This study presents a longitudinal analysis of three monolingual Iranian children's language development between ages 1;8-2;6, 2;2-3;2 and 2;4-3;4. The overall aims are to identify and establish the structural patterns in the acquisition of Persian, a pro-drop, inflectional and mostly verb final language. Structural patterns particular to Persian are identified in contrast to English and data drawn from the children's language progress are discussed in the light of recent theories of language acquisition. In addition, the study provides a comprehensive and systematic description of children's syntactic development in such a way as to be useful for clinical data analysis by Iranian speech and language therapists and includes some cross-linguistic comparisons with other research on language acquisition. The applicability of MLU (Mean Length of Utterance) measures to Persian is investigated and it is found that MLU measured in morphemes is most appropriate for evaluating the Iranian children's early language development up to value 4. In order to give a more detailed analysis of the children's language acquisition, the LARSP (Language Assessment Remediation and Screening Procedure) framework (Crystal, Fletcher and Garman, 1989) is adapted to Persian. Analysing Persian data with LARSP categories shows that there are many features common to both languages. Particular categories are identified. A PLARSP (Persian LARSP) profile is established based on the hypothesis that structures can be assigned to stages according to their number of elements at clause and phrase levels. The profile provides a framework for the analysis of language development in Persian and is employed in chapters 6 and 7 to set out the developmental picture of the children's language at approximately equal MLU values in the early stages, and age in the later stages. Close examination of the data points to the use of formulas by the children at early stages. Apart from the formulas, although the children show different strategies of language acquisition, the resulting distribution of categories is found to fit the data well, presenting an orderly progress down the chart according to MLU and age.