A Scientific Test of Chiropractic's Subluxation Theory

The first experimental study of the basis of the
theory demonstrates that it is erroneous

Edmund S. Crelin, Ph.D.

Chiropractic is defined in the dictionary as "a therapeutic system based upon the premise that disease is caused by the interference with nerve function, the method being to restore normal condition by adjusting the segments of the spinal column." [1] The International Chiropractors Association defines chiropractic as follows:

The philosophy of chiropractic is based upon the premise that disease or abnormal function is caused by interference with nerve transmission and expression, due to pressure, strain or tension upon the spinal cord or spinal nerves, as a result of body segments of the vertebral column deviating from their normal juxtaposition. The practice of chiropractic consists of analysis of an interference with normal nerve transmission and expression, and the correction thereof by an adjustment with the hands of the abnormal deviations of the bony articulations of the vertebral column for the restoration and maintenance of health, without the use of drugs or surgery. The term "analysis" is construed to include the use of X-ray and other analytical instruments generally used in the practice of chiropractic [2,3,7].

The definition given by the American Chiropractic Association is:

Chiropractic practice is the specific adjustment and manipulation of the articulations and adjacent tissues of the body, particularly of the spinal column, for the correction of nerve interference and includes the use of recognized diagnostic methods, as indicated. Patient care is conducted with due regard for environmental, nutritional, and psychotherapeutic factors, as well as first aid, hygiene, sanitation, rehabilitation and related procedures designed to restore or maintain normal nerve function [2,4,7].

According to chiropractic, the deviation of the body segments of the vertebral column from their normal juxtaposition that interferes with nerve transmission and expression is called subluxation. Two chiropractic descriptions of subluxation are:

In other words, subluxed vertebrae (spinal bones) are characterized by fixation and misalignment within the normal range of motion.

Daniel David Palmer, a tradesman who posed as a magnetic healer, "discovered" chiropractic in 1895. Magnetic healing was a popular form of quackery in the nineteenth century in which the healers believed that their personal magnetism was so great that it gave them the power to cure disease [7]. Palmer said:

I am the originator, the Fountain Head of the essential principle that disease is the result of too much or not enough functionating [sic]. I created the art of adjusting vertebrae, using the spinous and transverse processes as levers, and named the mental act of accumulating knowledge, the cumulative function, corresponding to the physical vegetative function—growth of intellectual and physical-together, with the science, art and philosophy—Chiropractic. . . . It was I who combined the science and art and developed the principles thereof. I have answered the question —what is life? [8]

The chiropractic philosophy originated by Palmer is the frame of reference of modern-day chiropractic thinking as exemplified in the most widely used chiropractic textbook [5] at the present time. Palmer put forth the concepts of Universal Intelligence, Innate Intelligence, and Educated Intelligence. Universal Intelligence is God. Innate Intelligence is the "Soul, Spirit or Spark of Life" or "Nature, intuition, instinct, spiritual and subconscious mind." It is the "something" within the body which controls the healing process, growth, and repair and "is beyond the finite knowledge." While Innate Intelligence utilizes the autonomic nervous system, the Educated Intelligence, or "conscious," utilizes "the cerebrospinal division for the volitional expression of its function."

Nature, or Innate Intelligence, has a great capacity to maintain or restore health if it is allowed normal expression within the body. However, mental, chemical, or mechanical stress can produce a greater or lesser displacement of the vertebrae, or vertebral disrelationship, and this displacement interferes with the planned expression of Innate Intelligence through the nerves. This interference then produces pathology. The chiropractor, by correcting the displacement, allows the Innate Intelligence to effect the cure [5,9]. The pathology that chiropractors treat by manual manipulation of the spine totals over ninety diseases, including gastrointestinal, genitourinary, respiratory, vascular, and emotional disorders; diabetes; deafness; eye disorders; cancer; arthritis; and infectious diseases such as polio, mumps, hepatitis, diphtheria, and the common cold [7,10].

No one, and this includes chiropractors, has ever experimentally determined how much vertebral displacement is necessary before a spinal nerve is impinged or encroached upon at the intervertebral foramen to produce pathology by interfering with the planned expression of Innate Intelligence. This study was performed to answer that question.

Of the 43 pairs of nerves that pass from the brain and spinal cord to the various parts of the body, only 24 pairs could ever be impinged or encroached upon by the displacement of one vertebra against another as the nerves pass out of the intervertebral foramina. There is a superior and an inferior articular process posterolaterally on each side of a vertebra. Anterior to each articular process there is a notch; therefore, when the processes articulate with those of adjacent vertebrae above and below to form the vertebral column, a series of holes is formed—the intervertebral foramina. [Note: "Foramen" is the medical term for an opening through a bony structure or membrane. The plural is foramina. The intervertebral foramina are the openings between the spinal bones through which the spinal nerves emerge from the spinal cord to connect to other parts of the body.]

Part of the anterior margin of an intervertebral foramen is formed by the intervertebral disc that joins the two bodies of adjacent vertebrae together. In addition to the way the bony parts articulate with one another, numerous ligaments and muscles, both long and short, serve to bind adjacent vertebrae together to restrict their movement. Although the displacement between adjacent vertebrae is small, the range of total motion of the entire vertebral column is considerable.

Throughout life the intervertebral foramina are quite large in relationship to the spinal nerves that pass through them [11]. The cross-sectional figures below illustrate the fact that there is ample room. Figure A compares the size of the 6th cervical spinal nerve to the opening between the 5th and 6th cervical vertebrae. Figure B compares the 9th thoracic nerve to the opening between the 9th and 10th thoracic vertebrae. Figure C compares the 3rd lumbar spinal nerve to the opening between the 3rd and 4th lumbar vertebrae. A tiny artery and an intervertebral vein usually accompanying each spinal nerve through the foramen. The remainder of the space of each foramen contains very flimsy, loose areolar tissue.

Materials and Methods

The vertebral columns of six individuals were studied. Three were infants, one a full-term newborn female that failed to breathe after birth; the other two, a male and a female, were full-term infants who died of a respiratory disorder within a week after birth. The remainder were adults: a 35-year-old male who died following a heart attack, a 73-year-old male who died of pneumonitis, and a 76-year-old female who died of infectious hepatitis. The vertebral column of each individual was excised within 3 to 6 hours after death. Shortly after death each cadaver was cooled to 40°F until the vertebral column was excised.

From a posterior approach the first cervical vertebra was disarticulated from the skull, and the fifth lumbar vertebra was disarticulated from the sacrum. Each spinal nerve was transected at a point 8 cm after it emerged from its intervertebral foramen. The ribs were also transected, leaving 5 cm of their proximal ends attached to the vertebral column. All of the deep musculature of the vertebral column was left intact except the bulk of the psoas major muscles and the caudal part of the erector spinae muscles. None of the ligaments and joint capsules of the vertebral column was disturbed. Therefore, the test of the displacement of individual vertebrae in. this experiment was actually a test of the passive action of the attached ligaments to limit any displacement. In a living individual there would have been the added resistance of the attached muscles contracting in a reflex manner to inhibit vertebral displacement, or subluxation.

As soon as the vertebral column was excised it was immersed in a physiological saline solution at body temperature to insure maximum flexibility of its joints during the testing that immediately followed. A careful inspection both before and after the testing revealed that each vertebral column was normal for the age of the individual from which it was excised.

A standard drill press was used for the tests. It had a rotating handle that allowed the forceful lowering of the chuck, to which was attached a Dillon force gauge certified to be accurate to within ±1% of full scale reading. It was a compression model with marked dial increments of 10 pounds up to a capacity of 1,000 pounds. The two pressure feet used were solid metal rods that could be screwed onto the bottom of the force gauge. The end of one of the rods that exerted pressure on the vertebral column was tapered and flat; the other was forked.

The drill press had a handle that allowed the pressure foot to be locked in position while exerting continuous compression on individual vertebrae of the column. Two metal vises were clamped to the platform of the drill press to support the vertebral column while it was subjected to a compressive force. The column was only lightly clamped by the two vises supporting it. This allowed five vertebral segments of the newborn column and three of the adult column, suspended between the vises, to move freely when force was applied. The pressure foot with a forked end was used to apply compression on both sides of each vertebra by fitting it over the transverse process; it was also used to apply compression to the back of each-vertebra by fitting the forked end over the spinous process. The pressure foot with the tapered flat end was used to apply compression to the front of each vertebra of a newborn column. However, a flat piece of metal the same width as each vertebral body of the adult column had to be interposed between the pressure foot and the body because almost as soon as force was applied the tapered end began to break the bone and pass into the body.

When the part of the vertebra to which the pressure foot was applied began to break or collapse, the force was stopped. After a couple of transverse or spinous processes broke early in the testing, I learned to determine by sight, sound, and feel just about the time it was going to happen again. Each vertebral body was quite compressible: it could be compressed to about two-thirds its anteroposterior width and still rebound to its original width when the pressure was released. If compressed beyond this point, it would remain in a collapsed condition.

A Dresser torque wrench was used to quantify the amount of torsional force applied when the vertebral column was twisted. The wrench face was marked in increments of 5 foot-pounds up to 140 pounds and certified to be accurate within ±1%. The adult column was held snugly in a vise with its anterior surface facing upward. The transverse processes were hooked under the jaws of the vise to prevent the column from turning when torsional force was applied to the portion of the column projecting beyond the vise. A chain clamp was firmly applied to each vertebra in turn, beginning with the first cervical and ending with the fifth lumbar. The chain clamp had a fitting into which the end of the torque wrench was inserted. A twisting force was applied both to the right and left. The maximum force applied was at the point when it was obvious that the tissues of the column were about to rupture. While the maximal torsional force was being exerted, the spinal nerves and their intervertebral foramina were observed. The entire newborn column was easily twisted manually both to the right and left and then held in the extreme position by clamping each end of the column in a vise while the spinal nerves and intervertebral foramina were observed.

An Ametek push-pull gauge was used to quantify the amount of force applied when the vertebral column was bent in all directions. The dial was marked in 2-pound increments up to 200 pounds and certified to be accurate to within ± 0.5% of full scale. The adult column was held in a vise in the same manner as it was for the application of a torsional force. The portion of the column projecting beyond the vise was attached to the push-pull gauge by a cord wrapped around it. Segments of the column were made to project beyond the vise and maximally flexed, extended, and laterally bent both left and right to the point that the tissues of the column were about to rupture. While the segment of the column was maximally bent in one direction, the spinal nerves and their intervertebral foramina were observed. As shown in the picture below, the newborn column was so flexible that it could easily be bent in a half-circle in flexion, extension, and left and right laterally [11]. It could be held in any position of maximal bending by placing it between the pressure foot and the platform of the drill press while the spinal nerves and their intervertebral foramina were observed.

A Mura volt-ohm-microampere meter was used when the first vertebral column, from the 35-year-old male adult, was tested. The meter was used to determine whether the border of the intervertebral foramen came into contact with the spinal nerve when compressive, bending, or twisting forces

were applied to the vertebral column. The wire from the positive pole of the meter was wrapped around the spinal nerve that was placed against one side of the intervertebral foramen; the wire from the negative pole of the meter was placed against the opposite side of the foramen. The meter was set at 1,000 ohms, and if the wires barely touched each other the recording needle would make a full swing across the face of the dial. The tests of the first vertebral column revealed that the relationship of a spinal nerve to the borders of its intervertebral foramen could very easily be determined with the naked eye at all times during the continuous application of force. Therefore it was not necessary to use the meter when testing the other columns.

All the spinal nerves emerging from their intervertebral foramina were exposed prior to the testing of each vertebral column. Gentle teasing with a pair of small forceps easily removed the flimsy areolar tissue surrounding the nerves to expose the borders of their relatively spacious intervertebral foramina (Figures 2 and 6). At any time during the testing when a constant force of compression, twisting, or bending was being applied to the column, the very soft and extremely flexible spinal nerves could easily be moved about. The cut ends of the nerves were grasped with forceps and held in all positions to determine by direct observation any encroachment or impingement smaller foramina might have on the nerves.


The range of maximum compressive force, or breaking point, that could be applied to the individual vertebrae of the cervical, thoracic, and lumbar regions of the newborn and adult columns before they became irreversibly collapsed is shown in the table below.

Range of Maximum Compressive Force before
Breaking Point of Individual Vertebrae






 Vertebral bodies  






 Transverse processes    



 Spinous processes  







While a continuous maximum force was applied to a vertebra by locking the drill press handle in position, the adjacent spinal nerves and their intervertebral foramina were examined and measured. There was a slight lateral displacement of an individual vertebra when maximum pressure was applied to one of its transverse processes, along with a slight increase in the size of the adjacent intervertebral foramen in relationship to its nerve.

There was slight displacement of an individual vertebra that resulted in a reduction in the size of the adjacent foramina when the highly unphysiologic maximum pressure was applied to its body or spinous process. However, the nerves passing through these foramina could be freely moved about while the force was being continuously applied, because in the adult columns the foramina were quite spacious in relation to their spinal nerves. There was never less than 1.5 mm of space completely surrounding the cervical nerves, 3 mm of space surrounding the thoracic nerves, and 4 mm surrounding the lumbar nerves. In the newborn columns there was also a relatively large amount of space surrounding the spinal nerves during the application of a maximum compressive force. There was never less than 1 mm of space surrounding the cervical nerves and 2 mm clearance surrounding the thoracic and lumbar nerves. Upon release of the compressive force the vertebrae of both the adult and newborn columns immediately returned to their original position, and the adjacent foramina immediately returned to their original size.

The greatest amount of twisting motion of the entire adult vertebral column occurred in the upper cervical region at a maximum torsional force of 35 to 45 pounds. The next greatest amount was in the upper lumbar region, and the least in the thoracic region. When the maximal torsional force of about 10 pounds was applied to the newborn columns, the degree of twisting motion was the same throughout their length and was comparatively much greater than that of the adult column, especially in the thoracic region.

Any reduction in the size of the intervertebral foramina during the application of torsional force to both the adult and newborn columns was insignificant in relation to the spinal nerves passing through the foramina. There was always a relatively large amount of space surrounding the nerves in the foramina, As the torsional force was gradually applied, careful observation revealed that the amount of sliding motion of the nerves was insignificant in relation to the foramina. My observations indicated that the nerves did not become unduly stretched when the column was maximally twisted.

The greatest amount of flexion of the adult columns occurred in the lower cervical and the mid-lumbar regions when a maximal bending force of 50 to 60 pounds was applied. There was only moderate flexion in the thoracic region of the column from the 35-year-old male and even less in the thoracic region of the columns from the two older individuals. The greatest extension of the adult columns was seen in the cervical region, with the next greatest in the lumbar region when a maximum bending force of 50 pounds was applied. A moderate extension occurred in the thoracic region of the column from the 35-year-old male, whereas that in the thoracic region of the columns from the two older individuals was hardly detectable. The greatest lateral bending was in the cervical region of the adult columns, with the next greatest in the lumbar region when a maximum force of 50 to 60 pounds was applied. There was only a moderate amount in the thoracic region.

Any reduction in the size of the intervertebral foramina during the application of a bending force to produce flexion, extension, and lateral bending of the adult columns was insignificant in relation to the spinal nerves passing through the foramina. This was true also on the concave side of the lateral bend, where the greatest reduction in foramen size occurred. Under all conditions a relatively large amount of space surrounded the nerves in the foramina. The nerves were observed as the column was bent, and the sliding motion was seen to be insignificant relative to the possibility that the nerves might be unduly stretched when the column was maximally bent.

When 30 to 40 pounds of pressure was applied by the drill press foot to the cervical end of the newborn columns, they became maximally curved in flexion, extension, and lateral bending to form a half-circle. No reduction in the size of the intervertebral foramina in maximum flexion and extension was significant, because there was always a relatively large space surrounding the nerves in the foramina.

The cervical end is at the top. A black piece of paper was placed behind the 5th to 9th left thoracic spinal nerves where they emerge from their intervertebral foramina to make them more visible.


On the convex side of the laterally bent newborn column there was a significant increase in the size of the foramina, whereas on the concave side there was a significant decrease, to the point that the borders of the foramina made contact with nerves passing through them. However, the nerves were not markedly impinged upon and could be made to slide back and forth within the foramina when they were grasped with forceps. The extreme degree of lateral bending needed to cause encroachment of the foramina on the nerves could not occur in an intact infant because the internal organs and the body wall with its ribs would not permit it.

This experimental study demonstrates conclusively that the subluxation of a vertebra as defined by chiropractic-the exertion of pressure on a spinal nerve which by interfering with the planned expression of Innate Intelligence produces pathology-does not occur. This is what should be expected when one recognizes that the vertebral column has been evolving for over 400 million years to support the body and protect the central nervous system. By a process of natural selection the vertebral column of mammals has evolved into one in which the articulations allow an overall range of motion so that individuals may function well for survival within their environment. At the same time the selective process has favored vertebral columns that have spacious intervertebral foramina in combination with the barest minimum of displacement between adjacent vertebrae-two factors that preclude impingement upon the spinal nerves as they pass through the foramina.


  1. The Random House Dictionary of the English Language. 1966. New York: Random House.
  2. Data Sheet on Chiropractic. 1970. Chicago: Department of Investigation, American Medical Association.
  3. International Chiropractors Review. International Chiropractors Association. March 1964, p. 2.
  4. Journal of the American Chiropractic Association. Nov 1963, p. 13.
  5. Homewood AE. 1962. The Neurodynamics of the Vertebral Subluxation. Published by the author.
  6. Harper WD. 1964. Anything Can Cause Anything. San Antonio, Texas: published by the author.
  7. Smith RL. At Your Own Risk: The Case against Chiropractic. New York: Pocket Books, 1969.
  8. Palmer DD. The Science, Art and Philosophy of Chiropractic. Reprint of 1910 edition. Portland, Oregon: Portland Printing House, 1966.
  9. Cohen WJ. Independent Practitioners under Medicare: A Report to Congress. Washington, DC: Department of Health, Education, and Welfare, 1968.
  10. Chiropractic Survey and Statistical Study. 1963. A report to the Board of Directors, National Chiropractic Association. (Mimeographed) Des Moines: Bratten and Associates, 1963, pp 32-35.
  11. Crelin ES 1973. Functional Anatomy of the Newborn. New Haven, Conn.: Yale University Press, 1973.
  12. Crelin ES. Anatomy of the Newborn: An Atlas. Philadelphia: Lea and Febiger, 1969.

This article was published with additional illustrations in the September/October 1973 issue of American Scientist, the journal of the Society of Sigma Xi. At that time, Dr. Crelin was Professor of Anatomy and Chairman of the Human Growth and Development Study Unit at the Yale University School of Medicine. He had published over 100 papers on the development, structure, and physiology of bones and joints and was the author of the first atlas of the anatomy of the human newborn ever published. In a subsequent study, Crelin and others dissected 15 freshly obtained cervical spines and concluded that (a) the ligaments that held them in place would not permit a range of motion that would cause impingement of the cord or spinal nerves and (b) for impingements to occur, ligaments would have to be ripped apart and bones broken.

This page was revised on January 21, 2010.

Links to Recommended Companies