This study focused on the quality of arrests resulting from calls versus officers' observations, for the period June 1978 to May 1979. It collected data on case dispositions as well as handling time, court time, and overtime for deputies and detectives. The crimes yielding the highest probability of arrest tended to be observation events, such as drunk driving, reckless driving, weapons violations, and felony drugs. Analysis of total felony arrests compared to observation felony arrests suggested a tendency to 'dig up' arrests rather than wait for them to happen. Cost analysis, however, revealed that observation arrests cost more than those resulting from calls and were less likely to go anywhere in terms of desirable dispositions. Thus, approximately one-third of the office's arrest effort was nonproductive. The study also examined costs and types of arrests for drunks and disorderlies, narcotics, weapons laws, and grand theft auto. Most arrests in all categories were from observation and by two-man cars. In the present fiscal environment, governments definitely will question the productivity of proactive police postures. The paper contains tables, 6 footnotes, and 13 references. For related materials, see NCJ-86875.