The authors remain convinced that process of fixation and replacement is accomplished in a similar manner in autogenous and homogenous bone grafts, but we are not at all sure that the differences in time of union and the number of failures can be explained entirely on the basis of tissue specificity. There is reason to believe that some of the cells of an autogenous graft are capable of survival and of instigating this process and thus accelerating it when host capillaries reach the graft. This could account both for the more rapid union of the autogenous graft as compared with homogenous graft and for the greater number of successes with autogenous grafts. Merthiolate-preserved bone does not compare favorably with autogenous bone grafts in that the process of fixation and replacement is definitely retarded, requiring prolonged protection with a higher percentage of failures (30.24 per cent. in this series).

The merthiolate bone bank is a very satisfactory method of preserving bone. We have no evidence of sensitivity to the merthiolate or excess tissue reaction to merthiolate preserved bone. The method is certainly easier and more economical of bone than the frozen bone back.

A homogenous bone graft which is adequately placed in a good bed and protected for a sufficiently long time may be expected to unite. Many of the failures listed above should not be ascribed to the bone bank or to homogenous bone grafts but to poor surgery. Still others can be attributed to the injudicious use of bone grafts; in these cases failure would probably have occurred irrespective of whether a homogenous or an autogenous graft was used. However, we have modified our original expectations of preserved homogenous bone graft and now feel that these grafts should be considered as useful aids in orthopaedic surgery rather than as universal substitutes for autogenous bone. Obliteration of cavities in bone created by sequestrectomy and saucerization for osteomyelitis, or local excision of benign tumors by homogenous bone from the bank, has given good results. Where the defect is large, bank bone is preferable to autogenous bone because of the great amount required. Homogenous peg or inlay grafts have Proved satisfactory in the treatment of minor non-unions. Likewise, bank bone has been very useful for internal splinting in arthrodesis and for treating certain fresh fractures. However, homogenous bone grafts should never be used in the treatment of major non-unions if there is any possibility of using an autogenous graft. It is our feeling that the use of bank bone should be reserved for those circumstances in which it is not feasible or advisable to use autogenous bone. These are:

1. When the available supply of autogenous bone does not fulfill the particular requirements;

2. When the taking of an autogenous bone graft will materially increase the hazard of the operative procedure:

3. In any condition where there is a chance that the graft will be lost because of infection;

4. During the course of an operation when it is decided that a bone graft would be useful and when no previous Plans for taking a graft had been made;

5. In those cases where the bank bone is used as an internal splint when the condition would not justify the taking of an autogenous graft.