84 C. A. L. Bassett 



its external electrical environment. On the other hand, the paucity of cells and the 

 relatively large amount of extracellular material in many connective tissues might 

 make a cellular control system unwieldy. If, as Weiss (1960) has proposed, action 

 potentials in peripheral nerves may cause reorientation of molecules in the axonal 

 membrane, why is it not possible that stress-induced potentials in macromolecules 

 may not accomplish similar results? Such a system would permit finite regulation of 

 extracellular material without specific cellular intervention and may well have been 

 the major control mechanism for eons before the evolution of a nervous system. 

 Certainly, the cellulose fibers in wood (another two phase material) have, without 

 neural regulation, a highly ordered structure similar in many respects to collagen in 

 bone (Bassett, in press). Finally, it seems possible that the electrical activity of 

 mechanically stressed organic macromolecules may influence directly their chemical 

 reactivity (e.g., "mineralizability"), particularly if, as has been suggested recently 

 (Little, 1964), the macromolecules behave as superconductors. 



Studies cited previously (Bassett and Becker, 1962; Becker et ai, 1964; 

 Cochran et ai, unpublished) have demonstrated that the amplitude and decay 

 characteristics of the electrical pulse, from bone, vary significantly, depending upon 

 the rate, magnitude and duration of deformation. The orientation of vessels, osteones, 

 lamellae or mineralized collagen bundles, in relation to the direction of the applied 

 force, also may affect the character of the pulse. Furthermore, it is probable that the 

 relative degree of mineralization and the state of hydration of any given region in 

 the osseous matrix will determine, in part, its electrical behavior (Becker et ai, 

 1964). As pointed out earlier in this report, it is impossible to measure the activity 

 of individual generators with dimensions in the Angstrom unit range. Therefore, the 

 pulses recorded in these studies represent a summation of billions of individual events 

 (some additive, some subtractive, some simultaneous, others not) occurring within the 

 specimen under investigation. Both equally and unequally biphasic signals have been 

 recorded. Truly uniphasic signals, however, have not been observed. The statement 

 by Shamos and Lavine (1964), "the reverse pulse is an electrical artifact caused 

 by . . . capacitane", ignores the fact that, on release, piezo-electric materials normally 

 produce a reverse pulse, if the initial, pressure-induced charge separation is permitted 

 to leak off while the material is still deformed. How, then, does an unequally 

 biphasic signal result? Three of many possible explanations can be advanced. First, 

 the generator seems to be a p-n junction which, in itself, may rectify the signal, since 

 current passes in forward bias more efficiently than it does in reverse bias. Second, 

 the rate and duration of loading (deformation) in the living vertebrate is such that 

 the signal may not decay fully before the next cycle begins (Fig. 4). Third, although 

 a specimen, gradually loaded, produces little or no detectable signal, a uniphasic 

 spike of normal magnitude results when it is rapidly unloaded (Bassett, in press; 

 Cochran et ai, unpublished). 



Although signal magnitude and decay time vary considerably, as a function of 

 different bone specimen characteristics, there is much greater uniformity in signal 

 polarity. Regions under compression, generally concave surfaces, are routinely nega- 

 tive with respect to regions under tension, generally convex. It is known clinically 

 and experimentally that the concave aspects of a bone will be buttressed with new 

 bone and the convex removed (Murray, 1936). These two observations lead 

 teleologically to the prediction that regions of negativity may be associated with 



