This graduate thesis deals with determinating phonotactic traits of early child's vocabulary of three children aged 1;05- 1;08 years. The aim of the paper is to examine in what type of a neighborhood do most of the words from the early child's vocabulary reside in. The question is do most of them reside in dense neighborhoods, as it was shown for other languages. Another question is what type of neigborhoods does language of adults towards children have, as well as language of adult speakers. In this study were included words from the Croatian corpus of child speech ( HKDJ- Kovačević, 2002), which contains transcripts from three children. Also, the most frequent words from the languge of adult Croatian speakers were taken into account. Neighborhood density was calculated for each word and morphophonological neighborhood was taken into analysis due to the specific characteristics of Croatian language. The results show that the early child's vocabulary contains more words that come from dense neighborhoods than words that come from sparse neighborhoods. The same is with the language of adults towards children. In concordance with expectations, vocabulary of adult speakers consists of more words that come from sparse neigborhoods. Given the results, it can be concluded that dense neighborhoods are a trait of early child's vocabulary in Croatian language.