The antennal morphology and sensilla ultrastructure of Tomicus yunnanensis, T. minor, and T. brevipilosus were studied by scanning electron microscopy and transmission electron microscopy. Eight common sensilla types were recorded: (1) sensilla trichodea (S.tr.) types 1 and 2 were located on the club and were innervated by five and eight dendrites, respectively; (2) sensilla chaetica (S.ch.) types 1 and 2 had no dendrites in the sensilla lymph lumen; (3) sensilla basiconica (S.b.), top-protuberated S.b. and fluted cones (Fl.c.) occurred on the club; and Böhm bristles (B.b.) occurred on the funicle. S.b. were the most abundant and were innervated by 10-14 dendritic branches. Top-protuberated S.b., a new sensilla type, were innervated by one dendrite. Fl.c. were innervated by five dendrites. Only three sensilla furcatea were visible on the antennae of female T. yunnanensis. The possible functions of these sensilla are discussed in relation to their morphology and ultrastructure. No statistical differences between sexes were found in the size and numbers of each sensilla type. Although the three species had similar antennal morphology and sensilla type, sensilla on sub-segments 3 and 4 of the antennal club of T. minor were much sparser than those of T. yunnanensis or T. brevipilosus. Concerning the antennae of T. yunnanensis, there were more S.tr. type 2, S.ch. type 2 and S.b. and the size of S.tr. type 1 and S.b. were significantly greater than those of T. minor.

Summary Research on tree response to pest attacks under drought and heat stress is crucial for predicting the effect of climate change on forest pests and tree health. In this work, we studied the influence of severe water stress on the resistance of Pinus yunnanensis to inoculations of Grosmannia yunnanensis, a pathogenic fungus associated with the aggressive bark beetle, Tomicus yunnanensis. Experiments were performed in two plots, one located at the top of a hill and the other at the foot of a hill, in Xiaojiu, Yunnan, China, from May to September 2010 and March to April 2012. Measurements of soil and needle water content and predawn needle water potentials confirmed that severe droughts occurred during these periods. Maximum water stress intensity even occurred in 2010 during the typical rainy season. Following low‐density inoculations, the reaction zone length and fungal growth were measured in the phloem. Sapwood blue staining and total sapwood occlusion were quantified following fungal mass inoculations. Our results suggest that severe water stress has a negative impact on a tree's resistance to G. yunnanensis. By comparing our results to previous work performed from 1997 to 1999 on the same tree species, and in the same locality under mild water stress, we concluded that drought had opposite effects on tree resistance above and below a certain threshold of predawn needle water potential. The threshold, corresponded to maximum efficiency for tree defence mechanisms, was visible in our results, and was estimated to be approximately ‐1.1 MPa to ‐1.2 MPa for P. yunnanensis under local conditions. We suggest that the existence of such a threshold of water stress intensity may be viable for any conifer undergoing biotic aggressions and, specifically, bark beetle attacks.

Abstract-Along with the development of modern radar technology, the radar signal system is more and more complicated. This makes the description way of the existing radar signal feature cannot effectively express and analysis it's characteristic. Therefore, in this kind of situation, a description method of radar emitter signal feature based on kernel density estimation is proposed. This method takes the probability density curve from kernel density estimation algorithm as radar emitter signal parameters characteristics stored into database, and on this basis a radar emitter signal recognition algorithm based on template matching is proposed.