Crosses were established between spine-free Willamette and spiny Willamette, varieties homozygous for the dominant gene S (Meeker, Latham, Leo, Tayberry), varieties and lines homozygous for the recessive gene s (8216B6, Glen Moy, 7515C5, Glen Prosen), varieties and lines heterozygous (Ss) for gene s (Glen Clova, 8210B1), spine-free selfed derivatives of spine-free Willamette and wild raspberries, including reciprocals and in some cases selfed progenies for comparison. Also included were crosses between 4 tetraploid spine-free derivatives of a spine-free Willamette derivative (8645) and parents for the dominant gene S and a cross between 8645 and 8422D1 (homozygous for s). The inheritance of spinelessness in progenies of spine-free Willamette was very variable. An hypothesis that spinelessness is caused by a mutation to a dominant gene remains tentative, because, among progenies expected to segregate, some were entirely spiny, others were entirely spine-free and some segregated for spinelessness. Several possible causes of the variable segregation were apparently eliminated, but the recessive gene s, which also confers spinelessness, appeared to have a major effect on segregation. The possibilities are discussed that variable segregation occurred either because the postulated mutant gene had pleiotropic effects on seed development similar to those of gene s, or because it is allelic to gene s. From a plant breeder's viewpoint the gene has limited value except for breeding purple raspberries.