Standard plate count was determined using standard plate count agar. One mL of raw milk sample was added into a sterile test tube containing 9 ml of sterile peptone water. After thoroughly mixing, the suspension was serially diluted up to 10⁻¹¹ and duplicate samples from the appropriate dilution (1 ml) was pour-plated using a 15 - 20 ml of cooled but still molten standard plate count agar solution and mixed thoroughly. The resulting plates were allowed to solidify and then incubated at 32˚C for 48 hours [16]. The plates with colonies ranging from 30 to 300 colony forming units (cfu) mL⁻¹ were selected for determination of standard plate count [16]. Standard plate count was determined as the total number of cfu per milliliter of milk sample was calculated using the formula provided by FDA [17].

N = ∑ C [ ( 1 × n 1 ) + ( 0.1 × n 2 ) ] d

where, N = Number of colonies per ml of milk sample

∑C = Sum of all colonies on plates counted

n1 = Number of plates used in lowest dilution counted

n2 = Number of plates used in highest dilution counted

d = dilution factor of the lowest dilution used.

TCC was determined using sterile violet red bile agar (VRBA). One ml of raw milk sample was added into a sterile test tube containing 9 ml of sterile peptone water. After thoroughly mixing, the suspension was serially diluted up to 10⁻⁹ and duplicate samples (1 ml) were pour-plated using a sterile 15 - 20 ml VRBA. After thoroughly mixing, the resulting plates were allowed to solidify and then incubated at 32˚C for 24 hours [18]. After incubation, typical dark red or purplish-red colonies appearing on the plates were counted as coliforms. For confirmatory test, five to ten typical colonies from each plate will be transferred into tubes containing 2% Brilliant Green Lactose Bile Broth and incubated at 37˚C for 48 hours [16]. Growth and gas production within incubation period was considered as sufficient evidence for the presence of coliforms [16]. Plates with 15 to 150 cfu·ml⁻¹ were used [19] for determining total coliform counts using the formula provided by IDF [20].

Yeast count (YC) and mold count (MC) were determined using sterile Potato Dextrose Agar (PDA). One ml of raw milk sample was added into a sterile test tube containing 9 ml of sterile peptone water. After thoroughly mixing, the suspension will be serially diluted up to 10⁻⁷ and duplicate samples of 0.1 ml were spread-plated on pre-dried surfaces of media containing PDA (Oxoid, UK). The plates were then incubated at 25˚C for 5 days [16]. Creamy to white/gray colonies were counted as yeasts whereas, filamentous (fuzzy) colonies of various colors (yellow, green, light brown) will be counted as molds [21] When difficulties were faced to differentiate some colonies whether they are yeast or mold, a microscopic examination using the oil immersion objective was carried out to identify whether the cells in the colonies are unicellular or multi-cellular. Plates with 10 to 150 colonies were used for determining yeast and mold counts using the formula provided by IDF [20].

Production system had significant (P < 0.01) effect on standard plate count (SPC). Thus, the average SPC for raw camel milk samples collected from the udder of milking camels in pastoral production system (5.86 ± 0.18 log10 cfu·ml⁻¹) was significantly higher (P < 0.001) than in agro-pastoral production system (4.85 ± 0.20 log10 cfu·ml⁻¹) (Table 1). Moreover, the mean SPC for raw milk samples collected from the storage equipments in pastoral production system (7.47 ± 0.13 log10 cfu·ml⁻¹) was significantly higher (P < 0.05) than in agro-pastoral production system (5.41 ± 0.21 log10 cfu·ml⁻¹) (Table 1). Such variation in between pastoral and agro-pastoral production system may be due to the difference in hygienic practices performed during milk production and postharvest handling. Moreover, such difference may be due to the variation in ambient temperature between two production systems. It has been reported that, the temperature of food including milk handling area highly determines the microbial count of food as temperature determines the microbial activities like multiplication of microorganisms in food like milk [23] ‎ [24].

n = number of samples taken; significance: *P < 0.05, ** P < 0.01, ***P < 0.001, SE. = Standard error.

Moreover, milk source had significant (P < 0.01) effect on SPC of raw camel milk sampled in the study area. Thus, the mean SPC for raw camel milk samples collected from the udder (5.86 ± 0.18 log10 cfu·ml⁻¹) were significantly (P < 0.001) lower than that for samples collected from the equipment of producers (7.47 ± 0.13 log10 cfu·ml⁻¹) in pastoral production system (Table 1). Moreover, in agro-pastoral production system, the mean SPC for raw camel milk samples collected from the udder (4.85 ± 0.20 log10 cfu·ml⁻¹) were significantly (P < 0.001) lower than samples collected from the equipment of producers (5.97 ± 0.22 log10 cfu·ml⁻¹) (Table 1). This might be due to further contamination of the milk during production and postharvest handling.

The mean SPC of raw milk samples collected from the udder of milking camel in the study area was 5.35 ± 0.19 log10 cfu·ml⁻¹ (Table 1). The finding is higher than 4.20 log10 cfu·ml⁻¹ reported by Tsegelem et al. [25] for milk samples collected from the udder of milking camel in Fafan zone. However, the finding was lower than the value (5.62 × 10¹ cfu·ml⁻¹) reported for raw camel milk samples collected from the udder of milking camel in Afar zone, Ethiopia [26]. The difference may be due to the difference in hygienic practices (like udder, hand, milking equipment washing and other practices) performed among milk producer households in different location.

The mean SPC of raw milk samples collected from the equipment of producers in the study area was 6.72 log10 cfu·ml⁻¹ (Table 1). The finding is higher than 4.80 log10 cfu·ml⁻¹ reported by Tsegelem et al. [25] for milk samples collected from the equipment of camel milk producers in Fafan zone. However, it was lower than the value (92.25 × 10⁹ cfu·ml⁻¹) reported for raw milk samples collected from the equipment of milking camel in Afar zone [26]. The difference may be due to the difference in hygienic practices followed during milk production and postharvest handling. The hygienic practices performed during milk production and postharvest handling may varies according to the production system, adapted practices, level of awareness, and availability of resources [27].

The mean SPC for raw camel milk samples collected from both in udder (5.35 ± 0.15 log10 cfu·ml⁻¹) and milk equipment (6.72 ± 0.15 log10 cfu·ml⁻¹) in the study area are higher than the upper acceptable limit (5 log10 cfu·ml⁻¹ of raw milk) given by Marshall [28] and O’Connor [29]. This may be due to poor storage temperature, long storage period after milking, health and hygiene of the camel, environment where milking is done as well as procedures used in cleaning and sanitizing the milking and storage equipment.

Similar to SPC, the production system significantly (P < 0.01) affect total coliform count (TCC) of raw camel milk samples collected both from udder and milk equipment. Thus, the mean TCC for raw milk samples collected from the udder of milking camels in pastoral production system (3.14 ± 0.18 log10 cfu·ml⁻¹) was significantly higher (P < 0.01) than in agro-pastoral production system (2.03 ± 0.15 log10 cfu·ml⁻¹) (Table 2). Moreover, raw milk samples collected from the

n = number of samples taken; significance: **P < 0.01, ***P < 0.001; SE. = Standard error.

storage equipments in the pastoral production system (5.21 ± 0.20 log10 cfu·ml⁻¹) was significantly higher (P < 0.01) than in agro-pastoral production system (4.20 ± 0.24 log10 cfu·ml⁻¹). Such significant difference in between two production system may be due to the difference in awareness on hygienic practices (like cleaning of utensils, washing milker’s hands, personal hygiene, quality of water used for hygienic practices, washing the udder, use of individual towels) performed during milk production and postharvest handling. Moreover, such differences might be due to the variation in health and hygiene of milking camel between two production systems.

Milk source had significant (P < 0.01) effect on TCC; and thus, the mean TCC for raw camel milk samples collected from milk equipment was significantly (P < 0.001) higher than that for udder milk samples (Table 2). This might be due to further contamination of the milk during storage and transportation. Moreover, lack of cooling system and longer time elapsed during milk storage and transportation in the study area could contribute to such increasing trends.

The mean TCC (2.59 ± 0.16 log10 cfu·ml⁻¹) for raw milk samples collected from the udders (Table 2) was lower than the value (4.4 log10 cfu·ml⁻¹) reported for raw milk samples collected from the udder of milking camel in Ab’aladestrict [26]. This variation may be due to herd hygiene, water sources, and sanitation for milking practices. However, it was comparable with the value (2.83 log10 cfu·ml⁻¹) reported for raw milk samples collected from the udder of milking camel in Switzerland [30].

The mean TCC (4.71 ± 0.23 log10 cfu·ml⁻¹) for milk samples collected from milk handling equipment of producers (Table 2) was higher than the value (3.84 log10 cfu·ml⁻¹) reported for raw milk samples collected from milk handling equipment of producers in Afar [31]. However, it was lower than the value (6.75 log10 cfu·ml⁻¹) reported by Morocco [32] and Benyagoub et al. [33] for raw milk samples collected from milk handling equipment in south west Algeria (6.75 log10 cfu·ml⁻¹). Such variations could be due to the differences in herd health management practices at farm, hygiene of milking procedures, type and quality of milk equipment used, personal hygiene of milk producers, udder heath, quality of water used for hygienic practices and others.

The upper acceptable limit of TCC for raw milk is 2.18 log10 cfu·ml⁻¹ [34]. However, the overall mean counts for raw camel milk samples collected both from the udder (2.59 ± 0.16 log10 cfu·ml⁻¹) and milk equipment (4.71 ± 0.16 log10 cfu·ml⁻¹) in the study area exceeded the upper acceptable limit. This might be due to poor herd/farmhygiene and health care management practices performed by smallholder milk producers. Gamal et al. [35] indicated that poor hygienic practices during milk production and postharvest handlings enable the milk to have high microbial counts than the upper acceptable limit and make the milk unsafe for its intended use.

The mean YMC of raw milk samples were significantly (P < 0.01) influenced by milk source and production system interaction (Table 3). Thus, raw camel milk sample collected from pastoral production system had significantly higher mean YMC than samples collected from agro-pastoral production system for both samples collected from the udder and milk equipment. Moreover, raw milk samples collected from milk equipment had significantly higher mean YMC than samples collected from udder of milking camel in both production systems.

n = number of samples taken; significance: **P < 0.01, ***P < 0.001, SE. = Standard error.

The average yeast and mould count (YMC) for raw milk samples collected from the udder of milking camels in pastoral production system (0.95 ± 0.12 log10 cfu·ml⁻¹) was significantly higher (P < 0.05) than in agro-pastoral production system (0.68 ± 0.10 log10 cfu·ml⁻¹) (Table 3). Moreover, raw milk samples collected from milk equipment in the pastoral production system (2.16 ± 0.20 log10 cfu·ml⁻¹) was significantly higher (P < 0.05) than in agro-pastoral production system (1.05 ± 0.23 log10 cfu·ml⁻¹). This may be due to variation in milk storage temperature, quality of water used for hygienic practices and health condition of milking camel as well as variation in knowledge on hygienic milk production and postharvest handling practices in between two production systems.

The mean YMC of raw camel milk samples were significantly (P < 0.01) influenced by milk source (Table 3). Thus, raw milk samples collected from the udder (0.82 ± 0.12 log10 cfu·ml⁻¹) were significantly (P < 0.01) lower than samples collected from the equipment of producers (1.61 ± 0.12 log10 cfu·ml⁻¹) (Table 3). This might be due to further contamination of the milk during storage and transportation. Moreover, lack of cooling system and longer time elapsed during milk storage and transportation could contribute for such higher count.

The mean YMC (1.21. ± 0.16 log10 cfu·ml⁻¹) for milk samples collected from the udders and equipment of producers (Table 3) was agrees with Alaoui et al. [36] who reported yeast and mould count of 1.6 log10 cfu·ml⁻¹ for raw camel milk in Morocco. This difference might be due to the variation in the hygienic conditions of milk environment, cleanliness of udders and teats before milking, hygiene of milk handlers’ hands, sanitary conditions of milk equipment and quality of water used for hygienic practices. Moreover, the presence or absence of cooling facilities during milk storage and transportation might contribute such difference.

According to Torkar and Vengust [37], the upper acceptable limit of YMC for raw milk is 1.7 log10 cfu·ml⁻¹. However, the mean YMC for raw camel milk samples collected from the udders (1.71 ± 0.12 log10 cfu·ml⁻¹) and milk equipment (2.65 ± 0.12 log10 cfu·ml⁻¹) exceeded the upper acceptable limit. This might be due to contamination from dust, air, containers, water used and poor personal hygiene. According to Frank (2001) the potential sources of contaminations of Yeast and Moulds are air, water and equipments and also occur during processing, packaging or storage of raw materials or finished products.

The presence of YMC in raw milk above the recommended upper limit is an indication of unhygienic condition during production and postharvest handling as well as failure to use cooling system during storage and transportation, and makes the milk unsafe for its intended use [37].