Pollination network results from the interactions between plants and their pollinators. Asymmetric specialization characterizes symbiotic interaction, and it is regarded among the structural characters of the network. To date, important progress has been made in recognizing asymmetry of the networks, but there was hardly any progress about the geographic variation and asymmetric specialization of the network as well as about the effects of plant life-form on the asymmetric specialization. Here, we summarized a number of concepts used for asymmetric specialization according to available literature and analyzed the geographic variation pattern of the networks as well as the relation of asymmetric specialization of the network to plant life-form and network size using linear regression based on data collected from published papers. We found that the network size is positively linearly correlated with network interaction and that it expressed an exponential increase with total richness of plant species. For all 25 networks investigated, the slope of linear regression (L_b) of the network ranged from 0.002 to 0.031 and it was negatively significantly correlated with plant richness (P), pollinator richness (A), species richness (R), interactions (I), and network size (M). L_b slightly decreased with altitude (P=0.053 for the partial effect of altitude), whereas no effect of latitude (P=0.618) was observed. The correlation coefficients between pollinators and herbs, trees, and shrubs were -0.197, -0.026, and 0.200, respectively, indicating that asymmetric degree was higher in herbs than in trees. This paper provides a novel method to investigate the asymmetry of plant-pollinator interaction network as well as an insight into the relationship between asymmetric degree and network size, latitude, and plant life-form.