As orthopedic investigations have become more intricate, bone specimens have sometimes undergone multiple freeze-thaw cycles prior to biomechanical testing. The purpose of this study was to determine if repeated freezing and thawing affected the mechanical properties of canine cortical bone. Six pairs of third-metacarpal bones were tested in three-point bending and six pairs of femurs were tested in torsion. One member of each pair was tested destructively at collection. The other member was tested nondestructively at collection and after each of five freeze-thaw cycles; followed by destructive testing after the fifth cycle. For destructive tests, the material properties (modulus, maximum stress, maximum strain and absorbed energy) of a specimen at collection were compared to those of the corresponding contralateral specimen which had undergone five freeze-thaw cycles. For repeated nondestructive tests, the modulus of a specimen at collection was compared to modulus of the same specimen at each of the five thaw intervals. During destructive testing, there was a significant (p = 0.02) decrease (20%) in maximum torsional strain. Other changes in bending and torsional destructive properties were not statistically significant. During repeated nondestructive testing, there were solitary significant (p < 0.05) increases (8% and 9%, respectively) in both bending and torsional modulus. However, these isolated changes were not correlated to the number of freeze-thaw cycles. The pattern of alterations in destructive and nondestructive biomechanical properties was most consistent with varying specimen dehydration at each thaw interval. Despite using accepted methods to maintain specimen hydration, repeated freezing, thawing, handling and testing of cortical bone increased the risk of moisture loss. Unless stringent efforts are made to insure proper hydration, the mechanical properties of canine cortical bone will be altered by repeated freezing and thawing, affecting the results of studies utilizing this technique.