Morphological characterization of wild Prunus scoparia Spach accessions in 11 provinces of Iran

Prunus scoparia (Spach) C. K. Schneid is among the most prevalent species which has the potential of being used as a dwarf rootstock for the cultivated almond. In the present study, the phenotypic variation of 521 wild accessions of this species naturally grown in 29 areas of 11 provinces in Iran was assessed. The accessions investigated showed significant differences based on the measured traits. The majority of the characters measured (90 out of 100) exhibited a coefficient of variation of higher than 20.00%, indicating considerable variation among the accessions. The range of nut-related characters was as follows: nut length: 9.72–22.87 mm, nut width: 5.81–15.54 mm, nut thickness: 5.67–12 mm, and nut weight: 0.18–0.99 mm. The range of kernel-related characters was as follows: kernel length: 6.83–19.23 mm, kernel width: 4.28–10.32 mm, kernel thickness: 2.16–7.52 mm, and kernel weight: 0.03–0.37 g. Kernel weight exhibited positive and significant correlations with nut length (r = 0.57), nut width (r = 0.54), nut thickness (r = 0.42), nut weight (r = 0.69), kernel length (r = 0.75), kernel width (r = 0.78), and kernel thickness (r = 0.58). Cluster analysis based on Ward’s method showed two different major clusters among all the accessions. Based on the bi-plot created using principal component analysis of population analysis, the studied 29 natural habitats formed four groups. The studied accessions showed considerable variation in terms of the measured traits within and among populations. This variation is due to cross-pollination, cross-incompatibility, natural hybridization, propagation by seeds, gene flow, and exchange of plant material between the study areas. By using crosses between accessions of different regions, it is possible to increase the amount of variability in different traits of wild almonds.


Plant material
The phenotypic variation of 521 wild accessions of P. scoparia collected from 29 regions of 11 provinces in Iran was evaluated for two consecutive years (2021 and 2022).Table 1 contains the geographical characteristics of the studied areas.The identification of the specimens was performed by Prof. Dr. Ali Khadivi.A herbarium voucher specimen with sediment number PS-2443 was donated to a public available herbarium of the Faculty of Agriculture and Natural Resources of Arak University, Iran.Permits required to collect the studied plant samples were obtained from the Ministry of Agriculture and Natural Resources of Iran.For correct sampling, a proper distance of at least 200 m between the accessions of each area was regarded so that the clone samples were not collected.

The characters evaluated
In total, 100 morphological traits related to flowers, branches, leaves, and fruits were recorded using 50 replications for each organ.Traits related to the size and weight of different organs were measured through digital calipers and electronic scales, respectively.To estimate the qualitative attributes in the form of code and rank, the almond descriptor (IPGRI) was used 14 .

Statistical analysis
The average data were used for analyses.To determine the significance between accessions, analysis of variance was done using SAS software 15 .The SPSS software 16 was used to determine the correlation between the traits as well as principal component analysis (PCA).Cluster analysis based on Ward's method and Euclidean distance and creating a scatter plot based on PC1 and PC2 were done using PAST software 17 .

Statement specifying permissions
For this study, we acquired permission to collect P. scoparia specimens issued by the Agricultural and Natural Resources Ministry of Iran.

Statement on experimental research and field studies on plants
All methods performed on plants (either cultivated or wild), including the collection of plant material comply with relevant institutional, national, and international guidelines and domestic legislation of Iran.
Peduncle length ranged from 0.77 to 3.95 mm, while peduncle width varied from 0.95 to 3.36 mm.Petal length varied from 4.30 to 16.37 mm, and petal width ranged from 2.56 to 15.01 mm.Sepal length ranged from 1.50 to 6.66 mm, while sepal width varied between 0.60 and 5.65 mm.
Tree height was moderate (1-2 m) and then low (< 1 m) in the majority of accessions (263 and 125 accessions) (Table 3).In breeding programs, low tree height is considered a useful trait for introducing dwarfing rootstocks 7,19 .
Tree growth vigor was high in most accessions (259) (Table 3).Trunk diameter was moderate and then high in the majority of accessions (238 and 155, respectively).The stem diameter in P. scoparia is very important for the production of gum and resin.It has been reported that most resin-producing plants form resin-producing ducts for self-defense 20 .One of the most important reasons that wild almond trees show high resistance to pests and diseases can be attributed to the unique feature of gum production in them 21 .In addition, hydraulic conductivity in most plant species is enhanced by increasing the thickness of stems and branches.The movement of water and nutrients needed towards the fruit is done better by the thick branches, which increases the growth and quality of the fruit 22 .
Annual branch color was light green in 356, green in 299, dark green in 64, and brown in 3 accessions.Also, current branch color in summer was predominantly light green (427 accessions) and then green (245), while it was dark green in 31, crimson in 12, and purple-green in 7 accessions.The leaves of P. scoparia normally fall in early summer and then the green branches continue photosynthesis to provide carbohydrates for root and branch growth and development for the rest of the growing season.In this case, the green branches compensate for the lack or absence of leaves.Considering that cytokinins and gibberellins are produced in the root 23 , the transfer of these substances to the branches, instead of forming new leaves, causes internode elongation in P. scoparia 3,4 .
The range of related characters of leaves on the annual branch (branch of the previous year) was as follows: leaf length: 8.11-54.35mm, leaf width: 1.23-10.15mm, petiole length: 0.60-9.55mm, and petiole width: 0.25-1.10mm.The range of related characters of leaves on the current branch was as follows: leaf length: 8.00-40.25 mm, leaf width: 0.74-10.15mm, petiole length: 0.46-10.31mm, and petiole width: 0.30-0.91mm (Table 2).In general, leaf area in P. scoparia is low, or in other words, its leaves are small, which indicates its greater adaptation to drought stress.Previous studies also reported that reduction of leaf area is an initial response of plant adaptation to drought conditions.Considering that the leaves of P. scoparia fall in early summer and the green shoots continue photosynthesis, this species can be a better choice as a rootstock.Also, the pubescence amount on the upper and lower surfaces of P. scoparia leaves is high, which is one of the responses to improve resistance to drought stress 24 .
Although the leaf size of P. scoparia is smaller, it is interesting that the dry matter content in this species is higher than that of domestic almonds.Dry and fresh weight in a plant determines its biomass production 25 .Interestingly, the cheapest and easiest method to track the performance and adaptation of plants to drought is to measure the accumulation of ash and mineral content 26 .It has been reported that plants with more dry matter www.nature.com/scientificreports/have higher yields under salinity-stress conditions 27 .Previously, positive and significant correlations between leaf ash content and yield 28,29 and also between leaf life span and dry matter content have been reported 30 .
The range of fruit stalk length and diameter was 1.85-5.53mm and 0.88-3.48mm, respectively.The range of nut-related characters was as follows: nut length: 9.72-22.87mm, nut width: 5.81-15.54mm, nut thickness: 5.67-12.00mm, and nut weight: 0.18-0.99mm.The suture opening of the shell was absent in 511 out of 521 accessions studied.The well-sealed shell is common in P. scoparia and is reported to be more resistant to fungus and insect infestation 31 .This offers new opportunities in breeding already not readily available in domesticated almond genetic resources.Shell thickness ranged from 0.40 to 1.76 mm.The presence of a relatively high variation in shell thickness among populations of P. scoparia offers the opportunity to select thin-shell nuts, which is important for the local production of this species as a nut crop 32 .
The range of kernel-related characters was as follows: kernel length: 6.83-19.23 mm, kernel width: 4.28-10.32mm, kernel thickness: 2.16-7.52mm, and kernel weight: 0.03-0.37g.Empty nuts were observed in 79 out of 521 accessions.Variations in kernel size and the occurrence of empty nuts can be due to variations in humidity and rainfall occurring in the natural habitats.When the almond species are subjected to drought stress, they will start to use the kernel moisture which then results in the shrinking of the kernel and the decrease in nut size 3,33 .Kester et al. 34 observed a highly significant effect of the environment on the occurrence of empty nuts, while Sanchez-Perez et al. 35 indicated a diminutive annual variation in this particular trait.
Many the almond characteristics are genetically controlled 36 .Also, the differences in the characters of accessions of different areas could be mainly because of the wider geographic regions and climatic zones covered in this study.Another reason behind these differences could arise from the variation in climatic conditions, especially in rainfall.The variation in annual precipitation is very common in the arid and semi-arid climate of Iran 33 .The pictures of leaves, flowers, and nuts of P. scoparia accessions studied are shown in Fig. 2.
PCA placed the traits in 26 components that explained 72.44% of the total variance.PC1 accounted for 6.10% of the total variance and showed significant correlations with petal length, petal shape, petal width, hypanthium diameter, sepal length, and sepal width.Nut length, nut thickness, nut width, nut weight, kernel weight, kernel width, and kernel length were placed in PC2 and explained 6.06% of the total variance.Tree form, trunk type, tree height, and trunk diameter were placed in PC3 and explained 4.58% of the total variance (Table 5).It has been reported that fruit-related traits are important for distinguishing accessions of almond species 9,33 .
In the scatter plot, the accessions were widely distributed on the plot level (Fig. 3).The results of the plot showed that the accessions have considerable variation so that a large number were placed outside the oval, which indicates their high differences with other accessions in terms of traits in PC1 and PC2.
The studied accessions showed considerable variation in terms of the measured traits within and among populations.This variation is due to cross-pollination, natural hybridization, cross-incompatibility, propagation by seeds, gene flow, and exchange of plant material between the study areas 39 .The traditional method of propagation and distribution of almonds is through seeds, which has caused the differentiation of traits and increased diversity over time.Also, the exchange of almond germplasm has been done in this country for millennia due to communication and interactions, which has caused interbreeding between populations.Thus, it is possible to justify the grouping of accessions of some distant populations [40][41][42] .Also, the dissimilarity between accessions of the species denotes the capability of generating new progenies and producing different associations or segregations of genes, thereby facilitating a partial removal of former linkages or the creation of new ones that can be applied in both classical and modern breeding methods.To generate new progenies in a subsequent generation (with new linkage groups or new population properties), it is a common practice to use distant genotypes [40][41][42] .Frost resistance is a major breeding goal for almond cultivars in many production areas because of their early flowering time during late winter and early spring.Some accessions of P. scoparia showed late flowering time.The possibility of use of almond related species with a very late-flowering date (high chilling requirements) to develop new cultivars with late-flowering would not only reduce frost damage, but reduce disease damage if flowering is delayed beyond the rainy season, and would allow more efficient use of increasingly scarce insect pollinators 43 .

Conclusion
A wide range of variations was detected within and among the populations studied of P. scoparia.The obtained results can be important for the management and protection of the gene pool.Also, these findings can be used to develop and introduce new rootstocks for almonds and other stone fruits.The traits, such as late flowering time, suitable trunk diameter, low tree height, suitable tree growth vigor, small leaf size, high nut weight, and high kernel weight are desirable traits that can be considered in almond breeding programs.Also, local cultivation

Figure 2 .
Figure 2. The leaves, flowers, and nuts of P. scoparia accessions studied.

Figure 3 .
Figure 3. Scatter plot for the studied P. scoparia accessions based on PC1/PC2.The symbols represent the accessions of each area in the plot (for an explanation of accession symbols, see Table1).
Figure 3. Scatter plot for the studied P. scoparia accessions based on PC1/PC2.The symbols represent the accessions of each area in the plot (for an explanation of accession symbols, see Table1).

Figure 4 .
Figure 4. Bi-plot for the studied populations of P. scoparia based on the morphological characters.

Table 1 .
Geographical description for collection sites of P. scoparia accessions studied in Iran.

Table 2 .
Statistical descriptive parameters for morphological traits used to study P. scoparia accessions.

Table 3 .
Frequency distribution for the measured qualitative morphological characters in the studied P. scoparia accessions.

Table 4 .
Simple correlations between the quantitative morphological variables utilized in the studied P. scoparia accessions.For an explanation of the morphological character symbols, see Table2.*,**Correlation is significant at P ≤ 0.05 and 0.01 levels, respectively.

Table 5 .
Eigenvectors for the main variables for the first three principal component axes from PCA of the morphological characters in the studied P. scoparia accessions.Bold values indicate the characteristics that most influence each PC.