An investigation was carried out within the framework of the University of Waikato Team Teaching Unit into the effects upon achievement of five different timing sequences in the introduction of reading and writing in a beginners’ French course. A review of previous research indicated that, in general, a pre-reading period had been of benefit in listening and speaking, but that a benefit to reading and writing which had been predicted by some advocates of the pre-reading period did not materialise. Ten classes in five intermediate schools in Hamilton took part in the experiment, each of the five treatments being applied to two classes located in different schools. The treatments varied from an immediate introduction of reading and writing to a 20-week delay in both. The scores obtained in a modern language aptitude test showed no significant pre-treatment differences between the five groups, but the margin was slight, and scores obtained on this test were used as a covariate in subsequent analyses. Beta coefficients were calculated with a view to weighting the part scores of the aptitude test so as to maximise its predictive power, but the multiple correlation coefficient obtained did not differ significantly from that derived from the raw scores, which were therefore retained. All other tests used - achievement tests in four language skills, delayed achievement tests in three, and an attitude questionnaire - were constructed for the purpose. Ideas were available from various sources, but batteries of tests appropriate to a very elementary level were not located and some of the techniques used, particularly in the administration of the speaking test, were exploratory. All test materials were pre-tested at least once, and selection was based on an item analysis. Subsequent checks on item quality and test reliability indicated that, apart from minor details, the tests were satisfactory. Tests in listening, reading and writing were taken at the end of the experimental year by 363 pupils, in approximately equal groups for each of the five treatments. The speaking test was also taken by 176 pupils. Analysis of covariance for each test revealed significant differences in the listening, reading and writing test scores, but not in those of the speaking test. The levels of significance varied from test to test, but inspection of scores indicated that in every test the third treatment, consisting of a ten-week delay in the introduction of both reading and writing, yielded the highest adjusted means, while those of the first and fifth treatments were the lowest. A follow-up test given to 99 pupils a year later revealed only one residual inter-treatment difference, between the reading scores of the first and fifth groups combined, and those of the other three groups combined. These findings resemble those of some earlier research in that a pre-reading period was found to benefit achievement in listening skills. They differ from these earlier findings in two ways: the speaking skill was not shown to be significantly improved by a pre-reading period, while reading and writing were improved, suggesting that some transfer of learning from oral to graphic skills may have occurred.