While the classical view is that the autonomic nervous system consists largely of antagonistic cholinergic and adrenergic nerves, about sixteen putative neurotransmitters in the autonomic nervous system have been proposed in the past few years, including various monoamines, polypeptides, purines and amino acids. Modulatory transmitter mechanisms have also been recognised, including prejunctional inhibition or enhancement of transmitter release, postjunctional modulation of transmitter action, and the secondary involvement of locally synthesised hormones and prostaglandins. The existence of more than one transmitter substance in some nerves is now widely recognised, and suggestions are made about the ways that this can lead to further peripheral control mechanisms at nerve terminals themselves. The cotransmitters always have synergistic actions on postjunctional effector cells, but two different mechanisms are postulated. (1) If both substances are stored in the same vesicles (for example, ACh or NA with ATP), release is closely parallel at all impulse frequencies. Upon release, the cotransmitter, in addition to having a direct action on postjunctional cells, may facilitate the action of the other transmitter and/or act as an inhibitor of its release. (2) If the two substances are stored in separate and different vesicle types (for example ACh or NA with some peptides), then differential release is possible at different impulse frequencies; the peptides released at higher frequencies modulate the role of the classical transmitters, by both prejunctional enhancement of its release and postjunctional facilitation of its action. The long-term effects by 'trophic' factors released by excitable cells including NGF, purine nucleosides and polypeptides, on development and regeneration were considered. The part played by peripheral neuroeffector control mechanisms has been underestimated. These are additional to central and ganglionic control mechanisms and are much more elaborate than originally thought.