The hellbender (Cryptobranchus alleganiensis) is a salamander that grows over a large range of body sizes (2-74 cm total length) making it an ideal organism for examining the effects of body size on morphology and performance. The goal of this study is to investigate the morphology changes over ontogeny and change in feeding ability. Cryptobranchus feeds on small aquatic insects as juveniles and shifts to crayfish as they get larger. Morphology can be expected to change as an organism grows larger, and because morphology and performance are closely linked, this morphological change can result in a change in feeding ability. Cryptobranchus alleganiensis are primarily aquatic salamanders that utilize both suction feeding and biting behaviors. I hypothesize bite force would increase with positive allometry reflecting a possible dietary shift during ontogeny in which larger Cryptobranchus favor crayfish. Because suction is the primary mode of feeding making it an important aspect of feeding throughout ontogeny, suction index was hypothesized to scale with isometry. Fourteen preserved specimens (11.9-34.5 cm SVL) were used to investigate the effects of scaling on suction potential and estimated bite force. Bite force was calculated using a 3D static equilibrium model and suction potential was calculated as suction index. Bite force scaled with positive allometry allowing the animals to bite harder relative to body mass with increasing body size, and suction index showed no effect of body size. Results of this study indicate that Cryptobranchus alleganiensis maintains suction performance across ontogeny allowing them to generate suction with similar ability ontogenetically, but increases its biting performance to cope with durophagous prey with a possible ontogenetic dietary shift.

The rotary Left Ventricular Assist Device (LVAD) is a mechanical pump surgically implanted in patients with end-stage congestive heart failure to help maintain the flow of blood from the sick heart. The rotary type pumps are controlled by varying the impeller speed to control the amount of blood flowing through the LVAD. One important challenge in using these devices is to prevent the occurrence of excessive pumping of blood from the left ventricle (known as suction) that may cause it to collapse due to the high pump speed. The development of a proper feedback controller for the pump speed is therefore crucial to meet this challenge. In this thesis, some theoretical and practical issues related to the development of such a controller are discussed. First, a basic nonlinear, time-varying cardiovascular-LVAD circuit model that will be used to develop the controller is reviewed. Using this model, a suction index is tested to detect suction. Finally we propose a feedback controller that uses the pump flow signal to regulate the pump speed based on the suction index and an associated threshold. The objective of this controller is to continuously update the pump speed to adapt to the physiological changes of the patient while at the same time avoiding suction. Simulation results are presented under different conditions of the patient activities. Robustness of the controller to measurement noise is also discussed.