I refer my readers to the works of Hedenius, which have saved me much literary work. He distinguishes in these cases between rheumatic and deforming spondylarthritis and gives the following scheme and summary of his views.

A. Spondylarthritis Rheumatica (Anchylopoetica).

1. Acute or subacute cases almost entirely localised in the spine.

2. Subchronic or chronic cases of anchylosis of the spine.

a. Only the spine is affected (Bechterew's type).

b. Both the spine and "girdle joints" [hip and (or) shoulder joints] are affected (spondylose rhizomelique, or Strumpell-Marie's type).

c. Rheumatic polyarthritis with primary or secondary localisation in the spine.

B. Spondylarthritis Deformans.

The former variety, which has lately been again described by von Bech-terew, Strumpell, Marie, and others, is generally of rheumatic or rheumatoid (toxic) origin, but it may also be caused by serious trauma or by the repeated occurrence of slight trauma. It most commonly attacks men in early middle age, and may be regarded as a more or less local polyarthritis anchylopoetica or adhesion of the small joints of the spine, and may exhibit all the clinical variations of this disease with regard to the course it takes. The pathognomonic symptom is anchylosis of the spine. The anchyloses occurring in the last stages of the disease produce some conspicuous phenomena (anchylosis of the spine and of the larger joints of the extremities, kyphosis, etc.) which have called particular attention to this form of polyarthritis and caused it at one time to be considered as a disease "sui generis."

Spondylarthritis deformans, which has been known and described for centuries, must not be confused with the above-mentioned disease. As far as we know at present it is not of bacterial origin, is not accompanied by fever, and is generally confined to the lower part of the spine. As a rule the disease is stationary and seldom shows a tendency to spread upward. Anchylosis of the spine is here not a primary, but a secondary phenomenon. The pathological processes are often without symptoms, but may, especially if they are confined to the sides of the vertebrae, cause considerable limitation in the functions of the spine, and also very troublesome pain in the nerves. This disease does not usually attack the extremities.

It is not uncommon, especially if the disease is somewhat advanced, to find intermediate stages between these two principal forms of the disease, just as both diseases probably may appear simultaneously in the same individual. It is an this account that the two diseases have so long been confused both anatomically and clinically.

I entirely agree with Dr. Hedenius that rheumatic spondylarthritis ought not to be treated by massage directly over the small joints of the spine. The articulations between the heads of the ribs and the vertebrae cannot be affected by massage, and deep friction cannot influence the articulation between the tubercles of the ribs and the transverse processes to any great extent. The whole region of the spine is very tender, and a patient's nervous system may suffer from this ineffectual treatment. In such cases we must content ourselves with prescribing baths and injections of fibroiysin, etc., and with careful passive and active movements of the spine.

As soon as a less painful and tender stage of the rheumatic spondylarthritis has been reached we may treat by massage to counteract atrophy of the muscles. We ought to do this also as far as is practicable in cases of spondylarthritis deformans, but, according to my limited experience of these cases, we must not hope to diminish the kyphosis merely by gymnastics and massage.