ldentification of summer nectar plants contributing to outbreaks of Mythimna separata (Walker) (Lepidoptera: Noctuidae) in North China

College of Plant Protection, Henan Agricultural University, Zhengzhou 450002, P.R.China

1. lntroduction

In the Asian and Australian continents, the oriental armyworm, Mythimna separata (Walker), is one of the most serious pests of cereal crops (Sharma and Davies 1983).Female moths lay few, mostly inviable eggs in the absence of carbohydrate (Quo et al. 1963; Quo and Liu 1964; Han and Gatehouse 1991). This species undertakes a seasonal,long-distance, multi-generational, round-trip migration between southern and northern China, and the larvae feed on wheat, maize, rice and other crops throughout the year.The availability and quality of nectar are important factors affecting armyworm population size (Li 1995). M. separata moths have been reported in spring time to feed on cultivated nectar plants such as peach Amygdalus persica L., plum Prunus salicina Lindl., apricot Armeniaca vulgaris Lam.,apple Malus pumila Mill., orange Citrus reticulata Blanco,and rape Brassica napus L. (Zhao 1988; Lin 1990). Wang et al. (2006) analyzed the relationship between major spring nectar plants and wheat damaged by oriental armyworm and found that in the four years when severe M. separata outbreaks occurred, the acreage of milk vetch was near its peak. Acreage of damaged wheat was related to the acreage of milk vetch during 1950–1992. We suggested that the unprecedented increase in the planting area of milk vetch, Astragalus sinicus, a primary spring nectar plant utilized by adult M. separata migrating to major wheat growing regions, was the key factor responsible for the severe armyworm outbreaks in China in 1966, 1972, 1976 and 1977.

During 1950–1985, the crop most seriously damaged by M. separata was wheat (by an infestation of first-generation larvae), but shifted to maize (by infestations of second- and third-generation larvae) after 1995 (Jiang et al. 2014).The fourth worst armyworm outbreak since 1950 took place in 2012 (Jiang et al. 2014), when a third-generation armyworm infestation decimated maize in northern and northeastern China in early August (Zhang et al. 2012).The summer nectar sources available during late May to early June (the migration period of first-generation adults)and mid-July to late July (migration period of secondgeneration adults) were likely important factors that lead to third-generation armyworm outbreaks. Armyworm moths are attracted to summer nectar plants such as cultivated locust Robinia pseudoacacia L. var. pseudoacacia, false indigo Amorpha fruticosa L., and scallion Allium fistulosum L. (Zhao 1988; Lin 1990). Female moths of M. separata feeding on wild nectar plants such as thistle Cirsium setosum and alfalfa Medicago sp. laid 1 747 and 645 eggs, respectively (Zhao 1988). Though these plants are commonly located in proximity to maize fields, little is known about their associations with adult M. separata summer migrations.

Extensive studies have been conducted to identify the nectar plants for Noctuidae moths. Black cutworm moths,Agrotis ipsilon (Hufnagel), were observed feeding on the blossoms of 20 of 53 plant species in northwestern Missouri,USA, where the greatest number of moths were observed on linden (Tilia spp.) and linden blossoms (Johanne 1991). Corn earworm, Helicoverpa zea; cabbage looper, Trichoplusia ni and celery looper, Anagrapha falcifera were reported to feed on a wide range of plants including Citrus, Salix, Quercus and Pithecellobium spp., based on an analysis of pollen attached to 392 moths captured in southern Texas and southern Oklahoma (Lingren et al. 1993). Two species of Helicoverpa moths were found to be attracted by safflower(Carthamus tinctorius) and cotton flower in Australia (Fitt et al. 1995). Xu et al. (1999) examined pollen attached to 101 moths of cotton bollworm, Helicoverpa armigera, by scanning electronic microscopy (SEM), and proposed that the host of H. armigera moth could be divided into two types:feeding hosts and oviposition hosts. By a combination of DNA barcoding and pollen morphology, Liu et al. (2016)found 28 species of pollen from 18 plant families on the tested moths of A. ipsilon. However, few papers have provided supporting evidence for the importance of nectar plants in pest epidemics.

We conducted this study to identify summer nectar plants for M. separata moths through field investigations, mothborne pollen identification and female ovarian development examination in M. separata from 2014 to 2016. Furthermore,we tried to determine the best of three moth traps used to capture migrating individuals and non-migrating individuals(based on female ovarian stage) which can be useful for monitoring M. separata migrations.

2. Materials and methods

The second-generation armyworm populations were the offspring of individuals that migrated from major wheat growing areas of China including Henan, Shandong,Jiangsu, Anhui, and Shaanxi provinces. According to the results inferred from a trajectory analysis, Pan et al. (2014)suggested that the third-generation armyworm population from northern China was composed of migrants from northwestern regions (Shaanxi, Gansu, Ningxia) (10%),Inner Mongolia (7%), and southwestern regions (Sichuan,Guizhou, Yunnan and Chongqing) (6%), while most (75%)originated from northern China and the adjacent southern region. Central and Northwest Henan Province, the adjacent southern region of North China, was the areas of focus for our field investigation and moth-borne pollen collection.

Using the Chinese Floral Database (http://db.kib.ac.cn/eflora/View/plant/Default.aspx), a search of candidate nectar plants inferred by moth-borne pollen species was conducted to obtain information of geographical distributions and flowering periods of potential candidate nectar plants. Our criteria for considering what is a summer nectar species for M. separata was that one nectar plant can produce sufficient and suitable nectar for female ovarian development, as well as the species having a flowering period from late May to late July, late May to mid-June, or mid-July to late July.

2.1. Field investigation on flowering of primary nectar plant, chaste tree

Nectar plants are divided into two groups in apiculture:primary and secondary. Primary nectar plants can produce sufficient nectar for a honey bee to provide surplus honey,while secondary nectar plants produce only enough nectar for honey bees to live and reproduce (Lin 1989).Chaste tree, Vitex negundo L. var. heterophylla (Franch.),is a primary summer nectar plant that grows on hills and small mountains in North China and Henan Province. We determined its flowering period in 14 counties and districts(Xingyang, Gongyi, Yanshi, Dengfeng, Mengjin, Ruyang,Ruzhou, Lushan, Yichuan, Jiyuan, Mengzhou, Qinyang,Huiji, and Yibin) in Henan Province and in Zezhou County in Shanxi Province from late May to mid-July in 2014 and 2015. The investigations in these 15 counties and districts(33.7–35.5°N, 112.3–113.5°E, 120 to 750 m elevation)were carried out to confirm whether the florescence of chaste tree coincided with the summer migration of oriental armyworm moths. Anthers of V. negundo var. heterophylla were collected from Shouyangshan Hill (34.76°N, 112.78°E,310 m elevation) in Yanshi County, Henan Province and preserved with a standard FAA (Formalin-acetic acidalcohol) solution so that the pollen could be compared with those adhering to M. separata moths collected from three types of traps.

2.2. Nocturnal observations on flower of chaste tree and behavior of feral moths

Adults of M. separata are nocturnal and their behavior around the chaste tree at night has not been reported. It is imaginable that chaste tree flower still open at night and is more convenient for M. separata moths to suck nectar.Observations were conducted on 20 June 2014 to confirm that blooms of chaste tree were available for moth visits between 20:00 and 21:30 at Mang hill (34.95°N, 113.52°E,about 130 m elevation) in the Huiji District of Zhengzhou City. The second site at Qiaowa Village (35.04°N, 112.50°E,about 260 m elevation) in Jiyuan City, Henan Province,was selected as the observation site of floral visitations of M. separata moths because of the high abundance of chaste tree and wheat fields. Nocturnal surveys were conducted using outdoor, rechargeable 10 W LED headlamp and insect net during 19:00–21:00 from early to mid-June 2015.

2.3. Moth collection and pollen identification

In 2015 and 2016, three types of traps were used not only to collect both male and female M. separata moths but also to compare trap effectiveness. The three trap types were blacklight (20 W lamp controlled by a photosensitive switch,Jiaduo Group Co., Ltd., Henan, China), vertical-pointing searchlight (1 000 W metal halide lamp; Shanghai Yaming Co., Ltd., Shanghai, China) and pheromone trap. Pheromone traps, composed of a noctuid trap (Pherobio Technology Co., Ltd., Beijing, China) containing a sex pheromone lure for M. separata (NewCon Inc., Ningbo, China), were fixed at approximately 1.6 m above ground level and lures were replaced monthly. From 2 June to 3 July 2015, armyworm moths were collected from one blacklight trap (placed at about 1.8 m above ground level) and four pheromone traps placed near Dongguolu Village (35.06°N, 112.67°E, about 140 m elevation), Jiyuan City, Henan Province. From 27 May to 25 July 2016, M. separata moths were collected from one vertical-pointing searchlight trap (operating daily from 20:00–04:00), six pheromone traps, and one blacklight trap (placed at about 1.6 m above ground level) in a suburb of Mengzhou City (34.92°N, 112.78°E, about 120 m elevation), Henan Province. The two cities are located in the northwestern part of the Henan Province, where a large number of M. separata moths are commonly observed from late May to mid-June.A limit of 35 moths were collected daily from each trap type due to a low number of workers and the lengthy processing time of pollen detection from each moth. The heads of moths were carefully excised, placed individually into 2-mL plastic tubes, and stored in refrigeration at roughly –20°C.Because the majority of pollen grains are typically found on the proboscises of noctuid moths such as Helicoverpa zea(Bryant et al. 1991), Pseudaletia unipuncta (Hendrix and Showers 1992), and H. armigera (Xu et al. 1999), only the proboscises of M. separata moths were examined under a dissection microscope for the presence or absence of pollen.Forceps (with needle-point 0.07 mm×0.02 mm) used for pollen transfer were cleaned after each proboscis examination to prevent cross contamination. The pollen grains from each proboscis were prepared for SEM and corresponding SEM micrographs were used for comparison with those in previously published research papers (Liu 1985; Zhong 1991)and two monographs (Li et al. 2011; Qiao 2014).

2.4. Determination of ovarian developmental stages in female M. separata moths

Migratory flight is correlated with the development of the ovaries and is curtailed when insects emerge gravid(Johnson 1969). Most immigrant females of M. separata moths caught by traps in the field were sexually mature,while most females in the emigrant population exhibited little or no ovarian development (Han et al. 1990; Jiang and Luo 2005). Thus, origin of this pest species can be determined by the ovarian development stage of females. There are five stages in ovarian development (GB/T 15798-2009 2009): stage I, undiscernible oocytes in ovarioles; stage II,immature oocytes in ovarioles discernible by white color;stage III, mature oocytes in ovarioles discernible by straw yellow color, a lack of distinct gaps among oocytes and no eggs having been laid; stage IV, distinct narrow to broad gaps observable between oocytes and a few to many eggs having been laid; and stage V, most eggs have been laid and few oocytes remain within ovarioles.

2.5. Analysis of distribution and flowering periods of candidate nectar plants

Pollen of closely related plants may have highly similar morphology, and thus should be distinguished using DNA-based techniques (Liu et al. 2016), or by differences in geographical distribution and flowering periods (Xu et al.1999). In our study, if the pollen attached to M. separata moth could only be identified to the genus level, then all species in this genus in China were treated as potential candidate nectar plants. The distribution and flowering periods of these candidate nectar plants were then obtained using the Chinese Floral Database. If flowering of a candidate nectar plant species occurs before mid-May or after August, then that species was rejected as a candidate summer nectar plant for adult M. separata. From late May to early June, some moths migrated northeastward from the first-generation outbreak region to the second-generation outbreak region, north of 39°N, and other moths from the same first-generation outbreak region migrated westward or southwestward to western China (Jiang et al. 2011).Distant regions such as Xinjiang, Qinghai, Xizang, South China and most southern regions of Southwest China(Yunnan and southern Sichuan), are far beyond the origin area. Furthermore, few armyworms were detected in these remote areas during the summer because they were outside of the summer migration pathway of the moths. Therefore,the locally distributed nectar plant species in these regions were not considered as nectar plants for adult M. separata.

2.6. ldentification of summer nectar plants for adult M. separata

Carbohydrates are an integral energetic resource for M. separata reproduction. When newly emerged female moths were fed with distilled water without any carbohydrates,its vitellogenesis was suspended (Cao 1995) and its calling behavior could not occur (unpublished data). Floral nectar is the primary source of carbohydrates and is crucial for M. separata moths in the wild to sustain long flights to reach target habitats and lay eggs. The presence of specific pollen on a moth is the direct result of that moth visiting the flower of that corresponding plant species. However, this plant may or may not have secreted sufficient and suitable nectar for a female to accomplish ovarian development. Among our observed female M. separata moths that carried one specific pollen, if five or more individuals’ ovaries were in each stage I, II and III of development, we considered the corresponding plant species as a suitable summer nectar source for adult M. separata. If many pollen grains were attached to the inner wall of proboscis, it would be inferrible that a large amount of nectar of corresponding plant species was extracted by the moths, the species was considered as a potential suitable summer nectar plant for adult M. separata. The importance of a nectar plant species for armyworm moths also can be evaluated by its relative abundance. The more places one plant species distributes, the more important it is for M. separata to suck nectar.

3. Results

3.1. Flowering date and time of chaste tree and nocturnal behavior of adult M. separata

In the Henan and Shanxi provinces, the flowering date of chaste tree was from late May to mid-July, where the majority of flowers bloomed between late May and mid-June. On trees in the small hills near the plains in Jiyuan City, flowers began to blossom in late May, and most flowers ended blooming in mid-June. In comparison, at the foothills of mountain in Dengfeng City and Lushan County, where the temperature is lower, flowers began to blossom in early to mid-June and a great deal of flowers still blossomed in mid-July. The florescence of chaste tree coincided with the summer migration of oriental armyworm moths. The flowers of chaste tree opened day and night, and their pollen grains bore many tiny holes in the exine (Fig. 1). The M. separata moths in flight, identified through insect net capture, exhibited rapid flight speed and blue light reflection under the LED lamp beam, different from the other moths.We found a total of twelve moths actively flying around chaste trees and two moths resting on V. negundo var.heterophylla near harvested wheat fields at Qiaowa Village in Jiyuan City during 19:00–21:00 from early to mid-June 2015.

3.2. Species of moth-borne pollen

Sixty-eight out of 430 M. separata moths collected in Jiyuan City in 2015 had pollen on their proboscises. Pollen grains were detected on the proboscises of 160 out of 807 moths collected in Mengzhou City in 2016. In these two locations,228 of 1 237 armyworm moths captured by the three types of traps were found to have pollen grains attached to the proboscis (Table 1). A total of fourteen pollen species were identified on M. separata moths (Table 1 and Fig. 2).Thirteen pollen species belonged to secondary nectar plants,while some species of the genus Vitex were revealed to be primary nectar plants. The proportions of armyworm moths carrying pollen of Toona sp., Ligustrum sp. and Syringa sp.were 59.65, 27.19, and 4.39%, respectively (Table 1). The proportions of moths that carried pollen captured in each type of trap: 15.45% of 369 moths trapped by blacklight traps, 17.06% of 557 moths captured by vertical-pointing searchlight traps and 24.44% of 311 moths caught by pheromone traps, respectively (Table 2).

3.3. Summer nectar plants for adult M. separata

Vitex pollen grains, found only on two moths, has a highly similar exine (Fig. 2-J) to that (Fig. 1-B) collected from the anthers of chaste tree which is most abundant species of Vitex. It is very definite possibility that the pollen species is V. negundo var. heterophylla, and chaste tree is a possible summer nectar plant suitable for adult M. separata in 2015,2016.

One hundred and thirty-four out of 136 moths carrying Toona pollen were captured in Mengzhou City (Table 1).The collecting date of adult M. separata carrying Toona sp.was from 27 May to 19 July, 2016 (Table 3). There are four Toona species in China: T. sinensis, T. microcarpa, T. ciliata and T. rubriflora. Chinese toon, T. sinensis, is widely planted in China because its buds are harvested as vegetables.In many regions, Chinese toon cannot blossom due to the frequent bud removal or over-harvesting by humans.Southwestern Henan, northwestern Hubei, and southern Shaanxi provinces, are regions of primary production of Chinese toon seed and also where the first-generation oriental armyworm occur. Furthermore, all these regions were in the summer migration pathway of M. separata moths. The flowering period of Chinese toon, recorded in the Chinese Floral Database, coincides with pollen observations of this study. Although red toon, T. ciliata, blossoms from April to June, partially coinciding with pollen occurrence in this study, few armyworms were detected during the summer in its areas of distribution (Fujian, Hunan, Guangdong,Guangxi, Sichuan and Yunnan). Moreover, the distribution of red toon is outside of the summer migration pathway of M. separata moths. T. microcarpa is an unlikely nectar source during the summer migration of adult M. separata because it blossoms from March to May. The growing area of T. rubriflora, the Fujian Province, is also outside of the summer migration pathway. Thus, we narrowed down the pollen source to T. sinensis for moths captured in Mengzhou City, where we caught 14, 15 and 10 females carrying pollen of T. sinensis with ovaries in stages I, II and III, respectively(Fig. 3). Notably, many T. sinensis pollen grains were present on the proboscises. These results suggest that T. sinensis is an important summer nectar plant suitable for adult M. separata.

Fig. 1 Flower (A, photograph taken at 21:30) and pollen (B) of Vitex negundo L. var. heterophylla (Franch.).

Table 1 Pollen species detected on Mythimna separata and the percentage of moths bearing each pollen species

There are 29 Ligustrum species in China. A comparison between Ligustrum pollen occurrence (Table 3) and florescence period of candidate nectar plants showed that four species of this genus may be the potential source of the pollen collected on moths. They are L. quihoui,L. lucidum, L. molliculum and L. ovalifolium. Seven, zero and three females carrying pollen of Ligustrum sp. had ovaries in stage I, II and III, respectively (Fig. 3). This suggested that Ligustrum sp. may be a summer nectar plant suitable for migrating M. separata. The percentages (and numbers in parentheses) of moths carrying pollen of Ligustrum sp.captured by pheromone, vertical-pointing searchlight, and blacklight traps were 54.84% (34), 16.13% (10), and 29.03%(18), respectively (Fig. 4 and Table 4). In contrast, 20.59,61.03 and 18.38% of moths carrying pollen of T. sinensis(Table 4) were collected by the same three types of traps,respectively. The much higher proportion of Ligustrum sp.pollen than that of T. sinensis was found in pheromone traps.

There are 19 Syringa species in China. The same analysis as performed in Ligustrum showed that four species, S. komarowii, S. reticulata var. mardshurica,S. villosa and S. pekinensis, may be the sources of observed pollen on moths. Syringa pollen was found adhering to three female and seven male armyworm moths. More than twenty Syringa pollen grains were attached to the inner walls of the proboscises of these moths (Fig. 2-A), indicating that a great deal of Syringa nectar was likely drawn by that individual.S. komarowii does not occur in Henan Province (Ding and Wang 1997). Moreover, six out of seven male armyworm moths with Syringa pollen were captured by pheromone traps. This indicates that the pollen found in this study was probably from Syringa plants near pheromone traps. Red clove (S. villosa), white clove (S. reticulata var. mardshurica)and Beijing clove (S. pekinensis) were suggested as possible sources of the pollen collected on moths.

3.4. Origin of female M. separata moth

Fig. 2 Scanning electron microscopy micrographs of pollen grains on the proboscis of Mythimna separata. A, proboscis and Syringa sp. pollen. B, Toona sp. C, Ligustrum sp. D, Syringa sp. E, Pinus sp. F, Delonix sp.? G, Lonicera sp. H, Helianthus sp. I, Artemisia sp. J, Vitex sp. K, Ailanthus sp. L, Betula sp. M, Punica sp. N, Euphorbia sp. O, Eupatorium sp. ? indicates genus could not be confirmed.

Table 2 Percentages of moths with pollen attached to their proboscises out of total numbers captured by the three types of traps

In 2015 and 2016, The highest proportion of females in ovarian stage V was captured in blacklight traps in both Jiyuan and Mengzhou cities. The vertical-pointing searchlight traps in Mengzhou City caught the highest proportion of females in ovarian stage I, whereas proportions of females in stages II, III and IV were successively lower(Fig. 5). The ovarian development of most female moths carrying the pollen of T. sinensis was in early or mid-stages(Fig. 3), suggesting that Mengzhou City is more likely a migration pathway than a recipient area of these mothsin 2016. These results suggest that the vertical-pointing searchlight trap is more effective for monitoring the long distance migratory flight of adult M. separata than the other two types of traps.

Table 3 Collecting dates of adult Mythimna separata carrying Toona, Ligustrum and Syringa pollen

4. Discussion

Fig. 3 Ovarian development of female Mythimna separata carrying pollen of Toona sp., Ligustrum sp. and Syringa sp.Stage I, undiscernible oocytes in ovarioles; stage II, immature oocytes in ovarioles discernible by white color; stage III, mature oocytes in ovarioles discernible by straw yellow color, a lack of distinct gaps among oocytes and no eggs having been laid;stage IV, distinct narrow to broad gaps observable between oocytes and a few to many eggs having been laid; stage V, most eggs have been laid and few oocytes remain within ovarioles.

Activity, especially flight, and oogenesis are competing energy sinks within an insect (Wheeler 1996). The spring migration of M. separata moths was accomplished in a series of night-time movements of 100–300 km instead of a single non-stop flight (Chen et al. 1989). In view of the importance of nectar, Wang et al. (2006) found that milk vetch, Astragalus sinicus, which blossomed during the spring migration of M. separata, was the key factor responsible for severe outbreaks of this pest from 1966 to 1977. The acreage of cultivated, herbaceous milk vetch exhibited unprecedented growth since 1966 and decreased rapidly after 1978. Meanwhile, the acreage of summer nectar plants of mostly woody shrubs or trees on hills and small mountains, was relatively constant (Wang et al. 2014).Unfortunately, the role of these shrubs and trees as nectar sources has not been reported.

Fig. 4 Trap effectiveness for monitoring of armyworm moths carrying pollen of Toona sp., Ligustrum sp. and Syringa sp.

Table 4 Percentages of armyworm moths bearing each of three pollen species of interest captured by the three types of traps

Few papers have provided supporting evidence of the importance of nectar plants in Noctuidae pest epidemics.We investigated the importance of nectar plants by using a combination of field investigations, moth-borne pollen identification and moth abundance of females at various stages of ovarian development in M. separata. During two mass migration periods of oriental armyworm moths,from late May to mid-June and from mid-July to late July,the temperature was often over 25°C. The pre-oviposition period, the window of time for individual moths to migrate,was less than seven days. The ovarian development period of stages I, II and III occur within two days, during which individual moths must visit as many flowers as possible to draw enough nectar for its migratory flight and vitellogenesis. Individual oriental armyworm moths may have a preference of feeding on one nectar plant species and occasionally feeding on more than one species. This may explain why only one pollen species was detected on the majority of trapped moths, 215 out of 228 moths,whereas at most 2 pollen species were detected on 13 out of 228 moths (Table 1). However, it is possible that pollen transfer to a moth’s proboscis was unsuccessful, removed during a visit to a different plant, lost in flight or lost in the course of experimental processing of captured moths.During the pre-oviposition period, if one type of pollen was found on the proboscis of five or more moths in each ovarian stage, I, II and III, and the corresponding plant was a primary or secondary nectar source in apiculture, then the corresponding plant would highly support the probability of vitellogenesis. This evidence supports that this plant serves as a suitable summer nectar source for adult M. separata.

Pheromone traps were commonly used in studies on the migration of Noctuidae moths (Hendrix et al. 1987;Hendrix and Showers 1992) instead of on the identification on pollen speices. Only a low proportion of male moths were found carrying pollen that gave direct evidence of long-distance migration, such as Pithecellomum pollen on eyes of 11 bollworm (Heliothis zea) moths (Hendrix et al.1987) and Calliandra pollen attached to the eyes of eight bollworm moths (Hendrix et al. 1987). Among 5 755 black cutworm and armyworm moths, only 14 moths (0.24%)were found carrying exotic pollens of Pithecellobium spp.or Calliandra spp., or both (Hendrix and Showers 1992). A high proportion of male moths were found carrying pollen of other plant families such as Compositae, Salicaceae, and Onagraceae, which provided important clues to potential nectar species visited by moths before the initiation of longrange movement. In order to identify nectar plant species and evaluate their relative importance, we used three types of traps (blacklight, vertical-pointing searchlight and pheromone) for capturing moths and then SEM to examine and identify the moth-borne pollen. Comparisons of trap effectiveness were conducted to reveal the origin of moths carrying different pollen species. It was supposed that the percentages of moth carrying different pollen species captured by same trap were same or similar, if origin of these moths was the same area and the importance of different corresponding nectars was similar. In this paper,T. sinensis, Ligustrum sp. and Syringa sp. are all secondary nectar plants in apiculture. Much higher proportion of moths carrying pollen of Ligustrum sp. (54.84%) than that of T. sinensis (20.59%) was found in pheromone traps(Table 4). This may be because the origin of moths carrying Ligustrum sp. pollen was originated from plants near Mengzhou City or near Jiyuan City. Privet, L. lucidum, a ubiquitous species in these two cities and also the most abundant plant among these four Ligustrum species in the Henan Province (Ding and Wang 1997), is likely a suitable nectar source for M. separata moths.

There were many pollen grains of the genera Toona,Ligustrum and Syringa adhered to proboscises of moths.The number of pollen grains attached to inner walls of proboscises may be a good indicator of the importance of a nectar species and should be investigated in future studies.

Field investigations on primary nectar plant chaste tree show that florescence of chaste tree coincides with the summer migration of oriental armyworm moths and the flowers of this tree opened day and night. These results seemed to support chaste tree, V. negundo var.heterophylla, as an important nectar plant for M. separata moths. However, Vitex pollen grains, found only on two moths. This may be because most chaste tree appeared to be more like shrubs with heights less than 2 m, and the availability of its nectar was low in most years. From late May to early and mid-June, a period often characterized by little precipitation in this region, we found the nectar of V. negundo var. heterophylla was viscid. We hypothesize that viscid nectar may be favorable to honeybees, but not to Noctuidae moths such as M. separata because of their long, thin proboscis. We surmise that intermittent,moderate to heavy rainfall around the hills and mountains in this region from late May to mid-July is favorable to local nectar plants to produce sufficient, suitable nectar to facilitate mass migrations of M. separata moths, and more importantly, cause severe outbreaks of armyworm. Most flowering Chinese toon, T. sinensis, with a height over 3 m, may have more stronger root system and secrete more nectar for moths to feed on. The effects of rain levels on the growth and flowering of nectar plants need further study in the future.

Migratory flight is continuous in one direction, covers long distances; and in its extreme form, clearly differs from flight near or within a breeding area, which soon end in feeding,mating, or oviposition (Johnson 1969). Accordingly, flights of oriental armyworm moth occurring during the pre-oviposition period is considered migratory, and flights occurring during the oviposition period is regarded as non-migratory. Under vertical-pointing searchlight trap in Mengzhou, the highest proportion of captured females were in ovarian stage I, which suggests that this trap was more effective for monitoring the long-distance migratory flights of adult M. separata than the other trap types.

Herbivore population dynamics are affected profoundly by nectar availability (W?ckers et al. 2007). In China, there are about 10 000 species of nectar plants, among which 85 species are primary nectar plants (Lin 1989). Given that primary summer nectar plants played a key role in severe second- and third-generation armyworm outbreaks, plant species in central, eastern, northern, northwestern, and northeastern China should be regarded as major candidates for nectar sources for M. separata moths if flowering periods coincide with the summer migration of armyworm moths.Identification of nectar plants in natural environments and determining their importance to moth migrations can help us predict population dynamics of armyworm in China. Further studies combining field and laboratory methods should be conducted in the future.

5. Conclusion

The chaste tree appeared a potential important nectar source for M. separata moths in the filed. However, this deduction was not supported by moth-borne pollen identification.Identification of moth-borne pollen according to its SEM micrographs revealed that 228 out of 1 237 armyworm moths captured by blacklight, vertical-pointing searchlight and pheromone traps. The three highest percentages of pollen-bearing armyworm moths were carrying pollen of Toona sinensis (59.65%), Ligustrum lucidum (27.19%) and Syringa sp. (4.39%). Analysis on the relationship between pollen and ovarian development showed that Chinese toon,T. sinensis, is a suitable summer nectar plant for adult M. separata. Privet, L. lucidum, and Syringa sp. are possible suitable summer nectar plants that need further study. The vertical-pointing searchlight trap was the most effective trap for monitoring the migratory flight of adult M. separata.

Acknowledgements

This work was funded by the project under the National Department of Public Benefit Research Foundation of China (201403031). The authors thank Prof. Xue Longyi and Senior Agronomist Sun Hongxia from Jiyuan Plant Protection Station, China, Senior Agronomist Gao Hongyan from Mengzhou Plant Protection Station, China for their help with armyworm moth collection. Special thanks to Dr. Li Jiamei from Henan Agricultural University for his direction on identification of pollen. Special thanks to Dr. Luo Lizhi and Dr. Jiang Xingfu from Chinese Academy of Agricultural Sciences for their direction on our study. Special thanks also to reviewers for their constructive suggestions.

Bryant V M, Pendleton M, Murry R E, Lingren P D, Raulston J R. 1991. Techniques for studying pollen adhering to nectarfeeding corn earworm (Lepidoptera: Noctuidae) moths using scanning electron microscopy. Journal of Economic Entomology, 84, 237–240.

Cao C. 1995. Effect of nutrition and neuro endocrine during adult stage on the vitellogenesis ovary development of the oriental armyworm, Mythimna separata walker. Ph D thesis,China Agricultural University, China. (in Chinese)

Chen R L, Bao X Z, Drake V A, Farrow R A, Wang S Y, Sun Y J, Zhai B P. 1989. Radar observations of the spring migration into northeastern China of the oriental armyworm moth, Mythimna separata, and other insects. Ecological Entomology, 14, 149–162.

Ding B Z, Wang S Y. 1997. Flora of Henan. vol. 3. Henan Science and Technology Press, China. (in Chinese)

Fitt G P, Dillon M L, Hamilton J G. 1995. Spatial dynamics of Helicoverpa populations in Australia: Simulation modeling and empirical studies of adult movement. Computers and Electronics in Agriculture, 13, 177–192.

GB/T 15798-2009. 2009. Rules for Investigation and Forecast of the Armyworm Pseudaletia (Mythimna) separata Walker.Standardization Administration of the People’s Republic of China. (in Chinese)

Han E N, Gatehouse A G. 1991. The effect of adult feeding on female pre-calling period and ovarian development in a migratory moth, the oriental armyworm Mythimna separata(Lepidoptera: Noctuidae). Bulletin of Entomological Research, 81, 395–400.

Han E N, Zheng Z Q, Song Z S. 1990. Difference between the responses of male oriental armyworm Mythimana separata(Walker) to sex pheromone traps. Journal of Nanjing Agricultural University, 13, 54–56. (in Chinese)

Hendrix III W H, Mueller T F, Phillips J R, Davis O K. 1987.Pollen as a indicator of long-distance movement of Heliothis zea (Lepidoptera: Noctuidae). Environmental Entomology,16, 1148–1151.

Hendrix III W H, Showers W B. 1992. Tracing black cutworm and armyworm (Lepidoptera: Noctuidae) northward migration using Pithecellobium and Calliandra pollen. Environmental Entomology, 21, 1092–1096.

Jiang X F, Luo L Z. 2005. Comparison of behavioral and physiological characteristics between the emigrant and immigrant populations of the oriental armyworm, Mythimana separata (Walker). Acta Entomologica Sinica, 48, 61–67.(in Chinese)

Jiang X F, Luo L Z, Zhang L, Sappington T W, Hu Y. 2011.Regulation of migration in Mythimna separata (Walker) in China: A review integrating environmental, physiological,hormonal, genetic, and molecular factors. Environmental Entomology, 40, 516–533.

Jiang Y Y, Li C G, Zeng J, Liu J. 2014. Population dynamics of the armyworm in China: A review of the past 60 years’research. Chinese Journal of Applied Entomology, 51,890–898. (in Chinese)

Johanne W W. 1991. Plant species identified as food sources for adult black cutworm (Lepidoptera: Noctuidae) in Northwestern Missouri. Journal of the Kansas Entomological Society, 64, 381–387.

Johnson C G. 1969. Migration & Dispersal of Insects by Flight.Methuen & Company Limited, London.

Li G B. 1995. Oriental armyworm. In: Institute of Plant Protection, Chinese Academy of Agricultural Sciences,ed., Diseases and Insect Pests of Crops in China. 2nd ed.China Agriculture Press, Beijing. pp. 697–720. (in Chinese)

Li T Q, Cao H J, Kang M S, Zhang Z X, Zhao N, Zhang H. 2011.Pollen Flora of China Woody Plants by SEM. Science Press,China. (in Chinese)

Lin C S. 1990. Physiology and Ecology of Oriental Armyworm.Peking University Press, China. (in Chinese)

Lin S Q. 1989. Nectar Plants. China Forestry Publishing House,Beijing. (in Chinese)

Lingren P D, Bryant V M, Raulston J R, Pendleton M, Westbrook J, Jones G D. 1993. Adult feeding host range and migratory activities of corn earworm, cabbage looper, and celery looper (Lepidoptera: Noctuidae) moths as evidenced by attached pollen. Journal of Economic Entomology, 86,1429–1439.

Liu B L. 1985. Pollen morphology of the family Verbenaceae in China. Bulletin of Botanical Research, 5, 23–62. (in Chinese)

Liu Y Q, Fu X W, Mao L M, Xing Z L, Wu K M. 2016. Host plants identification for adult Agrotis ipsilon, a long-distance migratory insect. International Journal of Molecular Sciences, 17, 851–863.

Pan L, Wu Q L, Chen X, Jiang Y Y, Zeng J, Zhai B P. 2014.The formation of outbreak populations of the 3rd generation of Mythimna separata (Walker) in northern China. Chinese Journal of Applied Entomology, 51, 958–973. (in Chinese)

Qiao B S. 2014. Color Atlas of Air-Borne Pollens and Plants in China. Peking Union Medical College Press, China. (in Chinese)

Quo F, Liu C L. 1964. Studies on the reproduction of the armyworm, Leucania separata Walker (Lepidoptera,Noctuidae). II. Effect of supplementary nutrition on fecundity.Acta Entomologia Sinica, 13, 785–794. (in Chinese)

Quo F, Wu T G, Tsai H L, Liu C L. 1963. Studies on the reproduction of the armyworm, Leucania separata Walker(Lepidoptera, Noctuidae). I. The biological characteristics of adults. Acta Entomologia Sinica, 12, 565–577. (in Chinese)

Sharma H C, Davies J C. 1983. The Oriental Armyworm,Mythimna Separata (Wal.) Distribution, Biology and Control:A Literature Review. Center for Oversea Pest Research,London.

Wang G P, Yang Q, Guo P. 2014. Summer epidemics of oriental armyworm in Henan Province since 2000 and geographical characteristics of outbreak region. In: Insect Research in Central China. China Agricultural Science and Technology Press, China. pp. 175–177. (in Chinese)

Wang G P, Zhang Q W, Ye Z H, Luo L Z. 2006. The role of nectar plants in the severe outbreaks of armyworm Mythimna separata (Walker) (Lepidoptera: Noctuidae) in China. Bulletin of Entomological Research, 96, 445–455.

W?ckers F L, Romeis J, Rijn P V. 2007. Nectar and pollen feeding by insect herbivores and implications for multitrophic interactions. Annual Review of Entomology, 52, 301–323.

Wheeler D. 1996. The role of nourishment in oogenesis. Annual Review of Entomology, 41, 407–431.

Xu G, Guo Y Y, Wu K M. 1999. Analyses of pollen adhering to cotton bollworm moths (Lepidoptera: Noctuidae). Scientia Agricultura Sinica, 32, 63–68. (in Chinese)

Zhang Y H, Zhang Z, Jiang Y Y, Zeng J, Gao Y B, Cheng D F. 2012. Preliminary analysis of the outbreak of the thirdgeneration armyworm Mythimna separata in China in 2012.Plant Protection, 38, 1–8. (in Chinese)

Zhao S J. 1988. Oriental Armyworm and Weather. China Meteorological Press, China. (in Chinese)

Zhong M J. 1991. Studies on the pollen morphology of Syringa L. in China. Journal of Zhongkai Agrotechnical College, 4,8–20. (in Chinese)