In this study, the experiment was carried out in a completely randomized design to discover the diversity among ۳۰۰ Ali Kazemi rice populations from three provinces (ten populations) in the
north of Iran using SPSS, Past, Graph Pad, and Excel software. The studies were conducted based on ۳۰ morphological traits. Quantitative traits, like panicle traits, grain length, ۱۰۰-grain weight, and number of spikelets, demonstrated high variation between populations. Cluster and ordination analyses were performed on morphological characters. The cluster analysis made it possible to divide the Ali Kazemi rice into three sub-clusters of different geographical regions, showing this local variety’s adaptability to the environment. Of the three populations,
Guilan appeared to be more variable than the Qazvin and Zanjan, as variability was observed among populations for some characters, such as awn length, number of spikelets, blade length, and panicle traits between populations. GLN۰۱ had the highest average plant height (۱۱۸۵.۶۰ mm). The populations of
Guilan also had the longest leaf blade and stem length. The highest value for grain size was also recorded from the
Guilan populations. For the ten Ali Kazemi rice populations, the ordinal analysis revealed the traits that cause the greatest variation between populations. Highly significant correlations were found between awn length and traits such as grain length (r= ۰.۳۹۵), panicle length (r= ۰.۴۷۷), panicle weight (r= ۰.۳۴۹), and ۱۰۰-grain weight (r= ۰.۴۴۰). The highest correlation was between the number of spikelets and some traits, such as plant height, leaf length, culm outer diameter, number of tillers, and panicle traits, indicating that a selection advantage for panicle traits occurs mainly for quantitative traits. Overall, the data collected in the present study could have valuable implications for improving Ali Kazemi rice cultivars in farmers' fields.