The study aims to describe and interpret the results of testing the speech of Lithuanian children in order to find out whether language-specific features of Lithuanian as a highly inflected language help children grasp the complex syntactic relations between the subject and the object in relative clauses. The investigation has been aimed to test the hypothesis to the effect that depending on the language type, object relative clauses are more difficult to acquire than subject relative clauses (Guasti, Cardinaletti 2003; Utzeri 2007; Brandt, Diessel, Tomasello 2008; O’Grady, Kim, Lee et al. 2011; Benţea 2012).The findings of the research support the hypothesis that relative clauses as objects (the OO type) are more difficult to grasp than those that function as subjects (OS). This claim has been statistically confirmed in the group of 3 – 4.5 year-olds, which had a big difficulty in perceiving the difference between relative clauses functioning as subjects or objects. These findings are in line with claims related to other languages, which demonstrate that children in early childhood, distinguishing between OS or OO, misinterpret the OO type more often (Guasti, Stavrakaki, Arosio 2008). However, when an unusual, strange situation is described, or an unfamiliar verb is used, children tend to assign the same syntactic function to the head noun and the relative pronoun; in such cases the OS clause is interpreted as the OO type. This tendency also supports the parallel-function hypothesis advanced by Tavakolian (1981). In later years children already perceive the difference between the OS and OO type relative clauses, but in a pre-school period misinterpretations of the OO type are quite numerous. Children of schooling age, on the other hand, have no difficulty in interpreting this difference correctly. The results of the present study confirm the assertion that older children grasp the functions of relative clauses and interpret complex structures more easily: they gradually realize that there are two different propositions conveyed in the main and the subordinate clause. It is assumed that children understand and start using relative clauses when their language processing skills have improved, and this happens while they are getting older.In addition to language processing skills, it is important to pay due attention to the frequency of usage principle. Our research findings show that subject relative clauses are more frequent in child-directed speech, and they also appear earlier in a spontaneous child language than object clauses; this is exactly what influences an easier perception of subject relative clauses. Semantic and pragmatic factors have to be mentioned as well: while acquiring a language, children master those grammatical structures where a particular form correlates with a particular meaning more easily (Diessel, Tomassello 2000).It might be assumed that an inflectional system of Lithuanian enables children to easier interpret the syntactic relationship between the main and the relative clause (see, e.g. Arnon 2010); however, the findings of this study allow us to claim that the structure of relative clauses is complex and multilayered, encompassing not only morphological but also semantic and pragmatic factors, which should be investigated more thoroughly in the future. It has also to be stressed that while researching child language it is important to evaluate the influence of adult language, which provides the necessary environment for the child to hear and to understand syntactic constructions of different complexity. Moreover, the assumption that a rich morphological system of a language makes it easier to grasp the complex syntactic functions of sentence elements is corroborated only in part – the findings of the study show that it takes time for Lithuanian children to master correct usage of relative clauses.