Selection cycle. Given the number of generations (i) and the population size at each generation (N_i), effective population size was calculated as: N_e = i/({1/N_1} + {1/N_2} +… + {1/N_i}) A mass action culture evolution in seawater based defined media was developed as a control to the squid evolution. The regime replicates the number of generations during each growth cycle in the squid following colonization and after each venting event (as determined empirically by direct plating of the cultures) and introduces similar bottlenecking seen in the squid during venting and colonization initiation of each new squid. V. fishery MJ11, WH1 and ES114 were conditioned at 28 degree C overnight in 1X ASW supplemented with 10 mM Hopes, 0.333 mM K_2 HPO_4, 18.7 mM NH_4 Cl, 0.0144% Case Amino Acids, and 0.53 mM Glucose. Six replicate lines for each ancestor were serially passaged at 6, 12 and 24 h for 15 days by transferring 2 mu l of culture into 2 mL of fresh culture evolution media. The culture evolution was incubated at 28 degree C with shaking. After the 15th day of culture evolution, each line was plated on SWT plates and colonies were randomly selected as population representatives. There were no colony morphology variants observed. Relative luminescence of the culture evolved population representatives was determined in SWTO liquid broth using the Tecan Infinite M200 plate reader equipped with luminescence detection (Tecan, Durham NC). Luminescence was quantified using a Turner 20/20 luminometer (Turner Designs, Sunnyvale CA). Quantitative luminescence was measured in squid (n = 3 – 5), 48 h post colonization, first directly on squid housed in individual vials and then following homogenization of squid in 100 mu l ASW which releases the bacteria from the light organ and oxygenates the bacteria, which would increase luminescence if the reaction were oxygen limited. The number of bacteria within the squid was enumerated by colony counts of dilutions plated onto LBS agar plates, following overnight incubation. The data from three experiments were examined for block effects by a one-way ANOVA and once we determined there were none, the data combined, presented as the mean, and analyzed by independent sample T-tests with 95% confidence intervals (SPSS Statistics V. 17.0). To quantify luminescence induction in vitro, overnight cultures of V. fishery were inoculated into SWTO broth at 0.5% and grown with shaking, and aliquots were removed at 30 min intervals to determine the luminescence and OD_600. For some experiments N-3-oxohexanoyl-L-homoserine lactone (C6-HSL) (Sigma) and N-octanol-L-homoserine lactone (C8-HSL) (Sigma) were added to SWTO at a final concentration of 120 nM. The experiment was repeated with similar results, and one representative experiment is presented. Induction of luminescence by AHLs in conditioned broth To assess the production of AHLs by the various strains, we used luminescence produced by V fishery ES114 following growth in broth conditioned with culture supernatant from the ancestral and evolved strains as an indication of AHLs present in the culture supernatant. Conditioned broth was prepared by growing bacterial cultures to a final OD_600 of 3.0 in SWTO, pelleting cells by centrifugation at 12,000 times g, and filter-sterilizing the cleared supernatant by passage through a 0.2^xm filter. The supernatant was combined 1: 1 with fresh SWTO, and 200 mu L of an overnight culture of V. fishery ES114 was inoculated in 15 mL in 125 mL flasks. Luminescence and OD_600 was measured from aliquots at 30 min intervals. As controls, luminescence produced by ES114 in unconditioned SWTO media, and broth conditioned with supernatant from either V fishery ES114 or CL24 (AinS LuxI) (Lupp and Ruby 2005). The experiment was repeated with similar results, and the data from one representative experiment is presented. The squid inoculate were made as previously described but in 1: 1 ratio of a single AVL and a DVL strain from population 6 confirmed by direct plating. Seven squid for each replicate competition were placed in inoculum for 12 h, and then removed and rinsed in fresh ASW. Light organ homogenates were plated as previously described, and then 50 colonies from each squid were patched on SWTO media to determine their visual luminescence phenotype. The relative completeness index RCI was calculated as the (final ratio, wild-type CFU/mutant CFU)/(initial ratio, wild-type CFU/mutant CFU). Development of a squid experimental evolution model Our experimental evolution design approximates the natural acquisition of the symbiont from the environment, its growth within the light organ, and its re-colonization of newly-hatched squid at each generation. Notably, beyond