Parapolybia wasps were collected in Japan from October 2009 to April 2010 in the forest park of Mito city, Mito (36°25'N, 140°22'E, alt. 130 m, site A), Ibaraki Botanical Garden, Naka (36°29'N, 140°26'E, alt. 60 m, site B), Hitachi Ohmiya (36°39'N, 140°16'E, alt. 160 m, site C), the Seminar House of Ibaraki University, Daigo (36°49'N, 140°23'E, alt. 190 m, site D), Ibaraki Prefecture; in October 2012 in Rissho University, Kumagaya (36°06'N, 139°22'E, alt. 50 m, site E), and Nourin park, Fukaya (36°06'N, 139°17'E, alt. 90 m, site F), Saitama Prefecture. Wasps were collected in the following four phases: phase I) the pre-hibernating phase, which occurs shortly after mating and colony break up, in which females temporarily gather at a nest (possibly at their natal nest) or on plant leaves from late August to mid-October (sites D, E, and F); phase II) the early hibernating phase, in which females alternate between hibernacula from late October to mid-November (sites A and D); phase III) the late hibernating phase, in which females remain in the hibernaculum from mid-January to early February (sites A, B, and C); and phase IV) the precolony founding phase, which occurs shortly after emergence of gynes from hibernation in April (sites A and B).

Most overwintering wasps were found in or near hollows of broadleaf trees such as Quercus serrata Murray, Quercus acutissima Carruthers, or Prunus jamasakura Siebold, except for cluster no. A-1, which was occurred under wooden stairs in the park. Wasps resting around the entrances of the holes were collected with forceps during the phase II to IV at sites A and B. For wasps deep inside a tree hollow, the tree was cut off or the entrance of the hollow was bored out, and wasps were extracted with long forceps and/or a wire during phase II at site D and during phase III at sites A and B. The collected wasps were transported in a cool box with ice and stored in that box until dissection later the same day.

The metasomas were removed from live wasps, or wasps killed by 60% ethanol, and dissection was performed on a glass Petri dish under a binocular microscope. The metasomas were examined for the presence of nematodes. Parasitic nematodes were categorized into the following three stages: juveniles (small individuals with of indeterminate sex); uteria (mature adult females with developed uterine sacs); and infective forms (adult females without such sacs) (Fig. 1). Voucher specimens of nematodes and wasps were deposited in the Natural History Collection of Ibaraki University.

In addition, wasps were examined for the presence of fat bodies, sperm in the spermatheca, and the degree of wing wear. Wasps with fat bodies, sperm in the spermatheca, and a lack of wing wear were classified as gynes, whereas individuals lacking fat bodies or sperm in the spermatheca, and with worn wings were classified as workers.

For Parapolybia indica, three pre-hibernating clusters (gathered at the natal nest) in groups of 14, 16 and 99 (129 total individuals) were collected. And an additional 551 wasps were collected in phases II-IV from 23 additional hibernacula in the phase II–IV (Table 1). Solitary hibernating wasps were found at sites A, B, C, and D (n = 5). The number of individuals found in a hibernating cluster varied from 2 to 192 (n = 18; mean ± SD = 30.44 ± 44.98, median = 11).

After dissection, thirteen individuals of Parapolybia indica were classified as workers (site A, n = 4; site D, n = 1; site E, n = 7; site F, n = 1). They were collected in phases I-II and had no parasitic nematodes in their metasomas. Seven wasps collected before the phase III had no fat bodies or sperm in the spermatheca, and their wings were slightly damaged (site A, n = 5; site D, n = 2). For such samples, the wings may have been damaged during collection, and/or they were recently emerged workers. The other Parapolybia indica (n = 660) and Parapolybia varia (n = 102) had fat bodies, sperm in the spermatheca, and a lack of wing wear and were classified as gynes.

The total number of individuals infected by parasitic nematodes was 42. Such individuals were collected from their hibernacula during phases other than the phase I (Table 1). No phase I wasps were found to be nematode infected. All infected individuals were Parapolybia indica and were clearly classified as gynes. The prevalence (%) of parasitic nematodes in each overwintering cluster varied from 3.5 to 50.0 (median = 10.5). No statistical relationship was found between the cluster size and prevalence (correlation efficient r = -0.14, R ver.2.15.2, R Core Team 2012). The prevalence (% of infected clusters) increased through the season: for each phase, 0.0 in phase I, 6.8 in phase II, 8.2 in phase III, and 10.4 in phase IV. The overall percentage of nematode infected wasps was 6.2% (Table 1).

A single overwintering cluster of Parapolybia varia was found at site D. It was without nematode infection. This cluster was found in the same tunnel as a cluster of Parapolybia indica. However, each cluster was monospecific and no species mixing was observed in the tunnel. Thus, in this study, only individuals of Parapolybia indica were infected by nematodes.

Parasitic nematodes were found only in the hemocoel. The stages of nematodes found in the wasp metasomas varied based on the phases of the wasps at collection. Infected wasps collected from phase II had a large number of juveniles and one or two uteria in their abdomen. Uteria and infective forms (Fig. 1) were found in the metasomas of wasps from phase III clusters. Infected wasps from phase IV contained only uteria.