Investigation of the actions of drugs on vascular smooth muscle is frequently complicated by indirect actions mediated through neural elements. However, valuable information on the direct actions of drugs on vascular smooth muscle can often be obtained from experiments on nerve-free preparations. Perfused placental vessels (von Euler, 1938;Panigel, 1962), spirally cut umbilical arterial and venous strips (Somlyo, Woo & Somlyo, 1965), chick amniotic membrane (Ferguson, 1940;Cuthbert, 1963), and rabbit ear vessels before innervation (Clark & Clark, 1943) have been employed in the past for this purpose. A nerve-free vascular smooth muscle preparation, responding uniformly to various autonomic transmitters could also be used to elucidate the nature of their storage and release.In the present study the isolated perfused human umbilical artery was used to study the direct action of drugs on the vascular smooth muscle. It was observed that with the method employed many vasoactive agents gave fairly uniform responses. Modification of these responses by other drugs could, therefore, be studied. The preparation exhibited spontaneous tone and could give a dilator response to drugs, without the previous addition of a spasmogenic agent. METHODSUmbilical cords were obtained from the Maternity Ward of Shree Sayaji General Hospital, Baroda, as soon after delivery as possible. After identifying the foetal and the placental ends, every cord was firmly tied at both ends and kept in a flask containing perfusion fluid of the following composition (g/l.): sodium chloride, 9.00; potassium chloride, 0.42; calcium chloride, 0.24, sodium bicarbonate, 0.2 and dextrose, 1.0. Tying ensured that any blood in the lumen of the vessels remained in a fluid state and could subsequently be washed out easily. In earlier experiments when the cords were not tied, many of them could not be used because of the intravascular clotting. The cords remained viable for about 5 hr after delivery. Experiments were, therefore, started soon after the collection of the cords.Before cannulation the cords were placed in fresh perfusion fluid at 34' C for 10 min. This considerably reduced the spasm of the vessels and thus facilitated cannulation. A piece of cord about 5 in. long was selected (excluding any portion with injury and pseudoknots), and this was cut at both ends with a sharp pair of scissors. One of the umbilical arteries was cannulated at the

Treatment with phenoxybenzamine and dichloroisoprenaline prevented the rise of blood pressure, contraction of the nictitating membrane and increase in cardiac contractile force produced by intravenous injections of bretylium and guanethidine in anaesthetized or spinal cats. Treatment with cocaine, imipramine or reserpine reduced the sensitivity to bretylium and guanethidine of the spinal cat. In the spinal cat treated with reserpine, sensitivity to the drugs could be restored by an infusion of noradrenaline. Chlorpromazine also blocked the pressor and nictitating membrane responses to bretylium and guanethidine. The drug effects were unaltered 1 hr after bilateral adrenalectomy. During the intravenous infusion of noradrenaline into spinal cats the pressor responses to bretylium and guanethidine were increased, whilst those to adrenaline and to noradrenaline were decreased. Guanethidine (3 to 5 mg/kg) injected intravenously into the cat caused a sudden relaxation of the rat isolated stomach-strip bathed in blood. In seven similar experiments bretylium (3 to 5 mg/kg) relaxed the strip only once; in the other six experiments there was either no effect (four experiments) or an increase in the tone of the strip (two experiments). It is concluded 'that the initial adrenergic effects of bretylium and guanethidine are mediated, at least in part, through a release of catechol amines from stores in the effector organ.

The isolated cremaster muscle preparation and spermatic nerve-cremaster muscle preparation of the guinea-pig were studied in vitro to determine their suitability as pharmacological test models. The preparation was contracted by acetylcholine, carbachol, succinylcholine and decamethonium (pD2 values, 4-2, 5-3, 7-3 and 7-4, respectively) through an action on a curare-sensitive cholinoceptor. Lobeline and DMPP were ineffective. Nicotine contracted the muscle, but there was tachyphylaxis. Tubocurarine and hexamethonium presumably competitively antagonized acetylcholine (pA2 values, 7-3 and 5-8); lobeline was a non-competitive antagonist (pD'2 value, 6-4). Atropine and mecamylamine exerted a dualistic action against acetylcholine (final pD'2 values, 5-3 and 6-7, respectively). Tubocurarine, succinylcholine and decamethonium exhibited their typical action when tested with spermatic nerve-cremaster muscle preparation; the latter two drugs also produced muscle spasm. Hexamethonium was a weak blocker of neuromuscular transmission. Atropine, mecamylamine, lobeline and DMPP exhibited neuromuscular blocking activity; however, directly evoked muscle twitches were also notably affected. The cremaster muscle preparations seem to add usefully to the list of currently used in vitro tests, with the added advantage that a mammalian skeletal muscle model is used for simultaneous quantitative studies.