I have something further to add on the lower opening of the central canal. Sixty years ago Stilling asserted that the central canal opened into the posterior median fissure of the cord in the region of the conus medullaris and that during life the entire canal was filled with cerebrospinal fluid. This statement, as well as Stillings' drawings showing this opening, was neglected or denied by most anatomists. Independently of this statement, I was able to demonstrate this lower opening by a new method in three spinal cords from young children. The method follows:

An anilin dye is injected into a cord, in the lower dorsal region, for instance, and if by chance the central canal is struck by the hyperdermic needle the dye will run down the central canal inside the cord and will appear on the outside of the cord in the region of the lower portion of the conus medullaris and filum. Plate IX shows a drawing of such a spinal cord from an infant, which shows the lower opening very nicely. If now that piece of cord which shows the dye on the outside be removed and serial sections be made, the dye will be found staining the central canal and the passage from the central canal to the posterior fissure of the spinal cord. Plates I, II and III are photomicrographs of sections of one of the cords of the Danvers series, showing how the central canal opens into the posterior median fissure and how the cord closes up again as we go below this opening into the filum terminalis. These sections of the cord were obtained from an adult aged 60 and are millimeter apart. If now in a given case, such a passage exists, or what is quite possible, a needle introduced for the purpose of making an injection, should injure the filum terminalis and make an artificial opening into the central canal, we would have a direct channel lined with ciliated epithelium, which might in a few minutes deliver a toxic material, injected into the subdural space, to the fourth ventricle. These findings, I think, show how it is possible for a poison or an infective agent to reach the meningeal surface of the cord and the central gray matter and leave the rest of the cord unaffected; that it is possible for disease to spread through the medium of the cerebrospinal fluid upward, through the central canal and by diffusion there from into the central gray matter. Poliomyelitis is a disease of childhood. An ascending paralysis may also occur in adults though more rarely than in children. One must recall the fact that this work shows

that the central canal was open in more than 7 per cent of adults. The percentage in young children is of course much higher and in the very early years of life is probably 100. In all cases of death from poliomyelitic disease, the condition of the central canal of the cord must in the future be investigated. I have been able to examine the cords from four cases of this disease in children and found the central canal patent in all of them. It is also not unusual to find the central canal containing acute inflammatory exudate which also speaks for this method of transmission. I have here photomicrographs of such a case.

I wish to add now briefly, some very interesting experiments in which it was sought to produce the condition found in acute poliomyelitis by injection into the spinal canal of dogs of small doses of dilute 2 per cent hydrocyanic acid (I c. c.). On making this injection a motor paralysis of the posterior extremities, usually more marked on the left side, is produced. Such animals have been kept alive for a varying number of days, the motor paralysis persisting, sensation remaining unaffected. When the cords of these animals are examined, lesions typical of poliomyelitis are found. We have chromatolysis of the anterior horn cells, acute inflammatory exudate throughout the gray matter and in the perivascular lymph spaces, the central canal filled with a like exudate and a replacement fibrosis in the anterior horns. In fact a pathologist would make a diagnosis from a section of such a cord as acute poliomyelitis (see Plates V, VI, VII).

I wish to show two other sections from the Danvers series which may be of some importance in explaining the pathogenesis of intra-medullary cord tumors. These sections are from the cord of case No. 1422 Danvers series. A section of the cervical portion of the cord shows that the central canal in this region is obliterated by a gliosis of the substantia gelatinosa centralis.

In the dorsal region, Fig. 12, the same process is seen, but the epithelial lining of the canal persists.

In the lumbar region, Fig. 13, this process is more marked, appearing as a small round neoplasm in the gray commissure imbedded in which may be seen part of the epithelial lining of the central canal.

FIG. 1.-Section from lower end of conus medullaris. Danvers Series No. 1624. Woman, aged 60, died of broncho-pneumonia.

FIG. 4-Section 400 microns below section shown in figure 3. One sees here that the central canal has recovered its posterior wall and continues into the filum as a terminal ventricle.