1. In two dogs, approximately one to two years and three to four months of age, an experimental comparison was made between the calcium accretion rate as defined by the Bauer-Carlsson-Lindquist equation, and the bone formation rate determined by double tetracycline labelling. 2. The overall calcium accretion rate was determined from the specific activity of the blood plasma, and the urinary and faecal excretion of isotope, following an intravenous tracer dose of Ca45. A time of five days after injection was used for the calculation of accretion rates, but data for shorter times of calculation are included. 3. Local accretion rates were obtained for different parts of the skeleton by determining the specific activities of bone samples at the end of the experiment. 4. The amount of isotope the uptake of which was not related to new bone formation (the diffuse component) was determined autoradiographically. 5. Local values for appositional growth rate and bone formation rate were obtained, using sections of undecalcified bone specimens, by measuring the linear separation between two tetracycline bone markers and the area of new bone enclosed by them. 6. In the older dog, the measurements for cortical bone showed that the accretion rate was two to three times as great as the bone formation rate: the observed diffuse component was sufficient to account for the greater part of this difference. Measurement of the bone formation rate for cancellous bone presented difficulties, but the approximate values obtained suggested that the accretion rate and the bone formation rate were of about the same order for this tissue. 7. In the younger dog, the bone formation rate could be determined only in cortical bone: at the sites studied, the values for the accretion rate and the bone formation rate did not differ by more than 20 per cent. It is suggested that this is due partly to the low specific activity of the diffuse component in this young animal, and partly to the relatively large amounts of new bone formed during the period of the experiment. 8. Despite the important differences between the rates of calcium accretion and bone formation that were found to exist in regions where there was only a small amount of new bone formation, there was a strong correlation between the two rates. The value of the accretion rate as a parameter of bone metabolism is clear.