Livestock Research for Rural Development 16 (9) 2004

Citation of this paper

Productivity and reproductive performance of the free-range local domestic fowl ecotypes in Tanzania


P L M Msoffe, M M A Mtambo, U M Minga*, J E Olsen**, H R Juul-Madsen***, P S Gwakisa *, S K Mutayoba**** and A M Katule*****


Department of Veterinary Medicine and Public Health, P.O. Box 3021, SUA, Morogoro
Department of Veterinary Microbiology and Parasitology, P. O. Box 3019, SUA, Morogoro
**Department of Veterinary Microbiology, The Royal Veterinary and Agricultural University,
Stigbøjlen 4, 1870 Frederiksberg, C., Denmark.
***Danish Institute of Agricultural Sciences, Research Centre Foulum, Tjele, Denmark.
Department of Animal Science and Prodcution, P. O. Box 3004, Morogoro, Tanzania
*****Tumaini University, Iringa, Tanzania.


A study was conducted to assess the productivity and reproductive performance of seven free-range local domestic fowl ecotypes in Tanzania named Ching'wekwe, Mbeya, Morogoro-medium, N'zenzegere, Pemba, Tanga and Unguja. Average weekly weight measurements and growth rates were evaluated for each ecotype and sex as was egg weight, fertility and hatchability.

Significant differences existed between ecotypes in all the five parameters studied. Ching'wekwe showed consistently low mean weekly weights, daily growth rate and mean egg weight contrary to Morogoro-medium and Tanga ecotypes. Egg fertility was low with only N'zenzegere and Unguja ecotypes exceeding 75%. Hatchability was also low ranging from 55% (Ching'wekwe and Morogoro-medium) to 74% (Pemba).

It was concluded that genetic and phenotypic diversity exists in the local domestic fowl ecotypes of Tanzania. The diversity constitutes a valuable resource for use in breeding programmes for improvement of the health and productivity of the local domestic fowls and in designing proper conservation strategies. Further studies are required to identify genetic markers associated with productivity and disease resistance within the local domestic fowl ecotypes. In depth studies on the performance of the Tanzanian medium ecotypes (Morogoro-medium and Tanga) is required to ascertain their suitability for promotion throughout the country.

Key words: Fertility, growth rate, hatchability, local domestic fowl ecotypes



Tanzania is endowed with a rich poultry genetic resource dominated by the free-ranging local domestic fowls (FRLDF). The FRLDF is a pool of diverse genetic resource, which is mainly kept in the rural areas of the developing world. The diversity of the FRLDF as expressed by phenotypes includes; adult body weight, egg weight, reproduction performance and immune responses to various antigens (Gwakisa et al 1994; Guèye 1998; Msoffe et al 2001). Studies have indicated that FRLDF in Tanzania exists in different phenotypic groups (termed ecotypes), which portray different productivity potential (Msoffe et al 2001). Genetic characterisation of the ecotypes strongly suggests that the ecotypes may represent different genetic entities (Msoffe et al submitted to Animal Genetics).


The productivity of the FRLDF in Tanzania and elsewhere has always been considered to be low mostly because of their low genetic potential (Katule 1988; Minga et al 1989; Kitalyi 1998; Mwalusanya 1998). Most of the previous efforts to improve the productivity of the FRLDF were therefore directed towards improvement of the genetic potential through crossbreeding with exotic breeds (Katule 1990; Kitalyi 1998). It is unfortunate that these efforts have not borne expected fruits and the productivity remains low. This in part may be compounded by the fact that very little is known about the genetic make-up of the FRLDF. However, more recent studies have reported wide ranges in production parameters among the FRLDF ecotypes indicating that the potential for high productivity is present but it has not been tapped through selective breeding (Minga et al 1996; Msoffe et al 1998).


This study was aimed at assessing the productivity and the reproductive performance of the FRLDF ecotypes under the intensive management system.


Materials and methods

Parental population and management


The parental free-range local domestic fowl ecotypes population used in this study were purchased from villages in Tanzania mainland and islands. The composition of the parental population including origin, number and brief description is given in Table 1.

Table 1. Composition of the parental free-range local chicken ecotypes




Brief description


N = 21

M = 4, F = 17

Chakwale, Kilosa district, Morogoro

Very short and compact chickens multi-variety plume, single combed


N = 20

M = 4, F = 16

Itumba, Ileje district, Mbeya

Medium sized birds, black bluish plume, single combed


N = 23

M = 5, F = 18

Msolwa, Kilosa district, Morogoro

Medium sized birds, multi-variety plume, single combed


N = 18

M = 3, F = 15

Mtamba, Mvomero district, Morogoro

Medium sized birds, multi-variety frizzled plume, single combed


N = 25

M = 5, F= 20

Southern region, Pemba

Small sized birds, multi-variety plume, mixed comb types


N = 18

M = 4, F = 14

Handeni, Handeni district, Tanga

Medium sized birds, multi-variety plume, single combed


N = 13

M = 3, F = 10

Western and Southern regions, Zanzibar

Small sized birds, multi-variety plume, mixed combed types

* Morogoro-medium ecotype; N = sample size; M = Males; F = Females


Hens bought were those that had at least hatched once but not more than three times, while actively breeding cocks were bought. The domestic fowls were transported to Sokoine University of Agriculture where they were maintained on floor pens in a deep litter system. Domestic fowls of the same ecotype were placed in the same floor pen each having two cocks and all the purchased hens. Throughout the study period all the domestic fowls were maintained on a commercial feed preparation (Layers mash). Both feed and water were given ad libitum. A comprehensive disease management strategy was followed to ensure the well being of the domestic fowls.

Productivity and reproductive performance of the local domestic fowl ecotypes


The productivity and reproductive performance of the local domestic fowl ecotypes were assessed by weekly live body weight measurements, daily growth rate, egg weights, the fertility and the hatchability of the incubated eggs.

Weekly weight measurements and growth rate for domestic fowls up to 20 weeks of age


Individual live weights were taken from hatching until 20 weeks of age using a sensitive digital weighing balance (Tefal ® France). A total of 194 females and 189 males were included. Sex differentiations were done retrospectively and mean weekly weight measurements were determined separately for each gender and ecotype.


The daily growth rate was calculated as:

Daily growth rate = W2 - W1 /28

Where W1 was the initial weight in the month; W2 was the final weight in the month and 28 was the number of days in the month.

Reproductive performance of the local domestic fowl ecotypes


The reproductive performance was assessed separately for each ecotype by recording the weights of all eggs produced, and the fertility and hatchability of the eggs set for incubation over one calendar year. Egg weights were determined after every three days and afterwards the average for each three-day measurement was calculated. For each ecotype, 100 three-day average entries were used to generate the data for analysis.


A total of 4072 eggs was incubated in 19 batches to determine the fertility and hatchability of the local domestic fowl ecotypes. Candling of the incubated eggs was done twice at 7 and 14 days after incubation. The percentage fertility of the eggs was calculated as follows:

% Fertility = (Te - Ie / Te) x 100


Where Te was the total number of eggs incubated and Ie the total number of infertile eggs.


The percentage hatchability of the eggs was calculated as follows:

% Hatchability = (He /Ve) x 100

Where He was the total number of hatched eggs and Ve the total number of viable eggs (after the first candling).

Statistical analysis


All the data on weekly weight measurements, daily growth rates, egg weights, fertility and hatchability were analysed using  Statistix ® analytical package (v.4.1). The data were subjected to descriptive statistics that gave the means, standard errors of the means, medians as well as the range of the data. One-way analysis of variance (ANOVA) was used and the means were compared by the least significant difference method (LSD) at 5% level of significance. Graphical presentations were made using the Microsoft Excel® spreadsheet programme.



Weekly weight measurements


Tables 2 and 3 show the mean weekly weight measurements for the local domestic fowls separated by ecotype and sex including the overall values for each sex category. Wide ranges in weekly weight measurements were observed in the overall values and males attained an average of 1000g bodyweight after 20 weeks of age; such an average was never attained in females.


Regarding hatch weight in females, the Ching'wekwe and Mbeya ecotypes showed the lowest and the highest mean values respectively. Comparing the means, it was seen that except for N'zenzegere and Unguja ecotypes all other means were significantly different from each other and from the extreme values (P<0.05). Throughout the study, the Ching'wekwe ecotype showed significantly low mean weekly weight measurements except in the fourth and fifth month when Pemba ecotype had a significantly low mean. Morogoro-medium ecotypes showed consistent and significantly higher (P<0.05) weekly measurements from the 4th week although Mbeya and Tanga ecotypes shared highest mean at different occasions (Table 2).

Table 2. Mean weekly weight measurements up to 20 weeks for hens (Mean ± SE)


Hatch weight

Week 4

Week 8

Week 12

Week 16

Week 20

Ching’wekwe (N = 33)

18.8 ± 0.5a

55.3 ± 0.7a

163.6 ± 5.7a

294 ± 4.7a

443.7 ± 14a

592.3 ± 7.8a

N’zenzegere (N = 33)

27.7 ± 0.3d

83.4 ± 1.9b

234 ± 4.3bc

382.4 ± 6b

557.5 ± 10c

731.5 ± 12.9bc

Mbeya (N = 24)

30.8 ± 0.4f

113.9 ± 2.8d

246.9 ± 11.9cd

440.8 ± 17.3c

592.1 ± 22.6cd

772 ± 19.2cd

Morogoro-medium (N = 28)

28.7 ± 0.2e

109.3 ± 2.2d

298.5 ± 8.4f

456.7 ± 15c

648.4 ± 19.9de

853.6 ± 22.3e

Pemba (N = 30)

24.7 ± 0.3c

93.6 ± 1.5c

255.4 ± 7.8de

310.7 ± 6.4a

458 ± 13a

754.7 ± 23.2c

Tanga (N = 23)

23.4 ± 0.4b

96.8 ± 4.8c

277.6 ± 12ef

441.4 ± 24.2c

615.1 ± 22.1de

829 ± 29.8de

Unguja (N = 23)

27.1 ± 0.3d

81.2 ± 1.9b

216.4 ± 4.8b

381.1 ± 15.4b

506.3 ± 12.3b

698.5 ± 16.3b

Overall (N = 194)

Range (Min – Max)

25.7 ± 0.3

(14 – 34)

89 ± 1.6

(47 – 134)

118 – 358

(118 – 358)

381.1 ± 6.5

(205 – 568)

540.9 ± 8.1

(316 – 776)

741.7 ± 9.2

(518 – 1036)

Means in column with common superscripts are not significantly different from each other (One way analysis of variance, P<0.05)

Mean hatch weights for males showed the Ching'wekwe and the Mbeya ecotypes to have the lowest and the highest values respectively. Comparison of means showed these two extreme values were significantly different from each other and to the means of all other ecotypes (P<0.05). A consistently significant lower weight measurement was seen in the Ching'wekwe ecotype throughout the study. Morogoro-medium and Tanga ecotypes showed consistently significant high weight measurements from 8th week onwards (Table 3).


Table 3. Mean weekly weight measurements up to 20 weeks for males (Mean ± SE)


Hatch weight, g

Week 4, g

Week 8, g

Week 12, g

Week 16, g

Week 20, g

Ching’wekwe (N = 21)

23.3 ± 0.2a

61.6 ± 0.4a

259.9 ± 4.3a

462.4 ± 6.3a

650.9 ± 15a

780.6 ± 17.5a

N’zenzegere (N= 21)

30.1 ±0.2c

119.7 ± 3.7b

339.1 ± 13.2b

571.3 ± 20.9b

844.4 ± 27b

1066.4 ± 18.3b

Mbeya (N = 33)

35.2 ± 0.3e

186.4 ± 6.1d

379.6 ± 3c

604.4 ± 10.7b

824.2 ±19b

1047.5 ± 22.7bc

Morogoro-medium (N = 28)

31.7 ± 0.2d

154.1 ± 5.8c

397 ± 6.3c

645.2 ± 16.9c

957.3 ± 26.2c

1229.2 ± 32.5d

Pemba (N = 24)

27 ± 0.1b

148.9 ± 5.1c

374.6 ± 9.3c

473.9 ± 17.2a

695.8 ± 20.8a

1031.4 ± 13.1bc

Tanga (N = 31)

29.5 ± 0.4c

144.5 ± 2.4c

379.4 ± 6.8c

657.4 ± 17.4c

919.1 ± 10.3c

1216.7 ± 17d

Unguja (N = 31)

29.7 ± 0.2c

113 ± 6.5b

344.4 ± 10.9b

579.5 ± 12.9b

826.6 ± 22.5b

115.4 ± 34.4c

Overall (N = 189)

29.9 ± 0.3

136.6 ± 3.2

357.9 ± 4.2

579 ± 7.5

826.5 ± 10.5

1088.8 ± 13.3

Range (min – max)

(22 – 36)

(60 – 238)

(237 – 870)

(275 – 870)

(560 – 1234)

(698 – 1524)

Means in column with common superscripts are not significantly different from each other (One way analysis of variance) P<0.05

It is also seen from the results (Table 2 and 3) that the average weight measurements for Ching'wekwe ecotype was consistently lower than the overall mean values. N'zenzegere, Pemba and Unguja ecotypes showed mean weight measurements close to the overall means. The other two ecotypes (Mbeya and Morogoro-medium) had mean weight measurements consistently higher than the overall mean values.


Daily growth rate


Results for mean daily growth rates for the seven local domestic fowl ecotypes including the overall mean (mean for all domestic fowls regardless of ecotype) are presented in Figures 1 and 2. The results show that in females, the overall mean growth rate during the first month was 2 grams/day with a sharp increase of up to 5.6 g/d in the second month. A slight depression in growth occurred during the third month, but a positive trend (a less sharp increase) resumed for the fourth and fifth month. In male, the overall mean daily growth rate for the first month was 3.3 g/d increasing sharply to 7.4g/ during the second month. In contrast to the females, there was a plateau on the third month then a gentle increase in the fourth and fifth month.

Figure 1. Mean growth rate (g/day) for local chicken ecotypes in Tanzania (females)


It was observed that each ecotype had a decrease in mean daily growth rate at a certain point in the study period. This period was between the second and third month in Morogoro-medium, N'zenzegere, Pemba and Tanga ecotypes but was between the third and fourth month in the Mbeya and Unguja ecotype. Between the fourth and fifth month, Ching'wekwe and N'zenzegere ecotypes showed a small decrease in the mean daily growth rate. The lowest mean growth rates were recorded in the Ching'wekwe ecotype females during the first, second and fifth month of the experiment. Females from Pemba and Unguja ecotypes showed the lowest mean growth rates in the third and fourth months respectively. The highest daily mean growth rate was shared between Mbeya, Morogoro-medium, Pemba and Tanga ecotypes at different times in the study period. However, it was the Morogoro-medium ecotype that showed high daily growth rate on more occasions (three out of the five month).


The daily mean growth rate in males showed a more or less similar trend as in females with only a few disparities (Figure 2).

Figure 2. Mean growth rate (g/day) for local chicken ecotypes in Tanzania (males).


Comparison of means between the various local domestic fowl ecotypes over the five-month study period resulted in significantly different values (P<0.05). For instance, for the greater part of the five-month trial (three months for females; four months for male), daily mean growth rate for the Ching'wekwe ecotype was significantly lower (in both male and females) compared to the other ecotypes. None of the ecotypes showed a consistently significantly higher mean daily growth rate. Of interest is the Pemba ecotype, which after a significantly lower growth rate in the third month, showed a significantly higher growth rate starting from the fourth month. Morogoro-medium and Tanga ecotypes showed comparable means from the second to the fifth month in females and the first two months and the fifth month in males. N'zenzegere and Unguja ecotypes showed persistently similar mean growth rate in males and only differed significantly in the fourth month for females. The Zanzibar ecotypes (Pemba and Unguja) showed significantly different daily mean growth rates in all occasions except during the second month (male) and fourth month (both male and females).

Egg weights, fertility and hatchability for the local domestic fowl ecotypes


The overall mean egg weight for all domestic fowl ecotypes was 42.5 ± 0.6g (ranging from 20.8g to 55g). Overall egg fertility was 70% (15% to 100%) while overall mean hatchability was 62% (11% to 100%). The mean values were greatly influenced by the range within the different groups. For instance the mean egg weight of 42.5g with a range of 20.8g to 55g indicates that most of the ecotypes had their mean egg weight towards the right side of the mean. Similarly for fertility and hatchability which showed even wider ranges.


Table 4 shows the average egg weights, proportion mean fertility and proportion mean hatchability by ecotypes. The mean egg weight for Ching'wekwe ecotype was the lowest (37.2g) while that for Mbeya ecotype was the highest (49.3g). The other ecotypes had mean egg weights falling between these two extremes albeit with a tendency towards the lower extreme. When the means were compared, it was revealed that the mean egg weights for Ching'wekwe and Mbeya ecotypes were significantly different from each other and from the other ecotypes (P<0.05). The mean egg weights for N'zenzegere and Pemba were different from each other and from the other ecotypes. Mean egg weight for Morogoro-medium, Tanga and Unguja ecotypes were similar but significantly different from the rest of the ecotypes.

Table.4. Mean reproductive performance of the local chicken ecotypes


Proportion mean fertility (Range)

Proportion mean hatchability (Range)

Mean egg weight


0.7 ± 0.03bc

(0.4 – 1.0)

0.55 ± 0.04a

(0.2 – 0.9)

37.2 ± 0.3a


0.75 ± 0.04bc

(0.3 – 1.0)

0.57 ± 0.05a

(0.2 – 0.9)

41.1 ± 0.3b


0.69 ± 0.03bc

(0.5 – 0.9)

0.64 ± 0.04ab

(0.4 – 1.0)

49.3 ± 0.2e


0.64 ± 0.06a

(0.2 – 0.9)

0.55 ± 0.06a

(0.2 – 0.9)

42.6 ± 0.6d


0.64 ± 0.03a

(0.4 – 0.9)

0.74 ± 0.05b

(0.3 – 1.0)

41.9 ± 0.2c


0.68 ± 0.03bc

(0.4 – 0.8)

0.63 ± 0.06ab

(0.1 – 1.0)

43 ± 0.3d


0.8 ± 0.03c

(0.4 – 1.0)

0.66 ± 0.03ab

(0.4 – 0.9)

42.6 ± 0.3d

Means in column with common superscripts are not significantly different from each other (One way analysis of variance, P<0.05)

The mean fertility indicated that Morogoro-medium and Pemba ecotypes had the lowest mean value (0.64) while Unguja ecotype had the highest mean value (0.8). In comparing the means, it was seen that Morogoro-medium and Pemba ecotypes had significantly lower mean values compared to the other ecotypes (P<0.05). However, the high mean value for Unguja ecotype was not significantly different from those of Ching'wekwe, Mbeya, N'zenzegere and Tanga ecotypes. There was a great variation between ecotype ranges. For example, the range for fertility for Morogoro-medium ecotype was 0.2 to 0.9 (20 - 90%) while that for Mbeya ecotype was 0.5 to 0.9 (50 - 90%). Except for N'zenzegere ecotype (0.3), the lowest value for fertility was 0.4 (40%) and the highest value was 1.0 (100%).


The data for hatchability showed two groups with significantly different means. The group with the lowest mean included the Ching'wekwe, N'zenzegere and Morogoro-medium ecotypes and that with the highest mean encompassed the Pemba ecotype. The other ecotypes had means that were not significantly different from the two groups. Again there were great variations between ecotype ranges with recorded data being as low as 0.1 (10%).


Weekly weight measurements


Sexual dimorphism could be the reason for the weights at hatching being higher for male chicks than for females. Similar observations were made by Katule (1990) and Wilson et al 1987), showing sexual dimorphism on weight measurements from the fourth and 13 weeks of age respectively.


The overall mean weekly weight measurements showed that after 20 weeks of age females had an average weight below 1000g. Katule (1990) had earlier shown that at 16 weeks of age, the average juvenile body weights for the local domestic fowls were 732g and 946g for females and males respectively. In the current investigations the overall mean weekly body weight measurements at 16 weeks were 541g and 827g for females and males respectively. The probable reason for this discrepancy would be the fact that the domestic fowls used in the previous experiment were not selected based on the ecotype criterion hence domestic fowls from different ecotypes or crosses with exotic domestic fowls could have been included.


According to the results the Morogoro-medium ecotype appeared to perform better in weekly weight measurements compared to the other ecotypes. The Morogoro-medium domestic fowl ecotype is the ordinary local domestic fowl of Tanzania with the widest geographical distribution and very few unique distinctive features. The Tanga local domestic fowl ecotype also by and large shares this description and it is therefore not coincidental that the two ecotypes although from different eco-climatic regions shared most values on the phenotypic parameters. These results are in agreement with those by Adedokun and Sonaiya (2001), who showed that local domestic fowls from different agro-ecological zones might show similarities in their production parameters Katule (1990) demonstrated that different genetic groups of domestic fowls showed consistently significant different weight measurements over a period of up to 39 weeks. Safaloah (1998) has made similar observations regarding body weights at eight weeks of age where different genetic groups showed statistically significant mean values. In that study, the groups compared were the Malawi local domestic fowls, Starbro broilers and Black Australorp. It can therefore be assumed that, the local domestic fowl ecotypes that showed significant variations in the productivity parameters may belong to different genetic groups (breeds, strains or varieties).

Mean daily growth rates


The overall mean growth rates varied between sexes and between the times of measurements. The overall mean growth rate in the second month was higher compared to what was observed in earlier studies (Wilson et al 1987; Mwalusanya 1998). In that study Wilson et al (1987) observed daily growth rates in Malian local domestic fowls for the first 10 weeks of life to be 4g/day. Mwalusanya (1998), reported mean growth rates per day of 4.1 to 5.11g/day for females and 5.06 to 5.66g/day for males in the Tanzanian local domestic fowls from three ecological zones. The discrepancies observed in the current study and the other studies may be due to the difference in the experimental settings. While the current study was based on-station, the other two studies obtained data from field observations. Lack of control of various factors such as feeds (availability and quality), and disease problems under the field situation may be responsible for preventing the local domestic fowls from expressing their full genetic potential. However, the results from the current experiment are lower compared to a study in Malawi by Safaloah (1998), who reported mean daily growth rates of 10.7g/day at eight weeks of age for the Malawian local domestic fowls. Nevertheless, as earlier stated, the genetic status of the Malawian local domestic fowl was not known.


Marked variations were observed when the mean daily growth rate was compared between ecotypes further suggesting the possibilities of ecotypes being distinct genetic groups. The absence of clear-cut and consistent differences does not mean that the domestic fowls are of a homogenous genetic group. Safalouh (1998), showed that the mean growth rate at eight weeks of age was not significantly different between the Black Australorp and the Malawian local domestic fowls.

Mean egg weight, fertility and hatchability


The overall mean egg weight in the current experiment was 42.5g. This was higher compared to 36.8g from Nigerian local domestic fowls (Adedokun and Sonaiya 2001), 37g from Bangladesh desi domestic fowls (Barua and Yoshimura 1997), 38.2g from Tanzanian local domestic fowls (Katule 1990) and 34.4g from Malian local domestic fowls (Wilson et al 1987). There are other studies that have reported higher mean egg weights compared to the current experiment. Mwalusanya (1998) reported overall mean egg weight of 43.6g from local domestic fowls from three ecological zones of Tanzania. In a different study, Zaza (1992) reported mean egg weight of 48g in Dandrawi domestic fowl (a local Egyptian domestic fowl).


It was observed that except for the Mbeya ecotype, mean egg weights for the other local domestic fowl ecotypes were lower compared to those reported in an earlier study for Kuchi and Singamagazi; two relatively large local ecotypes from Tanzania (Msoffe et al 2001). Katule (1990) had earlier shown that different genetic groups of domestic fowls laid eggs of significantly different weights. Such distinctions could not be seen when local domestic fowls from different agro-ecological zones of Nigeria (Adedokun and Sonaiya 2001) or different climatic zones of Tanzania (Mwalusanya 1998) were studied. This gives a strong indication that the differences observed in the current study are genetical in nature.


An earlier study by Msoffe et al (2001) reported mean egg weights for Ching'wekwe, Mbeya and Morogoro-medium ecotypes that were lower compared to values from the same ecotypes in the current experiment. The reason for this difference may be due to the fact that while only fifty eggs (taken once) were used to calculate the mean egg weight in the previous study, the current mean was an average of eggs produced over one calendar year. Therefore, the current data represent a more realistic picture on this parameter.


The overall fertility (70%) reported in the current experiment fell short of the figure (95%) on the same parameter reported by Wilson (1979). It is possible that the difference in the two experiments was attributed to the differences in the experimental settings. The low fertility in the on-station experiment was probably caused by confinement stress that prevented the male from expressing their optimal reproductive performance. There is also a possibility that there were some deficiencies in the commercial feeds that was given to the domestic fowls since no attempt was made to evaluate the nutritional status of the feeds. Another reason for the low fertility in the current experiment might be the fact that eggs had to be stored for up to one week prior to the incubation. Although all the necessary storage precautions were taken, there was still a chance that fertility was lost during storage especially due to diurnal temperature variations. Between ecotypes, fertility varied significantly possibly due to differences in the inherent ability of each ecotype to cope with the factors mentioned above.


The hatchability percentage obtained in the current experiment was rather low (overall hatchability of 62%) compared to some previous studies. For instance Wilson (1979) observed mean hatchability values of 90% in the Sudanese local domestic fowls. Barua and Yoshimura (1997) reported hatchability of 75% on the local Bangladesh domestic fowls. Similarly Mwalusanya (1998) reported hatchability in the free-range local domestic fowls of Tanzania to be over 80%. The overall hatchability in the current experiment was probably affected by the frequent electric power cuts that impeded the optimal performance of the egg incubator. Furthermore, it is possible that some chicks might have required some assistance during hatching (known to be provided by the brooding hen). Individual ecotypes showed a wide variation in hatchability ranging from 55% (Ching'wekwe and Morogoro-medium) to 74% (Pemba ecotype). Since eggs from all the local domestic fowl ecotypes were subjected to similar conditions (prior to and during incubation) then within the problems stated above the values obtained here may represent genetic variations between the different ecotypes.


The productivity and reproductive parameters assessed in the current study have suggested the presence of genetic variations within the local domestic fowl ecotypes. This is particularly supported by other studies that used only the ecological or climatic region as the criterion for differentiating the local domestic fowls and found the means of the different phenotypic parameters to be insignificant (Mwalusanya 1998; Adedokun and Sonaiya 2001). It is therefore more appropriate and meaningful to describe the local domestic fowl ecotypes based not only on their geographical origin, but also using observable physical parameters such as size of the adult bird, shape of the comb or the plumage characteristics (Msoffe et al 2001). This would increase the repeatability between experiments done at different intervals and will reduce the between laboratory variations. Finally it would be possible to standardise these phenotypic characters within countries or regions and come up with some criteria for characterisation of the local domestic fowls into breeds.


Further studies involving more parameters such as age at sexual maturity, number of eggs per clutch and per year, nutritional composition of the eggs, eggshell strength and coloration will be necessary before the ecotype status is elevated to breed. In the meantime, the ecotype status can be useful in planning for future improvement programmes through selective breeding within and between ecotypes.



The technical support of Mr. Maulid Mdaki is appreciated. This study was supported by DANIDA under the ENRECA project "Improvement of Health and Productivity of Rural Chickens in Africa". The authors are grateful for the support.



Adedokun S A and Sonaiya E B 2001 Comparison of the performance of Nigerian indigenous chickens from three agro-ecological zones. Livestock Research for Rural Development, 13:


Barua A and Yoshimura Y 1997 Rural poultry keeping in Bangladesh. World's Poultry Science Journal 53, 387-394.


Guèye E F 1998 Village egg and fowl meat production in Africa. World's Poultry Science Journal 54, 73-86.


Gwakisa P S, Katule A M and Rugaimukamu E A 1994 Divergent immune responses to Newcastle disease virus vaccination of rural chicken in Tanzania. Tanzania Veterinary Journal 43, 171-176.


Katule A M 1988 Cross breeding as a tool for genetic improvement of chickens in less optimum environments. In: Proceedings of the 15th Scientific Conference of the Tanzania Society of Animal Production, 27-30 September, Arusha, Tanzania, pp 86-111.


Katule A M 1990 Studies on the prospects of improving the performance of local chicken populations in Tanzania by cross breeding. Unpublished Thesis for Award of PhD Degree at Sokoine University of Agriculture.


Kitalyi A J 1998 Village chicken production systems in rural Africa: Household food security and gender issues. FAO Animal Production and Health paper 142 Rome pp 81.


Minga U M, Katule A M, Maeda T and Musasa J 1989 Potential of the traditional chicken Industry in Tanzania. In: Proceedings of the 7th Tanzania Veterinary Association Scientific Conference, Arusha, Tanzania, pp 207-215.


Minga U M, Katule A M, Maeda T and Mwanjala T 1996 The rural chicken industry in Tanzania: does it make economic sense? Tanzania Veterinary Journal, 16 (supplement 2), 25-28.


Msoffe P L M, Minga U M, Olsen J E, Yongolo M G S, Juul-Madsen H R, Gwakisa P S and Mtambo M M A 2001 Phenotypes including immunocompetence in scavenging local chicken ecotypes of Tanzania. Tropical Animal Health and Production 33, 341-354.


Msoffe P L M, Mtambo M M A, Minga U M, Gwakisa P S, Katule A M, Mutayoba S K, Mdegela R H, Mwalusanya N A, Yongolo MG S and Olsen J E 1998 Scavenging local chickens in Tanzania: Is there room for improvement in productivity? Tanzania Veterinary Journal, 18 (Supplement 4) 117-126.


Mwalusanya N A 1998 Productivity and nutritional status of local chickens under village management conditions. Unpublished Dissertation for Award of MVSc. Degree at the Royal Veterinary and Agriculture University, Copenhagen, Denmark. Pp 56.


Safaloah A C L 1998 Response of the Malawi local chickens to commercial feed up to eight weeks of age. Bulletin of Animal Health and Production in Africa, 46, 245-249.


Wilson R T 1979 Studies on the livestock of Southern Darfur, Sudan. VII Production of poultry under simulated traditional conditions. Tropical Animal Health and Production, 11, 143-150.


Wilson R T, Traore A, Traore A, Kuit H G and Slingerland M 1987 Livestock production in central Mali: Reproduction, growth and mortality of domestic fowl under traditional management. Tropical Animal Health and Production 19, 229-236.


Zaza G H M 1992 Poultry and rural development in Egypt. In: Proceedings of the Introductory Seminar on Poultry Development Policy, Barnveld, The Netherlands pp 115-118.

Received 21 June 2004; Accepted 5 July 2004

Go to top