We have earlier shown that sun dried, ground, non-heated cassava chips are highly digestible in growing pigs. In this study, we tested the possibility that sun dried cassava meal would also be well digested by dogs. Twelve adult dogs were subjected to a 3×3 Latin square design with three experimental diets and three periods (three weeks each). The control diet was a homogenized, commercial, extruded diet that was pelleted at environmental temperature. The two test diets were made in the same manner, but after grinding, the commercial diet was blended with either 10 or 30% of sundried, non-heated cassava meal. The diets were well accepted by the dogs. The inclusion of cassava in the diet did not affect the amount and water content of faeces, but it significantly elevated the faecal score in a dose-dependent fashion. The faecal score for the 30% cassava diet was equivalent to diarrhoea. It was suggested that the feeding high amounts of non-heated cassava lowers water binding of faecal matter. The apparent digestibility of the non-structural carbohydrates (nitrogen-free extract) in non-heated cassava, as calculated by the difference method, was found to be 79.8% for the 10% cassava level and 84.0% for the 30% level. It was concluded that high intake levels of sun dried, non-heated cassava are unsuitable for dogs.