second (Mäthger et al., 2003). The coloration of this and many other fish is determined by iridophores (light-reflecting cells) in the skin. They contain thin guanine plates, which are multilayer reflectors with refractive index higher than spaces separating them. Plates are connected with each other by microtubular structures. Any change in the distance between plates contained within the reflective cells will produce a change in the reflected color. This adaptive change in the distance between plates is regulated by underlying microtubule structures (Oshima and Fujii, 1987). Changes in the structure or length of microtubules change the distance between guanine plates and this leads to changes the color reflected by iridophores. It has been found that the sympathetic nervous system regulates the distance between plates and the body color by regulating the length (ploymerization/ depolymerization) of microtubules. Under stress conditions, the release of noradrenaline shifts the reflection toward red (longer wavelength), whereas in response to topical application of the neurotransmitter acetylcholine spectral reflections shift towards shorter wavelengths (Mäthger et al., 2004). Fluctuations in intracellular Ca²⁺  also change the structure of microtubules (Mäthger et al., 2003) and, consequently, the distance between plates and the reflected color by iridophores. In experiments it has been found that the neurotransmitter Ach (acetylcholine) increases the Ca²⁺ level (Mäthger et al., 2004). Studies on the color change (from blue to green) in the blue damselfish, Chyrsiptera cyanea, and in the neon tetra (Nagaishi and Oshima, 1989) have led the investigators to the conclusion that

The motile iridophores are solely under the control of the sympathetic adrenergic system, and that the co-transmitter, adenosine, may function to antagonize quickly the true transmitter-induced colored state of the cells. (Kasukawa et al., 1986)

Monocirrhus polyacanthus is a little Nandid fish in the Lower Amazon Valley. It resembles a dead leaf. Populations of this fish, known as “Peche de folha” by the Brasilian natives, consist of individuals of three main color groups (light gray, golden with only a few mottlings of dark-brown, and brown). All of them are capable of changing within one hour their color to darker or lighter tint according to the tones of the background (Cott, 1966b). At his time, O. v. Frisch, was marvelled: “How the fish brain can command the pigment cells so skillfully challenges comprehension” (Frisch, 1973).

When on a white background, the teleost fish medaka, Oryzias latipes, gradually acquires a lighter body color. This is result of two processes: initially, reduction of the size of skin melanophores and later the programmed cell death, apoptosis of dark pigment cells, melanophores (Sugimoto, 2002). The opposite is observed when medaka fish are kept on a dark background: size reduction and apoptosis of leukophores - white pigment cells. Experimental chemical denervation suppresses apoptosis of melanophores, suggesting that the process of melanophore apoptosis is also regulated by skin sympathetic innervation (Sugimoto et al., 2000). In other fish, adaptation to dark background for 2 weeks leads to an increase in the number and size of melanophores in the skin as well as to increased concentration of the pigment in those cells (van Eys and Peters, 1981).

Some teleost fish also change their body color/patterning as a way of communication in their social interactions. In Arctic char, Salvelinus alpinus, subordinate individuals, in the presence of dominant individuals, experience constant stress and activation of the hypothalamus-pituitary-interrenal axis via serotonergic neurons. This is associated with a submissive behavior and darkening of the skin as a result of increased secretion of the pituitary α-MSH (α-melanophore-stimulating hormone) and ACTH (adrenocorticotropic hormone). By contrast, release of neurotransmitters catecholamine, dopamine and norepinephrine, suppresses secretion of the above pituitary hormones and induces aggressive behavior and lighter body color. The darkening of the skin in subordinate Arctic charr fish has been interpreted as  being intended to reduce unnecessary fights and energy loss “in an established dominance hierarchy” (Höglund et al., 2002).

The clawed toad, Xenopus laevis, also modifies its body color so that it matches various environmental backgrounds. The hormone determining this adaptation is α-MSH (α-melanophore-stimulating hormone) synthesized and secreted by the pituitary under central neural control. Various neurotransmitters are involved in the release of this hormone by the pituitary. Using axonal tract tracing, R. Tuinhof et al. (1994) have identified details of the neural circuit (form retina to the hypothalamus), whose signals to the pituitary stimulate the release of the hormone and make the adaptive coloration possible (Tuinhof et al., 1994). Neurons of the suprachiasmatic nucleus project to the pituitary α-MSH (melanophore-stimulating hormone)-producing cells, where they release neurotransmitters regulating α-MSH synthesis (Tuinhof et al., 1994). Another serotonin center in raphe nucleus innervating the pituitary is also involved in the regulation of the secretion of the hormone (Ubink et al., 1999) (see also section Generation of Information for Adaptive Camouflage in Xenopus in chapter 2).

Larvae of salamanders of the family Ambystomatidae, when kept in total darkness, blanch after about one hour. When illuminated they darken in a dark background and brighten when the background is white. Predictably, removal of eyes prevents these camouflage responses in salamanders.

In arthropods neurohormones regulate color changes and movement of pigments during light-and-dark adaptation of their eyes (Pearse et al., 1987). It seems that this neural circuit might be less complex than some known behavioral circuits. It is important, in this regard, to remember that melanocytes (pigment cells) of the human skin derive from neural crest cells which, on the way of their migration to the ectoderm first have made a stop in the developing brain.

It is not surprising that other color-adaptable animals when finding themselves in a uniformly colored or monochrome environment such as snow-covered polar areas, over time (within a sufficient number of generations) switch from this adjustable coloration to a fixed white body color matching the white color of the environment.

Mimicry is a special form of camouflage where a species (the mimic) develops morphologies and behaviors resembling those of other species (the model). During the Batesian mimicry [after the British naturalist and explorer, Henry Walter Bates (1825-1892)], the mimic resembles a model that is not attractive or even is abominable to the predator, whereas during Müllerian mimicry [after the German zoologist Johann F. T. Müller (1821–1897)] two species, both abominable to their predators evolve similar morphology to deter the predator. The adaptation of the mimic to the model may be profitable to the mimic in the case of Batesian mimicry, or may be profitable for both mimicry partners in cases of Müllerian mimicry. A hypothesis presented by C. Darwin (1874) posits that Batesian mimics began to evolve at a remote past, when the model and mimic were much more similar. Even if right, the hypothesis would only account for a limited number of known cases of mimicry (Turner, 1977).

Juveniles of the predator coral fish, fang blenny (Plagiotremus rhinorhynchos), develop a striking resemblance to their prey, cleaner wrasse (Labroides dimidiatus) but only when in the proximity of about 1 meter from the prey; when removed from the prey/model, they lose the mimetic coloration and restore the normal color (Moland and Jones, 2004).

Environmental stress (heat shock or cold shock, e.g.) can modify the wing patterns in butterflies in such a way that they resemble certain wing pattern mutants (A. Kühn and K. Henke. 1936), implying that morphological changes attributed to mutations are not always or necessarily related to genes, as it has been for a long time taken for granted. A temperature-induced pattern modification that makes butterflies resemble other species has also been described (Nijhout, 1985).

There are many known cases of coloration and mimicry that do not offer any advantage to the carriers, what casts doubt on the role of the natural selection in the evolution of the mimicry. By considering the gradual evolution to be impossible and “saltational” origin unacceptable, Turner developed a two-step hypothesis, according to which, a mimicry starts with a major mutation achieving a sudden rough resemblance to the model, and is followed by the establishment of further (modifier) genes at other loci, which will improve the pattern (Turner, 1977). This hypothesis succeeded in overcoming the difficulty arising by the obvious impossibility of accumulation of small, nonadaptive mutations, but it contradicts Fisher’s assertion that carriers of major mutations always suffer severe, deleterious side effects. Besides, as Orr and Coyne argue, while differences in characters determined by a single gene are easily identifiable, it is very difficult to discriminate between many genes of small effect and the presence of major genes amid many modifiers. Hence, experimentation has “little power to detect the presence of major genes” (Orr and Coyne, 1992) and no reliable evidence exists  to demonstrate the existence of such major genes.

Suddenly occurring changes in the body coloration and patterning are beyond the explanatory power of neoDarwinian paradigm. Camouflage and mimicries usually involve activation and expression of such a great number of genes that it is impossible even to imagine how numerous individual changes in numerous genes could have been accumulated often without offering any advantage to the carriers. As T.H. Morgan would put it more than a century ago:

There is no need to question that in some cases animals may be protected by their resemblance to other animals, but it does not follow, despite the vigorous assertions of some modern Darwinians, that this imitation has been the result of selection. (Morgan, 1903)

With neural mechanisms proven to regulate camouflage (cryptic coloration), there is no visible reason for excluding the possibility of the involvement of the central nervous system in the evolution of mimicries. Any case of camouflage and mimicry implies, as a sine qua non, perception of the body color and neural representation of the model in the brain of the camouflant or mimic. While the neural processing of visual stimuli leading to the cascade of events that change the color and pattern to produce cryptic coloration is experimentally determined in numerous cases we are still ignorant on the mechanisms used by the mimics for transmitting the mimicry to their offspring.

It could be argued that the insect, the fish and clawed frog are aware of the fact that the mimicry and adaptive coloration contributes to their survival no more than an amoeba knows that its debris-engulfing routine is necessary for its survival. This argument is hardly relevant since the control of the CNS on adaptive changes of color in both fish and Xenopus takes place on an unconscious level, and for the color adaptation to take place is not necessary for these animals to “know” what they are doing. The unconscious instinctive “knowledge” is all they need to adapt their color or pattern to the background.

The conscious-unconscious dualism is an anthropocentric view rather than an evolutionary principle. No matter at what level of the CNS activity the adaptive coloration might be controlled and regulated, it is an adaptive, non-random phenomenon, based on the innate instinct that contributes to animal’s camouflage and survival.

Commenting on the insect mimicry, more than a century ago, Alfred Taylor wrote:

Here, in this common British butterfly, we have the whole problem set before us - vivid colour, the result of intense and long continued effort; grand display, the object of that colour; dusky, indefinite colour, for concealement; and the “instinctive” pose, to make that protective colour profitable. The insect knows all this in some way (Taylor, 1886)

Summarizing the evidence on the mechanisms of adaptive coloration presented in this section it may be said that no changes in genes are involved and no hypothetic mechanism has been ever presented to show how changes in genes may induce adaptive coloration. Vast majority of the known cases of adaptive coloration could reasonably be explained based on the experimental evidence on the neural mechanisms of coloration determined for a number of species.

Polyphenisms

Polyphenisms in Invertebrates

Populations of a marine bryozoan, Membranipora membranacea, are polymorphic for inducible spine type and consist of a constitutively spined type (6.2%), which produces spines in the absence of the predator stimulus, an unspined type (13.4%), and an inducible type (80%), which produces spine when detects the presence of its predator in the environment (Harvell, 1998). This is a complex case that comprises both polyphenism and predator-induced plasticity (in the case of the inducible type) triggered by the perception of the predator or its kairomones in the bryozoan’s brain.

Butterflies of the families Nymphalidae, Pieris and Papilionidae, produce pupae that display different body colors depending on the color of the pupation site. The peacock butterfly Inachis io (Nymphalidae), the white butterfly, Pieris brassicae (Pieridae), and Papilio polyxenes (Papilionidae), as pupae, display one of two alternative cryptic colors, green-yellow or brown-black, depending on the prevailing color at the pupation site. In order for the pupae to be able to adopt the body color that better matches the color of its surroundings, they first have to perceive the prevailing color of the pupation site. This perception takes place in the insect’s brain and in the insect’s brain is also synthesized the first signal of the signal cascade leading to production of the cryptic body color (green or brown). That signal is a neuropeptide, PMRF (pupal melanization-reducing factor), which is released from the brain into the hemolymph. PMRF is located throughout the entire central nervous system, but its release during the pre-pupal stage is controlled by neural stimulation from the brain.

When pupation of butterflies of the species Inachis io and Pieris brassicae takes place in a light-green background, the pupa synthesizes PMRF, which starts a signal cascade that leads to accumulation of lutein into cuticle and appearance of the green-yellow body color. When pupation takes place on a dark site, PMRF is not produced, lutein is not incorporated into the cuticle and the pupa appears black. Butterflies of the Papilionidae family, although using the same neurohormone for determining their cryptic body color, seem to rely on a different mechanism:

If the plasticity in pupal color in I. io and P. polyxenes is controlled by PMRF, then the neural control of  the hormone’s release would have to be different in the two species. In I. io (and P. brassicae) the hormone is released when pupation is on or in green vegetation, while in P. polyxenes it is released when pupation is on a brown surface. (Starnecker and Hazel, 1999)

Experiments on Graphium sarpedon nipponum (Frühstorfer, 1903) have shown that this butterfly determines its protective body color during the pupal stage, according to the color of the pupation site: green - on white background and reddish-brown - on black background.

Pupae of Papilio xuthus, however, use not the background color but the smoothness of the surface of leaves and twigs, on which they molt, as cues for determining their body color. This indirect cue also leads to adaptive coloration. On leaves and new twigs, which are smooth and green, they become green, while on dead branches, which have rough surface and brownish color, they become grayish-brown. For explaining the mechanism of the body colors in these cases Hiraga has presented the “tactile signals accumulation model”, according to which  body coloration is related to accumulation of tactile signals and perception in the pupal brain of the smoothness/roughness of the surface of the pupation site (Hiraga, 2005; figure 11.3).

Figure 11.3.  Adult butterfly Papilio xuthus and its pupae developing lighter on smooth twigs and darker on rough twigs (From Hiraga, 2005).

In some cases, social and olfactory stimuli also stimulate adaptive responses in the form of changes in phenotype and life history. So, e.g. exposure of female cockroaches of the species Nauphoeta cinerea to conspecific males or even to male pheromones alone, affects their time of reproduction, increases the number of offspring they produce, and increases the biases of producing parthenogenetic offspring in this facultative parthenogenetic species (Moore and Moore, 2003).

A proportion of eggs of Drosophila mercatorum develop parthenogenetically into adult flies. The proportion of unfertilized eggs, which develop parthenogenetically into adult flies increases with the decrease of the population density (Kramer and Templeton, 2001), suggesting that social or olfactory factors play a role of in determining the parthenogenetic development.

Whether a female individual in a bee colony will become a worker or queen is determined by the kind of food they are provided with. At a biochemical level it is also known that the fate of those individuals depends on the level of circulating juvenile hormone in the fourth larval stage, which in turn is determined by brain signals that stimulate (allatotropins) or inhibit (allatostatins) the synthesis of JH (juvenile hormone). The “caste-determining” effect of the food, thus, is determined in the CNS.

In North American field crickets Gryllus firmus and Gryllus rubens, formation of wings is determined by lack of juvenile hormone, as it may be concluded from the fact that in nonpolyphenic winged female crickets, Gryllus assimilis, formation of wings is prevented by stimulating JH secretion (Zera, A.J. et al., 1998), which is under strict control of brain neurohormones (allatotropins and allatostatins). A.J. Zera and coll. (1998) have produced analogous flightless female offspring with enlarged ovaries in the nonpolyphenic cricket, G. assimilis, by simply applying a juvenile hormone analog on adult females. In vivo experiments have demonstrated that flight-capable morphs of Gryllus firmus, synthesize greater amounts of total lipids and triglycerides necessary for flight, whereas flightless morphs synthesize greater amounts of phospholipids deposited in the eggs and ovary. However, the local administration of juvenile hormone in flight-capable morphs increases secretion of phospholipids similarly to the flightless morphs (Zhao and Zera, 2002). Moreover, experimental increase of the level of JH in flight-capable crickets not only increases the ovarian size, similarly to flightless morphs, but it also leads to the loss of flight muscles by histolysis (Zera and Cisper, 2001).

Populations of the common pond skater (a long-legged insect gliding over water), Gerris lacustris (L.) (Heteroptera: Gerridae), in Bavaria, Germany, show remarkable environmentally-induced discrete phenotypic differences in life history and morphology. Populations living in the field ponds are bivoltine (produce two generations annually) with predominantly long-winged individuals, while forest pond populations are univoltine with an increased proportion of short-winged flies (Pfennig and Poethke, 2006).

Seasonal Polyphenism in Insects

Each year the African butterfly B. anynana produces two different seasonal morphs: a wet-season generation of individuals with large marginal eyespots of several colors and a transverse band on the ventral side of the wings, and individuals with small eyespots of fewer colors and no band on their wings in the dry-season (figure 11.4).

Figure 11.4. Dry-season morph (left) and and wet-season morph (right) of Bicyclus anynana.