To screen the composite components of the plant volatiles that effectively attract Batocera horsfieldi (Hope) (Coleoptera: Cerambycidae) adults, GC-MS (Gas Chromatography – Mass Spectrometry) was used to analyze and identify the type and relative content of the volatiles of the host plants of B. horsfieldi adults: Viburnum awabuki K., Betula luminifera H. Winkl, Juglans regia L. and Populus tomentosa Carr. We measured the EAG (Electroantennogram) responses of the different antennal segments of unmated female and male B. horsfieldi adults to seven host plant volatiles (β-myrcene, (Z)-3-hexen-1-ol, nonanal, dichloromethyl ether, salicyaldehyde, 3-methyl-butanoic acid and trichloroethylene) and to their composite components. A Y-tube olfactometer was used to test the insects' behavioral responses. The results of GC-MS showed that the four host plants of B. horsfieldi had different chemical fingerprints. The EAG test results showed that the responses in the apical part of the unmated female and male B. horsfieldi antennae were stronger than those in the middle part and basal part (P < 0.01). For the seven volatiles, with five different concentrations (of 0.0001, 0.001, 0.01, 0.1 and 0.2 µL/µL), the unmated female B. horsfieldi had the strongest EAG responses to 3-methylbutyric acid of 0.0001 µL/µL and the unmated female and male B. horsfieldi had increasing EAG responses to salicyaldehyde after the concentration reached 0.1 µL/µL. Among the five composite components of plant volatiles, the one mixed according to a ratio of (Z)-3-hexen-1-ol: nonanal: trichloroethylene: β-myrcene (Z∶N∶T∶M)  =  57∶15∶17∶11 caused the strongest EAG response, and that response was distinctly higher than for the other four composite components (P < 0.01). The behavioral response results tested by Y-tube olfactometer were in accordance with the EAG response results. The composite component mixed according to the ratio of Z∶N∶T∶M  =  57∶15∶17∶11 had a stronger attractive effect for the unmated female and male B. horsfieldi adults.