This thesis provides evidence that learning difficulties some. black primary-schoolchildren may experience with certain textbooks, can be attributed, in part, to the visual text (imagery). These difficulties were established by eliciting responses from educationally dlscfdvantaged urban black primary learners to selected examples of visual texts using the Research Interview method. To further establish if these difficulties were attributable either to poorly executed/unrecognizable visual text, or to low levels of learned educational visual literacy skills - white primary-school children were also interviewed - as it was anticipated that they would be familiar with Western pictorial material due to their consistent exposure to books from an early age. The difficulties experienced by the black interviewees were attributed mainly to their level of learned pictorial perceptual skills and to a lesser extent to poorly/inadequately illustrated visual texts. It was found from interviews with the developers of visual texts - publishing personnel and illustrators - that the former were not entirely certain e.xactly which aspects of visual text were difficult for black primary learners to comprehend, while the latter were generaUy very uncertain. The procedure for visual text development by the developers (including textbook authors), was found to be problematic due to the lack of synthesis and consultative decision making in the process- between these persons. The limited time allocated to illustrators for producing visual text, as well as their professional isolation, were found to be factors which can give rise to ineffective and inadequate visual texts. Most publishers and authors, if they trial (field-test) materials, generally do not trial the visual text. The visual text is usually decided upon ultimately by the poblishers and produced after trialling and/or consultants have examined the written text. Consequently incongruent meanings and inconsistencies can result between written and visual text, which can affect the learning effectiveness of the composite text. Trialling (field-testing) of visual and written text together, was recommended to identify and address any difficulties experienced by learners prior to final publication of the textbook. Recommendations were provided for textbook selection committees, authors, teachers, publishers and illustrators.