Researchers commonly consider the Semitic root to be the major lexical prime in Hebrew, relating morphological families in the major word classes. Psycholinguistic evidence supports the role of the consonantal root in acquisition and processing of Hebrew, from children’s early ability to extract roots from familiar words to spelling and reading in Hebrew by adults. There is, however, little information regarding the actual distribution of roots in their canonical habitat of verbs in the Hebrew addressed to young children. To meet this lacuna, the authors analyzed verbs, roots, and binyan patterns in two types of linguistic input to children: (1) spoken -- child-directed speech to toddlers aged 1;8 -2;2 and (2) written – preschoolers’ storybooks and 1st- 2nd grade texts. Findings include type and token frequencies of input verbs, distributions of full and defective root classes, morphological verb families, and semantic relations between verbs sharing the same root. The picture that emerges questions established views of root-based morphological families, and proposes a novel model of early verb and root learning in Hebrew.