In recent years oceanogra.phers have been using temperature measuring 

 devices that can be read from 0.001C° to 0.0001C°. Mercury thermometers, 

 by comparison, have a precision of about 0.01C°. This advance in the 

 state of the art requires an evaluation of the assumption that temperature 

 times mass is a conservative property of seawater for adiabatic mixing 

 processes . 



Adiabatic mixing in the oceaji is usually discussed in terms of the 

 temperature -salinity (T-S) diagram (Figure la). The assumption is made 

 that when mixing water types A and B the resulting temperature and 

 salinity lie on the straight line joining the two end members. In 

 reality, for adiabatic mixing, the enthalpy (or heat content) is the 

 thermodynamic property of interest, and one should employ a specific heat- 

 content salinity diagram. The error incurred as a result of using a T-S 

 diagram can be determined from a knowledge of the specific enthalpy of 

 the waters mixed. An empirical equation for the enthalpy of seawater as 

 a function of temperature and salinity can be determined from the heat 

 capacity data of Cox and Smith (1959) and. from a knowledge of the heat 

 of mixing of seawater as a function of temperature. 



Okubo (1951) has estimated the heat of mixing of seawater by using 

 previously determined data on the heat of dilution for sodium chloride 

 solutions at 25°C. Clark, Nabavian, and Bromley (1960) determined the 

 heat of mixing of seawater at a salinity of 35^0 o and a temperature of 

 29.i4-°C. By mixing 0.5 gm of "JQPJoo seawater with 0.5 gm of distilled water, 

 they obtained a heat of mixing of -0.097±0.01 joules/gram. In both of 

 these cases the data are insufficient for determining the thermodsmamic 

 properties of seawater as a function of temperature. 



The preliminary results of a set of experiments for determining the 

 temperature dependency of the heat of mixing of seawater are presented 

 here. The calorimeter used (Figure 2) consisted of two copper cylinders, 

 one inside the other and silver soldered to common end plates. The 

 overall dimensions were 13.97 cm long by 11. U3 cm in diameter. The inner 

 cylinder had a series of holes running axially along the upper surface 

 for filling and, upon inversion, for mixing. It was also offset from the 

 axis of the outer cylinder to maximize thermal contact between the waters 

 in the two cylinders. Located on the mid-point of the inner cylinder were 

 two copper wells for positioning of thermistors. The two thermistors, 

 wired in parallel, were calibrated at four or more temperatures between 

 0° and 28°C, using both ac and dc bridges. The plots of logarithm of 

 resistance versus temperature were straight lines. Self -heating effects 

 in the thermistors, as used, were found to be negligible. 



The initial runs were made with an ac transformer bridge (Wayne Kerr 

 High Precision Comparator, Model Wo. b821). An audio -oscillator (Hewlett- 

 Packard, Model 201c) provided an excitation of h volts at 1 kHz. The 

 output was monitored with a tuned amplifier and null detector (General 

 Radio Co., Model No. 1232-a) and an oscilloscope (Hewlett-Packard, Model 

 No. 120B). a 10 kilohm, 5-decade resistance box (Leads & Northrup) was 

 used on the "standard" side of the bridge. Shielded and grounded leads 



428 



