The nature of the relationship between the teaching and learning of second languages in the classroom has rarely been the subject of empirical investigation. The teaching profession tends to regard this relationship as a relatively direct one. Teaching which is based on a language syllabus explicitly or implicitly assumes that, given sufficiently frequent presentation and practice, learning will take place in a linear, cumulative fashion, although actual teaching practices may intuitively respond to learning being different. Since teachers are concerned with establishing which methods bring about the desired learner outcomes, interest in the learner is generally restricted to observations of what it is he has learnt and what he still has to learn, rather than how he learns. Second language acquisition research, on the other hand, has tended to focus on the learner, without necessarily relating his behaviour to the learning context. It has also involved mainly informal or only partially formal learners. Relatively few studies have considered learners who were exposed to the second language only in the classroom. At the same time the results of studies with informal or mixed learners have often been assumed to apply also to classroom-only learners. In particular it has been suggested that second language development follows its own principles and therefore cannot be influenced by instruction. The purpose of the present study is to investigate the acquisition of a second language by classroom-only learners in relation to the teaching learners were exposed to. Subjects of the study are 42 child and 6 adult learners of German, all native speakers of English. The study examines the development of negation and interrogation. We will find that the relationship between learning and teaching is not always a direct one and will interpret this as the result of learners' organic, creative interlanguage construction. At the same time we will consider the operation of linear, imitative learning processes, which result in the use of formulaic language, as a more direct outcome of the teaching. We will conclude that the acquisition of a second language in the classroom involves both organic, creative and linear, imitative processes.