which were to be processed for spot heights. By careful selection from the 

 available printSs Hydro was able to come up with two good pairs with iden- 

 tical orientation and another which was only 5° different in direction from 

 the other two. They set up the grid system in the third so that it was aligned 

 in the saane direction as in the first two„ A grid system of 60 x 90 points 

 was finally settled upon, and Hydro began the laborious task of grinding out 

 5,400 spot heights for ea.ch of three pairs of stereo photogj^aphs of the sea 

 surface. 



With the data soon to be pouring out of HydrOj, the next important prob- 

 lem was to get it analyzed. We turned again to the DTMB UMVAC. By tbds 

 time the demands for time on their computer had gro-sTsm tremendously. They 

 wouldj however, be willing to run the analysis if it were first programmed. 

 Dr. Pier son investigated t!ie possibility of having the programming done at 

 NYU. The Engineering Statistics group of the NYU Research Division, under 

 the direction of Mr. Leo Tick undertook to do the task in about two months. 

 So, while Hydro was amassing the tables of numbers that were so import- 

 antj Mr. Eiiianuel Mehr was working out the problems of telling a mass of 

 vacuum tubes and wires (the UNIVAC) what to do with the numbers we in- 

 tended to feed it. By the first of April, Hydro was beginning to grind out tljie 

 data and I went out to see how they were doing and talk with thena about the 

 project, I was very much impressed with the careful and accurate job they 

 were doing. They felt that their observations were accurate to plus or 



minus one foot and reproducibility to about 2/5ths of a foot. As a check, I 



13 



