A variety of methods have been applied to the study of genotypic and phenotypic polymorphism in Cryptosporidium parvum. Results from these studies have consistently shown the existence of different genotypes and phenotypes within the species. A long-term goal of this work is the identification of markers for virulence in humans and animals and the elucidation of transmission cycles of C. parvum. Achievement of these goals will depend on the identification of highly polymorphic loci. Of particular interest are polymorphisms amenable to typing by polymerase chain reaction (PCR), as C. parvum cannot be expanded in vitro. Fingerprinting of isolates by restriction of PCR fragments or allele-specific PCR has given promising results. As originally observed by isoenzyme analysis, genetic fingerprinting has confirmed the occurrence in humans of unique C. parvum genotypes which are not found among calf isolates. This observation remains to be reconciled with the cross-infectivity of C. parvum to ruminant and nonruminant hosts and the important role that bovines play in the epidemiology of C. parvum and human cryptosporidiosis. Although PCR detection of C. parvum DNA from individual oocysts has been reported, the sensitivity of PCR detection when working with environmental or fecal samples is significantly reduced. Therefore, PCR is currently not used for routine diagnosis or environmental monitoring for C. parvum. Inhibitors present in environmental samples, mainly in water and soil, which can negatively affect PCR recoveries, have been identified, and several methods have been proposed to circumvent these problems. The further refinement of detection and genetic fingerprinting protocols will provide essential tools for indentifying environmental sources of oocysts and elucidating transmission cycles.