Abstract  [Aim]  To explore the differences in the types and contents of volatile compounds in pear and rapeseed flowers at the same flowering stage, and to identify the volatile compounds that influence bee host plant preference. The findings of this study will facilitate the development of inducers that can stimulate bees to pollinate pear trees, thereby providing a guide for the selective screening of bee species specialized for pear pollination. [Methods]  Headspace solid phase microextraction (SPME) and gas chromatograph-mass spectrometer (GC-MS) were used to analyze the volatile compounds in the flowers of Pyrus bretschneideri cv. Dangshansu and Brassica napus at different flowering stages. Using electroantennography (EAG) we isolated the volatile compounds from pear and rapeseed flowers that generated significant responses from Apis cerana and

A. mellifera ligustica. A Y-tube olfactometer was then used to verify the volatile compounds A. cerana cerana and A. mellifera ligustica showed significant foraging preferences towards. [Results]  A total of 52 volatile compounds were identified in the big bud stage, initial bloom stage, and full bloom stage of pear flower. A total of 61 of volatile compounds were identified from the three flowering stages of rapeseed flower. There were 12 shared volatile compounds in the three flowering stages of pear and rapeseed. In the three flowering stages of the pear, the relative content of acetoin and 1,3-di-tert-butylbenzene exceeded 9%. Additionally, the relative content of benzoic acid was significantly different between the three flowering stages (P<0.05). There were significant difference in the relative contents of acetoin, 1-octanol, 1,3-di-tert-butylbenzene, and kodaflex txib between the three flowering stages of rapeseed (P<0.05). An analysis of the common volatile compounds found in pear and rapeseed flowers at identical flowering stages showed there was a significant difference in the relative content of acetoin between the three flowering stages (P<0.01). Compared to paraffin oil, the antennae of A. cerana cerana and A. mellifera ligustica produced electrical signal responses to the 71 tested volatile compounds. Of these, A. cerana cerana exhibited the strong responses to acetic acid, propionic acid, 3-methyl-1-butanol, 1-penten-3-ol and pentanoic acid, 4-methyl-, ethyl ester, while A. mellifera ligustica showed strong responses to acetic acid, propionic acid, 2-methyl- propanoic acid, 2-isobutoxyethanol and pentanoic acid, 4-methyl-, ethyl ester. There was a significant difference in the relative response values of A. cerana cerana and A. mellifera ligustica antennae to 22 volatile compounds (P<0.05). EAG experiments showed that bees exhibited significantly higher response values to benzyl alcohol, diisobutyl phthalate, methyl 2-hydroxy-3-methylvalerate, octanoic acid, 3-methyl-1-butanol, 2-methylbutanol, linalool, undecane, dodecane, pentadecane, and hexadecane compared to

A. mellifera ligustica. Additionally, the response values of ethyl 3-phenylpropionateb, 4-methoxy ethylbenzoate and 1-penten-3-ol were also significantly higher than those of A. mellifera ligustica (P<0.01). The relative response values of

A. mellifera ligustica to 1,3-di-tert-butylbenzene, octanoic acid, ethyl ester, acetoin, propionic acid, hexanoic acid, and dimethyl disulfide were significantly higher than those of A. cerana cerana (P<0.05). Furthermore, the response values of

A. mellifera ligustica to 2-methyl-propanoic acid, 2-isobutoxyethanol were also significantly higher than those of A. cerana cerana (P<0.01). In the Y-shaped olfactometer experiment, the selection rates of phenethyl alcohol, methyl benzoate, longifolene, and linalool by A. cerana cerana were significantly higher compared to the control group (P<0.05). Additionally, the selection rate of dimethyl trisulfide and acetic acid was significantly lower than that of the control group (P<0.05). The selection rates of benzoic acid, methyl ester, 6-methyl-5-hepten-2-one, α-farnesene, 6,10,14-trimethyl-2- pentadecanone, ethyl-methylbutyrate, and 3-methyl-1-butanol in A. mellifera ligustica were significantly higher compared to the control group (P<0.05). Whereas, the selection rates of dimethyl disulfide, acetoin, acetic acid, 2-methyl-butanoic acid, 1-octene-3-ol, and 1-octanol were significantly lower than the control group (P<0.05). [Conclusion]  The acetoin released from pear flowers has a repelling effect on honeybees. In contrast, benzoic acid and methyl ester released from rapeseed flowers have an attractive effect on honeybees. This could be a key factor determining why bees display a preference for rapeseed flowers when pear and rapeseed flower simultaneously.