Economic analysis of the process of technical change has often involved macro-level studies of its causes and consequences. Relatively little attention has been given to the, more fundamental knowledge generation process itself. This stems in large part from the real difficulties of obtaining appropriate indicators of research output. The view that there exists a systematic relationship between research expenditures and knowledge increments has been taken up by numerous authors including Evenson (1968), Minasian (1969), Pakes (1978), Gri1iches (1979), and Kamien and Schwartz (1982). It follows naturally from the perception that, in general, science progresses by a sequence of marginal improvements rather than a series of discrete and essentially sporadic breakthroughs (see Burke [1978). Recent studies by Pakes and Gri1iches (1980), Hausman et a1. (1981), and Hall et a1. (1984) have sought direct estimates of the research input-output relationship for research performed by private firms in the non-agricultural sector. To date there appears to be no similar analysis of the public sector agricultural research process. The study reported here represents a first step in this direction. It develops some quantifiable indicators of agricultural knowledge production by the U.S. public sector research system and will also attempt to provide some clues as to the nature of the agricultural research spending-research output relationship.