The cell walls of rod-shaped Gram-positive bacteria are thick, multilayered networks that chirally twist as cells elongate. The underlying basis of twisting is not known, but probing the processes underlying this phenomenon may give insights into how cell wall material is inserted, how it evolves during cleavage, and the mechanics within the sacculus. In Bacillus subtilis, we see cell chains lacking hydrolases twist far slower than chains of wild-type cells, indicating that cell wall cleavage modulates the twisting rate. We see that when cells within chains separate, the two nascent ends rotate as they separate. Together, this suggests there is torsional stress within the cell wall that, when unreleased, perturbs overall chain morphology. Unlike Escherichia coli, we see that twisting does not arise from MreB's angle of motion, as its angle is identical in both fast-twisting wild-type cells and slow-twisting hydrolase-deficient cells. Rather, the circumferential insertion of glycans appears to establish this torsional stress, as increasing Rod complex activity by deleting ponA causes cells to twist faster than wild-type cells. Together, these experiments suggest the twisting of B. subtilis cells arises from radial glycan insertion, which somehow causes torsional stress in the wall that is later released by hydrolase activity.