The Practical Salinity Scale (1978) provides a precisely defined, unambiguous method that is universally applicable to all waters for determining salinity, or more precisely the practical salinity, from measurements of conductivity, temperature, and pressure.In the terminology of standards, it is highly desirable to have a scale that is also unique. For example, we would hope that measurements performed on a particular water sample at different temperatures would lead to the same value of the practical salinity. We have investigated this problem by examining the behaviour of natural seawaters from several of the world's oceans. We conclude that the PSS provides a unique scale for salinity measurements for these waters within the bounds in salinity.

The Practical Salinity Scale (PSS) 1978 is defined only for salinities within the range 2-42. We have investigated the relationship between mass-determined salinity, electrical conductivity, and temperature for salinities between 0 and 2 with the aim of developing an extension to the Practical Salinity Scale 1978. The paper presents our data, on the basis of which the following correction is propo...

The Practical Salinity Scale 1978 (PSS) defines the relationship among salinity, electrical conductivity, temperature, and pressure. Strictly speaking, the PSS is only applicable to waters with the same relative chemical composition as standard seawater. We have tested the predictions of the PSS for samples of Atlantic and Pacific waters. The results indicate that the PSS is valid for these natura...

A new portable salinometer has been developed which is based On a direct determination of the conductivity ratioR_{t} = (C_{x}/ C_{s})_{t}of sample(x)to standard(s)seawater in a dual-cell, continuous-flow system. The new salinometer requires only 10 ml of unknown and much less of standard, drawn from the source bottles through fine Teflon tubes, to obtain complete flushing and several repeat readi...

The paper describes a low-power barometer intended for remote weather stations, but also meeting the requirements for manned stations and airports, in which the condensation temperature of carbon disulphide (CS 2 ) is used to determine the barometric pressure (p) . A heated cylindrical bulb with a re-entrant well for a thermistor through the bottom and an internal radiation shield is 1/3 filled with CS 2 . Helical springs assist CS 2 migration on wetted surfaces and allow liquid and vapor to pass one another in the small diameter exit-condenser tube. A miniature Dewar flask gives thermal insulation and a 0.01-mm beryllium-copper diaphragm transmits the external pressure. The condensation temperature is read with a simple Wheatstone bridge and dc amplifier giving an output V_{0} = 0.5(p- 100 Pa) V. Pulsed power with the pulse length controlled by a second thermistor on the outlet tube is used for efficiency. Long term tests of a number of barometers have given power levels around 40 mW at 20\deg C and indicated maximum drifts of \pm50 Pa/year, \pml0 Pa/day, and \pm2 Pa short term.

Temperature is one of the most frequently measured parameters of the ocean because of its importance to the understanding and prediction of oceanic and meteorological events, and also because the measurement is required for the determination of salinity and density. The ocean temperature range is narrow,-2\degto35\degC, but measurement is complicated by the harsh ocean environment, the necessity o...

A portable sallnometer for use with small bottle samples is described. It is based on the continuous flow principle used in our precision laboratory salinometer. However, (1) it is not thermostatted, instead it has two cells in the same bath and measures the conductivity ratio of sample to standard seawater at the same temperature; (2) the water is drawn from the sample bottle by suction instead of driven by pressure; (3) the required sample is less than 1/10 that of the laboratory model so that standard water can be used continuously as reference; (4) there is no contamination of the rest of the sample, almost the whole of which is available for other measurements. The reading circuit operates on 100-200 Hz square wave current and determines the ratio of conductivity of the measured sample to standard seawater at the same temperature directly. Appropriate trims are provided to compensate for cell constant differences and deviation of the standard seawater from exactly 35 0/00 S.

This paper outlines the work done at the five laboratories which provided, at the request of the Joint Panel on Oceanographic Tables and Standards (JPOTS), the basic data for the new Practical Salinity Scale 1978, Interlaboratory agreement of a very high order was achieved. The new scale is in process of acceptance by all major oceanographic societies, at which time it will be officially recommended for all future reporting of oceanographic data.

The ratiosZ_{K,t}of electrical conductivities of potassium chloride (KCI) solutions of known concentration (K) to standard seawater at the same temperature have been measured at15\degC and24\degC for solutions withZ_{k,15}between 0.96 and 1.04. The "normal" concentration (N or K_{N}) givingZ_{N,15}= 1was found to beK_{N} = 32.4356gKCI/kg solution. The effect of temperature onZ_{N,t}was measured ov...

A new type of barometer, based on the sprinciple of the vapour pressure thermometer is described. Instead of measuring the equilibrium vapour pressure to determine the temperature the temperature at which the vapour pressure matches ambient pressure is measured to determine the barometric pressure. The barometer can operate at power levels of 20-25 mW or less and since it is based on a fundamental property of the pure substance should have exceptional stability, say 20 Pa ( atmos.) indefinitely. Current models give dc output variation of 0.5V/kPa over the normal barometric pressure range. A higher power version (200 mW) allowing much less sophisticated filling procedures is also described.

A new method of measuring the variation with temperature of the electrical conductivity of seawater is described. The sample water driven by low air pressure through a fine-tube heat exchanger is brought to thermal equilibrium with a thermostatted bath containing the cell. By operating two exchanger-cell pairs in different baths but supplied from the same sample the ratio of conductances (Ct1/Ct2) at any two temperatures t1 and t2 is obtained directly. Measurements were carried out on standard seawater, synthetic seawater (35‰ S), and a lot of Atlantic water (33.2‰) from below the freezing point to 35°C. In each case the results can be fitted by a 4th order equation with a maximum deviation of any measured point of less than 1 ppm S. The new equation for standard seawater can be introduced into a number of C, T, P → S conversion formulae.

This paper describes a semiautomatic apparatus for routine or precision comparisons of thermocouples of the same type in the temperature range 0–1100 °C. The couples being compared are welded together at the tips and placed in a tube furnace which is heated at rates varying from 10 to 100 °C./min. Measurements of carefully annealed thermocouples show that in the temperature range 300–1100 °C. platinum – platinum 10% rhodium thermocouple comparisons may be made to accuracies of ± 0.3 μv. (± 0.03 °C.) at heating velocities as great as 15 °C./min. while accuracies of ± 1.5 μv. at velocities of 100 °C./min. are feasible. The furnace temperature is varied by means of a motor-driven variac with automatic reversal at peak temperature. In addition to this standard comparison procedure, provision is made for comparing corresponding elements of the couples, for suppression of all or part of the measured e.m.f., and for measuring the whole e.m.f. of all couples when a comparison of different types is desired. The system can be adapted to XY recording with total e.m.f. plotted against e.m.f. differences.

Heat conductivity of natural rubber and GR–S was studied in the range from + 50 °C. to − 170 °C. and from 0 to 100% stretch. The apparatus used was a greatly modified version of one designed by Schallamach. The conductivity of both types of rubber at 0% stretch lies in the range between 3.5 × 10 −4 and 4.0 × 10 −4 cal./sec. cm. deg. C. Stretching increases the rate of change of conductivity with temperature of both natural rubber and GR–S, and decreases the conductivity of the latter. On lowering the temperature and raising it again natural rubber exhibits a complicated hysteresis phenomenon, while GR–S shows a hysteresis loop caused by a second order transition near the brittle point. The hysteresis phenomena of both types of rubber near the second order transition temperature shows considerable similarity to the changes in specific heat observed by Bekkedahl and coworkers. Above and below the transition region the heat conductivity decreases approximately linearly with temperature as might be expected from classical theory. The variation through the second order transition does not agree with classical theory, but may be explained qualitatively on the basis of a diffuse lambda type transition.