# Effect of food restriction on survival and reproduction of a termite Authors: * Silu Lin; Evolutionary Biology & Ecology, University of Freiburg, Germany,[lululindeisi@outlook.com](mailto:lululindeisi@outlook.com)] * Ido Pen; Theoretical Research in Evolutionary Life Sciences, University of Groningen, The Netherlands, [i.r.pen@rug.nl](mailto:i.r.pen@rug.nl)] * Judith Korb; Evolutionary Biology & Ecology, University of Freiburg, Germany, [judith.korb@biologie.uni-freiburg.de](mailto:judith.korb@biologie.uni-freiburg.de)] date: "1/9/2023" output: pdf_document: default --- ## Introduction This study examines how food availability affects the trade-off between maintenance and reproduction in termites, both on individual and colony levels. The data was collected by Judith Korb. Silu Lin and Ido Pen contributed to coding. ## Software requirements * R version 4.1.1 (2021-08-10) * R package: cmdstanr version 0.4.0 * R package: tidyverse 1.3.2 * R package: cowplot 1.1.1 * R package: readxl 1.4.1 * R package: brms 2.17.0 * R package: loo 2.5.1 * R package: rstan 2.21.7 * R package: bayestestR 0.12.1 * R package: tidybayes 0.12.1 * R package: bayesplot 1.9.0 * R package: logspline 2.1.17 * R package: ggpubr 0.4.0 * R package: BiocManager 1.30.18 * R package: gridExtra 2.3 * R package: dplyr 1.0.10 ## Files * Data.csv: fitness data to be analyzed. * Figure_and_analysis.R: R script containing the codes for generating figures and Bayesian analysis. * model_comparision.R: R script containing the codes for model selection based on WAIC scores. ## Explanations for variables in the data file * source_colony: the original colony ID which experimental colonies were collected from. * brood: eggs and larvae produced by the queen. * w_young: workers that were produced by the queen during the experiment. * w_old: workers that were present from the beginning until the end of the experiment. * Initial_colony_size: the number of workers at the beginning of the experiment. * queen_survival: the survival status of the queen at the end of the experiment, 1 indicates alive and 0 indicates dead. * colony: IDs of experimental colonies. * treatment: 0 indicates control (abundant food), 1 indicates less food. * colony_survival: the survival status of the colony at the end of the experiment, 1 incates alive and 0 indicates dead. * alates: the number of winged sexuals. ## Links between variables in the main text and the data file The left terms are the variables listed in the main text, the right were the variable names in the Data.csv file. * Queen survival = "queen_survival". * Colony survival = "colony_survival". * Worker survival = "w_old/(Initial_colony_size-alates)" * Queen fecundity = "brood + w_young" * Colony fecundity = "alates". * Treatment = "treatment". * Colony ID = "colony". * Incipient colony size = "Initial_colony_size".

Food availability affects the trade-off between maintenance and reproduction in a wide range of organisms, but its effects on social insects remain poorly understood. In social insects, the maintenance-reproduction trade-off seems to be absent in individuals but may appear at the colony level, although this is rarely investigated. In this study, we restricted food availability in a termite species to test how it affects survival and reproduction, both at the individual and colony level. Using Bayesian multivariate response models, we found very minor effects of food restriction on the survival of queens, individual workers, or on the colonies. In contrast, queen fecundity was significantly reduced while colony-level fecundity (i.e., the number of dispersing alates, future reproductives) increased under food restriction as workers gave up cooperation within the colony and became alates that dispersed. Our study shows that life history trade-offs can be mitigated by individuals’ social behaviours in social organisms.

C. secundus colonies were collected from dead Ceriops tagal trees in Darwin, Northern Territory, Australia (12°30’S 131°00’E) and kept in Pinus radiata wood blocks under lab conditions in Germany (28 °C, 70% relative humidity and a 12 h day/night cycle). Food availability was manipulated by adjusting the size of wood blocks. C. secundus senses a wood size of 1 termite per 2.5 cm³ as food shortage, while 1 termite per 10.0 cm³ reflects abundant food conditions. We set up 14 low-food colonies (1 termite: 2.5 cm³ of wood) and 21 control colonies (1 termite: 10.0 cm³ of wood) for this experiment. The experiment lasted for 18 months. In the end, we obtained fitness data for 34 colonies. We applied Bayesian multivariate response models to estimate the effect of food restriction on the survival and reproduction of individuals and colonies. R script can be found in the same data repository.

Data file (CSV) and R script.