Collocations are clusters of words that are acquired together and are subject to constraints in co-occurrence with their adjacent words. The inadequate acquaintance with collocations emerges into the formation of unacceptable collocations from the viewpoint of native speakers. The present study is a descriptive quantitative study of the translation of collocations in literary texts from English into Persin. The study sought to identify the most frequent types of unacceptable collocations in the Persian translated versions. For the purpose of the study, the four Persian translated versions of Hemingway’s “For Whom the Bell Tolls” were investigated precisely and all the unacceptable collocations were compiled and clustered into English patterns based on Benson’s theory, in an aim to decode the groups of patterns which are most frequently leading to bearing of unacceptable collocations in translation of English texts to Persian. A detailed SPSS analysis was conducted and the findings including frequency and percentage of each type of the unacceptable collocations were recorded. The most frequent types of unacceptable collocations spotted in the translated versions of the novel were as follows: 1. Adjective + Noun 23.3%, 2. Subject + Verb 11%, 3. Verb + Object 10.3% patterns.