A high moisture expansion of horse feedstuffs is known to be a possible risk factor for the development of esophageal obstruction or gastric distension, especially when they are not soaked prior to feeding. We analysed the water holding capacity (WHC) of commonly used feedstuffs and mixed feeds for horses, in order to identify feed materials with high and low water expansion capacities. The WHC gives the amout of fluid [g], that is absorbed by 1 g of material.21 dry feed materials commonly used as components of compound feed for horses and 21 pelleted concentrates for horses were analysed. For the determination of the WHC, the in vivo conditions were mimicked as far as possible. All samples were finely ground (64.9% <1mm). Each sample and buffer (pH 7.4) was incubated in a shaking water bath at 37.5°C for one, five and 60 minutes as well as 24 hours. After incubation, the samples were immediately centrifuged, the supernatant was discarded and the remaining sample was weighed. The WHC was calculated and given as WHC1, WHC5, WHC60 and WHC24 for the different durations of incubation. The chemical feed composition was determined by means of the Weender Analysis and Van Soest method.The method developed proved to be practicable, simple and highly repeatable.Significant differences appeared between the WHC of different feed materials but some groups (i.e. cereal grains etc.) with similar swelling behaviour could be defined. The highest WHC was found in feedstuffs with a high content of soluble fiber and mucilages. The WHC of feed materials was found to be positively correlated to the soaking times, but the biggest part of the water expansion took place within the first minute (WHC1 = 87.5±4.7% ofWHC60). Furthermore we detected a significant positive correlation of the ADF and NDF content on the WHC of mixed feeds(WHC1: r= 0.67 (ADF) and r= 0.81 (NDF), p< 0.0001). However an assessment of the WHC based only on the fiber content of the feedstuffs was not possible.Feedstuffs of the same botanical origin (with similar analytical composition), but processed differently(kibbled, rolled, flaked) showed significant differences of their WHC1. These results showed the complexity of the WHC, which is influenced by many different factors. Despite that, a practically relevant method for measuring the WHC of feedstuffs was developed for the assessment of their increase in volume following the absorption of fluid.