Summary Citrus huanglongbing (HLB) is the most devastating citrus disease worldwide. ‘Candidatus Liberibacter asiaticus’ (Las) is the most prevalent HLB causal agent that is yet to be cultured. Here, we analysed the flagellar genes of Las and Rhizobiaceae and observed two characteristics unique to the flagellar proteins of Las: (i) a shorter primary structure of the rod capping protein FlgJ than other Rhizobiaceae bacteria and (ii) Las contains only one flagellin‐encoding gene flaA (CLIBASIA_02090), whereas other Rhizobiaceae species carry at least three flagellin‐encoding genes. Only flgJAtu but not flgJLas restored the swimming motility of Agrobacterium tumefaciens flgJ mutant. Pull‐down assays demonstrated that FlgJLas interacts with FlgB but not with FliE. Ectopic expression of flaALas in A. tumefaciens mutants restored the swimming motility of ∆flaA mutant and ∆flaAD mutant, but not that of the null mutant ∆flaABCD. No flagellum was observed for Las in citrus and dodder. The expression of flagellar genes was higher in psyllids than in planta. In addition, western blotting using flagellin‐specific antibody indicates that Las expresses flagellin protein in psyllids, but not in planta. The flagellar features of Las in planta suggest that Las movement in the phloem is not mediated by flagella. We also characterized the movement of Las after psyllid transmission into young flush. Our data support a model that Las remains inside young flush after psyllid transmission and before the flush matures. The delayed movement of Las out of young flush after psyllid transmission provides opportunities for targeted treatment of young flush for HLB control.

Citrus flies are important quarantine pests in citrus plantations. Electronic traps (e-traps) based on computer vision are the most popular types of equipment for monitoring them. However, most current e-traps are inefficient and unreliable due to requiring manual operations and lack of reliable detection and identification algorithms of citrus fly images. To address these problems, this paper presents a monitoring scheme based on automatic e-traps and novel recognition algorithms. In this scheme, the prototype of an automatic motor-driven e-trap is firstly designed based on a yellow sticky trap. A motor autocontrol algorithm based on Local Binary Pattern (LBP) image analysis is proposed to automatically replace attractants in the e-trap for long-acting work. Furthermore, for efficient and reliable statistics of captured citrus flies, based on the differences between two successive sampling images of the e-trap, a simple and effective detection algorithm is presented to continuously detect the newly captured citrus flies from the collected images of the e-trap. Moreover, a Multi-Attention and Multi-Part convolutional neural Network (MAMPNet) is proposed to exploit discriminative local features of citrus fly images to recognize the citrus flies in the images. Finally, extensive simulation experiments validate the feasibility and efficiency of the designed e-trap prototype and its autocontrol algorithm, as well as the reliability and effectiveness of the proposed detection and recognition algorithms for citrus flies.