This study aims to explore the effects of different treatment conditions on chromosome root tip squashing, optimize the chromosome preparation method, conduct karyotype analysis of 13 Camellia species, and to provide a theoretical basis for the research on Camellia plants.

The effects of different pretreatment and dissociation conditions on chromosome slide preparation were compared to select a universally applicable method for Camellia species. Karyotype analysis of 13 Camellia species was conducted using the optimized method, yielding karyotype diagrams, parameters, and evolutionary trends for each species.

The optimized pressing method suitable for Camellia is to treat the root tips with saturated 1,4-dichlorobenzene for 6 h, then place in a 1 mol/L hydrochloric acid water bath at 60 ℃ for 15 minutes and stain with a modified carbol-fuchsin solution for 15 minutes for the conventional pressing operation. All 13 Camellia species exhibit a chromosome base number of x=15, the results of the karyotype analysis show that the karyotypes include diploid, tetraploid, hexaploid, and octoploid. Among diploids, the karyotype formula for C. semiserrata is 2n=2x=30=24m+6sm, and for the C. chekiangoleosa, it is 2n=2x=30=22m+8sm; for the tetraploid C. meiocarpa, the karyotype formula is 2n=4x=60=40m+20sm; among the hexaploid species, the karyotype formula for C. oleifera is 2n=6x=90=42m+44sm+4st, and for C. drupifera, it is 2n=6x=90=60m+26sm+4st; additionally, the karyotype formula for the octaploid C. gauchowensis is 2n=8x=120=94m+26sm. The range of asymmetry indexes for 13 Camellia species is 47.90%–64.63%, the relative length of chromosomes ranges from 2.78 to 13.59, the arm ratios range from 1.00 to 4.22, and the Lt/St ranges from 1.62 to 4.89. According to the Stebbins' karyotype classification system, the C. sinensis, C. chekiangoleosa, C. transarisanensis, and C. meiocarpa are of type 2A, C. boreali-yunnanica is type 1A, and C. osmantha is type 2C, while the remaining eight species are type 2B.

This study explored chromosome staining techniques widely applicable to Camellia, determined the chromosome numbers and ploidy levels of 13 Camellia species, and reported the karyotypes of C. transarisanensis, C. cordifolia, and C. osmantha for the first time. It provides an important reference for chromosome preparation and karyotype analysis of Camellia sinensis plants.

The qualitative and quantitative traits of 36 wild Camellia oleifolia resources in Ya'an, Sichuan Province were evaluated comprehensively to provide a theoretical basis for breeding suitable local C. oleifera cultivars.

The 17 quality characters and 19 quantitative characters of 36 wild C. oleifera resources were determined. The Shannon-Wiener diversity index analysis, significance analysis, principal component analysis and cluster analysis were used to comprehensively evaluate the wild C. oleifera resources with better comprehensive performance.

The results showed that the genetic diversity of 17 quality traits of 36 wild C. oleifera resources was relatively rich, and the diversity index of fruit shape and seed shape was the largest (1.390 5 and 1.412 3, respectively). The analysis of 19 quantitative characters showed that 36 wild C. oleifera resources have abundant genetic variability in quantitative characters, and the coefficient of variation ranged from 1.54% to 54.02%. The variation coefficients of seed number, fruit weight and fruit oil content were the largest, which were 52.04%, 46.07% and 41.52%, respectively. TQ3 had the largest seed number (7.07), TQ13 had the largest fruit weight (42.73 g), and TQ14 had the largest oil content (12.10%). The number of seeds, the quality of fresh fruit and the oil content of fresh fruit were of great significance in the evaluation of the excellent degree of C. oleifera resources. The comprehensive calculation and score based on principal component analysis showed that TQ14, TQ16 and TQ10 had the highest comprehensive scores and had high breeding potential. The 36 wild C. oleifera resources were divided into 6 categories by systematic cluster analysis. TQ14, TQ16 and TQ10 with good comprehensive performance were grouped into one category, and TQ20 with poor comprehensive performance were grouped into one category alone.

The quality and quantity traits of 36 wild C. oleifera resources had relatively rich genetic diversity, and TQ14, TQ16 and TQ10 wild C. oleifera resources had better comprehensive traits, which laid a foundation for the subsequent breeding of local C. oleifera varieties in Ya'an.