The structural relations between the Cattell-Horn-Carroll abilities and reading achievement outcome variables across child and adolescent development were examined in the "Kaufman Assessment Battery for Children, Second Edition", and the "Kaufman Test of Educational Achievement, Second Edition", co-normed sample. We estimated single-group models using a cross-validation technique and then tested those salient effects using multigroup structural equation models to test the equality of salient effects across stages of development to determine whether those effects were moderated by grade. Nonequivalent models were also used to test the directionality of reading decoding and reading comprehension. The direct effects of long-term retrieval and short-term memory on reading decoding were the same across grade groupings. Findings support moderation by grade for the direct effects of comprehension-knowledge and reading decoding on reading comprehension. In Grades 1-6, the direct effect of comprehension-knowledge on reading decoding was smaller than the direct effect in Grades 7-12. The direct effect of comprehension- knowledge on reading comprehension was much smaller in Grades 1-3 compared with Grades 4-12. The path from reading decoding to reading comprehension was much larger in Grades 1-3 compared with Grades 4-12. The nonequivalent models suggest that the directionality of influence is from reading decoding to reading comprehension and not vice versa. Results from this study support a differentiated view of reading development with increases in comprehension-knowledge and decreases in reading decoding in explaining individual differences in reading comprehension over time.