Multigenerational effects (often called maternal effects) are components of the offspring phenotype that result from the parental phenotype and the parental environment as opposed to heritable genetic effects. Multigenerational effects are widespread in nature and are often studied because of their potentially important effects on offspring traits. Although multigenerational effects are commonly observed, few studies have addressed whether they affect offspring fitness. In this study we assess the effect of potential multigenerational effects of parental body size and natal carcass size on lifetime fitness in the burying beetle, Nicrophorus marginatus (Coleoptera; Silphidae). Lifespan, total number of offspring, and number of offspring in the first reproductive bout were not significantly related to parental body size or natal carcass size. However, current carcass size used for reproduction was a significant predictor for lifetime number of offspring and number of offspring in the first brood. We find no evidence that multigenerational effects from larger parents or larger natal carcasses contribute to increased fitness of offspring.
Source of burying beetles
To generate the laboratory-bred population to use for the experiment, we captured adult N. marginatus at the Utah Wetland Preserve near Goshen, Utah, USA (this is public land, and no permits were required to trap or collect beetles), in August 2014 using pitfall traps baited with aged chicken. We created 41 independent genetic lines from wild-caught pairs by providing a 40g mouse carcass for each pair and allowing them to breed. We designated the offspring from the wild-caught pairs as the first parental generation. After eclosion and emergence of this first laboratory-bred generation, we placed individuals in small plastic containers (15.6 x 11.6 x 6.7 cm), fed them ad libitum raw chicken liver twice weekly, and maintained them on a 14:10 h light:dark cycle at 21°C until they were used in experiments. We designated the date of eclosion as the first day of life for all beetles used in the experiments.
The purpose of this experiment was to determine the effect on lifetime fitness of potential multigenerational effects of parental body size and natal carcass size in N. marginatus. We measured three fitness response variables: lifespan, total number of offspring, and number of offspring produced in the first brood. Number of offspring in the first brood was used in addition to lifetime number of offspring because number in the first brood may or may not follow total lifetime numbers , and in the natural environment, number of reproductive bouts may be fewer than that observed in laboratory experiments. Thus, number in the first brood may represent natural conditions better than total lifetime number of offspring in a laboratory environment. To generate potential multigenerational effects from parental body size and natal carcass size, we allowed large and small beetles to reproduce on large or small carcasses (first parental generation). To assess fitness effects of multigenerational effects, we then allowed female offspring from this first parental generation to reproduce throughout their lifetime. We then measured the three fitness responses for all females of this second generation and compared them among treatment combinations.
To determine what sizes constituted large or small beetles for assignment in the first parental generation, we used the distribution of pronotum widths from wild-caught beetles. We assigned beetles from the first parental generation with pronotum widths > 1 standard deviation above and < 1 standard deviation below the mean (derived from wild-caught population) as large and small, respectively. In the wild-caught population, the mean pronotum width of females was 6.67mm, with a standard deviation of 0.78mm (N = 50). The mean pronotum width of males was 6.87mm, with a standard deviation of 0.72mm (N = 50). Thus, the size range of large and small female beetles that we used in this experiment for the first parental generation was 7.44mm – 8.22mm and 5.11mm – 5.89mm, respectively, and the corresponding size range for large and small males was 7.60mm – 8.32mm and 5.42mm – 6.15mm, respectively. For this first part of the experiment, we used a fully crossed factorial design. There were four parental size treatments - large male with large female, large male with small female, small male with large female, and small male with small female. Each parental size treatment was crossed with both small (20g) and large (40g) carcass sizes, for a total of eight treatment combinations. We chose 20g and 40g carcass sizes based on a previous study that tested multiple carcass sizes from 5g to 50 g. The 20g and 40g sizes were both within the range of carcass sizes whereon N. marginatus experienced equally high reproductive success . We completed six replicates of each of the eight treatment combinations resulting in an initial sample size of 48 pairs for the first parental generation.
We began each first-generation experimental replicate by randomly choosing a genetically unrelated pair of beetles that fit into one of the parental size treatments. We placed the pair in a small brood container and randomly assigned them either a 20g (± 1.0g) or a 40g (± 2.0g) mouse carcass. We checked the brood containers daily for larvae, and after larvae arrived on the carcass, we removed the lid of the small brood container and placed the small container in an abandonment chamber (see  for details). We checked abandonment chambers daily for leaving adults or dispersal of larvae, and when the carcass was consumed and larvae dispersed into the soil, we removed and weighed the remaining parent(s). The larvae from each brood reached eclosion 4-5 weeks after dispersal. We weighed each newly-eclosed adult offspring, placed them in an individual container, and fed them ad libitum chicken liver until they reached sexual maturity.We used results from this first-generation experiment to test for effects of body size of parents (male and female separately) and carcass size on reproductive output and offspring traits. In the first-generation experiment female body size generated significant effects on offspring body size, but male size had no effect on offspring traits and there were no significant interactions between male and female body size . For this reason, we included parental size treatments as four independent treatments in analysis of the second-generation experiment (this paper) rather than as a factorial. This reduced complexity of the second-generation analysis by eliminating some two-way and three-way, and all four-way interactions. For additional information on the first parental generation methods and results from the experiment, see Smith and Belk .
For the second parental generation, we used the offspring from the first parental experiment described above to determine how the size of parental N. marginatus beetles and the size of carcass that they reproduced on affected lifetime fitness of their offspring. For the second parental generation, we randomly chose two female offspring from each of the first parental replicates for the experiment. These two females (from replicates of the eight first generation treatment combinations) were randomly assigned either large (40g) or small (20g) carcasses for reproduction, for a total of sixteen treatment combinations in the second experiment. We started six replicates for each of the sixteen treatments for a total of 96 replicates. Five experimental pairs failed to reproduce (one each for five of the treatment combinations), so the realized sample size was 91 pairs.
To determine fitness of second-generation beetles, sexually mature females (age > 21 days) were paired with a randomly selected, sexually mature, genetically unrelated male, and the pair was placed in a plastic container (14 × 13 × 17cm) filled with 10cm of moist soil and given either a 20g (±1.0g) or a 40g (± 2.0g) mouse carcass (based on the random assignment described above). The containers were kept in an environmental chamber at 21°C on a 14:10 h light:dark cycle. We checked for larvae daily, and after larvae arrived on the carcass, we removed the male from the container so that only the female cared for the offspring. After the larvae dispersed into the soil to pupate, we removed the female and placed her in a small container with ad libitum chicken liver for two days, and then set her up to mate again with a new, randomly selected, genetically unrelated, virgin male on the same size carcass (20g or 40g). This cycle continued for her entire life. In the second-generation experiment, number of successful reproductive bouts ranged from 1 to 5 with a mean of 2.9 bouts. We measured lifespan as the number of days the female survived after eclosion. After the offspring from each brood of this second-generation experiment eclosed as adults, we recorded the sex, weight, and pronotum width for each beetle. We summed the number of offspring from each brood to determine a female’s lifetime number of offspring.
To determine if parental size or natal carcass size influences fitness of offspring we used a mixed model analysis of covariance (ANCOVA) approach (SAS, Proc MIXED). We used three response variables to represent fitness of females from the second parental generation: i.e., total lifetime number of offspring, total number of offspring in the first brood, and lifespan. Predictor variables for each of the three models were first parental generation body size (four levels), natal carcass size (two levels), current carcass size (two levels), and pronotum size of second parental generation focal female (covariate). Initially we included all two-way and the three-way interactions of main effects in the model. However, interaction terms were not significant in any of the models, so, we report results from a reduced model that included only one, two-way interaction – natal carcass size by current carcass size. This interaction was of interest because of the significant effect of current carcass size on fitness measures, and the possibility of a multigenerational effect that natal carcass size might provide some “priming” for efficiency of use of a similar carcass size. Because we used two individuals from each of the first-generation broods to breed for the second generation, we used first generation parental brood ID as a random effect to account for the relatedness of individuals from the same brood. We used raw data for the analysis, and inspection of residual plots showed no departure from assumptions for the parametric model.
Data are arranged in columns. Column 1 is parental brood ID, column 2 is offspring female ID, column 3 is the parental size treatment, column 4 is the natal carcass size treatment, column 5 is the current carcass size treatment, column 6 is the pronotal width of the female offspring, column 7 is the number of successful reproductive attempts, column 8 is the lifetime number of offspring, column 9 is the total offspring mass, column 10 is the lifespan, and column 11 is the number of offspring in the first brood.
Key to treatment designations
Parental size treatment: 1=large male large female, 2=large male small female, 3=small male large female, 4=small male small female
Natal carcass size treatment: 1=20g natal carcass, 2=40g natal carcass
Current carcass size treatment: 1=20g natal carcass, 2=40g natal carcass