Seed storage behaviour of 101 woody species from the tropical rainforest of southern China: a test of the seed-coat ratio-seed mass (SCR-SM) model for determination of desiccation sensitivity

The Xishuangbanna tropical rainforest in Yunnan Province is the greatest biodiversity hotspot in China. However, the biodiversity of this region is under threat, making seed conservation through seed and/or germplasm banking particularly urgent and crucial. Seed desiccation sensitivity limits the possibility of seed banking of 47% of tropical rainforest species. Thus, knowing if a species has desiccation-sensitive seeds is an important first step in seed banking; however, often resources are limited, making it difficult to determine storage behaviour for all the species in a region. Prediction of seed sensitivity using the SCR-SM model based on seed-coat ratio (SCR) and seed dry mass (SM) might be an alternative for determining desiccation sensitivity of seeds of each species. Here, seed-desiccation sensitivity of 101 woody species from the Xishuangbanna tropical forest were analysed using this model, and physiological determinations were made for a total of 25 species. Seed storage behaviour for 59 species was used for model validation, and storage behaviour of 88% of these species was successfully predicted. Seed storage behaviour of 83% of the 59 species was successfully predicted using the 1000-seed weigth-moisture content (TSW-MC) criteria, which include seeds with 1000-seed weight >500 g and seed moisture content at shedding of 30-70%. The two predictive methods were subsequently used to predict seed desiccation sensitivity for another 42 species from Xishuangbanna whose storage behaviour was uncertain. Our results indicated that similar to 50% of the species in Xishuangbanna are likely to have desiccation-sensitive seeds.