Abstract: In their studies of multimodal text,many scholars have been devoted to the study of text-image relations.Some scholars regard verbal and visual as two semantic systems working together.Yet,both have their own specific forms but working together to fulfill their function.To others,text-image relations are similar to the logical-semantic relations between clauses.The relation of text to image can be found in the cohesion of the elements.Learner's English-Chinese dictionaries are a useful tool in English language learning.With the rapid development of modern science and technology,the modes used in learner's English-Chinese dictionaries have changed.Illustrations have been widely used.Following Halliday's(1985,1994) idea about logico-semantic relations and cohesion,and in reference to the models proposed by Martinec(1998), Royce(2007) and Martinec & Salway(2005),this paper is an attempt to explore the relationship between illustrations and definitions in learner's English-Chinese dictionaries,and illustrate,based on statistical analysis,the complementarities of text and image in learner's English-Chinese dictionaries.