series the evidence for learning is the positive correlation between years of formal study and proficiency. The SLOPE data indicated, however, that the classroom second language experience may also influence the acquired competence and we thus have no direst evidence that learning takes place at all. The hypothesis that the classroom contributes to acquisition only is sufficient to predict all the data covered above, as all proficiency tests, according to the Monitor Theory, involve some acquired competence.
There is, however, evidence that learning exists, and may increase proficiency. An interesting case of an advanced ESL performer ("P", discussed in Krashen and Pon, 1975, and in Chapter 1, this volume) supports the hypothesis that learning may increase performed accuracy by supplementing the acquired output. Our subject performed nearly perfectly in situations where monitoring was possible, but made errors in casual speech. She could, however, correct nearly all these errors and could also describe the grammatical rules she broke. This suggest that consciously learned competence was involved in those situations in which she made less errors.
The child relies primarily on acquisition. Thus, "intake" informal environments are sufficient. The class can provide only additional intake, and it appears to be the case that when children have access to rich intake environments, extra classes in second languages are not necessary (Fathman, 1975; Hale and Budar, 1970).
The Upshur, Mason, and Carroll studies are consistent with the hypothesis that intake informal environments can be quite beneficial for adult second language acquisition, and the distinction between intake and exposure-type informal environments disallows the Krashen et al. series as counterevidence to hypothesis I. The ineffectiveness of exposure type environments is confirmed by the lack of relationship between reports of time spent in the country where the target language was spoken and the results of an "acquisition" proficiency test. No studies provide counterevidence to a modified version of hypothesis II: formal environments are also beneficial. The need to decide between the original formulations of hypotheses I and