The data of Figure 4 suggest a method for defining what is meant by the elusive term "mature" reading- "mature" reading is the level at which reading competency equals auding competency. In Figure 4, the (+) on the auding curve marks the level of "mature" reading, that is, the point at which the probability of finding A > R is equal to the probability of finding A < R. The point is near the middle of the seventh grade level. Given the margin for error in these methods, it seems safe to assume that "mature" reading is achieved, in the typical case, sometime in the seventh or eighth grades. It is of interest to note that eye-movement research has indicated that the adult pattern of eye movements is achieved by the eighth grade (Tinker, 1965, pp.81-84). If, following Tinker, it is considered that effective comprehension causes efficient eye movements, rather than eye movements causing comprehension, then the data of Figure 4 suggest that, because reading comprehension becomes equivalent to auding comprehension around the seventh or eighth grade, eye movement patterns reflect the "smooth" process of comprehension achieved by auding.

At the college and adult level, the proportion of comparisons in which reading clearly exceeds auding (A < R) is only .5, suggesting that some college students and adults never achieve superior efficiency in gathering information by looking at print in order to language.

It seems most likely that the increase in reading over auding performance found in about half the cases at the adult and college level represents improved skill in extracting information from the stable visual display of print, rather than indicating an ability to comprehend some material by print that cannot be comprehended by auding.¹ This is suggested by Goldstein's (1940) work, which is the only study found in which reading rate was paced line-by-line to match the speed of presentation of material for auding. With this control of reading rate, auding exceeded reading comprehension.

In general then, the data of Tables 2 and 3 and Figure 4 support the first hypothesis derived from the model: Auding performance exceeds reading performance during the early years of schooling. Furthermore, data of Figure 4 indicate that near the seventh and eighth grade levels, auding and reading performance are comparable, suggesting a definition of "mature" reading as the reading skill level at which reading competency equals auding competency (seventh or eighth grade reading ability, in the "typical" case).

1. Note that the superiority of auding over reading performance in the early years of schooling does indicate the ability to comprehend language by one mode and not the other, whereas the superiority of reading over auding at the adult level is interpreted as a quantitative (amount learned), not qualitative (what can be learned), modality difference. The model rules out the latter possibility.