All organisms need to communicate with the biotic and abiotic components of their environments. Organisms also communicate within themselves. We saw during our discussions of homeostatic mechanisms and physiological adaptations that insects need to register conditions and changes within themselves and in their environments, and then respond to the conditions. The physiological basis of homeostatic mechanisms is that an organism can somehow measure aspects of the parameters that are regulated. For examples, changes in hemolymph salt concentrations and thoracic temperature need to be registered before corrective action can be initiated. Insects also respond to changes in their environments. Mechanisms of entering diapause seem a particularly good example. Here, insects register token stimuli, most often changes in photoperiod. These token stimuli are translated in neuroendocrine events, which in turn are translated into physiological changes. Nervous systems play crucial roles in perceiving environmental conditions, integrating environmental signals into meaningful information, and then producing responses in the forms of internal, physiological events. In another, speedier set of conditions, try imagining how an insect would fly if it could not see where it was going, or could not determine orientation with respect to its horizon. Within a flying insect, again, try imagining losing control of the small, direct flight muscles that control pronation of wings, or loss of coordination of the major flight muscles. Also consider various sorts of homeostasis during flight: internal temperatures are measured, then hemolymph flows adjusted accordingly; available fuels in hemolymph are oxidized, then replaced from stores quite distant from flight muscles. It would be difficult to regard any area of insect biology that does not depend importantly upon nervous systems.

Understanding insect nervous systems also has a certain practical significance because most insecticides work by poisoning one or another area of nerve function.

The nervous system is made of neurons. Every tissue is composed of cells. Fat bodies are mostly made of trophocyctes. Muscles are composed of muscle cells. Similarly, nervous systems are composed of nerve cells, which are also called neurons. The first figure shows the general structures of nerve cells. Nerve cells have cell bodies, called perikarya (pleural of parikaron). Most major organelles, including cell nuclei, endoplasmic reticula, Golgi bodies, mitochondria and so forth are found in perikarya. Long projections known as axons (pleural of axon) extend from cell bodies; axons carry nerve impulses to their ends, often called terminal arborizations. These arborizations are typically juxtaposed very closely to other nerves or to muscles. At these points, called synapses, electrical impulses are transduced into chemical signals, which are received at membranes of other nerves or muscles. Nerve cells receive stimuli at their dendrites (pleural of dendrite). Dendrites can be found directly on perikarya or at ends of axons.

We consider three kinds of neurons. The upper drawing shows a monopolar neuron, which have a single projection from parakarya. This single projection branches so that there is a dendrite at one end and terminal aborization at the other. Most insect neurons are monopolar. The middle sketch shows a bipolar neuron; these neurons are commonly associated with sense cells, and feature a dendrite at a distal end and a terminal aborization at the other. This arrangement is suited for direct transmission of environmental stimuli to the central nervous system. Multipolar neurons have several axons projecting from perikarya; this cell type occurs in some ganglia, and is also connected with stretch receptors.

Information from outside of an insect, such as light or mechanical stimulus, is most likely conducted from peripheral sense cells directly to central nervous system ganglia by axons without synaptic interruptions. These axons are called afferent axons. Other axons are involved in conducting motor commands from central nervous system ganglia, wherein lie the cell bodies, to muscles; these axons are called efferent or motor axons. Most afferent neurons synapse with other neurons that in turn synapse with still other neurons or else with efferent neurons. The neurons that form connections between afferent and efferent neurons are called interneurones. Nerve fibers are generally aggregates of several or many neurons.

Unlike vertebrates, in which a single, relatively large brain is encased in a skull, central nervous systems of insects are composed of variously distributed ganglia. The major association center is called the brain. The other important ganglia are subesophageal ganglion, frontal ganglion, up to three thoracic ganglia and up to eight abdominal ganglia.

Three major regions of brains are recognized: protocerebrum, deutocerebrum and tritocerebrum. The protocerebrum is the most complex part of insect brains; it has two lobes, and is continuous with the optic lobes. The deutocerebrum contains antennal lobes, from which project both afferent and efferent neurons. The tritocerebrum is a small, bilobed ganglion. Major nerve tracts connect the tritocerebrum to other ganglia. The tritocerebrum is linked to subesophageal ganglion by circumesophageal connectives and to the frontal ganglion by frontal commisure. This part of the brain has efferent and afferent connectives to certain mouthparts.

The subesophageal ganglion is the first ganglion in the ventral nerve cord. This ganglion has motor and sensory connections to salivary glands, mouthparts, and neck. Most insects have three thoracic ganglia, each with connectives to muscles and sensilla. Abdominal ganglia have connections to muscles, though usually fewer than thoracic ganglia.

Numbers of ganglia vary among groups of insects. The thoracic and abdominal ganglia are fused into a single compound ganglion in houseflies. This is an example of extreme fusion. The least fusion is shown in a thysanuran. The last abdominal ganglion is always a compound structure, derived from the ganglia of the last four segments.

The next figure sets forth the structure of a ganglion. The central nervous system is completely invested in what is called the nerve sheath, the most outer part of which is a non-cellular neural lamella and the inner part of which is called perineurium, composed of a single layer of cells. The perineurium is thought to produce and secrete the mucoprotein and mucopolysaccharide neural lamella. The perineurium serves as a blood-brain barrier, and as such it regulates the chemical environment of nerve cells and transports nutrients and other materials between hemolymph and nerve cells. The centers of ganglia are made up of axons and fibers of afferent, efferent and interneurones. These centers specifically do not contain cell bodies, and they are called neuropile. Within neuropiles, some axons and fibers are orientated such that they form fiber tracts. Cell bodies, or perikarya, are found around the periphery of ganglia. The next figure shows that inter ganglionic connectives are similarly arranged, with neural lamella and perineurium forming a sheath around the tracts and giant fibers running along in the middle. Within neuropiles and connectives, each neuron is surrounded by glial cells which form a protective barrier around nerve cells. Glial cells can have a number of arrangements, and are often folded around axons, in which case they are called mesaxon. Glial cells probably insulate nerve processes from each other to reduce uncontrolled electrical noise or "cross-talk" from one nerve to another. Glial folds do not occur around synapses.

An electric current is defined as the amount of charge passing through a conductor in some unit of time. In metallic electrical wires charge is carried by electrons, but in physiological electrolytic solutions current can only flow by the movement of ions because there are no free electrons. Movement of charged particles requires work be done, which requires expenditure of energy. Current will only flow through a conductor if there is a potential difference across it. A potential difference is the amount of work required to move a single charge from one place to another. For our purposes, the electrical events associated with nerve cells relate to movements of ions through nerve cell membranes. These ion movements are also electrical currents. The major ions in nerve physiology are sodium (Na) which is generally in higher concentration outside of axons, potassium (K) which is generally in higher concentration inside, and chloride (Cl). The important point here is that these ions have different permeabilities. These different permeabilities are usually spoken of as different conductances. In electrophysiology we think of an electrical current (movement of ions) in terms of the potential difference that drives the current and of the conductance of the driven ion. A sodium current is written like this:

I_Na = (V_Na) x (g_Na),

where V is potential difference and g is conductance. We recall from our discussions of osmoregulation that have talked about cells in various sorts of solutions. We saw that ions tend to move down their concentration gradients, from high to low concentrations. What we want to introduce now is one more feature of concentration gradients. When determining membrane equilibrium we need to consider both chemical gradient and electrical potential.

Let us imagine a two chambers separated by a membrane. If there is a high Na concentration on one side and a low Na concentration on the other, there will be a greater flux of Na in the down-hill direction, compared to the up-hill direction. This translates into a net down-hill Na flux. The Na flux would generate a net current flow. Now if we apply a potential difference between the two chambers, the potential difference can be twiddled such that the chemical concentration difference can be balanced by the potential difference. In theory an appropriate potential difference can be used to offset any chemical concentration difference. We can express this in slight more formal terms.

Nernst formulated a mathematical expression to relate a potential difference to a given concentration difference:

where R is the gas constant, T absolute temperature, C concentration, z valance of an ion and F the Faraday. All the constants can be treated accordingly, and for 37 oC we can write:

Here E is potential difference in millivolts and this is called the electrochemical equilibrium potential, or the Nernst equation. This is extremely useful for electrophysiologists because concentrations of ions can be measured and potential differences across membranes can be measured. Hence, electrophysiological events can be expressed in measurable terms. This becomes the basis for research on ionic mechanisms of nerve conduction.

The last complicating point is that resting potentials of nerve cells are made up from interactions of three ions, so the Nernst equation need be re-written to account for that:

where I stands for intracellular concentration and O stands for extracellular (outside) concentration. This is the Goldmann field equation, and for the giant squid axon, this model predicts the resting potential of membranes fairly well.

The physiology of maintaining a potential difference across membranes takes us back to the physiology of moving ions across membranes. One of the most well studied ion pumps is the sodium-potassium ATPase. In this system three Na ions are moved outwardly while two K ions are moved inwardly at the expense of 1 ATP. Hence, this pump requires extracellular K to function. Ions typically leak across membranes, and ion pumps are probably running constantly to maintain resting potentials in nerve cells.

Now let us change gears slightly and look at examples of nerve physiology in action. The next figure shows examples of what are called spike frequencies. The first example shows spiking of antagonistic motor neurons in a walking insect. We see a burst of action potentials, then a relatively quiet period, then another burst of action potentials. The second example shows responses of olfactory receptor neurons of a male moth to different levels of a sex attractant pheromone, tetradecenyl acetate. Again, at high levels of sex pheromone, we see very high spike frequencies.

These are examples of spiking in efferent and afferent neurons. Each spike represents an action potential. The next figure shows an action potential resolved at sub-millisecond level, along with a diagram of the ion dynamics associated with the action potential. Again, the resting potential is maintained by ion pumps. About -70 mV is a common sort of resting potential. Time points (1) and (7) shows the original ion distribution. As an impulse moves along an axon, the first phase of an action potential is caused by a sharp increase in Na conductance. This is due to opening of voltage-sensitive ion channels (3), so that Na ions can follow their concentration gradients across the axon membrane into the axon (4). A rapid change in potential difference across the membrane follows, perhaps as much as 100 mV. Na conductance is high for about 0.5 millisecond, whereupon K conductance rises, and K ions rush out of the axon. The sharp increase in outward K flux causes the falling phase of the spike potential (5). The high K conductance causes an overshoot, known as the positive phase. The negative after potential results from a local high accumulation of K outside the axon membrane. Na-K ATPase (6) eventually brings all ions back to resting potential concentrations.

A few points remain to be addressed. First, a very small number of ions is involved in an action potential, on the order of 100 ions, or less than one millionth of available K ions inside an axon. Second, action potentials usually move in only one direction due to refractoriness. This can be explained by considering conductances. Na conductance is very much higher than K conductance. Third, impulses must move along axons for information to flow. So we want to consider how impulses move along an axon. The local circuit theory is based on the idea that areas of an axon next to a point of action potential are negatively charged so that a local current flows away from a point of action potential on the inside, and towards it on the outside, as shown in the next figure.

The next figure is a model of a synaptic cleft. The important features include presynaptic vesicles, many filled with transmitter substance, usually acetylcholine in nerve-nerve synapses, and L-glutamate in nerve-muscle junctions. Gamma-aminobutyric acid (GABA) is the usual transmitter chemical in inhibitory junctions. Other features are presynaptic membrane, and postsynaptic membranes. Finally two proteins are important, acetylcholine esterase and acetylcholine receptor sites. This drawing shows the synaptic cleft, which separates pre- and postsynaptic membranes by 200 to 500 angstrom units.

Contemporary wisdom holds that depolarization of presynaptic membrane leads to fusion of presynaptic vesicles with membrane, and release of acetylcholine by a process of exocytosis. Vesicles contain on the order of a few thousand molecules of transmitter; each depolarization causes about 100 or so vesicles fuse with presynaptic membranes and release their molecules. Acetylcholine molecules move randomly as in free solution within synaptic clefts, and it is about equally possible for any given molecule to meet acetylcholine esterase or acetylcholine receptor sites. The enzyme hydrolyzes acetylcholine to acetate and choline. The acetylcholine molecules that bind to their receptor sites eventually cause a depolarization of the postsynaptic membrane called excitatory postsynaptic potential (EPSP), which is transmitted to the axon. Axon depolarizations and continuation of the nerve impulse follow.