Oxytocin, the most powerful uterotonic agent known, is released from the pituitary gland in large amounts during parturition in all placental mammals studied so far, including humans. Although parturition can proceed in its absence, oxytocin is thought to play an important role (see Russell & Leng, 1998). In the rat, pregnancy normally lasts for 21 days. About 24 h before the pups are born, increased production of prostaglandins by the uterus induces luteolysis, and ovarian progesterone production falls dramatically. This fall is an essential prelude to parturition; if prevented, then the rat pups will remain unborn. The fall leads to a further increase in prostaglandin production, and, directly or indirectly, to a host of changes that prepare the uterus and birth canal for parturition. In the last few hours of pregnancy, oxytocin receptors appear in high concentrations in the uterus, and establish a positive-feedback loop between the uterus and the hypothalamic oxytocin system. Uterine contractions, triggered by prostaglandins, excite the oxytocin cells, and oxytocin release triggers further prostaglandin production and further uterine contraction. Thus progesterone plays a critical role in the timing of parturition through its peripheral actions (see Leng & Brown, 1997). A paper in this issue of The Journal of Physiology (Brussaard et al. 1999) suggests that actions of progesterone at the oxytocin cells in the hypothalamus may also be important for parturition. Classically, progesterone acts through specific intracellular receptors to regulate gene expression. However, metabolites of progesterone can also have membrane actions, and in particular, allopregnanolone can act at GABAA receptors to potentiate the actions of GABA, depending upon the particular subunit composition of the receptor. GABA is an important neurotransmitter for oxytocin cells about 45 % of all synapses onto them contain GABA, and the total number of GABA synapses in the supraoptic nucleus is substantially higher in lactating animals than in virgins (El Majdoubi et al. 1997). The GABA innervation appears to play a role in patterning the pulsatile discharge of oxytocin cells that is observed both during parturition and during suckling-induced reflex milk ejection (Moos, 1995; Voisin et al. 1995). Brussaard et al. (1999) recorded GABAA receptor-mediated spontaneous monoquantal inhibitory postsynaptic currents (sIPSCs) from rat supraoptic neurones in hypothalamic slices in vitro. They found a higher incidence of sIPSCs in pregnant rats than in virgin rats, consistent with the observations of an increase in the density of GABA-containing synaptic boutons. Importantly, the sIPSCs were markedly prolonged in the presence of allopregnanolone. Taking into account the frequency and amplitude of sIPSCs, the action of allopregnanolone and the hypertrophy of oxytocin neurones in lactation (reflected in increased capacitance), Brussaard et al. (1999) inferred that the effective GABAA receptor-mediated synaptic current density was much greater in pregnant rats than in virgin or lactating rats. Thus the collapse of progesterone production at term may abruptly reduce the effectiveness of GABA inhibition, and thereby enhance the excitability of oxytocin cells. Clearly this may be important during parturition, but the effect may not persist for long. Indeed, within a day the duration of sIPSCs is significantly longer in the absence of allopregnanolone, which now has no significant effect. This seems to be due to a rapid switch in the types of a subunits inserted into the GABAA receptors. By mid-lactation, a massive change in expression of GABAA receptor subunit mRNAs is apparent. With competitive polymerase chain reaction Brussaard and colleagues found that, while the expression of both a1 and a2 subunit mRNAs was increased, the ratio of a1 : a2 subunit mRNA expression was changed 8-fold in favour of a2 subunit mRNA. (ABSTRACT TRUNCATED)