Bifidobacterium spp. are becoming increasingly popular in food products, in particular dairy products, due to their history of safe use and the accumulating clinical evidence for their human health-promoting properties as probiotics. If probiotic cultures are incorporated into fermented milks with a specific health claim, a minimum of 106 cfu g−1 are required and therefore reliable methods are essential to monitor their survival and viability following production and during storage. Since most commercially available probiotic products contain a number of bacterial species, it is important to use a medium which promotes the growth of the bacterial group of interest while also inhibiting the growth of starter and other probiotic bacteria present. Indeed, for any culture-dependent enumeration method, the basic prerequisite is its ability to elect and select the target microorganism from the background microflora. This review will detail the elective and selective media used to enumerate bifidobacteria in foods with emphasis on selective media, given the likelihood of water and food products having a bifidobacteria component that is not numerically dominant. Numerous selective agents used for the isolation of bifidobacteria from various environments and food samples are also discussed, as is their suitability for enumerating bifidobacteria from foods and water. It is apparent that no single medium is suitable for the enumeration of all species of bifidobacteria from water and food; however, it appears that the following media, supplemented with mupirocin, are suitable for the selective enumeration of bifidobacteria: de Man Rogosa and Sharpe agar (de Man et al., 1960) supplemented with cysteine hydrochloride, trans-galactosylated oligosaccharide propionate agar, trypticase phytone yeast or reinforced clostridial agar.