Introduction Understanding the composition and determinants of microbial populations of the equine hindgut is fundamental to our knowledge of feed fermentation, and preventing gut-related disorders. Literature characterising bacterial hindgut populations are numerous. However, other than demonstrating that Piromyces citronii and Caecomyces equi can be present (Gaillard-Martinie et al., 1995; Gold et al., 1988), studies of anaerobic fungal species in the equine hindgut have been limited. Anaerobic fungi are potent fibre degraders which have been extensively studied in the ruminant gut (Gordon & Phillips, 1998). The aim of this study was therefore to conduct a preliminary characterisation of the structure of anaerobic fungal populations in the equine hindgut, and assess how their size and composition was affected by animal and time. Material and methods Faecal samples were collected from animals (n = 4) undergoing a concurrent yeast feeding trial. A 2 x 2 randomised latin square design was used. Animals (horse B, H, R and S) were maintained on pasture before and during the experiment. Period 1 (P1) comprised a 3 week feeding period where Group 1 (horses R & S) received a treatment of 20g live Biosaff yeast daily and Group 2 (horses B & H) received a control treatment of 20g killed yeast. Period 2 (P2), also 3 weeks, followed immediately but treatments were reversed for Groups 1 and 2. During the last five days (d1-d5) of both P1 and P2 faecal samples from the first two defecations after 11 am were collected from each horse, and frozen at -20oC until analysis. Total DNA was extracted using a QiaAmp DNA Stool Kit (Qiagen, UK), with the DNA concentration and integrity verified using a spectrophotometer (Nanodrop, LabTech International, UK) and agarose gel analysis respectively. The presence and quantity of anaerobic fungi was assessed using a Taqman probe based QPCR assay targeting the 5.8S rRNA gene as previously described (Edwards et al., 2008). The population composition of anaerobic fungal positive samples was then assessed using an anaerobic fungal specific Automated Ribosomal Intergenic Spacer Analysis (ARISA) as previously described (Edwards et al., 2008). The ARISA profiles generated were then compared using a curve-based (Pearson) cluster analysis (Fingerprinting software, BioRad. UK). ResultsAnaerobic fungi were detected in all four animals. Horses H, R and S were positive for anaerobic fungi in only a few of the P1 samples in contrast to horse B, where anaerobic fungi were routinely detected during both P1 and P2. Consistent with this was the observation that horse B had a larger amount of anaerobic fungi (212 (s.e.m. 47) and 64.0 (s.e.m.24.5) ng anaerobic fungal DNA per g dried faeces (s.e.m. 47) for P1 and P2 respectively) compared to the other three horses (<3.96 ng anaerobic fungal DNA per g dried faeces). Interestingly the anaerobic fungal population composition of all four horses in P1 generally shared a high similarity (> 95%). Although differences were evident over P1 and P2 with horse B (Figure 1) it is suggested that these population composition changes are more likely to be due to a temporal effect, rather than the loss of the yeast supplement viability. All of the ARISA profiles were dominated by a 391 bp peak (Figure 2), which when sequenced and analysed by BLAST was found to have 99 % identity with an uncultivated anaerobic fungus from cow manure assigned to the Cyllmayces genus (Accession No. GQ850302).