**3. Historical perspective of the Nigerian cashew germplasm**

Available records showed that cashew was introduced to Africa the same period with India i.e. about 16th century through trade mission by the Portuguese explorers (Johnson, 1973; Mitchell and Mori, 1987). In other word, Nigeria shared similar history with India and the first introductions around 400 years ago were planted in the coastal area around Agege, Lagos, Nigeria (Fig. 3), similar to Goa in India (Archak et al., 2009). Spontaneous planting from this coastal area facilitated its spread to other parts of Lower Niger of the country (Woodroof, 1967; Venkataramah, 1976; Togun, 1977; Ohler, 1979). After the introduction and


Source: FAO (2011) http://faostat.fao.org/site/567/DesktopDefault.aspx?PageID=567#ancor

Table 2. Cashew and nut production (land area (ha) yield/ha and annual total output) in Nigeria from 1990 – 2009.

spontaneous spread, cashew trees thrive in the wild for about three centuries (1600-1900s), with no commercial value and exploited mainly for the afforestation and control of gully erosion in most eastern parts of the country with subtropical savannah ecology.

#### **3.1 First and second cashew germplasm introductions**

Historically, the development of the Nigerian cashew industry shared history with the political independence of the nation with the emerging regional governments (Eastern, Northern and Western) of self-rule in the 1950s striving for economic independence and development. While the northern region with vast savannah ecology focused on the cultivation of cereals and pulses, the east and west explored tree crops (cocoa, rubber, oil palm and cashew). This period coincided with the development and advancement of cashew nut processing technology by Indians in the 1950s and the demand for raw nuts was on the increase. These historical events led to the establishment of first set of large commercial cashew plantations in Udi and Oghe by the Eastern Nigerian Development Corporation (ENDC) and Iwo, Eruwa and Oke-ogun by Western Nigeria Development Corporation (WNDC) in the period 1953-1960. In addition to sourcing planting materials from the first cashew introductions that had spread across the country (wild and few Agriculture and Forestry Departments), significant percentage of the planting materials (cashew seeds) used for the establishment of these large plantations (ENDC & WNDC) were imported directly from India by the Ministry of Agriculture of these two regional governments, and this

spontaneous spread, cashew trees thrive in the wild for about three centuries (1600-1900s), with no commercial value and exploited mainly for the afforestation and control of gully

Historically, the development of the Nigerian cashew industry shared history with the political independence of the nation with the emerging regional governments (Eastern, Northern and Western) of self-rule in the 1950s striving for economic independence and development. While the northern region with vast savannah ecology focused on the cultivation of cereals and pulses, the east and west explored tree crops (cocoa, rubber, oil palm and cashew). This period coincided with the development and advancement of cashew nut processing technology by Indians in the 1950s and the demand for raw nuts was on the increase. These historical events led to the establishment of first set of large commercial cashew plantations in Udi and Oghe by the Eastern Nigerian Development Corporation (ENDC) and Iwo, Eruwa and Oke-ogun by Western Nigeria Development Corporation (WNDC) in the period 1953-1960. In addition to sourcing planting materials from the first cashew introductions that had spread across the country (wild and few Agriculture and Forestry Departments), significant percentage of the planting materials (cashew seeds) used for the establishment of these large plantations (ENDC & WNDC) were imported directly from India by the Ministry of Agriculture of these two regional governments, and this

erosion in most eastern parts of the country with subtropical savannah ecology.

**3.1 First and second cashew germplasm introductions** 

1990 50,000 0.60 30,000 1991 75,000 0.60 45,000 1992 90,000 0.61 55,000 1993 120,000 0.63 75,000 1994 135,000 0.63 85,000 1995 155,000 0.61 95,000 1996 175,000 0.63 110,000 1997 243,000 0.51 125,000 1998 243,020 0.63 152,000 1999 248,000 1.75 417,000 2000 259,000 1.80 466,000 2001 265,000 1.83 485,000 2002 273,000 1.88 514,000 2003 277,000 1.89 524,000 2004 292,000 1.90 555,000 2005 309,000 1.92 594,000 2006 320,000 1.99 636,000 2007 330,000 2.00 660,000 2008 330,000 2.20 727,603 2009 330,000 1.76 580,761 Source: FAO (2011) http://faostat.fao.org/site/567/DesktopDefault.aspx?PageID=567#ancor Table 2. Cashew and nut production (land area (ha) yield/ha and annual total output) in

Nigeria from 1990 – 2009.

Year Harvested area (ha) tonnes/ha Annual Production (tons)

constituted the first major cashew genetic resources in the country. Cashew aggressively spread to the other parts of the country from these two regions thereafter and today the crop now grow effectively in twenty-seven (27) out of the thirty-six (36) federal states (Fig. 3) of Nigeria with estimated 330,000 hectares of cultivated cashew land and raw nut outputs of 640,000 metric tons (FAO, 2011). The establishment of large plantations in the country ushered in era of establishment of cashew kernel processing plants for value addition locally. And by the late 1960s to early 1970s, Nigeria was exporting both the raw nuts and processed kernels to India. However, with limited expertise locally, the emerging industry encountered challenges in all areas of the cashew value chain that later led to the enactment of a law in 1971 mandating the Cocoa Research Institute of Nigeria (CRIN) to carry out research and development into production, processing and marketing for the Nigerian cashew industry.

And as a part of the first initiative to improve production, Cocoa Research Institute of Nigeria embarked on its first exploration and germplasm collection from the existing cashew in the wild, farms and plantations (including WNDC) across the country as early as 1973 (Sanwo, 1973), and assembled the collections at the Institute's field gene banks located in Ibadan (Western Nigeria), Uhonmora (Mid-Eastern Nigeria) and Ochaja (North-central Nigeria) thereafter. These cashew collections were later found to be of narrow genetic base (see Akinwale & Esan, 1989; Aliyu & Awopetu, 2007a, 2007b) and step was taken to broadening the genetic pool by introducing more materials from India, Tanzania and Mozambique around 1978 and 1980, which constituted the 2nd germplasm introduction.

#### **3.2 First cashew on-farm evaluation and selection programme**

The introduction of germplasm was simultaneously accompanied by on-farm (farmers' fields or plantations) evaluation of some of the selected materials especially at the Western Nigerian Development Corporation (WND)'s plots. The outcome of such preliminary evaluation resulted in the selection of the 25 half-sib progenies (genotypes) with potential high yielding (> 1000kg/ha) and were released as improved cashew cultivars tagged G-series by CRIN in the 1980s (see Akinwale & Esan, 1989). These G-varieties have been the main improved cashew materials that were distributed to farmers since 1988. This preliminary research intervention led to a dramatic improvement especially in terms of access to improved planting materials by farmers, which have hitherto sourcing materials from the wild, and laid a strong foundation for a investable cashew industry in the country and neighbouring nations.

#### **3.3 Third cashew germplasm introduction**

Introduction of the International Monetary Fund (IMF) structural adjustment programmes (SAP) and adoption of liberalization policy in late 1980s by the then government altered landscape for the cashew trade as well and ushered in increased competition within the African cashew industry with emphasis on quality. This paradigm shift brought into the fore the need for compliance with global quality standard in the competitive markets. Unfortunately, most of the African cashew exports, Nigerian inclusive fell short of the required global grading standards especially in size and quality of the nuts and kernels. The challenge prompted some affluent farmers to introduce new cashew materials (with characteristic bold nuts and high grade kernels) directly from Brazil towards the end of 1980s. In partnership with these farmers, the Cocoa Research Institute of Nigeria collected germplasm accessions from these introductions to expand the National Cashew germplasm base. These Brazilian materials now constitute the third major introduction into Nigerian cashew germplasm. In total, Cocoa Research Institute of Nigeria currently housed about 22 hectares of cashew germplasm field across 6 locations in the country.

### **4. Challenges for the cashew production in Nigeria**

Cashew production in Nigeria like other major producing nations is constrained by low yield and variable nut yields, nut quality and pests and diseases infestation. Incidentally, most of the existing farms were established with open pollinated seeds sourced from the wild and unimproved land races. Expectedly, large percentage of such trees exhibited significant variation in all traits, from growth form, yield to diseases and pests tolerance. For example, cashew trees have been found producing between 0kg and about 45kg nuts per tree with average yield per hectare sometimes between 0.4ton and 1.2tons/ha (Aliyu & Awopetu 2007a). Thus, broadening the genetic base through introduction of new alleles from exotic germplasm (Faenza et al., 1982) and systematic exploitation of heterosis (Masawe, 1994) of the germplasm through a recurrent selection can only be the best solution for tackling yield-related problems in cashew. The Cocoa Research Institute of Nigeria established with a national mandate to address these production challenges has initiated a comprehensive programme for the documentation and evaluation of the existing cashew germplasm in the country with the goal to identify better cultivars that combine higher yield with nut quality (size, colour, etc.). Unfortunately, such characterization and evaluation efforts were biased towards phenotypic and agronomic traits such as nut size, nut weight, sex ratio, colour of apple, size of the fruits, tree canopy, length of panicle and yield performance (Akinwale & Esan 1989; Mneney et al., 2001; Aliyu & Awopetu, 2007a) due to low capacity and limited molecular resources for tree crop research in the country. Although the traditional phenotypic method is useful, more often than not its efficiency could be masked by environmental effect, hence, the need to complement the phenotypic data with molecular method. A preliminary data from the on-going evaluation is summarized in the next section.

#### **5. Update on the genetic diversity of the Nigerian cashew trees**

In the last ten years, efforts have been concentrated on the fifty-nine (59) accessions (Tables 3) of Nigerian major cashew germplasm, comprising of three sub-populations (old land races, Indian and Brazilian). These trees were selected because of their reliable passport data and the presence of atleast three replicates for each accession. Available records showed that eleven (11) of these accessions were collected as clonal materials (CC-) from the farmers' fields along the Ochaja-Ankpa axis of the Kogi State (representing north central) in 1987/88. Furthermore, that these materials were probably from the remnants of the first introduction over three centuries ago (Sanwo, 1973; Akinwale & Esan, 1989). The second lot of twenty-three (23) accessions were mainly Indian introductions (CSI-) collected as open pollinated progenies from the Eastern- and Western- Nigeria Development Corporation (ENDC and WNDC) plantations in the 1970s, and were planted in the current locations (CRIN, Ibadan and Ochaja) between 1985 and 1986. And the third set of twenty-five (25) exotic accessions (CSO-) were from cashew materials recently introduced from Brazil by a private farm, Kosoni Ola Farms Limited, Oro, Kwara State and were planted in 1987/1988. These fifty-nine (59) accessions were planted at a spacing of 9.0m x 9.0m, with each accession represented by three entries.

base. These Brazilian materials now constitute the third major introduction into Nigerian cashew germplasm. In total, Cocoa Research Institute of Nigeria currently housed about 22

Cashew production in Nigeria like other major producing nations is constrained by low yield and variable nut yields, nut quality and pests and diseases infestation. Incidentally, most of the existing farms were established with open pollinated seeds sourced from the wild and unimproved land races. Expectedly, large percentage of such trees exhibited significant variation in all traits, from growth form, yield to diseases and pests tolerance. For example, cashew trees have been found producing between 0kg and about 45kg nuts per tree with average yield per hectare sometimes between 0.4ton and 1.2tons/ha (Aliyu & Awopetu 2007a). Thus, broadening the genetic base through introduction of new alleles from exotic germplasm (Faenza et al., 1982) and systematic exploitation of heterosis (Masawe, 1994) of the germplasm through a recurrent selection can only be the best solution for tackling yield-related problems in cashew. The Cocoa Research Institute of Nigeria established with a national mandate to address these production challenges has initiated a comprehensive programme for the documentation and evaluation of the existing cashew germplasm in the country with the goal to identify better cultivars that combine higher yield with nut quality (size, colour, etc.). Unfortunately, such characterization and evaluation efforts were biased towards phenotypic and agronomic traits such as nut size, nut weight, sex ratio, colour of apple, size of the fruits, tree canopy, length of panicle and yield performance (Akinwale & Esan 1989; Mneney et al., 2001; Aliyu & Awopetu, 2007a) due to low capacity and limited molecular resources for tree crop research in the country. Although the traditional phenotypic method is useful, more often than not its efficiency could be masked by environmental effect, hence, the need to complement the phenotypic data with molecular method. A preliminary data from the on-going evaluation is summarized in the next section.

hectares of cashew germplasm field across 6 locations in the country.

**5. Update on the genetic diversity of the Nigerian cashew trees** 

In the last ten years, efforts have been concentrated on the fifty-nine (59) accessions (Tables 3) of Nigerian major cashew germplasm, comprising of three sub-populations (old land races, Indian and Brazilian). These trees were selected because of their reliable passport data and the presence of atleast three replicates for each accession. Available records showed that eleven (11) of these accessions were collected as clonal materials (CC-) from the farmers' fields along the Ochaja-Ankpa axis of the Kogi State (representing north central) in 1987/88. Furthermore, that these materials were probably from the remnants of the first introduction over three centuries ago (Sanwo, 1973; Akinwale & Esan, 1989). The second lot of twenty-three (23) accessions were mainly Indian introductions (CSI-) collected as open pollinated progenies from the Eastern- and Western- Nigeria Development Corporation (ENDC and WNDC) plantations in the 1970s, and were planted in the current locations (CRIN, Ibadan and Ochaja) between 1985 and 1986. And the third set of twenty-five (25) exotic accessions (CSO-) were from cashew materials recently introduced from Brazil by a private farm, Kosoni Ola Farms Limited, Oro, Kwara State and were planted in 1987/1988. These fifty-nine (59) accessions were planted at a spacing of 9.0m x 9.0m, with each accession represented by three entries.

**4. Challenges for the cashew production in Nigeria** 



ENDC: Eastern Nigerian Development Corporation, WNDC: Western Nigerian Development Corporation, KFL: Kosoni-Ola Farm Limited.

Table 3. List of the fifty-nine cashew accessions and their pedigree analyzed for the yield and yield components variability over a period of ten years (1999-2009).

These cashew trees have attained full maturity (i.e. above 10 years old) at the commencement of the phenotypic evaluation in 1999/2000. Based on the data from previous studies (Masawe, 1994; Azevedo et al., 1998; Aliyu, 2006; Aliyu & Awopetu 2007a), only ten yield-related component characters (whole fruit weight – WWT, individual nut weight – NWT, total nut yield per tree – NYT, kernel weight – KWT, number of hermaphrodite flowers per panicle – HPP, percentage of pollen fertility - PPS, tree canopy size –TCS, days to optimum flowering- DFF, days to optimum fruit maturity - DFM and effective harvesting period (day) – HPD were selected for the phenotypic evaluation. The ten years data (1999-2009) were statistically analyzed using SYSTAT version 13.0 softwares (Systat Software, Inc. USA) and compared thereafter with three years data (Aliyu & Awopetu, 2007a).

Cluster analyses {Eucledian distance ward dendogram and principal component analysis (PCA} from the two studies showed no significant difference between the genetic groupings of the cashew accessions (see Aliyu & Awopetu, 2007a). The data grouped the accessions into five (5) major classes and eight (8) subsets that depict fusion based on source of introduction and/or breeding history (Figs. 4 & 5; Table 4). An overview of the phenotypic (genetic) variability of the accessions showed that the cluster analyses (Figs. 4 & 5) separated the trees into high yielding (clusters I, II & III) and low yielding (clusters IV & V) genotypes. And more than 50% of the accessions fell within the low yielding category. This trend does not only reflect enormous variability that could exist in a typical cashew field, but brought into the fore the level of redundancy in cashew farm in terms of yield performance. To facilitate the efficient utilization of better candidates identified from this evaluation exercise, a short qualitative description of their agronomic importance for each set derived from the cluster analyses (Figs. 4 & 5) is given below.

CSO11 " " " CSO12 " " " CSO13 " " " CSO14 " " " CSO15 " " " CSO16 " " " CSO17 " " " CSO18 " " " CSO19 " " " CSO20 " " " CSO21 " " " CSO22 " " " CSO23 " " " CSO24 " " " CSO25 " " " ENDC: Eastern Nigerian Development Corporation, WNDC: Western Nigerian Development

Table 3. List of the fifty-nine cashew accessions and their pedigree analyzed for the yield

These cashew trees have attained full maturity (i.e. above 10 years old) at the commencement of the phenotypic evaluation in 1999/2000. Based on the data from previous studies (Masawe, 1994; Azevedo et al., 1998; Aliyu, 2006; Aliyu & Awopetu 2007a), only ten yield-related component characters (whole fruit weight – WWT, individual nut weight – NWT, total nut yield per tree – NYT, kernel weight – KWT, number of hermaphrodite flowers per panicle – HPP, percentage of pollen fertility - PPS, tree canopy size –TCS, days to optimum flowering- DFF, days to optimum fruit maturity - DFM and effective harvesting period (day) – HPD were selected for the phenotypic evaluation. The ten years data (1999-2009) were statistically analyzed using SYSTAT version 13.0 softwares (Systat Software, Inc. USA) and compared thereafter with three

Cluster analyses {Eucledian distance ward dendogram and principal component analysis (PCA} from the two studies showed no significant difference between the genetic groupings of the cashew accessions (see Aliyu & Awopetu, 2007a). The data grouped the accessions into five (5) major classes and eight (8) subsets that depict fusion based on source of introduction and/or breeding history (Figs. 4 & 5; Table 4). An overview of the phenotypic (genetic) variability of the accessions showed that the cluster analyses (Figs. 4 & 5) separated the trees into high yielding (clusters I, II & III) and low yielding (clusters IV & V) genotypes. And more than 50% of the accessions fell within the low yielding category. This trend does not only reflect enormous variability that could exist in a typical cashew field, but brought into the fore the level of redundancy in cashew farm in terms of yield performance. To facilitate the efficient utilization of better candidates identified from this evaluation exercise, a short qualitative description of their agronomic importance for each set derived from the cluster analyses (Figs. 4 & 5) is given

and yield components variability over a period of ten years (1999-2009).

Corporation, KFL: Kosoni-Ola Farm Limited.

years data (Aliyu & Awopetu, 2007a).

below.

#### **5.1 Some agronomic values of the Nigerian cashew trees**

#### **5.1.1 Very high yielding with moderate quality kernels**

A mixture of two (2) Farmers clones and five (5) Indian accessions were the most prolific (producing >3000 nuts/tree/year) of all the fifty-nine accessions studied (Figs. 4 & 5; Table 4). These trees were characterized by heavy fruit clustering on the panicle and have potential to produce >2.5 tons of raw nuts per hectare annually. But the average weight of nuts is about 6.0g and individual nut weight ranging between 4.50 and 8.20g. Derivable kernels from these trees were mixture of W320 and W450 grades with the latter being prominent. The heavy annual production was consistent throughout the study and fruit maturity is mostly in the middle of the season i.e. March. The trees shared prolific fruiting (nut per tree) with accessions that constituted Cluster III (Figs. 4 & 5; Table 4). These cashew accessions are mostly suitable for immediate use as planting materials because of the superior yield characteristic, but little improvement work on the kernel quality is needed to select candidates with stable W320 kernels.

#### **5.1.2 Moderate yielding with high quality kernels**

Eleven (11) Brazilian accessions that combine moderate-high yielding, regular production with high quality nuts and kernels were identified from the analysis ((Figs. 4 & 5; Table 4). Weight of individual nut from these trees range between 8.50g and 14.0g, and was characterized by a mixture of W320 and W240 kernels, with the latter being in abundance. Furthermore, these trees were noted for early fruit production and short harvesting time. These materials are better than the G-series that were released and given to farmers since the late 1980s by the Cocoa Research Institute of Nigeria (CRIN) (see Akinwale & Esan, 1989). These eleven superior cashew accessions in addition to those described above (5.1.1) can be clonally propagated for release as improved cultivars to the farmers to meet the short term need and/or undergo further evaluation across different locations through a long term national and/or regional cashew improvement programme. In addition, these new materials can be used for the establishment of polyclonal seed gardens for both research and commercial uses.

#### **5.1.3 Cashew trees prolific fruiting with compact canopy**

Nine (9) of the accessions were (see list Table 4 & Fig. 4. ) characterized by trees with small and compact canopy, though they are highly prolific in fruiting but the total output per tree were significantly influenced by smallness of its nuts and kernels. The genetic attributes of these nine cashew accessions include small sized trees with compact canopy. These plant materials could be useful for future breeding of cashew cultivars that would be adaptable to high density planting in an effort to improve outputs per unit area.

#### **5.1.4 Cashew trees with low yield and poor agronomic qualities**

The characterization exercise also revealed that about 40% of the experimental lots (25 accessions, Table 4; Figs. 4 & 5) were of low yield and poor in agronomic traits. However, because of the large canopy nature of these trees, they are good genetic resources for the afforestation, land reclamation and erosion controlled programmes in the arid regions and areas threatened by gully erosions. The proportion of such trees in the gene pools should be reduced significantly. Further studies are needed to understand the poor correlation between tree size and nut yield in cashew (Masawe et al. 1999; Aliyu & Awopetu, 2007a).


Table 4. Inter- and intra-cluster fusion of the 59 cashew accessions derived from the genetic diversity analysis showing source of origin/or breeding history effect.

#### **5.1.5 Cashew trees with low yield and high quality kernels**

Seven (7) accessions that consistently produced fewer but extra fruits (both apple and nut) were identified from the study. They are predominantly Brazilian collections (Table 4 and Figs. 4 & 5) with low yield because these trees rarely produce more than 300-400 fruits per trees. With average nut weight of 16g, these trees can be a good source for the introgression of genes for high grade kernels in cashew breeding programme. Other attributes of these five accessions include high volume of apple juice yield with low astringency. This category of cashew trees are often referred to as Jumbo varieties by farmers and are inadvertently collected as planting materials from unapproved sources. Unfortunately, recent study on nut size and number trade-off in cashew (Aliyu & Awopetu, 2011) has shown poor yield in this category of cashew trees, thereby corroborating this result. Hence, such materials are not suitable for the establishment of investable cashew farms, but could be used as a good source of genetic resources for research and developments of better varieties/cultivars.

#### **6. Recurrent selection strategy for the development of hybrid cashew**

Following the conclusion of the evaluation exercise and grouping of the entries into their respective agronomic groups, the Institute is embarking on a long term recurrent selection strategy for the development of improved cashew varieties. Apart from multilocational evaluation of most of the superior materials identified in the characterization exercise, the breeding plan included exploitation of heterosis between the highly prolific, but small fruits

No. of Farmers Land races % of total accession (n=59)

5 0 2 11.9 21.7 0 18.2

0 11 0 18.6 0 44.0 0

4 4 1 15.3 17.4 16.0 9.1

0 3 3 10.2 0 12.0 27.3

7 0 1 13.6 30.4 0 9.1

0 0 4 6.8 0 0 36.4

7 0 0 11.9 30.4 0 0

0 7 0 11.9 0 28 0

% of Indian accession (n=23)

% of Brazilian accession (n=25)

% of Farmers Land races (n=11)

No. of Brazilian accession

Table 4. Inter- and intra-cluster fusion of the 59 cashew accessions derived from the genetic

Seven (7) accessions that consistently produced fewer but extra fruits (both apple and nut) were identified from the study. They are predominantly Brazilian collections (Table 4 and Figs. 4 & 5) with low yield because these trees rarely produce more than 300-400 fruits per trees. With average nut weight of 16g, these trees can be a good source for the introgression of genes for high grade kernels in cashew breeding programme. Other attributes of these five accessions include high volume of apple juice yield with low astringency. This category of cashew trees are often referred to as Jumbo varieties by farmers and are inadvertently collected as planting materials from unapproved sources. Unfortunately, recent study on nut size and number trade-off in cashew (Aliyu & Awopetu, 2011) has shown poor yield in this category of cashew trees, thereby corroborating this result. Hence, such materials are not suitable for the establishment of investable cashew farms, but could be used as a good source of genetic

diversity analysis showing source of origin/or breeding history effect.

resources for research and developments of better varieties/cultivars.

**6. Recurrent selection strategy for the development of hybrid cashew** 

Following the conclusion of the evaluation exercise and grouping of the entries into their respective agronomic groups, the Institute is embarking on a long term recurrent selection strategy for the development of improved cashew varieties. Apart from multilocational evaluation of most of the superior materials identified in the characterization exercise, the breeding plan included exploitation of heterosis between the highly prolific, but small fruits

**5.1.5 Cashew trees with low yield and high quality kernels** 

No. of Indian accession

Cluster Subcluster Accessions in subclusters

CSI58, CSI66, CC05, CSI36

CSO20, CSO12, CSO06, CSO05, CSO07, CSO01, CSO02, CSO15, CSO10

CSI05, CSI18, CSO03, CSO13, CC11, CSI00

CC04, CC09, CSO17

CSO24, CSO23, CSO25, CSO24, CSO22

I Ia. CC06, CSI62, CSI31,

II IIb. CSO14, CSO19,

III IIIc. CSI16, CSO16, CSO14,

IV IVd. CSO18, CSO11, CC07,

IVe. CC10, CSI06, CSI30, CSI01, CSI09, CSI61, CSI10, CSI27,

IVf. CC03, CC01, CC02, CC08

IVg. CSI51, CSI67, CSI23, CSI07, CSI13, CSI14, CSI11

V Vh. CSO09, CSO08,

#### **Legends for the characters and key.**

*NYT:- Total nut yield per tree per year (kg) (1-Low, 2-Moderate, 3-High, 4-Very High); NNT:- Nuts per tree/year (1- Low, 2-Moderate, 3-High, 4-Very High, 5- Extra-super high); HPP:- Hermaphrodite flowers per panicle (1- Low, 2- Moderate, 3-High, 4-Very High); PPS:- Pollen grain fertility (%) (1- Low, 2-Moderate, 3-High, 4-Very High); TCS:- Tree canopy size (m2) (1- Compact, 2- Moderate, 3- Large open, 4- Spread with extensive branches); NWT:- Nut weight (g) (1- Small, 2- Medium, 3- Large, 4- Extra large); WWF:- Whole fruit weight (g) (1- Small, 2- Medium, 3- Large, 4- Extra large);KWH:– kernel weight (g) (1- W450, 2-W320, 3-W240, 4-W180); DFM:- Days to optimum fruit maturity (1- Early, 2-Mid-season, 3- Late); HPD:- Harvesting period (1-Short, 2- Intermediate, 3-Prolong).* 

Fig. 4. Phenotypic variability of the between fifty-nine cashew germplasm accessions for the nine yield related components as obtained from the ten (10) years data (1999-2009).

Fig. 5. Principal component analysis (PCA) of the fifty-nine cashew accessions showing five clusters and their respective agronomic characteristics.

and very low productive and extra large fruit trees through controlled hybridization and *in vitro* embryo culture, where necessary. Less impediment is expected from the hybridization exercise because of the high cross-compatibility in cashew (Aliyu, 2007, 2008), albeit, that the environmental conditions are favourable.

CSI11

CSO02

CSO06

CSO05

CSO10

CSO15

CSO12


CSO01

CSI36 CSO07

CSO03

CSI63

CSO14

CC06



 Very high fruit clustering Very high fruits/tree Super tree yield Medium fruit size Medium nuts W320/450 kernels Consistent fruiting Mid-season fruiting Moderate trees with upright canopy



0

PCA 2 1

2

3

CSI58

CSI05

CSI31

 Very high fruit clustering Very high fruits/tree Moderate yield Small fruit size Extra small nuts W450/500 kernels Consistent fruiting Compact canopy Small trees Mixture early and midseasons fruiting Short maturity period

CSO13

CC05

**I** 

CSI00

**III** 

CSI62

CSI66

CSO20

CSO04

PCA 1 -2 0 2

CSO19

CSO08

**II** 

CSO09

1

 Moderate fruit clustering High fruits/tree High yield

 Large nuts + medium apples W240/320 kernels Consistent fruiting Early season fruiting Short maturity period Moderate trees

CSI67

CSO16 CSI27

CSI18 CSI10 CSI14

CSI13

CC11 CSO17

CSI07

CSI51

CC10

CSI30

CSI06

CSI23 CC07

CSI61

CSI09

Fig. 5. Principal component analysis (PCA) of the fifty-nine cashew accessions showing five

CSO25

CSO22

3

CC09

**IV**

CSO18

 Sparse fruit clustering Fewer fruits/tree Low yield

 Mixture of small and medium fruits Mixture of nut sizes Erratic fruiting pattern Mixture of W320/450 kernels Large trees with extensive canopy

CSO11

CSO21

CSO24

**V** 

CC02

CC03

CC08

CSI01 CC04

CC01

CSO23

 Sparse fruit clustering Fewer fruits/tree Low yield Extra large fruit (nut + apple)-Jumbo nuts

 W180 kernel Spread canopy Biennial fruiting Long maturity period Low apple juice astringency

and very low productive and extra large fruit trees through controlled hybridization and *in vitro* embryo culture, where necessary. Less impediment is expected from the hybridization exercise because of the high cross-compatibility in cashew (Aliyu, 2007, 2008), albeit, that the

clusters and their respective agronomic characteristics.

environmental conditions are favourable.

### **7. Application of protein-isoenzme electrophoresis as a useful tool for cashew characterization**

In the absence of a well equipped laboratory with recent markers such as Random Amplification of Polymorphic DNA (RAPD), Amplified Fragment Length Polymorphism (AFLP) and Simple Sequence Repeats (SSRs), protein-isozyme sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) has proven as a reliable alternative for the characterization of cashew germplasm. The PAGE analysis differentiated the accessions into six overlapping subclasses similar to the results of morphological characterization i.e. corroborating groupings based on phenotypic attributes of the trees (see Aliyu & Awopetu, 2007b). For example, out of fifteen (15) accessions of the Brazilian origin grouped together as a cluster I at 43.0% linkage distance by PAGE analysis (see Aliyu & Awopetu, 2007b), eleven (11) were clustered together as Cluster II in the 10 year phenotypic data, suggesting genetic congruity between phenotypic attributes and breeding history/or and source of origin of these cashew plants. Similarly, five accessions CSO24, CSO25, CSO22, CSO23 and CSO21 that are morphologically and agronomically similar for their extra-large (jumbo) fruits (nut and apple), late and irregular flowering and fruiting characteristics, and spreading and extensive branching pattern, clustered together on the PAGE analysis because they shared common isoenzymes and probably reinforcing theory of source of origin and/or increased nuclear DNA content as the underline genetic factor for the grouping. Beside the general characteristics, isoenzyme study further differentiates the accessions within a sub-cluster on the basis of apple skin pigmentation into red, yellow and orange. Similar to the results of the ongoing molecular (SSR) analysis, the protein-isozyme study equally revealed moderate genetic base for the Nigerian cashew germplasm. The result of the PAGE analysis (Aliyu & Awopetu, 2007b) has demonstrated that this biochemical method can be used as a reliable and effective alternative characterization tool in the assessment of genetic relationships and grouping into morphotypes in cashew.

#### **8. Microsatellite analysis reveals genetic redundancy in Nigerian germplasm**

In collaboration with the Leibniz Institute of Plant Genetics and Crop Plant Research (IPK), Gatersleben, Germany, one hundred and eighty-seven (187) cashew accessions from different populations are being evaluated using 10 simple sequence repeats (SSR) markers (Croxford et al., 2006) (Table 5 ). An overview of the phenogram derived from the molecular data showed that eight (8) of the ten (10) microsatellite loci except, mAoR2 and mAoR47 were polymorphic (Fig. 6) and could be efficiently used for characterization and management of cashew germplasm. Preliminary data from the study showed significant level of redundancy (homogenous group) (Fig. 6) within the Nigerian cashew germplasm and the need to develop a core collection to reduce the cost of management of cashew germplasm. The data also depicts narrow genetic base of the Nigerian cashew germplasm. Apparently, future efforts at enhancing genetic base of Nigeria cashew germplasm should focus on collaborations and extensive linkages with relevant research institutions from Asia and South America.

#### **9.** *In vitro* **embryo culture in cashew**

Embryos from immature nuts of cashew have been successfully cultured *in vitro* at the Cocoa Research Institute of Nigeria, to establish a routine protocol for the regeneration of plantlets from crosses between distant parents (genotypes). In the trial, 2-, 4-, 6- and 8-weeks


Table 5. List of ten (10) microsatellite markers with optimal annealing temperature (Ta) and allelic size ranges (Croxford et al. 2006) used for the molecular analysis of the genetic diversity in 187 accessions of cashew.

old embryos were evaluated on different media compositions ranging from pure Murashige and Skoog (MS) agar medium (Murashige & Skoog, 1962) to modified MS medium supplemented with 1 mM each of naphthaleneacetic acid (NAA), benzyladenine (BA) and gibberellic acid (GA3). The results showed that age of the embryo significantly influenced the rates of response and survival of the plantlets, with older embryos i.e. 6 weeks old and above performed better (Table 6) (Aliyu & Awopetu, 2005). Among the media composition tested, only modified MS medium supplemented with 1 mM of gibberellic acid (MS+GA3)


Percentage in parenthesis. LSD 0.05 = 0.332. DMRT: Duncan multiple range test.

a: WAPo: Week after pollination

b: Murashige and Skoog (MS) medium

c: MS medium + gibberellic acid

d: MS medium + naphthaleneacetic acid

e: MS medium + benzyladenine.

Table 6. Effect of age on the growth and development of cashew embryos in different *in vitro*  culture media compositions.

Locus Primer sequence (5'- 3') Repeat motif Ta (oC) Allelic size

R: GGAAGGGCATTATGGGTAAG (CA)10(TA)6 58.2 366-375

R: ATCCAGACGAAGAAGCGATG (AC)12(AAAAT)2 60.3 241-247

R: CCCCATCAAACCCTTATGAC (AT)5(GT)12 58.2 143-157

R: GTGAATCCAAAGCGTGTG (AT)2(GT)5AT(GT)5 58.2 178-181

R: CTTACAGCCCCAAACTCTCG (AT)3(AC)16 60.3 234-236

R: GGTTTCGCATGGAAGAAGAG (GA)24 56.1 124-159

R: GCGAAGGTCAAAGAGCAGTC (CAT)9TAT(CTT)7 60.3 197-206

R: GACCATGGGCTTGATACGTC (GAA)6(GA)3 58.2 172-178

R: GTGACACAACCAAAACCACA (GT)16(TA)2 58.2 191-203

Table 5. List of ten (10) microsatellite markers with optimal annealing temperature (Ta) and allelic size ranges (Croxford et al. 2006) used for the molecular analysis of the genetic

old embryos were evaluated on different media compositions ranging from pure Murashige and Skoog (MS) agar medium (Murashige & Skoog, 1962) to modified MS medium supplemented with 1 mM each of naphthaleneacetic acid (NAA), benzyladenine (BA) and gibberellic acid (GA3). The results showed that age of the embryo significantly influenced the rates of response and survival of the plantlets, with older embryos i.e. 6 weeks old and above performed better (Table 6) (Aliyu & Awopetu, 2005). Among the media composition tested, only modified MS medium supplemented with 1 mM of gibberellic acid (MS+GA3)

(WAPo)a MSb MS+GA3c MS+NAAd MS+BAe 2 0.00 (0.00%) 5.91 (9.85%) 0.00 (0.00%) 0.00 (0.00%) 4 12.99 (21.65%) 19.71 (32.85%) 14.79 (24.65%) 15.69 (26.15%) 6 37.20 (62.00%) 41.31 (68.85%) 36.09 (60.15%) 36.81 (61.35%) 8 45.69 (76.15%) 48.69 (81.15%) 43.20 (72.00%) 42.30 (70.50%)

DMRT (x̄) 7.99b 9.63a 7.88b 7.85b

Table 6. Effect of age on the growth and development of cashew embryos in different *in vitro* 

Percentage in parenthesis. LSD 0.05 = 0.332. DMRT: Duncan multiple range test.

(TAAA)2(TA)7(AAT) 5

mAoR2 F: GGCCATGGGAAACAACAA

mAoR3a F: CAGAACCGTCACTCCACTCC

mAoR6c F: CAAAACTAGCCGGAATCTAGC

mAoR7b F: AACCTTCACTCCTCTGAAGC

mAoR11c F: ATCCAACAGCCACAATCCTC

mAoR17b F: GCAATGTGCAGACATGGTTC

mAoR42c F: ACTGTCACGTCAATGGCATC

mAoR47 F: AAGAGCTGCGACCAATGTTT

mAoR48a F: CAGCGAGTGGCTTACGAAAT

mAoR52 F: GCTATGACCCTTGGGAACTC

diversity in 187 accessions of cashew.

Age of embryos

a: WAPo: Week after pollination b: Murashige and Skoog (MS) medium c: MS medium + gibberellic acid d: MS medium + naphthaleneacetic acid e: MS medium + benzyladenine.

culture media compositions.

R: CTTGAACTTGACACTTCATCCA

range (bp)

58.2 161-173

supported growth of young embryos of 2 weeks old. In order words, factors such as medium composition, age of embryo and sometimes the genotype influence success rate of *in vitro* culture of cashew embryos. Photoperiod, temperature, source of explants, browning, and contamination are other known factors capable of affecting *in vitro* propagation of cashew (Jha, 1988; Das et al., 1996). The study deed showed that older cashew embryos seem to be autonomous of growth regulators i.e. inclusion of these synthetic plant hormones into media composition only became critical for very young explants.

Fig. 6. An overview of the genetic diversity of 187 Nigerian cashew accessions from different populations derived from molecular analysis of 10 microsatellite marker loci.

#### **10. Cashew cytology and cytometry**

#### **10.1 Relative nuclear DNA content and genome size in cashew**

Information on genome size and ploidy needed for basic molecular breeding of this important commodity is rare. Recent flow cytometric analysis of fifty-four (54) cashew accessions from the Nigerian germplasm was carried out to determine the relative genome size, intraspecific variation and ploidy status of the species using *Solanum lycopersicum* cv. Stupicke as an internal standard reference. And because of the dearth of literatures on application of flow cytometry to cashew or its relatives, the study was preceded by the protocol optimization for the buffer system, sample size, internal reference standard and incubation time for isolated nuclei before analysis (Aliyu, 2011under review). From the analysis of the nuclear suspension in terms of fluorescence intensity, background yield (%), nuclear yield (nuclei s-1 mg-1) and coefficient of variation (%) of *G0*/*G1* peak, the cashew plant showed preference for Otto's buffer, leaf sample size of about 70mg and maximum of 20 mins incubation period. *S. lycospersicum* cv. Stupicke and *Glycine max* cv. Polanka are the ideal internal reference standards for the genome size determination in cashew (Fig. 7a)*.* The average relative nuclear DNA content in cashew is small with relative genome size of 2C = 1.01 pg i.e. about 988Mbp (1C = 490Mbp) recorded for the analyzed accessions (Aliyu, 2010). The relative nuclear DNA contents ranged from smallest (0.903 pg) to largest (1.285 pg) i.e. about 1.36 – fold range. Slight variation for relative genome size was recorded between the accessions (Fig. 7b) and it would be too hasty to conclude that such observation is an

Fig. 7. (a): Flow cytometric histogram showing relative DNA content between G1 nuclei of cashew (*A. occidentale*) and tomato (*S. lycopersicum*) (tomato as internal reference standard). (b): Histogram showing suspected intraspecific variation in relative DNA content between two accessions (IMA-10 and BRO-26) of cashew (*A. occidentale*) with different phenotypic characteristics (see Aliyu, 2010).

Information on genome size and ploidy needed for basic molecular breeding of this important commodity is rare. Recent flow cytometric analysis of fifty-four (54) cashew accessions from the Nigerian germplasm was carried out to determine the relative genome size, intraspecific variation and ploidy status of the species using *Solanum lycopersicum* cv. Stupicke as an internal standard reference. And because of the dearth of literatures on application of flow cytometry to cashew or its relatives, the study was preceded by the protocol optimization for the buffer system, sample size, internal reference standard and incubation time for isolated nuclei before analysis (Aliyu, 2011under review). From the analysis of the nuclear suspension in terms of fluorescence intensity, background yield (%), nuclear yield (nuclei s-1 mg-1) and coefficient of variation (%) of *G0*/*G1* peak, the cashew plant showed preference for Otto's buffer, leaf sample size of about 70mg and maximum of 20 mins incubation period. *S. lycospersicum* cv. Stupicke and *Glycine max* cv. Polanka are the ideal internal reference standards for the genome size determination in cashew (Fig. 7a)*.* The average relative nuclear DNA content in cashew is small with relative genome size of 2C = 1.01 pg i.e. about 988Mbp (1C = 490Mbp) recorded for the analyzed accessions (Aliyu, 2010). The relative nuclear DNA contents ranged from smallest (0.903 pg) to largest (1.285 pg) i.e. about 1.36 – fold range. Slight variation for relative genome size was recorded between the accessions (Fig. 7b) and it would be too hasty to conclude that such observation is an

Fig. 7. (a): Flow cytometric histogram showing relative DNA content between G1 nuclei of cashew (*A. occidentale*) and tomato (*S. lycopersicum*) (tomato as internal reference standard). (b): Histogram showing suspected intraspecific variation in relative DNA content between two accessions (IMA-10 and BRO-26) of cashew (*A. occidentale*) with different phenotypic

characteristics (see Aliyu, 2010).

**10. Cashew cytology and cytometry** 

**10.1 Relative nuclear DNA content and genome size in cashew** 

intraspecific variation. However, all the sampled accessions showed consistent histogram peak position throughout the measurement, suggesting stability of the genome ploidy and the likelihood that the cashew, *A. occidentale* are predominantly diploids. Slight variation in genome size distribution tends to correlate with the history and/or source of origin/introduction and ecological adaptation of the accessions (Aliyu, 2010). Further studies of the cashew genome size more importantly with a more sensitive dye like Propidium Iodide (PI) may provide clearer information on the absolute genome size, likelihood of intraspecific variation and genome stability of this tropical tree species.

#### **10.2 Chromosome counts in cashew**

Aceto-orcein squashing of meristematic tissues (root tips and flower buds) of some genotypes has been analyzed for the number, structure and behavior of chromosome in cashew. Although there are divergent opinions on the ploidy status of the *occidentale* species, the study revealed that most genotypes had 42 chromosomes (Aliyu and Awopetu, 2007c) and could be tentatively described as 2*x* = 42 for a diploid species. And that cashew karyotypes are usually symmetric and less divergent, with mainly metacentric and submetacentrics pairs (Aliyu and Awopetu, 2007c). There were similarity in the morphology, number and behavior of the chromosomes among genotypes from different sources of introduction, which could imply common progenitors and slow mutation rate (chromosomal) among the existing *occidentale*. Another possible explanation for the chromosomal congruity can be drawn from the outcrossing nature of the plants, thereby enhances exchange of DNA materials between close and distant sympatric relatives and overlap in phenotypic features between lineages. Such phenotypic similarity between genotypes permits cross-compatibility (free gene exchange) with improved adaptation.
