Ventriculocisternal perfusions were carried out on chronically prepared, unanesthetized goats. Measurements were made of steady-state rates at which inulin, fructose, creatinine, urea, K, Na, and labeled water (TOH) were removed from perfusion fluid at various hydrostatic and osmotic pressures. The ventricular system is virtually impermeable to inulin. Inulin is removed from subarachnoid spaces by bulk absorption at rates which vary linearly with hydrostatic pressure. Net absorption ceases at –15 cm H2O. Rate of net formation of cerebrospinal fluid (CSF) is equal to inulin clearance plus the difference between outflow and inflow rates; normally it averages 0.16 cm3 min–1 and is essentially independent of hydrostatic pressures in the range –10 to +30 cm H2O. Net rate of formation is linearly related to total osmotic pressure differences between plasma and CSF. The coefficient of osmotic flow is greater than that measured from diffusion of TOH, as in other porous membranes. Passive permeability characteristics resemble those of the vasopressin-treated toad bladder.

Cannulas were implanted in the lateral ventricles, cisterna magna and subarachnoid spaces of goats. Perfusion with synthetic cerebrospinal fluid (CSF) permitted study of exchange rates of test materials between CSF perfusion fluid and blood. Diodrast or phenolsulfonphthalein in low concentrations is removed from ventriculocisternal perfusate by a process of active transport resembling secretion by the proximal tubules of the kidney. The transfer maximum is 2–3 µg/min. and is reached at an inflow concentration of about 20 µg/ml. Active transport occurs from a volume of about 2 ml (total CSF volume is c 20 ml) located in the region of the 4th ventricle and cisterna magna. The choroid plexus of the 4th ventricle could be the site of active transport. After inhibition of active Diodrast transport, a passive component of transfer is revealed. Comparative studies with creatinine, fructose and inulin show that passive transfer takes place by diffusion as well as by absorption in bulk. Rates of passive transfer of these substances (per unit concentration difference) are comparable in magnitude with diffusion rates from the capillaries in 1 gm of skeletal muscle.

Respiratory responses to inhaled CO2 were measured in unanesthetized goats during repeated perfusions of the ventriculocisternal system through chronically implanted cannulas. [HCO3–] and pH were measured in carotid loop blood and cisternal outflow. Average steady-state alveolar ventilation increased fourfold when cerebrospinal fluid (CSF)-[HCO3–] was reduced from 30 to 15 mm/liter at constant, normal CO2 pressure or threefold when CSF pH changed from 7.32 to 7.21 at constant, normal CSF-[HCO3–]. Sensitivity was two- to sevenfold greater than reported for anesthetized animals. At constant CSF pH the ventilatory response to inhaled CO2 was 60% of the isobicarbonate response. Pco2 in cisternal outflow was shown to approximate that in cerebral tissue. HCO3– flux was measured as a function of CSF-[HCO3–] and concentration profiles between CSF and capillary blood were considered. Alveolar ventilation is a single linear function of [H+] in tissue fluid located two-thirds to three-fourths of the distance along the functional concentration gradient of HCO3– between CSF and blood at all values of Pco2 and CSF-[HCO3–] which we investigated.

Techniques have been devised for repeated perfusion of the ventriculocisternal system in chronically prepared, unanesthetized goats. The methods are designed for investigations of a) functional effects of prolonged changes of cerebrospinal fluid (CSF) composition and b) quantitative aspects of CSF formation and exchange of materials between CSF and blood. The present paper provides a technical background for subsequent papers dealing with specific problems. Goats tolerate perfusion rates of 1–3 ml/min at pressures adjustable between –10 and +30 cm H2O. Measurements are made of inflow rates, outflow rates, and concentrations of test materials entering and leaving the system. Composition of normal CSF has been determined and directions are given for preparation of sterile synthetic CSF. Brain weight is about 100 g and choroid plexuses 500 mg. CSF volume is 20–25 ml of which 8–12 ml is in the ventricles. Pressure-volume characteristics are described. Steady-state difference between inflow and outflow rates is a linear function of hydrostatic pressure; the slope is related to hydrodynamic resistance to bulk absorption. Preliminary observations are given to show effects of reduced Ca++ on nervous control of the cardiovascular system.

The amount of extra oxygen consumed by skeletal muscle following work (isotonic contractions) or isometric contractions was measured in the normally circulated gastrocnemius-plantaris muscle group of the dog. Work was varied by: a) having the muscle perform the same work a repeated number of times; b) varying the load against which the muscle-contracted, c) varying the amount of shortening of the muscle by limiting the duration of stimulation. The effect of frequency of stimulation was assessed by using two frequencies of stimulation in conditions a) and c). Finally, the extra oxygen consumed following isotonic work was compared with the extra oxygen consumed following isometric tension development. It was found that extra oxygen consumption of skeletal muscle following activity was not a direct consequence of the work done nor was it related linearly to the amount of shortening. The extra oxygen consumption appeared to be a response to the number of nervous stimuli delivered to the muscle. At a sufficiently low frequency each stimulus evoked the consumption of a constant quantum of oxygen. At higher frequencies, after a relatively small number of stimuli, oxygen consumption per stimulus decreased with successive stimuli.