Phenotypic evaluation of six cassava families (Manihot esculenta Crantz) from seed in Burkina Faso

141 Michel et al. Int. J. Biosci. 202 5 RESEARCH PAPER RESEARCH PAPERRESEARCH PAPER RESEARCH PAPER OPEN ACCESS OPEN ACCESSOPEN ACCESS OPEN ACCESS Phenotypic evaluation of six cassava families ( Manihot esculenta Crantz) from seed in Burkina Faso Sawadogo O. Michel *1 , Some Koussao 2 , Ouedraogo M. Hamed 1 , Tiama Djakaria 2 , Tiendrebeogo Fidèle 2 , Soro Monique 2 , Tonde Wendmanegda Hermann 1 , Sawadogo Mahamadou 1 1 Life and Earth Sciences Training and Research Unit, Joseph KI - ZERBO University, Ouagadougou, Burkina Faso 2Institute of Environment and Agricultural Research (INERA)/CNRST, Ouagadougou, Burkina Faso Key words: Cassava, Genotypes, Structuring, Burkina Faso http://dx.doi.org/10.12692/ijb/26.1.141-155 Article published on January 08, 2025 Abstract Phenotypic markers are important in plant genetic characterisation studies. They are used in the present study to assess the phenotypic structuring of cassava genotypes obtained by biparental crossing. The plant material studied consists of 56 cassava genotypes from the third generation of vegetative reproduction following germination of seeds from six families resulting from crosses. To evaluate these genotypes, an Alpha lattice experimental design was used with three replicates and three blocks per replicate. Blocks I and II each contained 19 genotypes and block III 18 genotypes. Data was collected on 10 qualitative traits on leaves, stems and roots. All the variables evaluated presented several modalities. The frequencies showed that: the greenpurple color (41%) was dominant for the apical leaf color characteristic. Stems color were predominantly light brown (30%). Green color (57%) was most common in the petioles. Genotypes showed more dichotomous ports (44%). In addition, the relative Shannon-Weaver diversity index (H’) was very high for all characters within genotypes (H’=0.90) and families (H'=0.66). The most polymorphic traits between genotypes were flowering ability (H’=1), stem color (H’=0.99), tuberous root texture (H’=0.97), apical leaf color (H’=0.96) and branching type (H'=0.93). The same index showed high intra-family diversity, family VI (H’= 0.83), family II (H’= 0.76), family IV (H’=0.69), family I (H’= 0.61), family III (H’= 0.53) and family V (H’= 0.52) showing high internal variability. ACH was used to structure the genetics into three phenotypic groups. This observed diversity can be used for cassava breeding in Burkina Faso. * Corresponding Author: Sawadogo O. Michel  michel.sawad[email protected] International Journal of Biosciences | IJB | ISSN: 2220-6655 (Print) 2222-5234 (Online) http://www.innspub.net Vol. 26, No. 1, p. 141-155, 2025 142 Michel et al. Int. J. Biosci. 202 5 Introduction Manioc (Manihot esculenta Crantz 1766) is a perennial shrub 1 to 5 m high (Allem, 2002; Alves, 2002). It belongs to the class Dicotyledones, family Euphorbiaceae, genus Manihot and species Manihot esculenta Crantz (Isendahl, 2011; Soro, 2022). It has a diploid chromosome number of 2n=36 and a highly heterozygous genome (Alves, 2002). It is one of the most important tuberous root crops, highly valued for its starch content in tropical countries (N'Zué et al., 2014). Cassava is grown all over the world, particularly in West Africa (Agré et al., 2015). Cassava can be grown in areas with rainfall ranging from 500 mm to 8000 mm (François, 1989). Depending on the variety, production can be spread over a long period of the year, making the tuberous roots available when needed (François, 1989). In recent years in Burkina Faso, climate variability has made farming very difficult. Crop diversification is very important to ensure food self-sufficiency. Tuber and root crops such as cassava can therefore be used to help achieve sustainable food security. In Burkina Faso, cassava production was estimated at around 17,081.25 tonnes in 2022 (FAOSTAT, 2024). As in all African countries, almost all cassava production in Burkina Faso is used for human and animal consumption (Amani et al., 2007). The tuberous roots are eaten raw or in the form of local dishes: boiled roots, grilled roots, placali, con'godê, attiéké and gari (Guira et al., 2017). In view of its food and nutritional potential, the quantities of cassava produced remain below national market demand, which in 2017 was estimated at around 124,917 tonnes of fresh tubers (Soro et al., 2022). In Burkina Faso, the major constraints to large-scale production are linked to several factors, namely: the long production cycle of six to 9 or even 12 months, the unsuitable quality of the soils used for its cultivation, which results in low root yields, the lack of suitable varieties, and the very narrow genetic base of cassava (Gmakouba et al., 2018). In order to meet consumer needs, production must be increased, and this requires efficient production technology based on the use of improved cassava varieties. Exchanges of genetic material between producers mean that they end up with duplicates of the same cultivar (Soro et al., 2022). The reproduction of cassava, which is generally done by cuttings, leads to the spread of its bio-aggressors, which become more and more numerous and infest new fields. Studies carried out by Tiendrébéogo et al. (2009, 2012) reported the presence of Cassava Mosaic Diseases (CMD) in certain areas of Burkina Faso. Cassava is often grown under rainfed and irrigated systems in Burkina Faso. This is due to the earliness of the rains in relation to the length of the vegetative cycle and the poverty of the arable land, which means that average yields in farming areas are low, less than or equal to 15t/ha (FAOSTAT, 2024). In response to this situation, a great deal of research has been carried out by INERA through the introduction and evaluation of six (06) improved varieties, catalogued and popularised, TMS 4(2) 1425; TMS 91/02312; TMS 92/0067; TMS 92/0325; TMS 92/0427; TMS 94/0270) with potential yield (40/ha) (Gmakouba, 2018; Soro, 2022; MASA, 2014). But of these, only TMS 94/0270, commonly known as V5, is the most widely produced for its very good attiéké quality. To meet this challenge, new cassava varieties need to be developed, with a view to broadening the genetic base so as to obtain varieties that are tolerant to FGD, rich in beta-carotene, and with yields of up to 40 tonnes per hectare. It is therefore essential to assess the agro-morphological diversity of this cassava collection (Manihot esculenta Crantz) in order to better exploit the potential of these genotypes. This study was therefore carried out with the overall aim of determining the structure of the 56 genotypes obtained by biparental crossing. Specifically, the aim was (i) to determine the variability of genotypes through phenotypic traits and (ii) to identify the traits that best discriminate between genotypes and families. Materials and methods Experimental site The trial was planted on 8 August 2022 and harvested on 10 January 2024 on experimental plots at the Environmental, Agricultural Research and Training 143 Michel et al. Int. J. Biosci. 202 5 Centre of Kamboinsé (CREAF/K). The center is located at longitude 001° 32.583‘W, latitude 12°27.326’ N and altitude 298 m (GARMIN GPSMAP 64S). According to Guinko (1984), the center has a North Sudanian climate, with two alternating seasons: a rainy season from June to October and a dry season from November to May. The site receives an average of 800 mm of rainfall per year. According to meteorological data from the Kamboinsé station in 2022 and 2023, the wet crop years of 2022-2023 and 2023-2024 recorded 1091.7 mm of rain in 57 days and 689 mm of rain in 48 days respectively. Over the same period, during the trial, average monthly temperatures ranged from 22.6°C (January 2022) to 36.2°C (June 2023). The extermes temperatures recorded ranged from 12.8°C in January 2022 to 46.8°C in May 2023. Fig. 1 below shows the rainfall and temperature variations recorded during the trial. Fig. 1. Umbrothermal histogram for the Kamboinsé station in 2022 and 2023 Plant material The plant material consists of 56 cassava genotypes. These genotypes come from six cassava families obtained by seed germination. The seeds are obtained by biparental crossing. These clonal hybrids are descended from high-performance parents with a high carotenoid content and a high dry matter content of between 25 and 29%. The parents were crossed in Nigeria by the cassava team at International Institute of Tropical Agriculture (IITA), and in Burkina Faso by poly cross where the female is known and the male parent is not. It was possible to obtain seeds in Nigeria thanks to good collaboration between IITA and INERA in Kamboinsé. These seeds were germinated at CREAF in Kamboinsé for this study. Genotypes are considered to belong to the same family if they are derived from the same biparental cross. The cuttings used in the trial were taken from the third generation of plants obtained after seeds germination. Table 1 below lists the genotypes studied and their families. Table 1. Distribution of genotypes by family Families Females Males Hybrids/Genotypes FAM I TMS 30572 Poly cross TMS 30572 (268) FAM I TMS 30572 (270) FAM I TMS 30572 (271) FAM I TMS 30572 (275) FAM I TMS 30572 (276) FAM I TMS 30572 (290) FAM I TMS 30572 (291) FAM I TMS 30572 (292) FAM I TMS 30572 (297) FAM I TMS 30572 (298) FAM I TMS 30572 (302) FAM I TMS 30572 (303) FAM I TMS 30572 (304) FAM I TMS 30572 (305) FAM II IBA070337 IKN130010 (IBA070337XIKN130010) - 1 FAM II (IBA070337XIKN130010) - 11 FAM II (IBA070337XIKN130010) - 12 FAM II (IBA070337XIKN130010) - 14 FAM II (IBA070337XIKN130010) - 15 FAM II (IBA070337XIKN130010) - 16 FAM II (IBA070337XIKN130010) - 18 FAM II (IBA070337XIKN130010) - 2 FAM II (IBA070337XIKN130010) - 5 FAM II (IBA070337XIKN130010) - 9 FAM III IBA070593 IBA011797 (IBA070593XIBA011797) - 3 FAM III (IBA070593XIBA011797) - 4 FAM III (IBA070593XIBA011797) - 5 FAM III (IBA070593XIBA011797) - 7 FAM IV IBA070593 IKN120210 (IBA070593XIKN120210) - 1 FAM IV (IBA070593XIKN120210) - 14 FAM IV (IBA070593XIKN120210) - 2 FAM IV (IBA070593XIKN120210) - 3 FAM IV (IBA070593XIKN120210) - 4 FAM IV (IBA070593XIKN120210) - 8 FAM IV (IBA070593XIKN120210) - 9 FAM V IBA070593 IIKN130010 (IBA070593XIKN130010) - 1 FAM V (IBA070593XIKN130010) - 4 FAM V (IBA070593XIKN130010) - 5 FAM V (IBA070593XIKN130010) - 6 FAM V (IBA070593XIKN130010) - 9 FAM VI TOGO 1 Poly cross TOGO 1 (277) FAM VI TOGO 1 (278) FAM VI TOGO 1 (279) FAM VI TOGO 1 (279) FAM VI TOGO 1 (281) FAM VI TOGO 1 (282) FAM VI TOGO 1 (283) FAM VI TOGO 1 (284) FAM VI TOGO 1 (285) 144 Michel et al. Int. J. Biosci. 202 5 FAM VI TOGO 1 (286) FAM VI TOGO 1 (289) FAM VI TOGO 1 (290) FAM VI TOGO 1 (292) FAM VI TOGO 1 (293) FAM VI TOGO 1 (294) FAM VI TOGO 1 (296) TOTAL 56 Experimental design The experimental design used was an ‘Alpha lattice’ with three replicates. Each replication contains of 3 blocks. Blocks one and two contained 19 genotypes and block three contained 18 genotypes. The distance between blocks was 1.5 m and between replicates 2 m. The spacing between rows and between points where cuttings were planted was 1 × 1m. The line measured 4 m, with 5 planting points and was made up of the same genotype. The planting cuttings were 15 to 20 cm long and had at least three nodes. They were planted obliquely so that 1/3 of them were buried in the ground folow polar direction of the stem. To optimise competition between cassava plants and weeds, four (04) weedings were carried out during the vegetative phase of the plants. Water was added once every fortnight during dry periods during the trial from November 2022 to June 2023 and then from November 2023 to December 2023 in order to keep the plants alive and vigorous. Data collection The morphological descriptors used are those proposed by Fukuda et al. (2010). These descriptors have already been used by N'Zué et al. (2014); Agré et al. (2015) and Gmakouba et al. (2018) in agromorphological diversity studies of cassava grown in Côte d'Ivoire, Benin and Burkina Faso respectively. Ten (10) qualitatives traits taking into account leaves, stem, flowers and tuberous root were observed during the present study. The various observations were made on all the plants of the genotypes in each block. They were recorded at successive periods: at three, six and nine months after planting and then at harvesting of the tuberous roots. Table 2 below shows the said characters and the various scores relating to them observed during data collection. Table 2. List of qualitative characteristics Organs Periods after planting (months) Characters Abbreviations Phenotypic classes/ scores Leaves 3 Color of apical leaves CAL (3) Light green; (5) Dark - green; (7) Purplish - green; (9) Purple 6 Shape of central leaflet SCL (1) Ovoid; (2) Elliptic - lanceolate; (3) Obovate - lanceolate; (4) Oblong-lanceolate; (5) Lanceolate; (6) Linear; (7) Pandurate; ... 6 Petiole color PEC (1) Yellowish - green; (2) Green; (3) Reddish - green; (5) Greenish-red; (7) Red; (9) Purple. Flowers 6 Flowering aptitude FLAP (0) Absent; (1) Present Stems 9 Stem Color SC (3) Orange; (4) Yellow; (5) Golden; (6) Light - brown; (7) Silver; (8) Gray; (9) Dark-brown. 9 Branching habit BRH (1) Erect; (2) Dichotomous; (3) Trichotomous. 9 Color of end branches CEB (3) Green; (5) Green - purple; (7) Purple. Roots At harvest Color of root pulp (parenchyma) CRP (1) White; (2) Cream; (3) Yellow At harvest Root shape ROS (1) Conical; (2) Conical - cylindrical; (3) Cylindrical; (4) Irregular. At harvest Texture of root epidermis TRE (3) Smooth; (5) Intermediate; (7) Rough Data analysis The data collected during the trial was entered into the Excel 2019 spreadsheet in the form of a « genotype × phenotype character » matrix. Data processing and frequency calculations were carried out using the same spreadsheet. Phenotypic character frequency distributions were calculated for all genotypes and within each family. The ShannonWeaver diversity index noted (H) (Shannon and Weaver, 1949), as described by Jain et al. (1975) was calculated with the aim of determining the phenotypic diversity of cassava genotypes and revealing the 145 Michel et al. Int. J. Biosci. 202 5 degree of polymorphism of the 10 qualitative traits analysed in order to detect the distribution of extra and intra family diversity. This formula was used to calculate the index. H     With: H = Shannon-Weaver diversity index; Pi = Frequency of each phenotypic class i of a given trait; n = Number of phenotypic classes for each trait; The Shannon-Weaver diversity index (H) was then converted into a relative phenotypic diversity index (H’) by dividing it by its maximum value Hmax (ln (n)) to obtain values between 0 and 1. H ln   The relative or equitability diversity index of (H') reaches its minimum value, which is zero (0) for monomorphic characteristics. The value of the index increases with the degree of polymorphism of the characteristics. It reaches its maximum value of one (1) when all phenotypic classes have equal frequencies (Belhadj et al., 2015; Gashaw et al., 2016; Ka et al., 2020; Sawadogo et al., 2022). Results Phenotypic characteristics of cassava genotypes and families Comparative analysis of the results for qualitative characteristics showed a diversity of colors and shapes between genotypes and within each family. Leaf characteristics The results of the morphological observations on the leaves showed variability for all the qualitative characteristics studied. As regards the coloration of the apical leaves, four modalities were observed: light-green (20%), dark-green (20%), green-purple (40%) and purple (20%). As for the families, they showed varying frequencies in the color of the young leaves. The families II and III had more genotypes with green apical leaves (60%) and (75%) respectively. The families I and V had a majority of genotypes with green-purple apical leaves (41%) and purple apical leaves (60%) respectively. The Figs 2 and 3 below show the distribution of frequencies of the different modalities linked to the coloration of the apical leaves of the families. Fig. 2. frequencies distribution linked to the coloration of the apical leaves of genotypes within each Fig. 3. Different colors of apical leaves Fig. 4. Frequency distribution of leaf petiole color for each family Fig. 5. Different colors of petioles 146 Michel et al. Int. J. Biosci. 202 5 Regarding the petiole color, the frequency distribution showed varying degrees of color that differed markedly from one family to another. Family II contains genotypes with a wide variety of petiole colors, six modalities observed: yellowish-green (10%), green (20%), reddish-green (20%), greenishred (10%), red (30%) and purple (10%). However, only two modalities of petiole color were found in family IV, yellowish-green (29%) and green (71%). The green color of petiole was dominant in family I (79%), family III (50%), family IV (71%), family V (60%) and family VI (56%). Figs 4 and 5 below show the different petiole color and the frequency distribution of leaf petiole color in each family. Fig. 6. Frequency distribution for central leaflet shape Fig. 7. Various shapes of the central leaflet Shape of the central leaflet Figs 6 and 7 show that the six families studied are made up of genotypes with various central leaflet shapes. The ovoid form (40%) and the obovallanceolate form (2%) are found respectively in the majority and minority of families. Figs 6 and 7 show that the six families studied are made up of genotypes of various central leaflet shape. The ovoid form (40%) is the most common in all families. On the other hand, oboval-lanceolate form (2%) is very rarely found in families. Family VI contains the genotypes with the most diverse leaflet shapes: ovoid (38%), Elliptic-lanceolate (25%), oboval-lanceolate (13%), oboval lanceolate (6%), lanceolate (6%) and linear (13%). On the other hand, the majority of genotypes in family I have the eliptical lanceolate shape (86%). Fig. 8. Frequency distribution for family stem color Stem characteristics Stem color Analysis of the stem coloration of the genotypes revealed four modalities. The families showed varying frequencies of cassava stem color. Families II (50%), III (75%) and VI (31%) contained more genotypes with orange stems. Family IV has a high proportion of silver colored stems (86%). Families I (57%) and V (60%) are dominated by genotypes with light brown stems. Fig. 8 below shows the distribution of frequencies for the different modalities linked to stem color in the families. Types of branching For the types of stem branching, the calculation of data frequencies showed three forms of branching in all families. Erect stems dominate in families I (64%) and III (38%). Dichotomous branching is the dominant modality in families II (40%), III (50%), V (80%) and VI (58%). Trichotomous type branching is weakly encountered in all families. However, the trichotomous form is found in family II (29%) and 147 Michel et al. Int. J. Biosci. 202 5 family V (7%). Figs 9 and 10 below illustrate the various types of branching present in the families. Fig. 9. Frequency distribution of branching types within families Fig. 10. Types of branching linked to the stems of different families Fig. 11. Frequency distribution of branch color at the top for each family Color of end branching As for the color of the terminal branches (twigs at the top) studied, two modalities were observed: green (30%) and green-purple (70%). Both modalities were observed in almost all families, with the exception of family I, where all genotypes had green-purple twigs at the top. The green-purple color (68%) of the shoots was more common in all families except family III (29%). Figs 11 and 12 illustrate the frequency distribution of shoot color at the top of genotypes within the different families. Types of branches present in the families. Frequencies of the different modalities linked to stem color in the families. Fig. 12. Different colors of twigs at the top A: Gr een color, B: Purplish-green color Fig. 13. Distribution of flowering aptitude frequencies for each family Fig. 14. Illustration of the presence and absence of flowers in cassava genotypes A: flower ing plant, B: non-flowering plant Flowering ability The results of the analysis recorded in Fig. 13 below show two modalities (presence, or absence) for the genotype flowering ability trait. The two modalities for this trait were observed in all families. Indeed, families III, IV and VII contain more genotypes that bore flowers (75%), (86%) and (56%) respectively during their vegetative cycle. However, families I (79%) and V (60%) are made up of genotypes the majority of which did not bear flowers during their vegetative cycle during the trial. As for family II, there 148 Michel et al. Int. J. Biosci. 202 5 were as many flowering genotypes (50%) as nonflowering genotypes (50%). Figs 13 and 14 show the frequency distribution of genotypes in families according to their ability to flower. They also show the differences in stem color between families. Fig. 15. Distribution of frequencies related the shape of roots for each family Fig. 16. Different shapes of tuberous roots Tuberous root characteristics Roots shapes The results of the assessment of tuberous root shape (Fig. 16) showed four modalities within families. Families I, II and VI are made up of genotypes with all four modalities (conical, conical-cylindrical, cylindrical and irregular) of tuberous root shape observed in the present study. The conical-cylindrical form (13%) is rarely found and the cylindrical form (38%) is strongly observed in all families. Family V is dominated by genotypes with cylindrical tuberous roots (80%). Family III is dominated by genotypes with irregular root shapes (78%). Cylindrical roots (38%) followed by irregular roots (36%) are the most common in all families, as shown in Fig. 15 below. Root texture The texture of the tuberous root discriminated between genotypes within families. Rough texture is the most common (45%), followed by smooth (30%) and intermediate (25%). Families I, IV and VI contain genotypes with all three root modalities (smooth, intermediate and rough). Families II and III are made up of genotypes with predominantly rough root texture. The genotypes in family V have more intermediate texture (80%) than smooth (20%) (Fig. 17). Fig. 17. Distribution of frequencies related the root texture for each family Fig. 18. Distribution of tuberous root pulp frequencies in families Fig. 19. The different colors of the root pulp Fig. 18 below shows the frequency distribution of tuberous root pulp color for each family. The creamy color (63%) of tuberous root pulp is frequent in all families except family V. Families III and V are made up respectively of genotypes with a creamy (100%) and yellow (100%) pulp color only. Family IV 149 Michel et al. Int. J. Biosci. 202 5 contains all 3 types of root pulp color: white (14%), creamy (57%) and yellow (29%) (Fig. 19). Analysis of diversity indices for traits observed on genotypes The relative Shannon-Weaver diversity index (H) calculated for the different characters linked to the leaves, stems, flowers and tuberous roots of all the genotypes and families revealed significant diversity between genotypes and also between families (Table 3). For all the traits studied, the relative ShannonWeaver diversity index ranged from H’=0.71 (petiole color) to H’=1 (flowering ability) for the genotypes. The average relative index for genotypes was around 0.90. For families, the relative Shannon-Weaver index varied from 0.41 (pulp color) to 0.85 (flowering ability), with an average of 0.66 between families for all the characteristics studied. Table 3. Variation in the relative Shannon-Weaver diversity index for the characteristics studied in the genotypes of the families Characters Modalities Ge notypes H’ H'ave FAM I FAM II FAM III FAM IV FAM V FAM VI CAL 4 0.96 0.65 0.65 0.41 0.78 0.49 0.71 0.62 SCL 6 0.8 0.28 0.45 0.38 0.58 0.59 0.88 0.55 PEC 6 0.71 0.37 0.95 0.58 0.33 0.53 0.55 0.53 CSE 4 0.99 0.71 0.74 0.41 0.3 0.49 0.99 0.61 BRH 3 0.93 0.81 0.99 0.95 0.99 0.57 0.81 0.85 CEB 2 0.89 0 0.97 0.81 0.86 0.72 0.95 0.72 FLAP 2 1 0.75 1 0.81 0.59 0.97 0.99 0.85 ROS 4 0.92 0.96 0.86 0.41 0.72 0.36 0.97 0.71 TRE 3 0.97 0.98 0.56 0.51 0.87 0.46 0.93 0.72 CRP 3 0.84 0.63 0.46 0 0.87 0 0.51 0.41 H'ave 0.9 0.61 0.76 0.53 0.69 0.52 0.83 0.66 CAL: color of apicales leaves, SCL: Shape of central leaflet, PEC: Petiole color, CSE: Color of stem, BRH: Branching habit, CEB: Color of end branches. FLAP: Flowering Aptitude, ROS: Root shape, TRE: Texture of root epidermis, CRP Color of root pulp, H’: Shannon-Weaver relative diversity index; H’ave: Shannon-Weaver relative average diversity index. Table 4. Variation in the relative index of leaf-related trait diversity Characters Modalities H' H'ave FAM I FAM II FAM III FAM IV FAM V FAM VI CAL 4 0.65 0.65 0.41 0.78 0.49 0.71 0.62 PEC 6 0.37 0.95 0.58 0.33 0.53 0.55 0.55 SCL 6 0.28 0.45 0.38 0.58 0.59 0.88 0.53 H'ave 0.43 0.68 0.46 0.56 0.54 0.71 0.56 CAL: color of apicale leaves, PEC: Petiole color, SCL : Shape of central leaflet, H’ : Shannon-Weaver relative diversity index, H’ave : Shannon-Weaver relative average diversity index. Relatifs index of diversity of the various characters (H') Relatifs index of diversity linked to the leaf The relative Shannon-Weaver diversity index was calculated for all leaf related characters in the six (06) families. The four modalities observed for apical leaf color, the relative diversity index varied from 0.41 (family III) to 0.71 (family VI), with an average of 0.62 within the families. Concerning the color of the petiole, with six modalities encountered, the relative diversity index varied from 0.33 (family IV) to 0.95 (family II) with an average of 0.55. As for the shape of the central lobe, the six modalities also observed gave a low relative Shannon-Weaver diversity index (0.28) obtained by family I and a higher one (0.88) obtained by family VI with an average for all families of 0.53. The relative Shannon-Weaver diversity index for all leaf-related characters was more polymorphic for family VI, with an index of 0.71, and family I was less polymorphic, with an average index of 0.43. The