One of the aims of science is not just to describe what happens, but also to explain it. I have tried to explain why the grasshopper has to go through the rather elaborate motor programme just described, but it would be nice to know know how the motor programme is produced, as well as what it looks like. And this means studying the connexions made by neurons within the nervous system.
We do have some information about how the jump motor programme is produced, but, to be honest, the most important parts of the programme are still a mystery. And there are problems even with the bits that we do know about ...
The job of most motorneurons is to transmit information from the central nervous sytem to the muscles. However, the fast extensor motorneuron is unusual amongst insect motorneurons because it makes synaptic output within the central nervous system, as well as at the extensor muscle. In fact, the FETi makes strong excitatory synaptic connexions to the excitatory flexor motorneurons. This is doubly unusual - not only is the motorneuron making central synaptic output, it is exciting its own antagonist.
Normally, one would not expect antagonistic motorneurons to excite each other, since each would then resist the muscular effect of the other. However, this is exactly what is wanted during the co-activation phase of the jump motor programme.
There is no equivalent FETi-FlTi connexion in the ganglia controlling the mid- or front-legs, and therefore it is highly likely that the connexion is a neural specialisation for the jump motor programme. Its function is presumably to help ensure that as soon as the extensor muscle starts to contract (driven by FETi), there is plenty of tension in the flexor muscle (driven by FlTi), so that the tibia stays flexed while energy for the jump is stored.
There are sensory receptors (campaniform sensillae) on the tibia in the joint region which detect cuticle bending. These receptors excite both FETi and FlTi motorneurons.
Extensor tension feedback
There is a lot of cuticle bending in the build up to the jump, so the feedback tends to keep the flexor and extensor motorneurons active during this time. Which is exactly what we want.
The interneuron called M makes inhibitory synaptic connexions to the excitatory flexor motorneurons. It produces a large burst of spikes at just the right time to shut down the FlTi motorneurons and release the leg for the jump. If M is made to spike early, which can be done by injecting positive current into it through an electrode, then the leg is released early, and the movement is less effective.
Thus the spike burst in M is in itself sufficient to cause the final trigger phase of the motor programme.
The circuits described above seem to contribute to generating the motor programme. However, they are certainly not all there is to it ...