Resource quality can have direct or indirect effects on female oviposition choice, offspring growth and survival, and ultimately on body size and sex ratio. We examined these patterns in Sirex noctilio Fabricus, the globally invasive European pine woodwasp, in South African Pinus patula plantations. We studied how tree position as well as natural variation in biotic and abiotic factors influenced sex-specific density, larval size, tunnel length, male proportion, and survival across development. Twenty infested trees divided into top, middle, and bottom sections were sampled at three time points during larval development. We measured moisture content, bluestain fungal colonization, and co-occurring insect density and counted, measured, and sexed all immature wasps. A subset of larval tunnels was measured to assess tunnel length and resource use efficiency (tunnel length as a function of immature wasp size).  Wasp density increased from the bottoms to the tops of trees for both males and females. However, the largest individuals and the longest tunnels were found in bottom sections. Male bias was strong (~10:1) and likewise differed among sections, with the highest proportion in the middle and top sections. Sex ratios became more strongly male biased due to high female mortality, especially in top and middle sections. Biotic and abiotic factors such as colonization by Diplodia sapinea, weevil (Pissodes sp.) density, and wood moisture explained modest residual variation in our primary mixed effects models (0-22%). These findings contribute to a more comprehensive understanding of sex-specific resource quality for S. noctilio and of how variation in key biotic and abiotic factors can influence body size, sex ratio, and survival in this economically important woodwasp.

Trees were felled and three logs (top middle, and bottom) collected from two P. patula pulp plantations in Mpumalanga, South Africa where moderate to high densities of S. noctilio-infested trees were known to be present. To locate trees infested by S. noctilio, we used the symptoms known to correlate with infestation, specifically the presence of resin drops and yellow or red foliage color (Talbot 1977, Dodds et al. 2010). At three time points throughout the S. noctilio larval life cycle, ten random S. noctilio-infested P. patula trees were selected from each of the two sites. In the sampled area, wasp adults emerge and lay eggs from late October/early November and sampling dates were selected to represent the early (March 2012), late (September 2012), and mid (June 2013) developmental stages. Since the mid-development larval stage was taken from a different larval cohort, only the early and late stages were used to estimate sex-specific patterns in survival and size distribution. Each main stem was visually divided into thirds and a 90 cm section was cut from the approximate middle of each third to represent the bottom, middle and top of each tree. Trees that showed no evidence of active tunneling on the cross-sectional face (top or bottom) of at least one cut log section were discarded and a new tree selected. Diameters of the top and bottom of each log were measured and wood volume calculated using mean log diameter, assuming a perfect cylinder. Four moisture measurements from each log were taken immediately after felling using a Delmhorst RDM-3 moisture probe (species setting: P. radiata [P. patula was not available]) and averaged to represent log moisture. To prevent wasp development prior to log dissection, logs were stored at 4°C at the Forestry and Agricultural Biotechnology Institute (FABI), University of Pretoria.

Tunnel sampling and larval extraction

Using a hand-operated hydraulic MAC-AFRIC™ 10-ton hydraulic log splitter, logs were carefully split into small fragments (~200 mm long by 5 mm wide) to extract all S. noctilio larvae and pupae. Up to eight randomly selected tunnels per log were meticulously traced back to their origin at the phloem-xylem interface and measured for length and maximum width. Only tunnels within the middle two thirds of log length (60 cm) were selected for measurement to exclude tunnels that continued outside the sampled area and so could not be fully measured. Tunnels were measured as they were uncovered during splitting using digital calipers; lengths were summed across split fragments to provide a total length for each selected gallery. All measured tunnels were separated by at least 10 cm throughout their length. All larvae and pupae were removed from the log, counted, measured, and sexed. Larval sex was determined by the presence of hypopleural organs in cuticular folds on either side of female larvae between the first and second abdominal segments (Gilmour 1965). Also called mycetangia (reflecting their function in collection and storage of arthrospores of the fungal symbiont, A. areolatum) these organs are absent in the first instar. Since all larvae encountered in this study were second instar or greater, this did not represent a source of error. Pupal sex was determined by the presence or absence of an ovipositor, visible through the pupal skin. Measurements included length (from the tip of the head capsule to the tip of the tail spine at the end of the abdomen), head capsule width, and wet and dry body mass. In addition, ten larvae per log section were randomly selected and dissected to confirm the presence or absence of the parasitic nematode, Deladenus siricidicola (Bedding and Akhurst 1974) used as a biocontrol agent in South Africa and elsewhere. Lastly, the number of dead larvae as well as identity and abundance of other insects in the log sections were recorded.

Note:  27 logs did not have S. noctilio and will not have rows corresponding to these logs in the "Immatures" table.