of intercomparison of protected thermometers was carried out in the 

 field measurements. The thermometers were used in pairs, and one 

 of each pair was shifted periodically so that a given thermometer 

 eventually was paired with a number of other thermometers. From 

 a total of 3,036 intercomparisons, the standard deviation between the 

 corrected readings of a pair of protected thermometers was ±0.009° 

 C. Of these comparisons, 2,735 involved thermometers having a 

 range of —2° to +8°, with a standard deviation of ±0.009° C, 182 

 comparisons between thermometers of range —2° to +20° or greater 

 gave a standard deviation of ±0.013° C, and 119 comparisons were 

 between thermometers with a range of +3° to +13° and gave a 

 standard deviation of ±0.007° C. As most of the observed tempera- 

 tures listed in the Table of Oceanographic Data are means of the 

 corrected readings of a pair of thermometers and, as many of the 

 thermometers had recent ice point determinations, it is considered 

 that the tabulated observed temperatures are good to ±0.01° C. 



As in the past years all salinities were measured with the Wenner 

 salinity bridge with the exception of the surface water samples from 

 stations 8728, 8757, and 8758, which because of their low salinities 

 were measured on a Hytech inductive salinometer. In the field 

 measurements, the bridge was standardized with sea water from an 

 oil-sealed carboy. Copenhagen standard water of batch P36 was 

 measured as an unknown twice during each salinity run. At the end 

 of each survey, these measurements were used to correct the tentative 

 value of the salinity of the oil-sealed carboy which had been used as a 

 substandard of salinity and to determine the corrections to the 

 salinities for the survey. The corrections indicated were as follows: 

 First survey, +0.0l75°/oo; second survey, +0.0l5°/oo; third survey, 

 + 0.025°/oo; fourth survey, +0.0l75°/oo; and the postseason survey, 

 + 0.0045%o. The corrections for the first four surveys are large 

 because an incorrect Wenner bridge X-dial value for Copenhagen 

 standard water was used. 



Necessary corrections have been applied to the tabulated values 

 appearing in the table of Oceanographic Data. Temperature control 

 of the samples in the electrolytic cells of the Coast Guard's Wenner 

 bridge limit the precision of the individual measurements to 0.005°/oo. 

 In view of the precise chlorinity determinations of February 1960 and 

 the check runs on the calibration curve, it would appear that the salin- 

 ities measured with Coast Guard Wenner bridge were measured with 

 an accuracy of about ±0.01°/oo 



In November 1963, in conjunction with salinity determination ex- 

 periments for the U.S. Naval Oceanographic Office, salinity values 

 were obtained that suggested that a new calibration curve be deter- 

 mined. The methods described on pages 34-37 of Bulletin 46 of this 

 series were employed. A measured value of B, 200.2 ohms, and four 



45 



