Abstract: Convergent evolution has been long recognized, although it may be difficult to determine exactly what had happened in a particular case. For the purpose of phylogenetic classification, we would like to detect convergences from resemblances among groups. Analyzing habitat data of Chinese Liliaceous plants given in Flora Reipublicae Popularis Sinicae vols. 14 and 15, the authors has discovered a fact that the fruit character, capsule or berry is usually correlated with ecological preferences of plants in the group. The genera with capsule. e. g. Tofieldia, Chionographis, Heloniopsis, Ypsilandra, Veratrum, Tricyrtis, Eremurus, Anemarrhena, Chlorophytum, Diuran-thera, Thysanotus, Hosta, Hemerocallis, Gagea, Lloydia, Tulipa, Fritillaria, Lilium, Nomocharis, Notholirion, Allium, Aletris have their members mostly growing in open places, such as grassland, meadow, thicket, forest margin …… etc. and the members of the genera with berry, e. g. Tupistra, Aspidastra, Clintonia, Smilacina, Maianthemum, Disporum, Streptopus, Polygonatum, Disporopsis, Trillium, Smilax, Heterosmilax are mostly growing in forest. Liriope, Ophiopogon, Peliosanthes and Paris sect. Euthya, though mostly (Liriope?) inhabit forest, have their seeds cha-racteristic of brilliant greenish blue or red fleshy seed coat, and exposed after dehiscence of capsule. In this way, such berry-like seeds are clearly involved in the correlation with environment and achieve the dispersal advantages exhibited by berry. The reason why the forest-living genus Petrosavia has follicle-like capsule is not clear. Its saprophytic or mycotro-phic character is likely to correlate with forest habitat. Utech(1979) has carefully studied floral vascular anatomy of Theropogon and Convallaria of tribe Convallarieae and found that the two genera are similar in berry fruit but different in pattern of vascularization. He comes to the conclusion that this similarity is the result of convergent evolution. Considering Utech's conclusion as well as the correlation between habitat and fruit character pointed out in this paper it may be argued with some certaint that the convergence had happened in fruit character. So we consider that whether the character of berry or capsule can be used as a tribal taxonomic character as in most Liliaceous classification system should be given a further closer scruting in each case.