ABSTRACT
Pre-war rice statistics in Sri Lanka are very poor in quality or nonexistent. This is largely because of the British emphasis on plantation export crops at the expense of paddy production. Until the maha 1947/48 season there is no continuous yearly series. Periodic agricultural census figures and independent estimates prepared by ministries provide us with benchmark figures, but do not relate to one another and can only be considered to be very rough estimates. The 1946 Census of Agriculture marks the beginning of data collection based on plotwise enumeration, and during the maha 1947/48 season, the Department of Census and Statistics began the regular collection of paddy area and yield statistics using village headmen and Divisional Revenue Officers (DRO) as data collectors. Although these early statistics were based on eye estimates only, they set a precedent for future, improved data collection procedures. Rice yield statistics, in particular, were found to be underreported. Prior to 1949, most village headmen routinely reported paddy yields of 14 bushels/acre (0.72 mt./ha.). In 1949, the Department of Agriculture ran crop-cutting experiments that showed mean yields of 27 bushels/acre (1.39 mt./ha.).
