This work explores the fitness consequences of autotomy and regeneration in adult crayfish and shows how measuring potential fitness costs of regeneration can be complicated by likely compensatory changes that buffer these costs. Furthermore, this chapter also details the process of chelae regeneration in juvenile crayfish and reveals that molting and limb regeneration can be de-coupled; an unexpected finding.

1. 1. Material and methods
      1. Adult crayfish maintenance, autotomy, and mating procedures
         1. Colony

Crayfish (Red swamp crayfish, Procambarus clarkii, Girard 1852) were maintained in a 100-gallon tub in the laboratory with a circulatory water pump and filter under a controlled day/night light cycle (12/12 hours). The environment was enhanced with orange mesh and PVC pipes of different sizes to provide shelter and reduce cannibalism. Individuals were obtained from laboratory stocks at the University of Kentucky and purchased via Carolina (https://www.carolina.com/). Animals were fed weekly ad libitum with salmon pellets (soft, moist salmon feed pellets, size 5/32", from Rangen Inc.).

1. 1. 1. 2. Autotomy

Autotomy of one cheliped was induced by cutting the limb at the merus/ischium articulation (Chapter 3, Figure 4.1A) using a sharp blade or scalpel. The remaining limb part is either readily autotomized during or after the cut effectively preventing hemolymph loss. All experiments were conducted in laboratory at the University of Kentucky. All crayfish were measured to the nearest tenth of millimeter using digital calipers along seven measurements: Total Body Length (TBL), Cephalothorax Total Length (CeTL), Cephalothorax Width (CW), Cheliped Total Length (ChTL), Propodus Length (PL), Palm Width (PW), and Dactyl Length (DL; Figure 3.1).

1. 1. 1. 3. Mating

Adult female crayfish (Red swamp crayfish, P. clarkii) were selected from our colony and isolated in individual tanks (20x15x10 cm) as part of a flow-through system under a controlled day/night light cycle (12/12 hours). Their environment was enriched with one PVC pipe providing shelter and mimicking the natural behavior of this species when in berry. After producing eggs, a female P. clarkii retreats to a burrow and stays there until her crayling reached their fifth molt and are ready to live on their own (McClain et al. 2007).

Each mature female was individually presented to three random males at a time in a mating arena (circle, radius = 25 cm). The water temperature was kept between 23-25˚C using a water heater (300W submersible fish tank water heater adjustable temperature) prior to adding the crayfish to improve mating success. Mating was attempted for a period of 30 minutes after which the female was removed from the mating arena; attempts were repeated with different males until a successful mating was observed or after a month passed, whichever came first. The mating was successful if females were seen held by one male in the mating position (chelae locked over her head by the male chelae and abdomens facing each other) for at least 10 consecutive minutes.

1. 1. 2. Juvenile rearing, maintenance, measurements, and papilla stage histology

Female adult crayfish were removed from our colony and individually kept when in berry (i.e., with visible eggs on the underside of the tail; McClain et al., 2007). Once hatched, we separated the craylings and raised them individually in cups from which the bottom had been replaced with a fine nylon mesh and placed in a plastic tub with all other juveniles and air pumps under a controlled day/night light cycle (12/12 hours). Juveniles were monitored and fed salmon pellets (soft, moist salmon feed pellets, size 5/32", from Rangen Inc.) daily. Once juveniles reached at least 3 cm (body size) autotomy was induced similarly as in adults (see above). After autotomy individuals were checked twice daily to assure accurate molt detection since crayfish eat their ecdysis to recover calcium bicarbonate (McClain et al. 2007). We measured cheliped sizes prior to autotomy and regenerate length and papilla stage every day following autotomy (by taking a picture of the individual underside and using ImageJ; Schneider, Rasband & Eliceiri, 2012). The only exception to this process was when a molt was seen in the cup (because the individual was too soft to handle).

1. 1. 3. The effect of limb loss and regeneration on adult female crayfish reproductive success

After inducing autotomy of one randomly chosen cheliped, each female was mated and then isolated (see above). When females were observed in berry the eggs were gently removed using a dull blade or pen cap onto a paper towel for counting. The width of twenty randomly chosen eggs were then measured to the nearest tenth of millimeter using electronic calipers and compiled to determine the average egg size of the clutch. Control individuals went through the same procedure minus the induced autotomy.

1. 1. 4. Data analysis

A Principal Component Analysis (PCA) was run with the seven body measurements (see above and Figure 4.1 and 4.2). Furthermore, Welsh two-sample t-test were run for each of the seven body measurement variables to statistically compare regenerating and unmanipulated crayfish size.

The effect of limb loss on crayfish fitness was analyzed using linear mixed models following a Gaussian distribution with the number of eggs or the average egg size per female as the dependent variable and the regenerative status (yes/regenerating or no/unmanipulated) as the independent variable. Because no size difference was found between regenerating and unmanipulated crayfish (see results), we used Total Body Length (TBL, in mm) as a fixed factor in our models to control for the effect of female size (clutch size being highly correlated with female size; McClain et al. 2007). Finally, p-values were calculated using the Satterthwaite’s method (Kuznetsova et al. 2017).

All analyses were conducted in R (Team 2017) using the following packages: ggplot2 (Wickham 2009b), reshape2 (Wickham 2007), cowplot (Wilke 2017), multcomp (Hothorn et al. 2008), psych (Revelle 2017), nlme (Pinheiro et al. 2017), betareg (Cribari-Neto and Zeileis 2010), gridExtra (Anguie 2017), lme4 (Unknown 2010), lmerTest (Kuznetsova et al. 2017), and factoextra (Kuznetsova et al. 2017).