Purpose of investigation: The fetal-lamb model is a fundamental tool for developing clinical applications for the treatment of human fetal pathology (e.g., open fetal surgery). Accurate estimation of gestational age is important to avoid size-related problems during surgery. Methods: To evaluate the effect of twinning on ovine fetal growth, we followed seven twin pregnancies in Comiso ewes from mating through parturition. Fetal growth indexes (muzzle-occipital and mean abdominal diameters, abdominal circumference, femoral and humeral length) were measured weekly using perimammary ultrasonography and analyzed with a linear quadratic regression model based on natural logarithms of each parameter and fetal age. Results: The model explained > 90% of the variability observed, with determination coefficients of 95% (femoral length, abdominal circumference), 94% (abdominal diameter, humeral length), and 89% (muzzle-occipital diameter). Conclusion: Mean birth weight was lower than that reported for singleton fetal lambs, as it is in bigeminal pregnancies in humans, despite the uterine and placental differences between these two species. With the exception of slightly earlier growth deceleration, curves for head and long-bone growth resembled those for singleton ovine fetuses. Ovine fetal growth patterns (like those of humans) in singleton and twin pregnancies are similar.