Abstract: To reveal the precursory characteristics of the coal and gas outburst in the trigger period,four large model tests were completed and deep analysis of data continuously monitored by microseismic system was carried out. The study concentrates on the events distribution,bispectrum estimation and their evolution over time. The research results are shown as follows. Microseimic signals related with outburst precursory events concentrate on the 6th frequency band whose frequencies are from 1 250 to 1 500 Hz. They mainly distribute on the active period and occur at intervals. Five orders of energy magnitude estimated by bispectrum include 103 ,104 ,105 ,106 and 107 mV3 with obviously normal dis- tribution of the related events. The microseimics of outburst precursory events which really control the developing trends in the trigger period are the events with 107 mV3 of energy magnitude order. Their always exists an adjusting pe- riod before coal and gas outburst. They are characterized as intensive events with low energy and a short time of inac- tive period without any microseismic events. The longer the adjusting period,the longer the outburst lasts and the high- er the outburst intensity is. The intensified precursory events with low energy and inactive period are the important in- formation to early warn the coal and gas outburst.