1888.] The Spinal Cord and the Cortex Cerebri. 23 



One of the non-polarisable electrodes was then brought into conta.ct 

 with the cat end of the cord and the knotted ligature, while the other 

 was connected with the longitudinal surface of the cord 2 cm. from 

 the cut end by means of soft thread cables soaked in saline solution 

 and tied loosely round the cord. In one experiment the connexion 

 was with one lateral column only. Mass movements of the elec- 

 trodes upon the spinal cord were suitably guarded against, though it 

 was found that the cord might be shaken without producing any 

 effect in the electrometer. 



On connecting these electrodes with the electrometer a consider- 

 able electromotive difference was found to exist between the contacts, 

 the excursion of the mercury being so great, i.e., beyond the field of 

 the microscope, that its amount could not be estimated in terms of 

 the micrometer eyepiece. The cut surface was always negative to the 

 longitudinal surface, and the amount of the difference as estimated by 

 the compensation method was about 0'02 D. It appeared to be highest 

 when the section passed through the dorsal region without involving 

 the lumbar enlargement. A difference between the surfaces of the 

 cord has been previously observed by du Bois-Reymond. 



The cortex cerebri was now exposed and the exciting circuit pre- 

 pared. The inductorium previously employed was again used with 

 one Daniell cell in connexion with the interrupter of primary coil 

 and the Helmholtz side wire. The exciting electrodes had platinum 

 points 2 mm. apart. 



The demarcation current having been compensated, and the electro- 

 meter placed in connexion with the non-polarisable electrodes, the 

 motor area for the lower limb was excited. The results of the obser- 

 vations made upon four monkeys and several cats may be summed 

 up as follows : 



(1.) The application of the exciting electrodes to the cortex was 

 without exception only followed by a movement in the electrometer 

 when the area of representation of the lower limb was touched, and 

 this even when owing to prolonged excitation of the arm area the 

 upper limb was in violent epileptic convulsion. We found that when 

 the exciting electrodes were moved over the surface of the brain the 

 observer at the electrometer only gave notice of a movement in the 

 instrument when the person exciting had crossed the margin of 

 representation of the limbs. This shows that electromotive changes 

 in the cord sufficient to affect our instrument occurred only when the 

 motor area of the lower limb was excited. All error due to escape is 

 thus set on one side, while at the same time this remarkable fact con- 

 firms the localisation of function. 



(2.) The excitation of the motor area for the lower limb was 

 accompanied and followed by characteristic movements of the 

 mercury (figs. 4 and 5). The excitation by means of the interrupted 



