Abstract: Viscosity is inversely related to temperature. The circulatory system of Antarctic icefish may have been designed to prevent high blood viscosity at low temperatures by taking advantage of the increased solubility of oxygen at low temperatures, allowing use of hemoglobin-free blood. This necessitates a high-output, high-velocity, low-pressure, low-resistance circulation. High-velocity flow requires adequate viscosity to minimize loss of laminar flow and increased friction. This creates an interesting design problem: in other animals, hemoglobin determines blood viscosity via the hematocrit, whereas in icefish, blood viscosity is produced largely by antifreeze glycoproteins. The effect of inappropriate blood viscosity on maximal cardiac output is seen in experiments with a related fish, Pagothenia borchgrevinki. In this species, acclimation to a particular temperature involves tailoring blood viscosity to cardiac power, which varies with the availability of oxygen and temperature. The factorial scope for cardiac output—i.e., the ratio of maximal to basal cardiac output—is greater in acclimated than unacclimated fish despite the similar availability of oxygen. Experiments also suggest that blood viscosity determines the maximum tolerable temperature in Antarctic fish. Those experiments demonstrate that blood viscosity is actively controlled. It is part of what the physiologist Claude Bernard called the milieu intérieur. The hemoglobinless phenotype requires simultaneous customization of the heart, vasculature, and blood, including its viscosity. Simultaneous, coordinated acquisition of multiple unique features, as required by the absence of hemoglobin, is inconsistent with Darwinian evolution, which postulates that species develop by small, incremental changes over time.