Perumytilus purpuratus

García-Huidobro, M. R., Varas, O., George-Nascimento, M., Pulgar, J., Aldana, M., Lardies, M. A. & Lagos, N. A., 2019, Role of temperature and carbonate system variability on a host-parasite system: Implications for the gigantism hypothesis, International Journal for Parasitology: Parasites and Wildlife 9, pp. 7-15 : 9-10

2.4. Parasitism effects on body size of P. purpuratus

For all P. purpuratus collected from the field, the effect of parasitism on body size was evaluated through an allometric scaling relationship. The mussel shell length (l), width (w), and height (h) were measured with a digital caliper (± 0.1 mm), and the total weight was measured covariance (ANCOVA) were used to compare total weight, soft tissue weight, volume, shell weight and CaCO 3 weight between sites (Quintay and Concepcion), and parasitized condition, with the maximum shell lengths used as covariates. In the case of shell thickness, ANCOVA analysis incorporated the shell weight as covariate while correcting for shell surface. Previous to these analyses, all variables were log transformed. We show the Least Square Means (LSM), the predicted response of parasitized and non-parasitized, at the mean value of the covariate for each morphological variable. The corresponding bivariate regression analyses and their parameters estimation is presented as supplementary material. Finally, growth and metabolic rate of mussels were compared between experimental temperature treatments and parasite condition using two-way ANOVA. In all cases, a Tukey post-hoc analysis was performed to determine significant differences between groups ( Zar, 1996). All analyses were performed using Minitab 14 º ver 13.3.2. with an analytical balance (Mettler Toledo ± 0.0001 g). From these measurements, the volume of each mussel was approximated to an ellipsoid volume: [(4π/3) × (l) × (w) × (h)] ( Calvo-Ugarteburu and McQuaid, 1998). Later, individuals were dissected to determine the condition of parasitism, and the soft tissue weight and shell weight were measured using the analytical balance. The shell thickness was estimated using the relationship between shell weight and shell surface area ( Briones et al., 2014). The shell surface incorporated the possible effect of the valve curvature (using shell depth or width as a proxy variable) on the surface estimation, and was based on the formula of shell surface proposed by Reimer and Tedengren (1996): [l (h 2 + w 2) 0.5 π / 2], where l, h, and w were the maximum shell length, height, and width, respectively. The CaCO 3 content of the mussel shells was estimated by calcination at 500 ̊C for 4 h to remove the organic components of the CaCO 3 matrix (e.g., Watson et al., 2012). Calcinated shells were stored in desiccators for 1 h and then weighed.

For P. purpuratus exposed to temperature treatments, growth and metabolic rate (oxygen consumption) were quantified at the end of the experimental exposure. Growth rate (mm d −1) of P. purpuratus was estimated from changes in the maximum shell length between the start and end of the experimental period (e.g., Osores et al., 2017). The oxygen consumption was measured (as oxygen consumption, in mg O 2 h − 1 g −1) by using a Presens Mini Oxy-4 respirometer (e.g., Benítez et al., 2018). The experimental animals were individually placed in respirometer chambers filled with 70 ml of seawater and oxygen-saturation through air bubbling (15 min). The measurements were performed at a controlled temperature of 12 and 18 ̊C using an automated temperature chiller. In each chamber, dissolved oxygen was quantified every 15 s over the course of at least 60 min. Oxygen sensors were previously calibrated in anoxic water using a saturated solution of Na 2 SO 3, and in water 100% saturated with oxygen using bubbled air. The same chambers, without animals, were used for controls, and the oxygen concentration did not decay more than 3%. Oxygen decay due to back-ground noise was deducted from the individual measurements.