The mammalian carotid body (CB) is the principal arterial P_O2 detector that maintains blood homeostasis via the reflex control of ventilation. Thus in response to a fall in blood P_O2, chemoreceptors (type 1 cells) located in the CB release neuroactive agents onto afferent nerve terminals, resulting in an increased neural discharge in the carotid sinus nerve (CSN). The cell bodies of the chemoafferent neurons reside in the petrosal ganglion and their central projections terminate in the nucleus tractus solitarii (NTS) in the brainstem. The neurotransmitter mechanisms that operate at synaptic sites between petrosal afferent terminals and CB type 1 cells have generated considerable interest and debate over the last ~60 years. Undoubtedly, much of the complexity arises from the broad diversity of neurotransmitters or neuromodulators expressed by type 1 cells, as well as from the fact that these neuroactive agents can potentially act on several receptor subtypes located either presynaptically on type 1 cells and/or postsynaptically on petrosal nerve endings. Among these agents are dopamine (DA), noradrenaline, acetylcholine (ACh), ATP, serotonin (5-HT), substance P and λ-aminobutyric acid (GABA). Several studies indicate that DA, for many years the leading transmitter candidate and best-studied indicator of the secretory status of type 1 cells, is not the main excitatory neurotransmitter, at least in the rat CB. There is evidence that DA may play a modulatory role in inhibiting release from type 1 cells via an autocrine or paracrine mechanism. I will review recent evidence, based mainly on the use of a co-culture model of rat type 1 cell clusters and petrosal neurons, for the idea that co-release of ATP and ACh onto petrosal endings is the principal mechanism leading to CB excitation by natural stimuli. I will also discuss possible mechanisms by which this principal pathway can be modulated by both inhibitory and excitatory ‘presynaptic’ influences. Firstly, I will discuss recent evidence that 5-HT release from type 1 cells during hypoxia may act in a positive feedback manner via autocrine or paracrine mechanisms to increase chemoreceptor gain. This appears to involve G protein-coupled 5-HT receptors (probably 5-HT₂ serotonergic receptors) and PKC-mediated inhibition of K⁺ channels. Secondly, I will discuss an opposing negative feedback, PKA-mediated inhibitory pathway due to released GABA acting on presynaptic GABA_B receptors on type 1 cells. The theme that is emerging is the presence of a slower presynaptic ‘push-pull’ integrated feedback system, superimposed on the faster postsynaptic events mediated by ionotropic receptors. While other pathways still remain to be elucidated, the reasons for this complexity remain speculative. Nevertheless, the system appears well suited for mediation of synaptic plasticity in the CB, and may well contribute to the well-known altered sensitivity of the organ during adaptation to chronic hypoxia.