In the present study, conducted with French-speaking children, we examined the reliability (group study) and the prevalence (multiple-case study) of dyslexics' phonological deficits in reading and reading-related skills in comparison with Reading Level (RL) controls. All dyslexics with no comorbidity problem schooled in a special institution for children with severe reading deficits were included in the study (N = 15; Chronological Age [CA]: 111 ± 8 months; RL: 80 ± 3 months). For the group study, the 15 dyslexics were matched pairwise on reading level, non-verbal IQ, and gender to 15 younger RL controls (CA: 85 ± 4 months). For the multiple-case study, the RL control group included 86 average readers (CA: 83 ± 4 months; RL: 85 ± 5 months). To assess the relative efficiency of the sublexical (or phonological) and lexical reading procedures, we relied on two comparisons: pseudowords vs. high-frequency regular words (the comparison mainly used in languages with a shallow orthography); and pseudowords vs. high-frequency irregular words (the comparison mainly used with English-speaking dyslexics), pseudowords and irregular words being either short or long. The dyslexics' skills in the domains supposed to explain their reading deficit were also examined: phonemic awareness, phonological short-term memory and rapid naming. In the group study, the dyslexics lagged behind the RL controls only when they were required to read long pseudowords. The results of the multiple-case study indicated that the prevalence of this deficit was high (the accuracy scores of all but two of the 15 dyslexics being more than 1 SD below the RL control mean), and that deficits in phonemic awareness were more prevalent (seven cases) than deficits in phonological memory (one case) and in rapid naming (two cases). Three unexpected results were observed in the group study: the difference between regular words and pseudowords (to the detriment of pseudowords) was not greater for the dyslexics; the difference between irregular words and pseudowords (to the benefit of pseudowords) was more significant for the RL controls; and there were no significant differences between the groups in reading-related skills. To explain these results, the severity of the dyslexics' reading deficit and the remediation they have benefited from must be taken into account. In addition, the fact that the outcomes of the comparison between pseudoword vs. regular or irregular word reading were not the same will make it possible to understand some discrepancies between studies carried out either in English or in a language with a shallower orthography (French, for instance)