The focus of the study reported here is on two issues: whether the two most elaborate systems of natural observation, that is, those developed by Escalona and Watts could be applied to mother-child interaction recorded in a teaching situation; and whether there are relationships between maternal-child behavior so observed, and a more extended definition of competence than those originally applied by the various independent investigators. This study, therefore, is considered to have used elements of the natural observation measures and applied them to the videotape from the Instructional Strategies in Infant Stimulation (ISIS) project to further examine the social roots of competency. The sample consisted of 53 families with 22 boys and 31 girls who participated in the ISIS project. They were low income families residing in or near Gainesville, Florida. It is held that, generally, the project demonstrated that the use of observation systems developed originally for naturalistic observation in the home is functional for the analysis of videotapes of structured teaching situations. Further, it is asserted that the mix of theoretical orientations contributes to the richness of this understanding of the meaning of the social transactions between babies and mothers. The project is seen as having demonstrated that within a social class there are variances in maternal-baby behavior which influence infant performance. (Author/JM)