A comparative analysis of dendrometric, macromorphological, and micromorphological characteristics of Pistacia atlantica subsp. atlantica and Pistacia terebinthus in the middle Atlas region of Morocco

Abstract The genus Pistacia, with its species having notable ecological, economic, and medicinal implications, demonstrates remarkable environmental adaptability. The central objective of the study is to analyze interspecific variations between Pistacia atlantica subsp. atlantica and Pistacia terebinthus across three distinct bioclimatic zones in the Middle Atlas region of Morocco. The methodology includes collecting dendrometric measurements and conducting macromorphological examinations on these two taxa, with a detailed analysis of 27 qualitative and quantitative variables. A micro-morphological analysis of leaves, using scanning electron microscopy (SEM), is employed to explore specific features such as size and stomatal density, as well as qualitative aspects like epidermal cell shape and trichomes. Dendrometric measurements have revealed that the canopy surface and the number of trunks per tree can serve as distinctive features between the two species. Regarding the sex ratio of Pistacia atlantica subsp. atlantica, 59% of the examined trees are males, primarily associated with the jujube tree in arid zones and the dwarf palm in humid areas. In contrast, female Pistacia terebinthus exhibit a similar percentage, predominantly associated with oak groves and cade juniper in their distribution areas. Principal component analysis of biometric measurements emphasized a significant disparity between the two species, representing 60.25% of the total variance. The use of SEM unveiled new features facilitating the identification of the two species. By leveraging the macromorphological and micromorphological variability of pistachio trees, we can qualify those best suited to diverse bioclimates. In this regard, we suggest incorporating them into reforestation and rehabilitation programs aimed at restoring our declining ecosystems.


Introduction
The Mediterranean flora in general, and the Moroccan flora in particular, is now much better understood than it was 20 years ago.Systematic research, whether of a traditional or modern (molecular) nature, has refined our knowledge of various taxa (families, genera, or species) [1].Our research delves into the present condition of iconic Moroccan ecosystems, with a specific focus on two members of the Anacardiaceae: PA (Pistacia atlantica subsp.atlantica) and PT (Pistacia terebinthus).
With around ten species of pistachio trees, some of them share similar physical characteristics, posing challenges in accurate identification when observed in their natural habitats.This challenge is particularly evident for the PA and PT, given their morphological resemblances.Overcoming this potential confusion and ensuring precise identification necessitates a comprehensive study of these species [2].
The genus Pistacia comprises ten species characterized by alternate, pinnate-compound leaves [3].Also, from a taxonomic perspective, establishing the limits of Anacardiaceae has been challenging; that is to say, not long ago, Anacardiaceae and Bursuaceae were grouped under the family Terebinthaceae [4].According to Zohary [5], PA has been classified into four distinct eco-geographical varieties: cabulica, latifolia, kurdica, and finally, atlantica, originating from the Maghreb region [6], while Álvarez et al. [7] have elevated the first three varieties to the status of subspecies; moreover, other authors considered some of them as separate species [8][9][10][11].
However, it is worth noting that the classification proposed by Rechinger [12] was not accepted by Iranian botanists [13].PA is naturally found from the Mediterranean basin to Central Asia.It is a dominant tree in its natural spaces and plays a significant economic role in these regions.The leaves are imparipinnate with 5-9 lanceolate leaflets, measuring 26-70 mm in length and 8-22 mm in width.The petioles are winged and are extra-axillary, ranging from 7 to 15 cm in length [14].
PT is a shrub or small tree and typically range in height from 2 to 6 m, occasionally reaching up to 12 m.Their deciduous leaves, either impari-or paripinnate, measure between 10 and 19 cm in length and 6-19 cm in width, eventually becoming leathery.Petioles are usually rounded, rarely flattened, and the rachis lacks winged extensions.Leaflets, numbering from 3 to 11, are typically 3.5-8 cm long and 1-3.1 cm wide, with an average ratio of about 2.7:1.They are arranged opposite to subopposite and vary from ovate to narrowly ovate, often with a pointed tip and a fine pubescence or smooth surface.The terminal leaflet, if present, tends to be smaller than the lateral ones, measuring between 1 and 6 cm long and 5 and 20 mm wide.Staminate panicles, reaching up to 8 cm long, feature diffuse branching and large bracts with long white trichomes.Pistillate panicles, up to 15-23 cm long, become more robust in fruit and may have a slightly hairy or smooth surface.Initially reddish, the drupes eventually ripen to a blue color, completing the characteristic features of these plants [15].
PT is native to Eastern Mediterranean countries, where it is commonly found on rocky sites with open vegetation.It avoids the driest and coldest locations [16].
Morphological characterization of a plant represents a critical and significant trait as it encompasses growth and developmental profile while also playing a role in formulating conservation strategies.When distinguishing between taxa or for taxonomy purposes, it is essential to consider the various levels and degrees of intraspecific and interspecific variations [17].
Indeed, these two taxa are characterized by their intraspecific diversity, broad geographic distributions, and therapeutic uses by different ethnic groups in Asia and North Africa [18].
On the other hand, our comparison also encompasses the phytodermic and stomatal characteristics of the two taxa under study to explore potential variations between them.It should be pointed out that the morphology, distribution, and behavior of stomata exhibit responsiveness to a diverse range of signals, ranging from intracellular signaling to global climate changes [19].
Precise identification of the two taxa is essential for establishing a combined strategy for conservation and improvement.Indeed, species identification involves three distinct types of characterization: morphological, biochemical, and molecular.In this study, we specifically focused on the first type, directing our attention to macromorphological and micromorphological characterization, which form the fundamental basis for species classification.

Sampling area
Trees of PA and PT used in this study were sampled from three regions (Figure 1) located in different bioclimatic zones of the Middle Atlas in Morocco (Table 1).It is important to emphasize that the populations of PA and PT are scattered or isolated.Therefore, we sampled our trees at various locations in the same study area based on the presence of both species.Regarding PT, we noticed that this taxon is distributed in humid areas, with a few isolated individuals found in the early stages of the semi-arid climate, particularly in our Amghas zone.In other words, as aridity increases, PT disappears regardless of altitude.

Dendrometric characteristics of trees
Due to the uneven distribution of trees, an initial visual count was conducted on each slope of every station.Subsequently, various parameters were assessed, including plant sex, circumference, height, canopy, associated flora, number of trunks, and tree vigor.The selection of trees was made randomly and based on the presence of the two studied taxa.We chose only trees that had reached a flowering stage in order to know the sex of each tree (Table 2).Height was measured using a geometric principle based on the Bucher's cross, while circumference was measured with a tape measured at 1.30 m from the ground [20].
Although the exact calculation of the crown area is theoretically impossible [21], it can be approximated by utilizing directly measurable values such as the crown height and diameter using the following formula: where S represents the crown area at the ground level in square meters, d1 is the ground diameter of the northsouth crown, and d2 is the east-west crown diameter at the ground level [22].Statistical analysis of data: to assess the consistency or disparity of the collected dendrometric parameters, various statistical indicators were utilized.The standard deviation (σ) and coefficient of variation (CV) were calculated among these parameters, providing insights into the dispersion and distribution of individuals around the mean.Furthermore, a two-factor analysis of variance was applied, accompanied by the use of Tukey's test for both variance analysis and mean comparison.

Morphological analysis 4.1 Sampling of the plant material
Due to the scattered distribution of individuals of the studied species, sample collection was carried out subjectively by selecting the most accessible subjects in the field.Consequently, 30 trees were measured in each of the three  study stations.Given the deciduous nature of PA, periodic field trips were conducted throughout the year 2022 to collect leaves, inflorescences, and fruits at different stages of development.Morphological characteristics were chosen as the main analysis parameters, as they are frequently used by taxonomists to differentiate between taxa.These characteristics provide crucial information about the variability of adaptations and are essential for a better understanding of natural population differentiation, as indicated in ref. [22].

Measured parameters
The reference samples for each institution were sorted at the Herbarium of the Dher el Mehrez Faculty of Science, specifically the "LBEAS" laboratory (Figure 2).
At each location, a random selection of 20 trees was made.Subsequently, each plant was sampled, including 20 leaves and 20 fruits.The leaves and fruits were naturally dried and stored in the laboratory.Biometric and morphological measurements were conducted following the method described by the descriptors of the pistachio tree [26].
In total, 27 morphological characteristics were measured, consisting of 13 qualitative traits and 14 quantitative traits.The qualitative traits are represented in Table 3, while the quantitative traits are presented in Table 4.

Micromorphological characteristics of leaflets 5.1 Observations under a photon microscope
For optical microscopy, observations were conducted on epidermal impressions obtained using the epidermal sampling method.The technique involves applying a sufficiently thick layer of transparent nail polish to both sides of a fresh leaf.Once the nail polish has completely dried in ambient air, the peel of the nail polish, reproducing the topography of the leaf, is gently removed with tweezers and spread onto a slide previously moistened with a drop of water.Observations were made at magnifications of ×10 and ×40.The slides are stored in the Laboratory of Biotechnology, Environment, Agriculture, and Health at Sidi Mohamed Ben Abdellah University.Comparison of Pistacia atlantica and Pistacia terebinthus  5

Scanning electron microscopy (SEM) observations
The sample preparation method for SEM observation depends on the information sought, including the topography, microstructure, nature of the sample, and its presentation form, such as biological samples or powders [27].
The epidermal surfaces of the leaves were examined using a scanning electron microscope.Three specimens were taken from each site, each comprising a 5 mm 2 section of the dry leaf surface (abaxial and adaxial surfaces).SEM images were digitally recorded at various levels of magnification with a JSM-IT500HR at the Regional University Interface Center (CURI) of Sidi Mohamed Ben Abdellah University in Fès.Observations and measurements focused on ten qualitative and quantitative characteristics of trichomes and stomata.Qualitative features included the shape of the epidermal cells, stomatal shape, stomatal complex type, and trichome type.Quantitative features encompassed stomatal density on the upper and lower surfaces, as well as the length and width of stomata.

Dendrometric characteristics 6.1.1 Quantitative measurements
Comparison of the conformation of samples from PA and PT trees.Sixty trees were studied in three selected zones, with 20 trees per zone.
According to our dendrometric measurements summarized in Table 3, a significant phenotypic difference was observed in the field between PA and PT, specifically in terms of canopy surface and the number of trunks per tree.For each species, there was no significant difference in the tree height and circumference between the three zones.Indeed, the results for height, circumference, and canopy surface were consistently higher (1,399 cm, 299.3 cm, 5,483 cm 2 ) for PA than for PT in all three zones.
For PA, 59% of the examined trees were male, and the identical percentage was noted for female trees of PT (Figure 3).It is crucial to underscore that no instances of hermaphroditism were detected in any of the populations under study.

Qualitative measurements
As regards the tree trunks of the two taxa studied, we found that PA trees are mainly individualized, with rare cases of two trunks (Figure 4).
With regard to the vegetation associated with PA, the analysis highlights the presence mainly of Ziziphus lotus L. Regarding tree vigor, 50% of the trees of both species showed low vigor, while only 20% of the specimens  represented vigorous trees (Figure 5).In terms of trunk, all measured PT trees have branching (not-individualized) trunks, ranging from 2 to 7 stems in some cases.
7 Morphological traits of leaves, inflorescence, and fruits

Inter-species variability
The leaves of PA observed in different locations showed an average length of 13.39 ± 2.32 cm, varying from a minimum of 7.2 ± 2.84 cm to a maximum of 19.8 ± 2.16 cm.Leaf width was between 5.4 ± 1.78 cm and 15.4 ± 1.54 cm, with an average width of 10.65 ± 1.96 cm.For leaves of PT, length ranged from 5.5 ± 3.49 cm to 19.9 ± 3.18 cm, with an average length of 13.83 ± 3.15 cm.The calculated width had an average variation of 11.61 ± 2.45 cm, ranging from a minimum of 7.1 ± 2.39 cm to a maximum of 17.71 ± 3.06 cm (Table 4).
With regard to leaf morphology, both species are mainly characterized by an odd-pinnate arrangement of their leaflets.In the case of PA, the average number of leaflets was 8.56 ± 1.92.The most frequently observed number of leaflets was 7, representing 21.6% of the cases.In contrast, PT had an average of 8.13 ± 1.72 leaflets per leaf.The number of leaflets most frequently encountered was 7, with a frequency of 31.6% (Figure 6).
The comparison of the ratios between the length and width of the terminal leaflet in the PA and the PT within three distinct stations reveals significant differences.The average length/width ratio for PA was 3.03 + 1.32, which was significantly higher than that of the PT estimated to be 2.05 + 0.82.The analysis of the dimensions of the fruits (Figure 4) reveals a marked distinction between PA and PT.The seeds of PA had higher values, with an average of 7.99 ± 0.89 for the length, while width and thickness were 6.05 ± 0.66 and 5.90 ± 0.86, respectively.For PT, the average width, length, and thickness were 6.64 ± 0.35, 5.67 ± 0.37, and 5.41 ± 0.40, respectively (Figure 7).
The length, width, and thickness of the almond of PA generally were greater than those of PT.The analysis of the dimensions of the almonds reveals a marked distinction between the two species examined.For PA, the length, width, and thickness of the almond were 5.283 ± 0.96, 4.79 ± 0.363, and 4.3 ± 0.29, respectively; and for PT, they  were 4.996 ± 0.57, 4.42 ± 0.6, and 4.2 ± 0.556, respectively (Figure 8).Morphometric analysis also revealed a significant difference in the length of the inflorescence rachis between the PT and PA.Indeed, the length of the rachis of the inflorescence is greater in PT than in PA.Also, PA has a higher number of primary lateral inflorescential branches compared to PT (Table 4).

Inter-population variability 8.1 Principal component analysis (PCA)
The 14 quantitative morphological characters of specimens from various populations of PA and PT were subjected to PCA.The main emphasis of the interpretation of the results was placed on the first two axes, namely axis 1 and axis 2. The information provided by these two selected axes represents 60.25% of the total variance (Figure 9).To identify the variables responsible for the variations along the selected axes, we used the correlation between the variables and the factors (Table 5).Only the first two factors (axes) present variables with factor weights ≥0.6.It is obvious that most of the measured variables are strongly negatively correlated with axis 1 and axis 2. However, the width of the terminal leaflet is negatively correlated with axis 2, while the length of the inflorescence rachis is positively correlated with axis 2. On the other hand, the parameters of the fruits are strongly correlated with axis 1.It is clearly observable that the populations of PA and PT are distinguishable, which indicates morphological differences in the characteristics of the leaves, flowers, and fruits.The traits (variables) responsible for this distinction are those that have high correlations with the axes concerned (≥0.6).The size of the terminal leaflet exhibits significant variations.
In PA, 73.3% of the terminal leaflets resemble those of basal leaflets, whereas in PT, this proportion is 48.3%.Furthermore, 73.3% of the terminal leaflets in PA have an elongated lanceolate form, whereas in PT, it is 71.67% and they have an oval rounded shape.Concerning the apex of the leaflets, 83.3% of the acuminated forms are observed in PA, while 56.66% of the mucronulate forms are noted in PT (Table 6).Also, the shape of the basal leaflets differs between the two species.In PA, 88.3% of the basal leaflets are attenuated, whereas in PT, this proportion is 53.3%, with 33% adopting an oblique form.Regarding the edge of terminal leaflets, it is flat in PA (85%) and PT (91.66%).
The petiole exhibits differences.In PA, the petiole is rounded and straight on the upper surface (66.6%), while in PT, it is flattened (93.3%).The leaf color varies, being green and dark green in PA (48 and 38.3%, respectively) and similarly in PT (60 and 30%, respectively).However, the color difference was observed in November, the time when the leaves are about to fall.Young leaves of PA are pale when young, while those of PT are reddish.The central leaf veins are glabrous at 100% for both species (Table 6).

Flower and fruit parameters
In PA, the flower buds have a narrowed oval form, while in PT, 90% have an enlarged oval shape.The flower color is reddishbrown in 63.3% of PA and 86.66% of PT.Inflorescence abundance varies, with 80% moderately abundant inflorescences in PA and 46.67% sparsely abundant in PT.The fruit shape of both species is round (l/w < 1.5) at 100% for PA and 66.67% for PT.The apex of fruits is rounded in PA (75%) and flattened in PT (65%) (Table 6).According to Table 7, it is observed that stomatal density values in PA range from 150 to 440 stomata per mm² on the abaxial side, with an average of 280.03 stomata per mm².Conversely, on the adaxial side of the leaf, where the number of stomata is lower, the variation ranges between 5 and 85 stomata per mm², with an average of 28.03 stomata per mm².
For PT, stomatal density values vary between 144 and 480 stomata per mm² on the abaxial side, with an average of 295.4 stomata per mm².On the adaxial side of the leaf, the number of stomata varies between 17 and 165 stomata per mm², with an average of 83.02 stomata per mm².It is noteworthy that stomatal density also varies based on the origin of the leaves of the two species, with the highest recorded values in Ait Naamane, followed by Afourgagh, and finally, Amghas.
The measurements of the stomatal size on the adaxial and abaxial sides of the leaves of PA and PT revealed significant variations.In the Afourgagh area, the stomata for PT had an average length of 27.56 μm on the abaxial side and 22.34 μm on the adaxial side.The average width was 19.43 μm on the abaxial side and 10.48 μm on the adaxial side.In Amghas, the stomata of PA has an average length of 22.06 μm on the abaxial side and 18.94 μm on the adaxial side, with average widths of 8.607 and 8.863 μm, respectively.Ait Naamane PA had smaller stomata, with an average length of 14.47 μm on the abaxial side and 12.68 μm on the adaxial side.The average widths were 9.584 and 8.724 μm, respectively.
For PT, in Afrougagh, the average length of the stomata was 16.86 μm on the abaxial side and 21.46 μm on the adaxial side.The average widths were 8.343 and 12.32 μm, respectively, in Amghas, the stomata of PT had an average length of 10.53 μm on the abaxial side and 9.184 μm on the adaxial side, with average widths of 7.942 and 7.634 μm, respectively.Biplot (axes F1 et F2 : 60.25 %)  Ait Naamane PT had stomata with an average length of 11.77 μm on the abaxial side and 11 μm on the adaxial side, with average widths of 7.52 and 7.64 μm, respectively.These results highlight the variability in the stomatal size between different zones and the two Pistacia species studied (Table 8).

Types of trichomes
Figure 9 shows the two main types of trichomes observed in all examined leaves, namely glandular and uncinate.Trichomes are primarily present either on the abaxial or adaxial surfaces, with some cases showing the presence on  both surfaces, while in other cases, they are absent.In the present study, trichomes were found on both surfaces of PA leaves in the three study zones: Amghas, Afourgagh, and Ait Naamane.Meanwhile, in PT, trichomes were absent in all samples tested from the three leaf sources.Qualitative results regarding trichomes are presented in Table 9.Three distinct categories of stomata have been identified in all examined leaves, including anomocytic, paracytic, and actinocytic stomata.More than 90% of the leaves exhibited anomocytic stomata, and sometimes a single epidermis had two or more types of stomata (Figure 10).Additionally, in both species, stomata are primarily located on the abaxial surface of the leaves, although they are occasionally present on both surfaces (Table 9).

Dendrometric characteristics
In this study, most dendrometric parameters of the pistachio tree are not influenced by different zones, and the same observations were made concerning acacia and were endorsed by studies conducted by Fonton et al. [28].However, statistical analysis revealed two discriminant features for identifying PA and PT.First, the canopy of PA covers a larger area than that of PT in the three study zones.El Zerey-Belaskri et al. [29] showed that PA trees are characterized by dense foliage compared to other species of the Pistacia, confirming our findings regarding the canopy area.Second, the PT tree has a multi-trunk arboreal stem starting at less than 50 cm from the ground.On the other hand, the PA tree has a well-individualized stem, whether for young shoots or older subjects.This observation was confirmed by Quezel and Santa [30].The trunk of the PA tree is primarily singular, although in rare cases, it may branch out from the base.On the other hand, similar studies conducted in the literature [2,30] showed that PA trees are robust and vigorous, reaching heights of 15-20 m; however, intermediate trees are also observed.The trunks of adult trees can exceed 1 m in diameter and 6 ms in circumference, with a striated dark gray bark.
Our dendrometric results are partially inconsistent with those of Aloui [31].He showed that the climatic effect is significant on the tree circumference and canopy diameter but is not a significant factor in tree height.
Regarding the sex ratio, our results align with the study of Mabberley [32].Flowers are generally unisexual in Anacardiaceae, and the genus Pistacia is distinguished by its dioecious reproductive system.Some exceptional cases of sexual types have been reported in the literature; Ozbek and Ayfer [33] observed two hermaphrodite trees in Turkey.
11.2 Morphological traits of leaves, inflorescence, and fruits Among the distinctive features, several are used in the classification of species in the Pistacia genus, including the size of the terminal leaflet, the number of pairs of leaflets, the shape of the apex of the terminal leaflet, the presence of leaflet wings on the rachis, and finally, the shape of the fruit [5,8].The description provided in previous studies [5,8,13,34] suggests that the leaves of P. atlantica possess leaflets with finely winged rachis [35].The rachis is broadly winged, with wings extending along its entire length [13].Regarding the terminal leaflet, it is stated to be as large as the basal leaflets [34].In our study, the size of the terminal leaflet varies, and the shape of the apex can be acute or acuminate, contrary to the obtuse shape reported by Zohary [5].
Regarding our observations on inflorescences, the parameters we studied did not uncover any distinctive characteristics.However, during the sampling process, we noticed that some male PA produced flower buds starting in January.Based on our sample set and the studies conducted by Yaaqobi et al. [36] on the flowering and germination of PA seeds, we considered these cases to be exceptional occurrences.On the other hand, flowering is early in trees PA and late in PT and this phenomenon is more pronounced in areas where aridity decreases.It is also crucial to emphasize that flowering occurs before vegetative development, with a tendency for males to flower before females.This observation aligns with Delph [37], demonstrating that males invest more resources in flowering early in the season before leaf production.Thus, resources allocated to the development of vegetative growth, including photosynthetic tissues (leaves), may be limited or unstable [38].
On the other hand, in the mountains of the Bulgarian Rhodopes, some monoecious plants of PT were found [39].Some trees were entirely monoecious, while others had only staminate flowers or bore only pistillate inflorescences.To assess the stability of the monoecious type, a study was conducted to investigate the morpho-anatomical development of floral buds and the floral biology of the aforementioned genotypes of PT.The study of monoecious plants of PT in the Mediterranean region is important for the opportunity it provides to increase the biodiversity of agroforestry systems, for rootstock platforms, and for obtaining new interspecific hybrids as potential rootstocks [40].
The statistical analyses indicate significant variations both in terms of size and shape of the seeds between the two species in the three stations studied.This observation has been validated within the subspecies of the PA.Indeed, Behboodi [13] described the leaves and fruits of the PA tree found in Iran.The drupes are slightly broader than long in P. atlantica subsp.mutica; they are globular, with equal length and width in P. atlantica subsp.cabulica; and they are depressed globular, measuring 5-8 mm × 8-10 mm in P. atlantica subsp.kurdica.Previous studies conducted on the fruits partially align with our findings.Indeed, the fruit is a syncarpous drupe, morphologically variable, characterized by a thin and oleaginous exocarp [5,6,29].It is small in size (≤0.8/0.7 cm) [29], with large and oval drupes that turn from red to dark green when mature (August to September).Its single seed is oval and yellow.Previous results report an average of 37 fruits per infructescence in adult individuals [41].
The measurements taken on the size of the fruits and almonds of PT and PA trees reveal a significant issue of empty seed formation, estimated to be 77%.This result has been explained in previous studies [42,43]; the authors suggest that the production of fruits with non-viable seeds can be triggered by specific conditions affecting a plant during the pollination period or during the process of fertilization and early fruit growth [44].

Micromorphological characteristics
Stomata represent the primary mechanism by which higher plants adapt to their constantly changing environment [45].
Homeostasis, a fundamental biological property of living organisms, requires the regulation of the internal environment while interacting with the surroundings [46].Adaptive homeostatic ranges are significantly broader in plants than in animals [47], but the overall conservation of homeostasis remains equally crucial in both kingdoms.
The results obtained from measurements of qualitative and quantitative characteristics of stomata and trichomes agree with some studies but differ from others.Statistical analyses reveal significant differences for the measured variables among the studied species.According to the statistical results, the most discriminating stomatal characteristics recorded during this study are size and stomatal density.
Examining the data, the density is higher in the abaxial epidermis, and it appears that increased aridity is associated with a decrease in the number of stomata.This trend is observed on both the adaxial and abaxial surfaces, and similar results were reported by Gindel [48] in studies conducted in Israel.PT exhibits a higher stomatal density than PA.Additionally, PT develops stomata of smaller size than PA.The few stomata on the adaxial surface are primarily located near the main and secondary veins of the leaflet.These observations support the earlier findings of Doghbage et al. [49].
Our conclusions are inconsistent with the phytodermological study of Pistacia atlantica subsp.atlantica and Pistacia lentiscus conducted by Smail-Saadoun [50].This study demonstrated that the adaptation of these two species to water scarcity involves a complete absence of stomata on the upper surface of the leaves and the presence of paracytic mesoperigenous stomata on the lower surface of the leaf.About 90% of the samples exhibit anomocytic forms, and in rare cases, actinocytic and paracytic stomata are  observed in both species (Figure 10a and b).Most epidermal cells have an irregular shape, except for 10% of the cells in PT from the Afourgagh area, which have a polygonal shape.Furthermore, the guard cells were not aligned at the same level as the adjacent epidermal cells, and the subsidiary cells did not clearly surround the guard cells.Our results are in agreement with those of Al-Saghir et al. [51] and contradict the findings of the study conducted by Lin et al. [3], which reported that all pistachio species (P.atlantica, P. exicana, P. chinensis, P. integrrima, P. khinjuk, P. texana, P. vera, and P. weinmannifolia) had actinocytic stomata.
According to Al-Saghir [52], amphistomy is considered the primitive characteristic in Pistacia species, with occurrence on a single surface (hypostomy or epistomy) seen as an evolution.P. vera is identified as the most primitive species, while P. terebinthus is considered more recent.In our study, PA exhibits almost the same stomatal density on the abaxial surface compared to PT in Afourgagh and Amghas, but it is lower on the adaxial surface, where stomata are localized along the main and secondary veins in all three studied zones.The researchers [10] suggest that stomatal loss in Pistacia species could be an adaptation to climate changes.This adaptation is also supported by The outcome of Simon [53] considered that the regulation of evaporation at the stomatal level can occur either through their sunken position or by the presence of thick edges that protect the pore [53].Mucilage deposited on stomatal pores could also contribute to reducing evaporation.Wax deposits would serve a similar function, especially in species with low hair coverage.
The information regarding the presence or absence of trichomes on Pistacia leaves is contradictory.Al-Saghir et al. [51] reported the presence of trichomes in all Pistacia species, while other authors indicated that Pistacia leaves are devoid of hairs [3].The results of the present study reveal the absence of trichomes on the leaflets of PT.Many immature leaves lose their trichomes as the leaf blade opens and expands [54] showed that trichomes produce essential oils that appear to play a protective role against herbivores and pathogens.

Conclusions
This study has unveiled crucial aspects for the identification of two species within the Pistacia genus, examining dendrometric, macromorphological, and micromorphological characteristics.Parameters such as canopy surface and trunk count have been identified as distinctive indicators between these two species.Additionally, features including leaf and terminal leaflet dimensions, as well as the size and shape of the terminal leaflet, fruit and almond size, stomatal size and density, and the shape of epidermal stomata and trichomes, were recognized as the most discriminating.Given the broad distribution of these species in North Africa, particularly PA subjected to diverse climatic conditions, it is crucial to conserve these resources and utilize them optimally.This holds particular significance within the context of research and development projects, especially those aimed at restoring heavily degraded forest environments in the Middle Atlas region of Morocco.Furthermore, the potential presence of genetic variability underscores the need to delve deeper into research in this area, considering the comparison of populations from different geographical regions through the use of molecular markers.

Figure 1 :
Figure 1: Map of the study area and sampling stations (ARC-GIS program).

Figure 3 :
Figure 3: Number of male and female individuals in the three zones; PA: Pistacia atlantica Desf and PT: Pistacia terebinthus.

Figure 6 :
Figure 6: Frequencies of the number of terminal leaflets.

Figure 9 :
Figure 9: Correlation circle and individual projection onto factors 1 and 2.

15 Figure 10 :
Figure 10: (a) Stomata on the abaxial surface of PA; (b) stomata of the abaxial surface of PT; (c) micrographs of stomata on the abaxial surface of PA; (d) micrographs of stomata on the adaxial surface of PA; (e) size of stomata on the abaxial surface of PA; (f) size of stomata on the adaxial surface of PA; (g) micrographs of stomata on the abaxial surface of PT; (h) Micrographs of stomata of the adaxial surface of PT; (i) size of stomata on the abaxial surface of PT. (j) size of stomata on the adaxial surface of PT. (k) form of the epidermis of PA; (l) main leaf veins of PA; and (m) and (n): ciliated and glandular trichomes of PA.

Table 2 :
Dendrometric characteristics measured for trees in distinct populations

Table 6 :
Frequencies (%) for qualitative traits measured in different populations of Pistacia atlantica Desf and Pistacia terebinthus L.

Table 7 :
Stomatal density in different study areas Characteristics Stomatal density of PA on the abaxial side Stomatal density of PA on the adaxial side Stomatal density of PT on the abaxial side Stomatal density of PT on the adaxial Comparison of Pistacia atlantica and Pistacia terebinthus  13

Table 8 :
Size of stomata in the abaxial and adaxial sides

Table 9 :
Qualitative variables measured for stomata and trichomes