The aim of this study was to map the camel milk value chain and establish the predisposing factors for increase in microbial counts in milk along the chain. Isiolo County was chosen for the study. Data collection was done through key informant interviews, structured interview schedules, observation and microbial analysis of milk samples. During milk sampling, milk temperature, environmental temperature, time and volume of milk from which the sample was taken were recorded. Along the value chain, microbial counts in milk increased significantly from log 4.91 ± 1.04 CFU/ ml at production to log 7.52 ± 1.32 CFU/ml at Nairobi market for total viable counts and log 3.68 ± 1.28 CFU/ml at 10 10 production to log 6.42 ± 1.13 CFU/mlattheNairobimarketfor coliform counts. At production, milking persons neither washed their hands nor cleaned the camels’ udder before milking, and plastic, non-food grade containers were the only form of receptacles used for milk along the chain. The relationship between microbial counts and time taken to transport milk along the chain was significant while the volume of milk in the receptacle had no effect on microbial counts. The milk was held at a temperature of between 28 and 32.5 °C before delivery to secondary collection centres from 10:15 am to 6:30 pm for cooling. Training on milk quality for milk handlers at the collection centre had no effect on microbial counts. Affordable access to low-cost food grade plastic containers as well as cooling milk in the individual receptacles within two hours of milking, without bulking and refilling again into the receptacles for transportation, as is the practice, would reduce microbial counts. Similarly, training on milk quality should start at production where milk contamination is initiated. Finally, milk value addition would improve milk shelf- life enabling access to distant markets. This would greatly improve the livelihoods of the pastoral camel milk producers. Keywords: Camel milk, Isiolo, Microbial load, Predisposing factors Introduction milk-producing areas and urban centres such as Milk production in Kenya is dominated by cattle and Isiolo, Nairobi and Nakuru (Musinga et al. 2008). The camels which contribute 86 and 10% of total milk bulk of the milk stays at ambient temperature for a production respectively (GOK 2010). Kenya has a long time before reaching either the cooling centre or camel population of over 3.3 million (KNBS 2010), the market (GOK 2010). The milk is commonly han- and are particularly important in North Eastern dled in plastic containers which are not just difficult region of the country where a large community of to clean but also the poor quality of water used to Somali and related ethnicities are more familiar with clean them enhances microbial levels in milk (Mwangi camel milk (FAO 2011). Along the camel milk value et al. 2000). The milk handlers also pool the milk chain, only 5% of the milk is processed while the rest from different sources without carrying out quality is consumed at household level and hotels in the tests (Younan and Abdulrahman 2004). These factors result in increased microbial counts along the chain * Correspondence: firstname.lastname@example.org as reported by Matofari et al. (2013) with the presence of Department of Dairy and Food Science and Technology, Egerton University, pathogenic microorganisms such as Salmonella spp., P.O Box 536-20115 Egerton, Nakuru, Kenya Staphylococcus aureus and Streptococcus spp. reported Department of Pure and Applied Sciences, Technical University of Mombasa, P.O Box 90420-80100, Mombasa, Kenya (Matofari et al. 2013; Odongo et al. 2016; Kashongwe et Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Nato et al. Pastoralism: Research, Policy and Practice (2018) 8:16 Page 2 of 8 al. 2017). The main problem, however, faced by the traders market in Nairobi city, a distance of 275 km to the south along this chain is milk spoilage (Wayua et al. 2012). The of Isiolo town (Noor et al. 2013). occurrence of these factors, and their contribution to in- crease in microbial counts in milk along the chain have Methods not been reported. Empirical studies of these factors are Study design and field data collection therefore important in prioritizing interventions to not A cross-sectional study was set up to collect data from only reduce milk spoilage along the chain but also reduce milk handlers. Primary and secondary data was collected health risks to which camel milk consumers may be to map the camel milk value chain. Primary data was exposed. collected by observation and use of structured interview schedules administered to actors along the chain, from Study area producers, collection centre operators, transporters and This study was carried out between February and traders. A total of 90 milk handlers were interviewed August 2015 within Isiolo County and along the value face-to-face. The questions were translated into Swahili chain to Nairobi (Figure 1), Kenya. Isiolo County lies be- or the preferred language of the respondent where ap- tween longitudes 36° 50′ and 39° 30′ East and latitudes propriate. Key informant interviews were held with the 0° 5′ and 2° North and has a total area of 25,605 km secretary of the Camel Owners Association and the (Noor et al. 2013). Isiolo County is a typical arid and chairpersons of the Anolei women’s group and Tawakal semi-arid region (Noor et al. 2013) with average annual group to validate the findings. All the participants in this rainfall of 418 mm (Mati et al. 2005) and a camel popu- study voluntarily provided data and information. lation of 39,084 (KNBS 2010); most camel owners prac- tice pastoralism, with camel milk being an important Milk sample collection and laboratory analysis source of livelihoods (Wayua et al. 2012). Most of the Milk samples were collected from the camel during marketed milk ends up either in Isiolo town, or Eastleigh milking, and along the value chain from primary Figure 1 Map of Kenya showing Isiolo County and Nairobi city Nato et al. Pastoralism: Research, Policy and Practice (2018) 8:16 Page 3 of 8 collection sites near the camel herds, secondary collec- 37 °C for 24 h. All the colonies on the plate were counted tion centres in Isiolo town, transporters, and traders to represent enterobacteriaceae counts, while pink/red both in Isiolo and Nairobi. In Isiolo, 76 milk samples colonies were counted as coliforms (lactose fermenters). were collected from the interviewed actors in Isiolo and 55 milk samples collected from Nairobi traders. The size Statistical data analysis of the containers from which milk was sampled ranged Data was subjected to statistical analyses using SAS ver- from 3, 5, 10 to 20 l. Sampling was done in 60-ml sion 9.0 (SAS Institute Inc., Cary, NC, USA). The statis- sterile sampling bottles and transported in a cool box at tical analyses involved determination of frequencies and 4 °C to the laboratory in Isiolo County referral hospital for arithmetic means, correlation between factors that affect samples from Isiolo and University of Nairobi’s Food microbial loads, and regression between independent and microbiology laboratory for samples from Nairobi. The dependent variables. Shapiro-Wilk test was carried out to analysis was done within four hours after sample collection. determine normality of data when n < 30. The microbial The samples were analysed for total viable counts, coliform counts were transformed to log and reported as mean ± counts and enterobacteriaceae counts. During sampling, standard deviation CFU/ml of milk. All the statistical tests the time of sample collection was recorded as well as the were done at a significant level of p ≤ 0.05. Information environmental and milk temperature. All the milk samples from the respondents which was validated by the key in- were voluntarily provided by the participants. formants was used to map the value chain. Determination of total viable counts (TVC) was done as described by Messer et al. (1985) for the standard Results plate count. A milk sample measuring 1 ml was diluted Description of the pastoral camel milk value chain sixfold (1:10) using buffered peptone water for all the The camel milk value chain is summarized in Figure 2. samples except the samples from Nairobi which were di- The major milk producing areas in Isiolo were Burat2, luted tenfold (1:10). One millilitre from each dilution Kulamawe, Shaab, Harakote, Ngaremare, Laglava, was delivered onto a sterile petri dish. This was done in Dahayal, Shambani, Mlango, Gotu, Malbi Shilmi, Kobi duplicate. It was then followed by pouring about 20 ml Fora and Elhat. The main stakeholders involved in the of molten (approx. 45 °C) Plate Count Agar (Oxoid) pre- value chain were camel herd owners, cooperative pared according to the manufacturer’s instructions. The societies, transporters, traders, County veterinary office inoculum were then gently mixed by alternate clockwise and non-governmental organizations. The camel herd and anticlockwise movement for three minutes and left owners employ herders to look after the camels. For the for 30 min on the bench to solidify. The plates were then health of the camels, government extension officers were incubated at 32 °C for 48 h followed by counting of col- available from the County offices to provide vaccination onies. Plates with colonies between 30 and 300 from the and treatment. Non-governmental organizations also lowest dilution were counted and reported as colony- offer support in animal health. The herd sizes ranged forming units per millilitre of milk (CFU/ml). Determin- from about ten camels to hundreds of camels. ation of coliforms and enterobacteriaceae followed the The lactating camels were milked twice in the morn- same procedure except that the media used was MacCon- ing, usually at 6:00 am and 8:00 am, and the milk was key agar (Oxoid) and the temperature of incubation was first bulked at the boma/kraal (fenced enclosure). Milk Figure 2 Pastoral dairy value chain for camel milk from Isiolo County Nato et al. Pastoralism: Research, Policy and Practice (2018) 8:16 Page 4 of 8 for the herd owner’s home consumption was separated delivered it to the secondary collectoin centres using wheel- from that to be marketed. The milk to be marketed was barrows. Transport of milk to Nairobi city was done only then collected by transporters or traders from the pri- using passenger vehicles with milk kept in plastic mary collection centres which were close to the herd. containers (Fig. 4)thatwereusedtodeliver milk to Most of the marketed milk was bulked at the group- the collection centre. owned secondary collection centres run by Anolei women’s group and Tawakal group. The Anolei women Microbial loads of milk along the chain group handled up to 2500 l per day while Tawakal han- Total viable counts (TVC) of milk from individual camels dled 500 l per day. On arrival at these collection centres, increased significantly from 4.91 ± 1.04 at production the milk was tested for alcohol stability and bulked in a to 6.49 ± 0.77 at secondary collection centres and 7.52 ± 1.32 chill tank at 4 °C. Milk that failed the test was not at the Nairobi market. Coliform count (CC) at produc- bulked but cooled in its container in a separate cooler. tion was 3.68 ± 1.28 which also increased significantly The milk would then be sold as suusa, kulel or orawa- to 6.42 ± 1.13 at Nairobi market. Similarly, enterobacteria- kandi (types of fermented milk, suusa being sweet-sour, ceae counts also increased significantly from 3.89 ± 0.98 at orawa-kandi being bitter-sour and kulel tastes between production to 5.29 ± 1.10 at secondary collection centre, suusa and orawa-kandi). The milk bulked in the chill and 6.49 ± 1.13 at the Nairobi market. Milk bulked at the tank would be emptied again in plastic receptacles the camel boma had TVC of 5.09 ± 0.60 and CC of 3.44 ± 1.52, following day at 5:00 am and loaded on buses for trans- both of which increased significantly at the secondary col- port at ambient temperature to Nairobi, a journey of lection centre and Nairobi market. about seven hours. In Nairobi’s Eastleigh estate, milk At the primary collection centre, TVC, CC and entero- traders receive the milk for sale to home consumers, ho- bacteriaceae were 5.09 ± 0.60, 3.14 ± 0.93 and 3.19 ± 0.77, tels and restaurants within Nairobi. Occasionally, some respectively. These increased significantly during trans- milk is transported to other towns such as Nakuru. Sec- port of milk to the secondary collection centre and to ondary milk collection centres run by individuals handle Nairobi market. All the counts increased significantly between 20 and 100 l per day. Other local operators cool from the secondary collection centre to the Nairobi mar- the milk and sell it to traders within Isiolo, hotels, res- ket. The microbial load along the chain from production taurants and home consumers. to Nairobi market is summarized in Figure 3. Different modes of transport were used for milk transpor- tation along the chain. Delivery of milk from production to Predisposing factors for increased microbial counts in milk primary collecton centres, usually a distance of < 1 Km, General hygiene (udder hygiene, milking personnel hygiene, was dominated by transport on foot at 67%, followed by milking container type and cleanliness and excessive use of donkeys at 22%, and vehicles at 11%. Delivery of milk human handling of milk) to the secondary collecton centres was dominated by mo- During the study period, there was neither hand washing torcycles at 42%, with vehicles at 24%. The rest of the milk nor udder cleaning before milking the camels. All the (34%) was first delivered by vehicles to the Milk stage milk containers were made of non-food grade plastic within Isiolo town where middlemen bought the milk and with non-hygienic design (Figure 4). Figure 3 Microbial loads along the camel milk value chain Nato et al. Pastoralism: Research, Policy and Practice (2018) 8:16 Page 5 of 8 Although this is not a common occurrence, it demon- strates the challenges along the value chain that would lead to high microbial count in milk and eventually milk loss through spoilage. The mean ambient temperature along the chain from production to secondary collection centre was 30.8 ± 1.04 °C with a minimum of 28 °C and a max- imum of 32.5 °C while the milk temperature had a mean of 30.7 ± 1.4 °C with a minimum of 28 °C and a maximum of 35 °C. The milk at the secondary col- lection centre is then cooled to 4 °C in a bulk tank. Figure 4 Milk in plastic containers at the reception of a milk It is then returned to the plastic containers and trans- collection centre in Isiolo ported to Nairobi where the milk temperature rises again to 17.8 ± 2.7 °C. The ambient temperature dur- The containers were previously used for vegetable oil. ingmilk samplinginNairobi was22.5°C. At the secondary collection centres, group members There was a strong correlation between ambient and were involved in all activities from milk reception to milk temperature (adj. R = 0.90181, p < 0.0001), and record-taking, predisposing the milk to further microbial therefore, only data on milk temperature was used for contamination due to excessive human handling. modelling in this study. For microbial loads, correlation The pastoral milk production area had no water sup- between TVC and CC was strong with adj. R = 0.54233 ply, apart from water that was brought by milk traders and p = 0.0004 while that between TVC and from Isiolo town. This water was hardly enough for enterobacteriaceae had adj. R = 0.9028 and p < 0.0001. cooking or drinking, let alone any form of hygiene. All Only data on TVC was therefore used in further the milk containers were cleaned in either Isiolo town or statistical analysis. Nairobi city using running water and soap. The con- The correlation between time and milk temperature tainers were then smoked for disinfection using wood was equally strong (adj. R = 0.84158, p < 0.0001). from especially acacia trees. Although these factors influ- However, due to their importance in microbial growth, ence microbial loads in milk, they were the same in the each was subjected to simple regression analysis with study area, and their effect on microbial load in the milk TVC. The relationship between TVC and time was could not be determined from this survey. found to be significant (adj. R = 0.1473, p = 0.0045) while that between TVC and temperature was found not Effect of holding time and temperature on microbial counts to be significant (adj. R = 0.0527, p = 0.073). Milking was done between 6:00 am and 8:30 am and the milk delivered to the primary collection centre where the milk was picked by transporters or traders to the subse- Effect of milk volume on microbial counts quent nodes along the value chain. The mean pick-up Along the chain, from production to the secondary time was 10:58 am ± 28 min. The average delivery time to collection centre, the relationship between TVC and the the secondary collection centre was 11:47 am ± 1 h volume of milk contained in a receptacle was not signifi- 40 min with a range from 10:15 am to 6:33 pm. On the cant with adj. R = 0.05 and t value = 40. However, larger day of sampling, 200 l of milk from Kulamawe which receptacles had marginally lower microbial counts is 80 km from Isiolo town had not been delivered by compared to smaller receptacles. 7:00 pm because the vehicle had mechanical problems, which was said to be not unusual. The collected milk is bulked in a chill tank and cooled overnight to 4 °C. At Effect of level of training of milk handlers on microbial 5:00 am the following morning, the milk was filled into counts plastic containers and loaded onto buses for transporta- At the production level, none of the operators had any tion to Nairobi where it is delivered at about 11:30 am. training on milk quality. These were the same operators However, during the study, 2500 l of milk from one sec- who transported milk to the primary collection centre. ondary collection centre had not been dispatched to At the secondary collection centre, 44% of the operators Nairobi because of disagreement on pricing between had training on milk quality. The mean TVC of the milk the owners of milk at the collection centre and the handled by those who were trained on milk quality was agents in Nairobi. Due to limited capacity of refriger- 6.68 ± 0.74 while those who had not been trained had a ation, half the volume of milk was not refrigerated but mean of 6.18 ± 0.71. The counts were however not sta- still transported to Nairobi with a delay of 24 h. tistically different with p = 0.06154. Nato et al. Pastoralism: Research, Policy and Practice (2018) 8:16 Page 6 of 8 Discussion to Nairobi the following day in the morning. Non- The pastoral camel milk value chain governmental organizations such as the Netherland The pastoralists often move from one place to another Development organization (SNV) have supported the in search of grazing land and water for their animals. groups to acquire aluminium churns for milk handling. Milking of lactating camels occurs at the boma (kraal) Despite this, milk is still transported in unhygienic plas- where the pastoralists would have pitched camp. The tic containers since aluminium churns are a challenge milk would then be taken to the nearest primary collec- for transport of milk by bus to Nairobi due to milk spill- tion centre which is usually temporary and only exists age and weight. The bus operators therefore only accept till the pastoralists move with the herd to another graz- to transport milk in plastic containers. The use of food ing area. This therefore creates a challenge in milk col- grade plastic containers would be an alternative to alu- lection because the traders and transporters sometimes minium churns to overcome this challenge. have challenges in tracking the herders. The lack of roads makes it even more difficult for the milk to be col- Milk holding time and temperature lected. Motorbikes therefore are the most effective Time and temperature are two of the most important ex- means of transport from the hinterland to the secondary trinsic factors affecting microbial growth, and generally, collection centres. Vehicle transport is used to transport milk should be cooled to 4 °C within two hours of milking milk from distant milk production areas such as (Jay et al. 2005). In this study, between production and Kulamawe and Gotu. Similarly, only vehicle transport is secondary collection, milk was held at an ambient available to deliver milk to Nairobi. In Nairobi, the milk temperature of between 28 and 32 °C for up to 12 h. This is received by milk agents who then apportion the milk substantially increases microbial counts of the milk and to the traders. The traders quickly take away the milk accounts for significant increase in microbial loads from for sale at the roadside or to restaurants. The whole production to the secondary collection centre. There was value chain is informally controlled by a few people in indeed a significant relationship between time and TVC Isiolo and their associates in Nairobi. Any activities to between production and the secondary collection centre improve the quality of the milk along the chain will have despite the result having a low adj. R which can be to involve these actors, who incidentally, benefit at the attributed to transforming microbial counts to log .Since expense of producers. the secondary collection centres have cooling facilities, microbial growth could be curtailed if milk was delivered Predisposing factors for microbial loads along the chain less than two hours after milking. Cooling the milk to 4 °C General hygiene (udder hygiene, milking personnel hygiene, overnight at the secondary collection centre, then milking container type and cleanliness and excessive transporting it to Nairobi at ambient temperature, a human handling of milk) journey of seven hours, compounds the problem of Poor udder hygiene has been linked to high microbial increased microbial counts. In this study, regressing counts in milk (Gleeson et al. 2009). The microbial load microbial counts against temperature was not found to at production continues to increase in subsequent nodes have a significant relationship because of a small range in up to the market as was also reported by Matofari et al. the milk temperature from production to secondary (2013). The non-food grade plastic milking containers as collection centres. well as failure to clean hands and the camel’s udder con- tribute to milk contamination, and our work (in press) Relationship between volume of milk and microbial counts has isolated Escherichia coli from both the udder surface This relationship between microbial load and volume of and the hands of the milking persons. The identified milk in the receptacle was not found to be significant. challenge was the lack of appropriate milk containers Smaller containers have a large surface area to volume ra- and unavailability of water for maintenance of udder and tio and were expected to have a higher microbial count. milking persons’ hygiene. However, milk in large containers was likely pooled from At the Anolei and Tawakal secondary collection cen- different sources, which is considered a risk factor for in- tres, group members are involved in most activities from creased microbial load (Younan and Abdulrahman 2004). milk reception to record-taking. The collection centre was therefore a beehive of activities, and the inordinate Level of training of operators human presence and excessive manual handling in- Training of milk handlers has always been advocated as creases the risk of milk contamination. To reduce exces- a way of improving milk quality (GOK 2010). In this sive human handling and exposure of milk by pooling it study, results of microbial counts in milk at the second- into chill-tanks and transferring the milk back to the ary collection centre indicated that there was no signifi- plastic receptacles, the alternative is to cool milk in their cant difference in microbial load of milk handled by original containers in a cold store, before transporting it individuals trained on milk quality vis a vis those Nato et al. Pastoralism: Research, Policy and Practice (2018) 8:16 Page 7 of 8 without training. It was also observed that the milk industry, and it is therefore required to streamline it. This handling practices of individuals trained on milk quality would greatly improve the livelihoods of the pastoral were not different from those without training. Training camel milk producers. of milk handlers along the chain while neglecting milk Acknowledgements handlers at production does not result in reduced micro- We wish to thank our key informants and all the actors along the value chain bial load. This is because milk contamination begins at for the cooperation they accorded us during the study. We also thank Isiolo County Referal Hospital and University of Nairobi for providing us with production and the microbial load increases exponen- laboratory facilities. tially along the chain due to lack of a cold chain, in- appropriate milk containers and non-observance of good Funding hygienic practices. Training should enable the milk han- This study was facilitated by the RELOAD project funded by the German Ministry of Education and Research (BMBF) within the framework of GlobE dlers and actors along the value chain to adhere to a set initiative, grant number 031A247A-D. The funding organization however was of Standard Operating Procedures that limit microbial not involved in the design of the study. Data collection, analysis, growth. It was also observed that milk containers are la- interpretation, and writing of the manuscript were wholly the responsibility of the authors. beled by owners, and even after the milk is bulked for cooling, it is returned to the same containers for trans- Authors’ contributions port to Nairobi city. These activities create congestion SMN collected the data and drafted the paper. JWM and BOB designed the experiment. CH interpreted the results. All authors read and approved the and excessive handling of the milk at the collectoin final manuscript. centre where everybody wants to track their milk (or containers). The actors at the collection centres Authors’ information therefore not only need training on milk quality but SMN is a PhD (Food Science) candidate at the Department of Dairy and Food Science and Technology, Egerton University, P.O Box 536-20115, also support to strengthen institutional capacites of the Egerton, Kenya, and also an Assistant Lecturer at the Department of Pure collecton centres so that operations are left to elected di- and Applied Science, Technical University of Mombasa, P.O Box, 90420- rectors and trained employees. However, as discussed 80100, Mombasa, Kenya. JWM (PhD, Egerton University) is a Professor of Food Safety/Food Biotechnology at the Department of Dairy and Food earlier, there are milk traders/middlemen in Isiolo and Science and Technology, Egerton University, P.O Box 536-20115, Egerton, Nairobi who benefit from the current arangement at the Kenya. BOB (PhD, Wegeningen UR University, Netherlands) is a Professor of expense of producers, and they should be brought on Animal Science at the Department of Animal Science, Egerton University, P.O Box 536-20115, Egerton, Kenya Kenya. CH (Dr. Sc. Agr, University of Hohenheim) board for any effective change to be achieved. is a Livestock Scientist and the Managing Director, German Institute of Tropical and Subtropical Agriculture, SteinstraBe, 19, 37213 Witzenhausen, Germany. Conclusions and recommendations The study found that practices at production and along Competing interests The authors declare that they have no competing interests. the chain contribute to high microbial counts in milk. The long time that milk is held at ambient temperature Publisher’sNote while awaiting collection at primary collection centres, Springer Nature remains neutral with regard to jurisdictional claims in published or during transport, contributes to increased microbial maps and institutional affiliations. load in the milk. Most importantly, training of milk han- Author details dlers should include all actors starting with the herders Department of Dairy and Food Science and Technology, Egerton University, at production. This should be complemented by policy P.O Box 536-20115 Egerton, Nakuru, Kenya. Department of Animal Science, to support provision of water not only to support camel Egerton University, P.O 536-20115 Egerton, Nakuru, Kenya. German Institute of Tropical and Subtropical Agriculture, StreinstraBe, 19, 37213 Witzenhausen, husbandry but also hygiene during milking. Supporting Germany. Department of Pure and Applied Sciences, Technical University of the producers, transporters and traders to acquire low- Mombasa, P.O Box 90420-80100, Mombasa, Kenya. cost food grade plastic containers with a capacity of 10 Received: 7 November 2017 Accepted: 25 April 2018 to 15 L of milk will not only reduce microbial counts but also be convenient during handling and transporta- tion. Most importantly, design of cheaper and more References practical cooling facilities based on solar energy or evap- FAO. 2011. Global food losses and food waste. Extent, causes and prevention. International Congress “Save Food”, Düsseldorf, Germany. http://www.fao. orative cooling using charcoal housing would greatly re- org/docrep/014/mb060e/mb060e00.pdf. Accessed 15 October 2016. duce milk spoilage. Processing of the milk would add Gleeson, D., B. O’Brien, J. Flynn, E. O’Callaghan, and F. Galli. 2009. Effect of pre- value by improving its shelf-life and therefore access dis- milking udder preparation on the microbial counts on teats prior to cluster application. Irish Veterinary Journal 62 (7): 461–467. tant markets. In addition, the Kenya Bureau of Standards, GOK (Government of Kenya). 2010. Kenya Dairy Master Plan. Nairobi, Kenya: a government agency established under the Kenyan law Ministry of Livestock Development. http://kdb.co.ke/press/publications/ (Standard Act, CAP 496), and responsible for quality and reports/5-kenya-national-dairy-master-plan/file. Accessed 16 June 2016. Jay, J.M., M.J. Lossner, and A.G. Golden. 2005. Modern Food Microbiology. New standards, should intervene and regulate this industry. York: Springer Science+Business Media Inc. Similarly, the Kenya Dairy Board (established under the Kashongwe, B.O., B.O. Bebe, J.W. Matofari, and C.G. Huelsebusch. 2017. Associations Dairy Act, CAP 336) has a direct mandate over the dairy between milking practices, somatic cell counts, and milk post-harvest losses in Nato et al. Pastoralism: Research, Policy and Practice (2018) 8:16 Page 8 of 8 smallholder dairy and pastoral camel herds in Kenya. International Journal of Veterinary Science and Medicine 5: 57–64. KNBS (Kenya National Bureau of Statistics). 2010. Population and housing census: Livestock population by type and districts, 2009. http://www.opendata.go.ke/ datasets/census-vol-ii-q-11-livestock-population-by-type-and-district- 2009?selectedAttribute=Camels. Accessed 14 Dec 2015. Mati, B.M., J.M. Muchiri, K. Njenga, F. Penning de Vries, and D.J. Merrey. 2005. Assessing water availability under pastoral livestock systems in drought-prone Isiolo District, Kenya. Working paper 106. Colombo, Sri Lanka: International Water Management Institute (IWMI) http://www.iwmi.cgiar.org/Publications/ Working_Papers/working/WOR106.pdf. Accessed 6 November 2017. Matofari, J.W., P.L. Shalo, M. Younan, J.N. Nanua, A. Adongo, A. Qabale, and B.N. Misiko. 2013. Analysis of microbial quality and safety of camel (Camelus dromedaries) milk chain and implications in Kenya. Journal of Agricultural Extension and Rural Development 5 (3): 50–54. Messer, J.W., H.M. Behney, and L.O. Leudecke. 1985. Microbiological count methods. In Standard Methods for the Examination of Dairy Products, ed. G.H. Richardson, 133–150. Washington D.C.: American Public Health Association. Musinga, M., D. Kimenye, and P. Kivolonzi. 2008. The camel milk industry in Kenya. Results of a study commissioned by SNV to explore the potential of camel milk from Isiolo district to access sustainable formal markets. Final report prepared by Resource Mobilization Center (RMC), Nanyuki, Kenya. http://fr.afraca.org/ download/general_rural_finance_publications/Camel-Milk-Value-Chain-The- Case-of-Isiolo-District-A-Study-Report-by-Muli-Musinga-et-al.pdf. Accessed 14 Dec 2014. Mwangi, A., S.M. Arimi, S. Mbugua, E.K. Kangethe, A.O. Omore, and J.J. McDermott. 2000. Assurance of marketed milk quality in Kenya. Paper presented at Faculty of Veterinary Medicine, Biennial Scientific Conference, 30– 31, August, 2000. Nairobi: University of Nairobi. https://cgspace.cgiar.org/ handle/10568/1583. Accessed 8 Aug 2014. Noor, I.M., A.Y. Guliye, M. Tariq, and B.O. Bebe. 2013. Assessment of camel and camel milk marketing practices in an emerging peri-urban production system in Isiolo County, Kenya. Pastoralism: Research, Policy and Practices 3(28). https://doi.org/ 10.1186/2041-7136-3-28. Odongo, N.O., P.O. Lamuka, G.O. Abong, J.W. Matofari, and K.A. Abbey. 2016. Physico-chemical and microbiological postharvest losses of camel milk along the camel milk value chain in Isiolo, Kenya. Current Research in Nutrition and Food Science 4 (2): 80–89. Wayua, F.O., M.W. Okoth, and J. Wangoh. 2012. Survey of post-harvest handling, preservation and processing practices along the camel milk chain in Isiolo District, Kenya. African Journal of Food, Nutrition, and Development 12 (7). Younan, M, and Abdulrahman. 2004. Milk hygiene and udder health. In Milk from the camel: Handbook on products and processing, ed. Z Farah and A Fischer, 67-76. Zurich/Singen, Switzerland vdf. Hochschulverlag.
– Springer Journals
Published: May 30, 2018