Purpose: Substance P is a well known neurotransmitter and has been known to mediate pain. Recently, it has been unveiled that substance P is involved in the recruitment of mesenchymal stem cells to wound sites. The purpose of this study was to exam bone formation when a combination of substance P and silk fibroin was used in a bone defect model. Methods: Twenty rabbits were used and 40 calvarial defects were formed. They were divided as 4 groups (unfilled control, silk only, silk+$10{mu}g$/ml substance P; Sub10, and silk+$100{mu}g$/ml substance P; Sub100). All animals were humanely sacrificed 4 or 8 weeks after grafting. The specimens were analyzed by micro-computerized tomography and histological analysis. Results: When compared to the unfilled control to silk only group, there was significant difference in bone mineral density (BMD) and the attenuation coefficient (AC) at 4 weeks ($p$=0.037 and 0.038, respectively). When compared Sub10 group to Sub100 group, there was significant difference in BMD and AC at 8 weeks ($p$=0.004 for all). Residual graft amounts were $52.1{pm}15.8$%, $15.2{pm}9.2$% and $9.0{pm}3.3$% for silk only, Sub10, and Sub100 groups, respectively. When comparing the residual graft amount of silk only to sub10 or sub100, the differences were statistically significant ($p$ <0.001). Conclusion: The silk fibroin scaffold showed higher BMD and AC than the unfilled control. The combination graft with substance P and silk fibroin scaffold showed a faster graft degradation than with a silk fibroin scaffold only.