Session: 1234567891011121314151617         Page 4748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101 of 755

And that was -- if it couldn't be demonstrated in the laboratory, or if you couldn't quantify it in terms of somebody's behavior, then I felt that we were only giving out part of the information. Certainly, you've got to have the signal before somebody can listen to it. But in Salt Lake City, or a couple of hundred miles north of Salt Lake City in the mountain states, you can almost hear the grass grow. In the New York area, you could go as far as White Plains and not hear the station because of man-made noise, or signal interference and things of that kind. So that the maps reflected the behavior of the audience.

And you discovered this by using the black box that you had developed?

No, that was totally unrelated. Ideally, the black box in profusion would have given you the information much better than we got it as we did. And I did not develop the technique of -- the early technique for the listening areas. That was done by John Churchill who had the title, as I recall, of Chief Statistician and predated me in the company by at least a year or so.

But we were trying to keep track in the Research Department of expenditures by advertisers, by categories. By expenditures, I'm referring solely to advertising expenditures so that if the Sales Department wanted to know whether the Colgate Company was spending money in various media, we could tell them the amount and by month where they spent it and for which products and so forth. In those days, some of that was reported to Publishers' Information Bureau --PIB, I think it was called. And we had to go through books and pull it out. Today that's all on a database and you can get it without any trouble at all. But ultimately we had people who did nothing but keep track of the expenditures in advertising,