Secondary Logo

Journal Logo

SECTION I: SYMPOSIUM: C.T. Brighton/ABJS Workshop on Computer-assisted Orthopaedic Surgery

THE CLASSIC: On a Method of Investigating the Deep Ganglia and Tracts of the Central Nervous System (Cerebellum)

Clarke, R, H; Horsley, Sir Victor

Author Information
Clinical Orthopaedics and Related Research: October 2007 - Volume 463 - Issue - p 3-6
doi: 10.1097/BLO.0b013e31814d4d99
  • Free

Robert Henry Clarke (1850-1926, Paris) (Fig 1) attended Cambridge University, received his medical training at St. George's Hospital in London, and then served as a house surgeon to Dr. Alexander Patterson at the Western Infirmary, Glasgow. After several years, he returned to St. George's. He collaborated with and joined Sir Victor Horsley (Fig 2) at University College, London in the late 1880s. He developed pneumonia in the early 90s, however, and traveled to Egypt for recovery. Cerebral localization of function had become an important issue following the work of Hughlings Jackson and others, who had stimulated the cerebral cortex with small amounts of galvanic current to plot body movements. Apparently during his time in Egypt Clarke developed the idea for a stereotaxic frame which could be clamped to an animal's head to minimize movement between the brain and probing instruments. He presented the idea to Horsley sometime after 1895. However, the first instrument was apparently not made and used experimentally until about 1906, and it is that device (Fig 3) which is reported in our Classic this month. Clarke died in Paris after a long illness, the nature of which has never been established, although he claimed it stemmed from taking a single aspirin tablet.

Fig 1
Fig 1:
Robert Henry Clarke is shown. Reprinted with courtesy of the Wellcome Library images collection, London.
Fig 2
Fig 2:
Sir Victor Horsley is shown. Reprinted with courtesy of the Wellcome Library images collection, London.
Fig 3
Fig 3:
The stereotaxic apparatus of Clarke and Horsley is shown. Reprinted with courtesy of the Wellcome Library images collection, London.

Sir Victor Alexander Haden Horsley (1857-1916, Egypt) received his initial education at Cranbrook School, then studied medicine and received his surgical training at University College London, where he received his first appointment in 1884. He was particularly interested in neurosurgery and neurophysiology, and was reportedly the first surgeon to remove a spinal tumor through a laminectomy. He authored a number of publications in various aspects of neurophysiology, through which research he became acquainted with Robert Clarke. He is perhaps best remembered for his role in developing a practical stereotaxic apparatus, although he and Clarke worked in many areas of neurophysiology and published together. For unclear reasons, the two close friends and collaborators parted ways sometime after 1907, leaving some bitterness with both. Horsley volunteered for surgery duty in Mesopotamia during the First World War, and there he died of heat stroke in what is now modern Iraq in 1916.

The apparatus described by Clarke and Horsley in our Classic established the principles for not only stereotaxis, but modern-day computer-assisted surgery, including navigation and robotics.

Richard A. Brand, MD

In the course of an investigation into the structure and functions of the cerebellum, the former by Marchi's method, the latter by observations of symptoms following artificial lesions and by excitation in normal animals, we found further progress checked by the want of a method of producing localized lesions in the deep structures-that is, nucleus dentatus, etc.-of the cerebellum without injuring neighbouring parts, and some such method was so essential for our purpose that it necessarily became for some time the principal object of our research. We were fortunate enough to find in electrolysis with insulated (glass) needles a method of producing accurately defined circumscribed excitation or lesions admirably adapted to our requirements. But to render it available for practical experiment it was necessary to find a satisfactory means of localizing deep centres and conveying the stimulation or electrolytic needle to them. The topographical methods best adapted to meet the first requirement, and the instruments to give effect to the second, had to be devised, and involved a good deal of time and experiment, which it is hardly necessary to say have not reached fluality, but we have arrived at what we consider a good working method, which has the advantage of being applicable not only to the cerebellum, but to all parts of the brain. It being designed for excitation as well as for electrolysis, it admits of a combination of stimulation and electrolysis of deep centres by which accurate localization results can be obtained in the study of the central ganglia of the brain.

In connexion with these developments of method we have accumulated a large number of facts and observations which we are collecting for publication, but the completion of which must take some time. Meanwhile, we think it desirable to publish a very brief summary of the methods we have adopted and the results so far definitely ascertained.

Having solved the difficulty of producing a circumscribed lesion, the next point was its application. As our first object was the investigation of the cerebellar nuclei, it was necessary first to localize them.


We have tried to meet this requirement as follows: First the brain is theoretically subdivided into cubic millimetres by parallel lines 1 millimetre apart in each plane. The planes employed we determine as follows: The whole brain is divided into 8 segments by three section planes roughly bisecting it in each dimension. (1) The Sagittal Section Plane, that is, the middle line, is arrived at by measuring the centre of a series of transverse diameters of the cranium. (2) The Frontal Section Plane is a section perpendicular to the last passing through the centres of both external auditory meatuses. (3) The Horizontal Section Plane is perpendicular to the two previous and cuts the frontal plane at the same distance above the external auditory meatus on each side and resting in front on the nasion. These three section planes are the bases of measurement. Parallel to each plane the brain is assumed to be cut into a series of slices or lamellae 1 mm. thick extending on both sides of the section plane and counting from it as zero. Each lamella is divided into square millimetres by lines corresponding to the other two planes, indicated by letters and numbers like the latitude and longitude of a map (which, in fact, a lamella resembles), and working charts are made by cutting frozen sections of the lamellae, covering them with a glass plate ruled in millimetre squares with fine lines, and photographing them. By this means any cubic millimetre in the brain can easily be identified, recorded, and referred to.


A mechanical contrivance being required to direct the needle to any deeply-seated cubic millimetre, for example, part of a nucleus, etc., the principle on which it is constructed is as follows: A point in a regular cube of known distance from three surfaces representing three planes, can be identified by perpendiculars of correct length dependent from these surfaces. The desired point being the only spot where they can meet, if one of these perpendiculars represent a needle, and if it is introduced in one surface at the point where the perpendiculars to the other surfaces would meet and directed forwards parallel to these two surfaces, it will engage the deep point (representing a nucleus, for example). Now the eight segments into which the brain is divided by the three section planes are regular cubes so far as these surfaces representing three planes go; our instrument, therefore, is constructed to direct a needle through a particular point on one surface of a segment parallel to the other two till it reaches the required point within the segment. To carry this out the instrument consists essentially of four parts.

  1. The Frame.-An oblong rigid structure adjusted as accurately as possible to coincide with the three section planes and firmly fixed to the head in this position.
  2. The carrier supports the needle holder and moves on guides, which are fixed to the frame in two planes.
  3. The needle holder, made of vulcanite, supports and insulates the needle and moves on the carrier, its movement being perpendicular to the guides on the frame and completing the movement in three dimensions, which is necessary to reach any point. The carrier can be moved on the guide to any part of the surface of a segment, and the needle, by the perpendicular movement between the carrier and holder, can be directed to any point within it.
  4. The needle consists of a fine steel wire nickel-plated 24 to 26 standard wire gauge, insulated to within 1 millimetre of its point by a fine glass vaccine tube about two sizes larger. The wire needle is connected for electrolysis by its proximal end through a terminal with either the anode or cathode of a battery of dry cells, the current of which is regulated by a dead-beat ammeter and rheostat, or it can be equally well employed for excitation, being attached in that case to a coil and dry cell.


Currents of 2 to 5 milliamperes are employed for two or three minutes; either anode or cathode produce a satisfactory lesion, those derived from the former being smaller and better defined. The lesion consists of a central coagulum surrounded by a necrotic zone and a very narrow zone of cedema which passes rather abruptly into healthy tissues. Lesions of any size, from a pin's point upwards, can be obtained by regulating the current, the time of its application, and the size of the needle (that is, area of metal exposed).

Application of Combined Method.

As already stated, the method of localization and instruments are equally applicable for stimulation or electrolysis, and the facility with which either can be substituted for the other possesses great advantages. When a particular result has followed an excitation and it is desired to secure an absolutely accurate record of the situation of the electrode where the effect was produced, it can be precisely defined by merely changing the leads and producing a very small electrolysis at the same point. This is subsequently verified in hardened sections by photographing it under a millimetre glass plate and comparing it with the original chart. If it is desired to mark by degeneration the tract by which the impulse travelled, the same result can be repeated in an aseptic operation by stimulation-then a sufficient lesion produced by electrolysis-the animal kept for three weeks-the brain treated by Marchi's method and the degenerated tract photographed. The whole process thus furnishes a very complete record and illustration of the structure and function of the area submitted to experiment.


We have used all forms of electrical excitation with electrodes of varied arrangement-that is, for “unipolar” and bipolar stimuli, the latter giving the most reliable results.

The precise orientation of the points stimulated was secured, as above stated, by punctiform electrolysis without moving the electrode.


The cortex cerebelli is, in our opinion, inexcitable compared to the cortex cerebri or nuclei cerebelli. The response to excitation is that of a sensory receptive rather than that of a motor efferent centre.

The apparent localization of function in the several parts of the cortex cerebelli is dependent on the association between the latter and the nuclei cerebelli.

White Substance and Communicating Fibres.

The fibres leading from the cortex of the cerebellum to the nuclei are excitable in increasing degree as the electrodes approach the nuclei in which they end.

The effects produced on the muscles are the same as those obtained by direct excitation of the nuclei.


(a) Nucleus Dentatus.

Excitation of the nucleus dentatus evokes an exceedingly constant motor result, namely:

  • (1) Conjugate deviation of the eyes, to the homolateral side, the homonymous eye tending to move more and earlier than the other.
  • (2) Head moves towards the homolateral side.

(b) Nucleus Tecti.

  • (1) Rotation and deviation of eyes to homolateral side. Frequently skew deviation.
  • (2) Head moves moves towards the homolateral side, but less actively than from excitation of the nucleus dentatus.

(c) Vestibular Nuclei, including Deiter's nucleus, Bechterew's nucleus, and the subdivisions of the nucleus of origin of the vestibular nerve (Sabin).

This important nuclear station we regard as intermediate between the cerebellum (that is, cortex and nuclei) and the spinal cord.

Direct excitation of these foci evokes movements of the face, trunks and limbs of definite character which have been regarded hitherto as of cerebellar origin.


The present preliminary communication is too brief to state the symptomatic results of the lesions produced in the degeneration experiments. They will be given in the detailed paper.

© 2007 Lippincott Williams & Wilkins, Inc.