Prop noise and electronic interference may give an irregular trace but do not seem 

 to affect the accuracy. Even with a straight line trace, there is no reason to expect a greater 

 agreement of IT and bucket temperature than .5°C. 



From our limited experience, we believe the IT readings are essentially the surface 

 water temperatures for the air/sea temperature range and conditions under which we oper- 

 ated during the December 1963 and April 1964 flights. We have no other direct evidence to 

 present. 



Flights used for contiguous coverage should be on the same day or within a few days 

 of each other. However, the necessity of obtaining clearance in certain restricted areas may 

 make this difficult or impossible to achieve. We had to alter our flight path slightly for this 

 reason. 



Before we rely on monthly flights, the changes that occur in shorter periods should 

 be investigated. Particularly during certain months of known instability and in areas as the 

 North Carolina Cape - Gulf Stream confluence. We are not able to do this under present 

 manpower conditions . 



We suggest temperature be shown to the whole degree C . [or KJJby 1 degree intervals 

 on a grid along the flight path. No attempt to interpret the data by isothermal lines should 

 be done with initial results. Let the recipients do their own interpretation. 



The power supply seems marginal and barely adequate. A recording device for 

 machine interpretation would be preferred. If strip recorders are used, then the number 

 and frequency of in-flight time or position and bucket temperatures should be standardized. 



-59- 



