between overall SLOPE scores and measures of exposure. Despite our findings that the SLOPE, as administered, is primarily an acquisition measure (because it yielded a "natural" difficulty order and allowed no monitoring time), no relationship was found between the measure of exposure and SLOPE scores.

These results confirm the suspicions voiced above about using "exposure-type" measures of informal linguistic environments, and underline the claim that active involvement is necessary for acquisition

    Table 2. SLOPE performance and measures of exposure 
    and formal instruction
_________________________________________________________________________

                      Years in English-           Years of formal
                      speaking country             English study
_________________________________________________________________________
                       r              p            r             p
SLOPE scores         0.014            ns          0.42         p < 0.001
_________________________________________________________________________

Partial correlations were used, as years in English-speaking country and years of formal study were correlated, r = -0.24, p < 0.01. Ordinary correlations were computed, however, and were quite similar to those reported above; for SLOPE and exposure, r = 0.003, and for SLOPE and formal study, r = 0.40.

to take place. Thus, if the SLOPE is a test of acquired competence only, it must be concluded that the question asked in the Krashen et al. series is a measure of time spent in "exposure-type" environments only, and this apparent counter-evidence to hypothesis I disappears entirely. No studies in the literature survey, however, are counter-evidence to the hypothesis that an "intake-type" informal environment may be quite efficient in increasing adult second language proficiency.

The significant correlation in Table 2 between years of formal instruction and SLOPE scores supports the hypothesis that the classroom can be of value, and in fact generally is of value, in language acquisition as well as in language learning.

While all studies described here are consistent with a revised version of hypothesis II, that in general formal instruction increases second language proficiency, none of the studies gives evidence to indicate that "learning" does indeed take place in formal situation in addition to acquisition. Note that "learning" occurs in the Upshur and Mason studies only under the "self-study" hypothesis, while in the Krashen et al.

48