458 The Study of the Mechanics of the Spine 
been reached. In this literature one reads of hypomochhons, tangents, 
horizontal and lateral axes, centres of motion, etc., and mental confusion 
must result from a careful study of these theories. 
One way of approaching the question has been left practically un- 
touched except for a single experiment of Bradford’s (7). This line of 
investigation would be to see what the normal spine would do in a scolio- 
sis artificially produced and again to inquire if there is anything in the 
normal movements of the spine to account for this phenomenon of rota- 
tion in connection with lateral curvature. Perhaps the most popular 
theory to-day is that of Meyer (8), proposed in 1865 and then falling into 
discredit until taken up by Albert (9) in 1899. This theory maintained 
that rotation occurred because the human spine consisted of two columns, 
the column of bodies and the column of arches, and that these two col- 
umns possessed a different degree of elasticity. In lateral curvature, 
therefore, rotation occurred because these two elements of the human 
spine reacted in different degrees to side bending and one lagged behind 
the other in side yielding. In some experimental work upon the cadaver 
undertaken by the writer at the Harvard Medical School (10) certain 
phenomena were observed which seemed to have a bearing on the occur- 
rence of rotation in lateral curvature, and Prof. Thos. Dwight suggested 
that it might be worth while to investigate the question whether the spine 
did not follow certain general laws of mechanics and did not behave as 
would a flexible rod under similar conditions. The writer is indebted to 
Prof. Dwight not only for his anatomical material but for his continued 
advice and criticism throughout the work. 
With the aid of Prof. I. N. Hollis, of Harvard University, two general 
laws governing flexible rods were formulated as follows: 
(A) Although a straight flexible rod (e. g., a quadrilateral rod of 
rubber or lead) may be bent in one plane without twisting, if such a rod 
is already bent in one plane it cannot be bent in another plane without 
twisting. 
(B) Although a straight flexible rod may be twisted without acquir- 
ing a side bend, a flexible rod already bent in one plane cannot be twisted 
without acquiring a side bend. 
The following experiments were then undertaken : 
1. A quadrilateral rod of rubber was fixed at its lower end and then 
bent forward away from the observer. The top of the rod was then bent 
to the left and the front of the rod was observed to twist to the right. 
2. A similar rod of lead followed the same rule under the same condi- 
tions. 
3. The backbone of a fish followed the same rule under the same con- 
ditions. 
