ABSTRACTThe relation between performance in phonemic segmentation and reading and writing ability is discussed. Not much is known about how segmentation is carried out and which word properties influence performance. Therefore, effects of word properties (length, CV structure, syllabic structure, meaning) were investigated in two experiments. Strong indications were obtained that an onset‐rime distinction is relevant for the process of segmentation. The meaning of a word appears to have no influence. The decentration hypothesis can therefore be abandoned as an explanation for segmentation difficulties.Effects of length and syllabic boundary can be explained by the (disruptive) effect of consonant clusters, which are not only difficult to segment themselves, but also adversely affect the processing of segments earlier in the word. This leads to the conclusion that a simple, strictly serial model for segmentation cannot be adequate. The results furthermore suggest that an articulatory rather than a phonological code is the object of segmentatio