The retrieval of columnar endocervical cells from the squamocolumnar junction has generally been considered to be a measure of the adequacy of a Papanicolaou smear; this implies that, if endocervical cells are absent from the smear, the examination for cervical cancer is less than optimal and should be repeated. A study was undertaken to determine if women with serial Papanicolaou smears without endocervical cells showed an increased rate of development of cervical atypia in subsequent smears. The smears of 18,914 women were evaluated for the presence or absence of endocervical cells and for the subsequent development of an abnormal smear over a four-year study period. No differences were found in the rates of atypia between women with and those without endocervical cells on serial Papanicolaou smears. Women with prior Papanicolaou smears without endocervical cells were much more likely to have a subsequent Papanicolaou smear without endocervical cells. Although no difference was found in the incidence of cervical atypia in the two groups during this short study period, these results should be considered to be preliminary.