After chilling, the eggs were transferred to 13.1°C for 14days to stimulate hatching. This protocol has previously been used successfully to stimulate hatching of N. filicollis eggs (Van Dijk and Morgan, 2009; Oliver et al., 2014, 2016) and so there is no reason to suspect that in the current study eggs did not receive sufficient heat stimulus to hatch. However, in the absence of a compelling alternative explanation for so few eggs hatching despite long periods of chilling, this remains a plausible explanation. Alternatively, it is possible that many of the eggs were not viable despite a visual assessment indicating that many L3 were moving within the unhatched eggs. Due to incubator malfunctions prior to the commencement of the current study the eggs were held at 20°C for longer than planned (i.e. 116 days) prior to chilling, and some were then chilled for as long as 224 days. In contrast, Van Dijk and Morgan (2009) incubated their N. filicollis eggs at 20°C for only 42days prior to chilling. It is feasible that these extended time periods may have resulted in depletion of procollagen c proteinase
reserves in the developed L3 such that they were unable to break out of the eggs once they were returned to 13.1°C. This would explain the low hatching percentage despite the larvae being mobile within the eggs. However, further experimentation will be required to determine whether either of these possible explanations will account for the low egg hatching percentage in this study.
The concept of amalgamating time and temperature into chill units has been used previously in plant science (Alburquerque et al., 2008; Rose and Cameron, 2009; Kuden et al., 2013) and is analogous to the degree day concept often used in invertebrate phenology studies (Nealis et al., 1984; Lobinske et al., 2002; Lewis et al., 2003). The application of chill units in the current study implies that eggs accumulate chilling (i.e. progress towards meeting the requirements for hatching) at a rate determined by temperature. This assumption is supported by the current results in that hatching commenced earlier when eggs were chilled at lower temperatures (Fig. 1) and different combinations of temperature and time (i.e. CU) resulted in similar hatching percentages (Fig. 2).
However, the results also indicate that there is not a single constant chill accumulation threshold which applies to all eggs. A proportion of eggs hatched with relatively little chilling and this proportion continued to increase with the CU accumulation up to a maximum of about 1000 CU. This suggests that within the N. filicollis population there is considerable biological plasticity in terms of chilling requirements, with some individuals requiring relatively little while others require extensive chilling before they will hatch. This could be an adaptation
to survival in a changeable environment, where hatching of a single cohort of eggs is spread over long time periods, potentially several years. Hence, if one season is adverse for larval survival outside the egg a proportion of the cohort is retained until subsequent seasons, thereby spreading the risk of localised extinction. This is similar to the bet hedging behaviour of N. battus described by Van Dijk and Morgan (2008, 2010). Based on the fitted relationship, it would require about 1000days at 10°C to reach maximum egg hatch whereas at 2°C this would be achieved in 111 days. An anomaly in the data is that while there was consistently low egg hatch after storage of eggs at 8°C, some of the samples stored at 10°C showed relatively high egg hatch (Fig. 1). This is out of context and it is unclear why more eggs would hatch after storage at 10°C than at 8°C. It was, however, noted that there appeared to be more fungal contamination in some of the wells exposed to 8°C and this
may have influenced the number of larvae hatching.
Extrapolating these results into the field suggests that while some eggs will hatch after limited chilling, this New Zealand isolate of N. filicollis requires a large number of chill units in order to achieve maximum levels of egg hatching. In some years this could mean that more than one season of chilling would be required to accumulate enough chill u