The study was conducted in the Sudano-Sahelian zone of Cameroon (Figure 1). It is an agro-ecological zone between 8˚36" to 12˚54" north latitude, and 12˚30" to 15˚42" east longitude. It stretches from Adamaoua to the shores of Lake Chad over approximately 100,000 km², which is more than one-fifth of the area of Cameroon IRAD [18] .

Data were collected randomly from 61 breeders in 13 localities, including 7 localities in the far north (Guidiguis, Kalfou, Dziguilao, Mokolo, Tchatibali, Doukoula, and Touloum) and 6 localities in the northern region (Nyakira, Bibemi, Mayo -Lopé, Ngong, Guider and Mafa-Tchéboa). This information was collected

using a formal questionnaire developed for this purpose according to the approach established by the African Union Interafrican Bureau For Animal Resources (AU-IBAR [19] ). The Chi-square test was used to test the adequacy between the plumage and the different factors listed during the study in order to show the dependence between these random variables.

Table 1 shows that the activity is dominated by men (73.77%) who in their majority have not crossed the average course. They are farmers (54.10%), retired civil servants (18.04%), traders (14.75%) and teachers (4.91%). The activity is considered incidental and therefore little consideration is given to it. Almost all breeders, because of their level of education, say they know nothing about genetic diversity and nevertheless its importance. As shown by the work of Dongmo et al. [20] , Massawa et al. [12] , Teye and Adam [21] , the majority of operators in the guinea fowl sector have a very low level of education. Beyond this observation, Teye and Adam [21] reported that the activity is just secondary, which further complicates their interest in animal welfare.

The purpose of the data in Table 2 is to verify whether the guinea fowl breeders surveyed have the required knowledge on sustainable breeding that takes animal welfare into account. The results show that almost all farmers say they have never heard of sustainable agriculture or its merits. The notion of animal welfare,

♀: Female; ♂: Male; FSLC: First School Leaving Certificate; F4: Form 4; GCE: General Certificate of Education; OL: Ordinary Level; AL: Advanced Level; *: not significant.

Ts: Totals; Fr: Frequency; Abs: Abstention; Khi²: Khi² test; *: not significant.

translated into their respective dialect, made them laugh each time with the same reactions of astonishment. For some, we can at least speak of the well-being of bulls, horses and donkeys that pull the plow for agriculture. These animals, they say, must have chaff and cake frequently to provide them with the energy they need to pull the plow. On the other hand, poultry in general and guinea fowl in particular do not provide any effort in rural activities and could not benefit from any favor.

Despite this reasoning, many recognize that the environment for rearing guinea fowl seems to have changed: More and more guinea fowl are fleeing into the brushwood that punctuates the villages where they seem to find peace, animals that are increasingly becoming more vulnerable to pathologies which once only affected chickens, the temperature which seems to have increased making it difficult to channel and control the animals, the predators which seem to be increasingly present around the concessions, the low agricultural yield which makes it difficult to distribution of cereals (red millet, sorghum, corn, etc.) to animals, the virtual disappearance of termites which were the main source of protein for guinea fowls. In this regard, they accuse the abusive use of pesticides and fertilizers.

These results confirm those obtained by Teye and Adams [21] who reported in their work on the constraints to guinea fowl production in northern Ghana that the vast majority of guinea fowl breeders do not have knowledge of livestock management breeding. Nevertheless, Ioan et al. [16] confirm the remarks of the breeders because they showed that a significant loss of biodiversity occurs, deeply affecting nature and living things, causing changes in the natural habitat. These changes are due to agricultural production systems that overuse chemicals and deforestation that profoundly alter the habitat of poultry.

A total of 671 adult guinea fowls were counted according to their plumage among 61 breeders in the 13 localities (Table 3). Eight major plumages have been identified. The dominant color is pearl grey with a frequency of 38.89% followed by Violet (black color) (13.85%), royal purple (12.81%), Pie (9.53%), Bronze (7.30), Buff (6. 85%), Lavender (6.11%) and White (4.61%). The contingency test on the representation of the different colorations according to the localities gives the asymptomatic significance values (p-value) all above the significance threshold of 0.05.

Pg: Pearl grey; Rp: Royal purple; Buf: Buff; Wht: White; Pie: Pie; Vio: Violet; Brz: Bronze; Lav: Lavender; #: Number of breeder; Ts: Totals; Khi²: Khi² test; *: not significant; **: Significant.

These plumage colorations encountered are part of the twenty-two variations of colors recognized and reported by the Guinea Fowl International Association (2009). Issoufou [22] in Niger listed nine plumage colors (Brown, White, Lavender, Pearl Grey, Coral Blue, Brown Pied, Pied, Bronze Pie and Bronze). Pearl grey plumage is most common in all localities as observed by Issoufou [22] in two regions of Niger and by Agbolosu et al. [23] in Ghana. The GFIA [24] also reported that white albino varieties with pristine plumage are quite rare.

A total of 154 animals were lost by operators with a frequency of 22.95% (Table 4). Depending on the plumage, White comes first with a frequency of 51.61%, followed by Lavender (36.58%), Buff (36.95%), Royal Purple (20.93%), Bronze (20.40%), Pie (20.31%), pearl-grey (18.77%) and Violet (black) (17.20%). The locality of Bibemi comes first with 16 guinea fowl lost out of 45 for a percentage of 35.55% followed by the locality of Guidiguis with 17 guinea fowl lost out of 75 for a percentage of 22.6%.

The Khi² test of guinea fowl losses according to plumage and locality gives asymptomatic significance values (p-value) all above the significance level of 0.05. We conclude that there is no significant difference between animal losses and localities at the 5% threshold. The same factors responsible for the loss of guinea fowl are present in all localities. Dao [25] in Burkina Faso also presented in his work on the socioeconomic and phenotypic characterization of local guinea fowl populations in Burkina Faso that guinea fowl mortality is observed in all localities. Dahouda et al. [26] in Benin reported the same finding during their work on the constraints related to the breeding of guinea fowl. However, these authors did not show that these losses are more oriented toward guinea fowl with light plumage as evidenced by our results.

Table 5 shows that in the traditional way of rearing guinea fowl in a village environment where the animals are completely free, the main causes of the loss of animals by farmers are, among other things, predation, theft, flight of animals and pathologies. Predation is the major constraint with a frequency of 35.72%. The predators frequently cited by operators in guinea fowl farms are, among others, dogs, pigs, domestic cats, snakes, monitor lizards, and tiger cats. Furthermore, the results show that the guinea fowl most targeted by the above-mentioned predators are white plumage with a frequency of 56.25% followed by Buff plumage (41.17%) and Lavender plumage (40%). On the other hand, the predation is weak for the black (violet) plumage (25%), the royal purple (33.34) and pearl grey plumage (34.69). These results show that the brightness of the coloration seems to favor predation and therefore seems unfavorable for the genetic resources that carry these characters. Knowing that most predation takes place at night, guinea fowl with dark or barred plumage seem to benefit from the providence of nature.

Pg: Pearl grey; Rp: Royal purple; Buf: Buff; Wht: White; Pie: Pie; Vio: Violet; Brz: Brown; Lav: Lavender; #: Number of breeder; Ts: Totals; Khi²: Khi² test; *: not significant; **: Significant.

Pg: Pearl grey; Rp: Royal purple; Buf: Buff; Wht: White; Pie: Pie; Vio: Violet; Brz: Brown; Lav: Lavender; Ts: Totals; Khi²: Khi² test; *: not significant; **: Significant.

In addition, pearl grey guinea fowl qualified as wild type, royal purple, black and bronze guinea fowl seem to have a more developed wild instinct, which is evidenced by the frequency of recorded escapes. According to breeders, the escape of animals is favored by several factors such as:

- The number of livestock: it has a stabilizing effect on guinea fowl. The greater the number of livestock, the less there are guinea fowl escapes. As a result, small holders regularly lose animals to large numbers.

- The presence of a dog in the house: regardless of their plumage, guinea fowls are frightened by the sounds of dogs and tend to flee the house.

- The age of the guinea fowls: The oldest guinea fowls tend to be less frequent in the concessions.

- The stress generated by the brutal capture of one of theirs for sale or consumption.

The Khi² test shows that there is a significant difference between predation and bright plumage.

These results confirm those obtained by Gluckman and Mundy [27] who showed that the striped feather pattern helps birds blend into the environment. For Hill and McGraw [28] , the dark plumage facilitates camouflage. McAinsh et al. [29] reported in Zimbabwe that losses due to predation are the main constraint to guinea fowl rearing. In descending order, the predators reported by breeders are snakes, hawks, shrews, dogs, pigs, ducks, cats and lizards. Dahouda et al. [26] presented the same predators in Benin and Burkina Faso. However, none of his studies show a difference in predation between different animal colorations.

Table 6 shows that the choice of guinea fowl color for breeding depends on the criteria accepted by public opinion and which seem to find justification. Of all the guinea fowl breeders interviewed, 27.87% choose the color according to the docility of the animals in breeding and their resistance to infections. In addition, 24.6% say they choose animals according to their thermal resistance and for the remaining 19.67% of breeders, their choice of plumage is mainly guided by the prolificacy of laying. Thus, on the basis of these four criteria, we find that white, Buff and lavender guinea fowl seem more docile in breeding. Similarly, white and buff guinea fowl seem to resist heat stress better than other phenotypes. Since color plays a role in the absorption and reflection of incident solar radiation, birds with black phenotypic characteristics may be more susceptible to heat stress under intense solar radiation from the Sudano-Sahelian zone. Birds with light plumage (White, Buff and Lavender) on the other hand, may be more tolerant in the same conditions. This absorption and reflection of light rays by the plumage could also condition the docility of light-coloured animals which feel less threatened by thermal stress or, on the other hand, reinforce the wild instinct of animals with dark plumage which seem to escape for less stressful horizons.

Pg: Pearl grey; Rp: Royal purple; Buf: Buff; Wht: White; Pie: Pie; Vio: Violet; Brz: Brown; Lav: Lavender; Ts: Totals; Khi²: Khi² test; *: not significant; **: Significant.

On the other side, guinea fowls with dark plumage seem to be more sought after in breeding for their supposed prolificacy in laying eggs and their resistance to infections. Paradoxically, resistance to heat stress seems to have developed in guinea fowls with dark plumage, an ability to resist certain infections compared to light coloured guinea fowls. The Khi² test shows that there is a significant difference between the plumage and the docility of the animals on the one hand and the resistance to heat stress on the other hand.

These results confirm those obtained by Ayorinde et al. [30] who showed that plumage coloration and genetic variations seem to affect the performance of birds, in particular their productivity and their capacity for resistance under traditional agricultural conditions. They showed a significant difference in feed intake and conversion between white, grey, black and ash varieties. Beaumont and Chapuis [23] have also shown that man has always looked for animals that are not very sensitive to stress, because they are then more efficient and easier to raise. Conversely, Boko et al. [31] in the study on the actogram of local keets subjected to heat stress in Benin showed that temperature and humidity affect the performance of keets despite their resistance and thus, independently of their coloration. Although Boko et al. [32] and Sanfo et al. [33] showed that guinea fowl are in general less susceptible to infections, these authors did not show a correlation between guinea fowl plumage and their susceptibility to infections. If the resistance of guinea fowls with light plumage to heat seems logical because of the reflection of light by their plumage, other factors such as the docility of the animals, the laying ability and the resistance to infections such as expect our results must be the subject of in-depth studies to confirm or refute these results.

Table 7 shows that plumage influences the choice of guinea fowl. Among the selection criteria related to the plumage, we have the traditional rites, the color of the meat, the weight of the animals in adulthood and the offerings. It emerges from this table that 14.76% choose the color of the plumage for the traditional rites, 31.14% make the choice for the color of the meat, 27.87% for the weight and 26.23% for the offerings in sacred places or for important people.

Thus, White and Buff guinea fowl are preferentially requested for traditional rites with 30.7% and 25% respectively and for offerings. In addition, light guinea fowls (White, Buff and Lavender) urged them as their choice of meat, which is similar to that of chicken. Nevertheless, according to the breeders, the white guinea fowl have a reduced size compared to the others, which would partly explain their low frequency in the herds. On the other hand, dark guinea fowl seem to be more in demand for their weight: Pearl grey (42.5%), Royal purple (50%), Piebald (40%), Black (50%), and Bronze (25%).

While the choice of animals for criteria such as weight and meat quality seems objective and justifiable, the choice of colors for traditional rites and offerings is more subjective. The white color associated with purity and innocence seems to be more in demand for rites. Among the Mafa people of the far north of Cameroon, for example, a pair of white guinea fowl is sacrificed on the wedding night if, after verification, the young bride is still a virgin. This rite, long perpetrated and today in the process of disappearing, is a symbol of blessing, prolificacy and above all a recognition towards the young bride who is considered to have no defilement and therefore deserves the confidence of her husband for a life of happy couple. Among the Dagombas and Gonjas in northern Ghana, Teye and Adam [21] also reported that pure white guinea fowl is used for religious sacrifices and to perform certain funeral rites.

Pg: Pearl grey; Rp: Royal purple; Buf: Buff; Wht: White; Pie: Pie; Vio: Violet; Brz: Brown; Lav: Lavender; Ts: Totals; Khi²: Khi² test; *: not significant; **: Significant.

Contrary to the ancestral ritual among the Mafa people and the Dagombas and the Gonjas who have a preferential choice for the white guinea fowl, Anthony [34] mentioned that in the Afro-Brazilian community descending from the Yoruba (Nigeria and Benin), the sacrifice of any guinea fowl carries the transformative energy of destiny. This sacrifice of the guinea fowl in the ancestral tradition of this people appears to be decisive for the dynamic balance of the cults, the reproduction of the religious system based on initiation.

Moreover, these results confirm the work of Kgakole et al. [35] . These authors recently showed in their work on the growth performance of lavender, pearl grey, royal purple and white guinea fowl in Botswana that royal purple and pearl grey guinea fowl had the highest weights in adulthood, unlike white guinea fowl. Sanfo et al. [17] showed that native guinea fowl varieties have different asymptotic weights depending on the phenotype. Black and pearl grey guinea fowl weigh more than lilac and white guinea fowl. Although Khan et al. [36] showed that the nutritional and organoleptic qualities of guinea fowl meat can vary according to genetics, gender, age and rearing conditions, they did not specify that these organoleptic qualities are in favor of the white guinea fowl as evidenced by our results. Furthermore, Houndonougbo et al. [37] showed that guinea fowls of different colorations and of the same age reared under the same conditions showed a variation in weight. This observed weight difference could be explained by genetic differences between the ecotype.

Table 8 shows that the dominant/dominated relationship in poultry in general and in particular in guinea fowl depends above all on the number of intra-specific individuals. The number of guinea fowl with the same plumage coloration seems to have a group effect on the other plumage on the one hand and on all poultry on the other hand. The pearly grey, royal purple, black and pied guinea fowls in large numbers in most of the flocks seem to have a dominance in terms of food, mating or strategic positions in the henhouse or on the tree branches. This

Pg: Pearl grey; Rp: Royal purple; Buf: Buff; Wht: White; Pie: Pie; Vio: Violet; Brz: Brown; Lav: Lavender; Ts: Totals; Khi²: Khi² test; *: not significant; **: Significant; +: dominant; -: dominated; ±: more and less dominant; /: not observed; ♂: male; ♀: female.

dominant-dominated relationship is even more visible when food resources are even more limited. In addition to the group effect, the stay of the animal in the environment is essential to assert itself. Thus, according to breeders, new animals are easily dominated regardless of their plumage. These results could partly explain the flight of animals for livestock with large numbers because animals with small numbers according to their plumage seem to seek refuge in livestock where they are well represented. Our results are in line with those of the work of Jens et al. [38] who showed in their manual on the technique of poultry production in a village setting that dominance relationships are frequently observed within animals. For Julien [39] , long-term dominance relationships have a negative impact on the well-being of the animal genetic resource, hence the need to improve the framework of animal genetic resources. Although no work has been carried out on the dominance relationship between guinea fowl on farm, these results obtained based on the observations and experience of guinea fowl breeders must be studied in other studies with new approaches to confirm or invalidate these results because dominance relationships are a constraints to breeding.