L1 syndrome results from mutations in the L1CAM gene located at Xq28. It encompasses a wide spectrum of diseases, X-linked hydrocephalus being the most severe phenotype detected in utero, and whose pathophysiology is incompletely understood. The aim of this study was to report detailed neuropathological data from patients with mutations, to delineate the neuropathological criteria required for L1CAM gene screening in foetuses by characterizing the sensitivity, specificity and positive predictive value of the cardinal signs, and to discuss the main differential diagnoses in non-mutated foetuses in order to delineate closely related conditions without L1CAM mutations. Neuropathological data from 138 cases referred to our genetic laboratory for screening of the L1CAM gene were retrospectively reviewed. Fifty-seven cases had deleterious L1CAM mutations. Of these, 100 % had hydrocephalus, 88 % adducted thumbs, 98 % pyramidal tract agenesis/hypoplasia, 90 % stenosis of the aqueduct of Sylvius and 68 % agenesis/hypoplasia of the corpus callosum. Two foetuses had L1CAM mutations of unknown significance. Seventy-nine cases had no L1CAM mutations; these were subdivided into four groups: (1) hydrocephalus sometimes associated with corpus callosum agenesis (44 %); (2) atresia/forking of the aqueduct of Sylvius/rhombencephalosynapsis spectrum (27 %); (3) syndromic hydrocephalus (9 %), and (4) phenocopies with no mutations in the L1CAM gene (20 %) and in whom family history strongly suggested an autosomal recessive mode of transmission. These data underline the existence of closely related clinical entities whose molecular bases are currently unknown. The identification of the causative genes would greatly improve our knowledge of the defective pathways involved in these cerebral malformations.