The objective of this dissertation is to describe and analyze the experimental practices of agronomists from the Olericulture Section of the Campinas Agronomic Institute with onions in the period from 1937 to 1960, and the experimental ordering and standardization procedures that constitute onions with defined and known characteristics, filtering and fostering relationships. The Campinas Agronomic Institute was one of the most relevant Brazilian research institutes at the time. In the 1930s, it had already adopted experimental models based on statistics developed in England and which were quite recent. Furthermore, the time frame allows us to monitor the beginning of research on vegetables in a systematic way in the country, since until the creation of the Vegetable Culture section in 1937, the IAC, founded in 1889, did not have sectors responsible for research aimed at the introduction and improvement of crops aimed at supplying the domestic market and not for industrial or export-oriented production.