<p> This study deduced the actual state of change in plot size, street width, and number of sides facing streets in the agricultural lands to determine how streets act as buffer space. The effectiveness of these indicators as evaluation indexes to describe characteristics of agricultural lands that remain in use was examined. Concretely, the following categories of items were reviewed by comparing agricultural lands in 1983 and 2016.</p><p> I. Characteristics of plot size, street width, and number of sides facing were reviewed.</p><p> II. A typology of plot transformations and characteristics of changes in plot size, street width, and number of sides facing streets were reviewed relative to patterns of plot transformations.</p><p> In general, the 194 agricultural plots that were 1000–5000 m² in 1983 fell to 95 plots in 2016, but their proportion of absolute number increased from 57.0% to 70.9%. Agricultural lands connected to district streets increased from 69 to 84 plots, with an increase in ratio from 20.3% to 62.7%. Agricultural lands that do not touch streets decreased from 36 to 4 plots between the two time points, and the ratio decreased from 10.6% to 3.0%. That is, the ratio of the agricultural lands connecting to larger streets and the ratio of agricultural lands with more sides facing streets increases, and the buffer space between agricultural lands and urban area was greater in 2016 than in 1983. Moreover, in 2016, the ratio of agricultural lands smaller than 1000 m² that connected to district streets was higher than the ratio of agricultural lands smaller than 1000 m² connected to fine streets, and the plot size of agricultural lands tended not to relate to the width of connecting streets relative to agricultural lands in 1983.</p><p> Eight plot transformation types were found. The number of agricultural lands connecting to a larger street and the number of sides facing streets grew. Moreover, the area of many agricultural lands among the surviving land was 1000–3000 m² in 1983, and those that faced more than three streets were easy to continue. However, being disconnected was not a direct factor for the dissipation or continuation of agricultural lands. Moreover, 77.6% of agricultural lands continuing in 2016 faced streets larger than district streets, but it was found that many of them were connected to fine streets in 1983.</p><p> Thus, many agricultural lands in use in 2016 connected to larger streets than district streets without connection to the width of connecting streets in 1983. This leads to the conclusion that the plot size, street width, and number of sides facing streets are valid as evaluation indexes for characteristics of agricultural lands that have continued to be in use.</p>