• 1The structure of the head in the nematode family Subuluridae has been studied comparatively and the results obtained used as the basis of a new classification.
  • 2The three structures at the base of the buccal cavity are shown to be not teeth, as they are commonly described, but modifications of the anterior end of the pharynx. These modifications are of two basic types: one which is lobed and spirally arranged and one which bears an elaborate arrangement of tooth-like structures and flanges on the inner surface.
  • 3It is shown that at least three parts of the head are of value in systematic analysis: the shape of the mouth opening, the cross-sectional shape of the buccal cavity and the structure of the pharyngeal portions.
  • 4The various structures are analysed throughout the family, and are shown to vary along three evolutionary lines which show a high degree of parallel evolution leading to complex mouth openings with lip-lobes in all three. It is suggested that this trend towards a complex mouth is associated with increasing efficiency of the mouth as a holding organ.
  • 5It is shown that the so-called ‘double’ buccal cavity, commonly reported from Subulura species, has always been reported because of a misinterpretation of the pharyngeal portions, due to the reduction of their musculature leading to fully cutieular portions.
  • 6The nine genera of which the family is composed, two of which–Prima-subulura and Tar subulura–are new, have been grouped into three subfamilies, of which one–Allodapinae–is new, on the basis of the analysis of the head structure, in conjunction with other previously better known characters.
  • 7The genera are grouped as follows: Subulurinae with three genera, Subulura, Oxynema and Labiobulura; Allodapinae subfamily nov. with two, Allodapa and Aulonocephalus; Parasubulurinae with four, Parasubulura, Primasubulura, Tarsubulura and Leipoanema.
  • 8As far as possible nominal species have been referred to these various genera on the basis of published descriptions as well as a study of specimens.
  • 9Since most of the rearrangements suggested in this paper are based on the structure of the head, which has been almost completely ignored heretofore, many nominal species cannot be referred to a genus and a list of thirty such names is given. These names are considered to represent species incertae sedis.