J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

Journal ofNeurology, Neurosurgery, and Psychiatry, 1977, 40, 417-433

Proprioceptive pathways of the

RICHARD J. SCHNEIDER, A. T. KULICS, AND THOMAS B. DUCKER From the Division of Neurological Surgery, University of Maryland School of Medicine, Baltimore, Maryland, USA

SUMMARY In the Macaque, surgical lesions were made in the dorsal , in the dorso- , and through half of the spinal cord. The somatosensory and motor capacity of the animal were examined neurologically and electrophysiologically. The exact lesion was then confirmed pathologically in detail. The results of these experiments indicate that limb position information from the distal limb and proximal limb are relayed to the brain in two different fashions. Distal limb position information, especielly the cortical representation of the limbs' volar surface as it moves in space, is drastically impaired by dorsal funiculus or posterior white column lesions. Proximal limb position may or may not be impaired by similar lesions, for this information while initially in the dorsal or posterior white columns is sorted out (as it ascends in the spinal cord) to the dorsolateral funiculus or white columns. For example, in the lower thoracic spinal cord, both distal and proximal hind limb sensation are impaired -by posterior Protected by copyright. white column damage; in the cervical cord, only distal sensation is impaired by the&same lesion, and proximal information is spared. We refer to this neuroanatomic rearranging as 'fibre sorting', and we believe that it is clinically significant in spinal cord disease.

'rhe role of the dorsal funiculus or the posterior tion between the dorsal funiculus and the lateral white columns of the spinal cord in somatic sensi- funiculus is not so clear-cut as diagrams picture bility has for some time been an issue of debate. it, the general interpretation accepts this dissocia- Dissatisfaction has arisen with the accepted and tion as fundamental. widely disseminated viewpoint, as stated by Animal experiments have not given such clear Mountcastle and Darian-Smith (1968): 'What re- results. Thus Ferraro and Barrera (1934) found mains in the mechanoreceptive sphere after large that macaques with posterior or dorsal funicular fiber or dorsal column lesions is the capacity to lesions rostral to the cerwical plexus showed few recognize that a mechanical stimulus has oc- deficits in the ability to their hindlimbs. u5e They http://jnnp.bmj.com/ curred; it is no longer possible to specify it as to attributed this to the preservation of propriocep- intensity, location or shape.' This interpretation is tive sensation fibres from the hindlimb ascending supported by much of the clinical literature. The in the spinocerebellar tracts in the lateral funi- results of neurological examinations of humans culus, and thus escaping their lesions. Similar with various spinal cord pathologies are often results from a series of experiments on macaques interpreted as Walshe (1970) does: 'The . .. done by Gilman and Denny-Brown (1966) led sensory afferent path which subserves touch and them to hypothesise that lesions of the posterior pressure, postural sense and also vibration sense white columns produced a deficit in the ability to on September 24, 2021 by guest. turns upwards in the dorsal columns and reaches explore and orient to extrapersonal space, but the dorsal column nucleus of the medulla on the spared manipulation like grooming which involved side of entry.' Though recognition has been given intrapersonal space. Moreover, electrophysio- by Walshe to the fact that the dissociation of func- logical and anatomical studies have shown that This work was initiated at the University of Pittsburgh with NIH somatosensory information can, under certain grants MHI11682 and NB07712, and completed at the University of conditions, be transmitted by other than lemniscal Maryland with funds from the Bressler Foundation. pathways (Albe-Fessard, 1967), and that pathways Address for reprint requests: Dr R. J. Schneider, Neurosurgery Division, 12th Floor, University of Maryland Hospital, Baltimore, other than those in the dorsal funiculus may con- Maryland 21201, USA. tribute more heavily to somatosensation in certain Accepted 15 November 1976 species such as carnivores.(Bowsher, 1965). 417 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

418 Richard J. Schneider, A. T. Kulics, and Thomas B. Ducker These results indicate a more limited role for 1964; Andersson et al., 1965; Landgren et al., the funiculi gracilis and cuneatus. Semmes (1969), 1.965; Albe-Fessard, 1967) and cortical levels in re-examining the concepts of epicritic and pro- (Powell and Mountcastle, 1959b; Schneider, 1972; topathic sensation, called attention to the possi- Dreyer et al., 1974). bility that the phylogenetically newer 'lemniscal In the light of these various lines of experi- system' was specially involved in exploration and mental results, we proposed to determine whether manipulation or active touch. Wall (1970) con- it is necessary to postulate a special role for the cluded that the dorsal funiculi were necessary for dorsal white column in somatic sensibility in order external search behaviour when internal search to explain the existing behavioural and neuro- for identification of a somatosensory event failed logical evidence, or whether a reinterpretation of to produce a satisfactory answer. He wrote: 'Im- previous results considering such processes as pulses ascending in the dorsal column-medial fibre sorting could explain these findings. system are involved in controlling the analysis of messages arriving over the other - Methods tosensory pathways.' This restricted the role of the dorsal funiculi to that of a governor in the Preoperative and postoperative neurological and cybernetic sense. Melzack and Bridges (1971), behavioural examinations were performed on each after a series of experiments on cats with dorsal of 18 adolescent monkeys (Macaca mulatta) which funiculus lesions, augmented Wall's control sys- were confined in a restraint apparatus (Carlson, tem view by concluding that the dorsal funiculi 1972). Preoperatively, the examinations were played an important role in the choice of an ap- carried out from one to several hours before the propriate behaviour programme. Dubrovsky et al. surgical lesions. The surgical procedure was at-

(1971), delimited this role in behavioural pro- tendant upon completion of a full examination Protected by copyright. gramme selection by concluding, from their work without questionable responses in any category. on cats, that the dorsal funiculus was necessary Postoperatively, several examinations were made for planning and execution of serial acts in which at different periods of elapsed time from the sur- accurate timing, precise forelimb projection, and gical procedure-usually within the first week, and space orientation are necessary conditions. again on the second and fourth weeks. Behavioural Concurrent with this behavioural research, a observations were also made on the animals pre- body of anatomical and electrophysiological data operatively and postoperatively, with special was growing which supplemented and modified attention to the immediate postoperative period. traditional concepts of the posterior white column. The neurological examinations tested the fol- Thus, it was demonstrated that only 25% of the lowing six categories of response: (1) grasp myelinated fibres entering these columns from reactions-that is, the true grasp reflex and in- the hindlimbs reach the gracilis nuclei (Glees and stinctive grasping-which are distinguished by the Soler, 1951). Analyses of peripheral afferent fibres fact that in ascending order of complexity lie the (Iggo, 1963) showed that in several different types coarse grasp or traction reaction, the true grasp mechanoreceptive information was being trans- reflex, and instinctive grasping (Denny-Brown mitted centrally. Later electrophysiological experi- and Chambers, 1958; Twitchell, 1965); this last- http://jnnp.bmj.com/ ments showed that fibres conveying several of named reaction is a contactual orienting reaction these modes of mechanoreceptive information whose voluntary performance has also been called present at more caudal levels of the dorsal white spontaneous grip or spontaneous grasp (Ferraro columns disappeared, or were severely diminished, and Barrera, 1934); (2) tactile and contact placing at more rostral levels in cats (Petit and Burgess, (reflexes of the hindlimb); (3) limb reactions to 1968; Burgess and Clark, 1969) and in monkeys light touch of the skin, brushing of the hairs, (Whitsel et al., 1969). Furthermore, a segregation pricking and pinching (with pin and toothed of submodalities was seen in the nuclei of the forceps); (4) muscle tonus; (5) proprioceptive on September 24, 2021 by guest. dorsal funiculus between proprioceptive and ex- reactions-ability to coordinate limb movements teroceptive afferent information (Kuhn, 1949; in space, to locate and secure extrapersonal Gordon and Paine, 1960; Gordon and Seed, 1961; objects, and to maintain normal as opposed to Perl et al., 1962; Gordon and Jukes, 1964; Bieden- catatonic postures; (6) plantar and patellar tendon bach, 1972), and this dissociation of submodalities reflexes. Forelimb tactile placing was not reported was reflected by the cell types in these regions in our experimental results because the various (Kuypers and Tuerk, 1964; Hand, 1966; Bieden- lesions did not offer us the chance to evaluate bach, 1972). A similar dissociation has been shown enough animals where fibre sorting in the at thalamic (Jasper and Bertrand, 1964; Mallart, fasciculus cuneatus could have occurred. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

Proprioceptive pathways of the spinal cord 419 The standard neurological tests were undertaken anterograde and retrograde tract degeneration especially to evaluate any damage to motor path- were determined by the Marchi method. Recon- ways in the lateral funiculus which might influence structions of spinal cord lesions were drawn from our behavioural observations or electrophysio- slide projections of histological sections. These are logical recording. We felt that corticospinal tract displayed in Figs 1, 2, and 3. Cortical micro- damage would have implied damage to the dorsal electrode penetrations were reconstructed to per- part of the lateral funiculus which would also mit analysis of the cytoarchitectural area traversed have interrupted somatosensory projections other and laminar location of the neurones isolated in than the dorsal white columns, such as the dorsal each penetration (Schneider, 1972; Dreyer et al., spino-cerebellar tract, and Morin's pathway (Ha 1974). and Morin, 1964; Ha, 1971), the spino-cervico- lemniscal tract. Similarly, the integrity of the Results anterolateral projection was assessed by pricking and pinching stimuli, since damage here would Basically, there are four spinal cord lesions: the imply ventral involvement as dorsal funiculus, the lateral funiculus (in the well as interruption of mechanoreceptive afferent dorsal aspect only), the dorsal and dorsolateral fibres within the anterolateral pathway. funiculus combination, and the combination lesion The operative procedure was as follows. extended into the cord hemisection. The dorsal Animals were anaesthetised with sodium pento- funicular lesion cut the posterior white tracts, barbitone 30 mg/kg ip (five animals) or 25 mg/kg funiculus gracilis in the low thoracic area, and iv (13 animals). Intravenous catheters were in- both gracilis and cuneatus above. The lateral serted for supplemental anaesthesia during sur- funicular lesion spared the motor tracts and the gery. A laminectomy was then performed anterolateral quadrant but attempted to cut the aseptically. The was split and retracted posterior spinocerebellar tracts and Morin's path- Protected by copyright. with suture ties under a microscope; the pia mater way. The dorsal and dorsolateral lesion (see Fig. was also retracted. A Bard-Parker gauge 11 blade, 4) included all those tracts to various extents. cut off, marked in millimetre gradations, and held Finally, the hemisection definitely included the in a haemostat was used to make the lesion. The motor system. dura mater was then reapposed with 5-0 gut or silk, gel-foam was placed over the cut spinous LESIONS OF THE DORSAL FUNICULUS processes, and muscle, fascia, and skin rejoined. Perhaps the signal manifestation of an animal Procaine penicillin (one million units) was ad- with a complete dorsal lesion is his ministered immediately after the operation, and propensity to place the most distal aspect of the for several days thereafter. Animals were observed extremity (hand or foot) on the floor with its after surgery, and any sign of infection was dorsal rather than ventral surface down. We call treated promptly. Three control laminectomies this dorsal placing. This observation has been were performed using the same procedures; emphasised by others (Ferraro and Barrera, 1934; permanent neurological defects and electro- Gilman and Denny-Brown, 1966). It occurred physiological changes were absent from these even when the limb was oriented appropriately, http://jnnp.bmj.com/ preparations. and it was most noticeable in animals with only Microelectrode recordings were obtained from a unilateral lesion. but it was definitely present in the primary somato-sensory cortex of the monkeys bilateral lesions. approximately one month after spinal surgery Lesions in the dorsal funiculus alone were either (Schneider, 1972; Dreyer et al., 1974). At the end in the thoracic area or cervical area. When the of each electrophysiological experiment, the animal had a low thoracic lesion, both distal and animal was deeply anaesthetised with pentobarbi- proximal proprioceptive function were markedly tone, and sacrificed by intracardial infusion with impaired, and the animal had little use of the on September 24, 2021 by guest. 0.9% saline followed by 10% neutral buffered lower limb. In high thoracic lesions, the monkey formalin. Appropriate portions of the brain were demonstrated normal posture and a steady gait. imbedded in celloidin, and coronal or sagittal It would be active and climb the cage wall. But sections were taken at 30 um; the sections were in the feet, grasp was impaired, palpatory be- stained with cresylecht violet. The spinal cord in haviour was diminished, dorsal placing of the the area of the lesion, as well as selected sections foot did occur, and distal malalignment was rostral and caudal to it, were also imbedded in noticeable. Thus, distal proprioceptive function celloidin. The extent of the lesion was determined was impaired but proximal proprioception ap- by examining Mahon stained sections, while peared spared. In midcervical lesions, the same J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

420 Richard J. Schneider, A. T. Kulics, and Thomas B. Ducker Protected by copyright.

Fig. 1 Figs. 1, 2, and 3 Drawings made from projections of stained tissue sections from each animal in this experimental series. Total extent of lesions was determined by overlapping consecutive sections. Striped area implies total damage, circles imply partial damage. Rostral and caudal drawings were made at levels beyond which lesion size diminished with no additional pathology. Rostral drawings in DCX:12 were determined by ascending degeneration pattern, only two drawings were done for DCX:8, and was

DCX:J5 http://jnnp.bmj.com/ drawn from written account of total extent of lesion. deficit was present in the feet and, in addition, dentate ligaments, with sparing of the cortico- the forelimbs were now markedly affected. The spinal tracts, the animals had surprising difficulty. arms flailed during ambulation, and were held in In both low cervical lesions, the animals upon a sling-like posture at rest. The animal would use recovery from surgery would climb their cage the entire forearm to palpate the surroundings. shelf and cage walls, but showed a preference for

There was some recovery with time but skilled lying on the cage floor, with hindquarters slightly on September 24, 2021 by guest. manipulation was dependent on visual cues. In elevated because of an increased tonus in the hip the very high cervical lesions in one animal fore- extensors. The limbs ipsilateral to the lesion had limb, proximal proprioceptive function appeared no trouble localising objects in space, but upon to recover, but distal position and stereopsis reaching for such objects their grasp was markedly remained disordered. diminished in vigour. The orientation of their feet at rest was often at unusual rotational angles L.FSIONS OF THE LATERAL FUNICULUS to the body-that is, at right angles to the In lateral funicular lesions, in the dorsolateral direction the body was facing. Forelimbs most (posterolateral) part of the cord, behind the often showed a posture of shoulder adduction, J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

Proprioceptive pathways of the spinal cord 421

ROSTRAL

CAUL L

CAUDAL Protected by copyright.

Fig. 2

elbow and wrist flexion when the animals were in refused to stand and would lie in postural hyper- a sitting position, somewhat like the animals with extension (Gilman and Denny-Brown, 1966). high dorsal funiculus lesions. The affected hind- When the lesion was further extended to cord limbs manifested a slight resting tremor, and hemisection, there was additional flaccid paralysis showed plantar flexion of the toes at rest, which of the ipsilateral extremity plus self-mutilation of http://jnnp.bmj.com/ was greater at the interphalangeal than at the the affected limb. metatarso-phalangeal joints. During ambulation Tables 1 and 2 give the results of the neuro- the toes tended to be widely splayed providing a logical examinations done on animals whose greater contactual circumference. With vision tractotomies included pathways as verified by diverted, distal limb contact with objects caused . The scheme of these tables includes peculiar reactions. Among these were staring in only residual changes at one month after surgery the direction of the surface touched, repeated since this was the time when electrophysiological collision with the surface when vision remained mapping was commonly performed. Except with on September 24, 2021 by guest. elsewhere directed, and what may have been a pinch (which was to study pain sensation), entries startle reaction to encountering the object. refer to the limb ipsilateral to the lesion. To make the tables more understandable, follow MORE EXTENSIVE LESIONS a typical entry, DCX: 4. This was an animal with Combining lesions of both dorsal and dorso- a lesion of the dorsal funiculus of the left side lateral funiculi from the sixth cervical to the (see comment), at the level of the third thoracic seventh thoracic segment had a profound effect segment of the spinal cord. One month after sur- on the animal, for it made no attempt at move- gery, his left forelimb pulled away from grasp ment in and around the cage. The monkeys eliciting stimuli, and his left hindlimb failed to J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

422 Richard J. Schneider, A. T. Kulics, and Thomas B. Ducker Protected by copyright.

Fig. 3 react to such stimuli. He would palpate and orient Several of these results should be emphasised. to things with his left forelimb but not with his The hindlimb grasp reflex is almost uniformly http://jnnp.bmj.com/ left hindlimb. The left hindlimb did not place absent after dorsal funicular transection and after when stimulation was delivered to hairs or lightly the combination lesion, with or without motor to the skin surface, but did place when deeper involvement, when above the level of entry for pressure to the foot was applied. The left forelimb afferent fibres from the affected limb. Instead, the but not the hindlimb reacted to hair displacement hindlimbs usually display no reaction, and the and light tapping on the skin surface. There was a forelimbs show tactile avoiding, a response which normal withdrawal to pinch by the right limbs, appears in the early postoperative period in the normal muscle tone in the left limbs, normal limb hindlimbs, and diminishes thereafter. Partial dor- on September 24, 2021 by guest. movements on the left side when reaching for sal funicular lesions (DCX: 8, DCX: 9, DCX: 15) objects or orienting in space, and a normal may or may not spare some grasp reflex. In pure patellar but no plantar reflex in the left hindlimb. seventh cervical dorsolateral transections, the fore- The animal had a tractotomy which extended to limbs were unaffected, while the hindlimbs could the right side where it included the dorsal and show a diminution which might be traceable to dorsolateral funiculi. It is thus listed in Table 2 as the weakened grasp rather than an absence of the well. (The comments column lists the side ex- reflex. In order to abolish forelimb instinctive amined when both sides were so used, or when grasping, most of the forelimb plexus should be ambiguity might exist.) below a dorsal funicular lesion or hemisection. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

Proprioceptive pathways of the spinal cord 423

A B

C Protected by copyright.

,- ".

11o. I.,.,..w....:.v

.s or 0 http://jnnp.bmj.com/

r .o

_ 'St-.AP .ew - E F on September 24, 2021 by guest. Fig. 4 A. Complete right dorsal funiculus lesion at C5 segment with left dorsal funiculus involvement (DCX:16). B. Ascending pattern of axonal degeneration (DCX:16). C. Spinal cord h1emisection at C7 segment, with partial sparing in medial regions of dorsal and ventral funiculi (DCX:2). D. Dorsal funiculus lesion at C3 segment with some dorsolateral funiculus involvement (DCX:6). E. Dorsolateral funiculus lesion at C7 segment involving mainly proposed area of projection of spino-cervico-lemniscal tract fibres with some sparing of dorsal spinocerebellar tract fibres (DCX:17) (see Fig. 3). F. Dorsolateral funiculus lesion involving spino-cervico-lemniscal tract and presumably most dorsal spinocerebellar tract fibres (DCX: 18).

B J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

424 Richard J. Schneider, A. T. Kulics, and Thomas B. Ducker

C)V) C)7C) 0 o Cd eV o o _ 0o be:X ; Y CU YXCU o Q CUQ> CU Cidm, od.:>0 m m o m ,C,~~ ~~~ 0u C.6o LL *2 -o 0 - -+ 0 + - a'a-t

.0 O 0 + a- + 0 0 O-- + + 0

_ + -i- -4- -i+ + O t- 0 0 0

._ a+ + + + + + +a+ + + 0 ±t + + a- II + + + +

.0 Protected by copyright. + + + + + a-L_ a a- + + -a-+ +

+ + + + ± +_ -t- -+ a- °0 + a- E ,0 11 -a t- 0 + + o 0 + a--+ 0 -ai o 0 0 0 0 0 ~C t- + 0 + o 0 + + + 0 O 0 0 0 0 +0

+ + + a- + a- + 0 + 0 S It 0 http://jnnp.bmj.com/ 0 0 0 0 0 0 f+t+ + t. + t + + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Q 0o 0 0 O0 0 0

a- + 0 a- a+ 0 0 + + + 0 Cs .2U)

-Co on September 24, 2021 by guest. o 0 0 0 0 0 0 0

z + a- -i- --a +--i+ + + !-a-- QZ)

L- 5D < 00 e 00 W0 C 't - C- o m i;L

,It in) Uc O) ON 2 2 ri x x x x x x x x x x x x OZ ) c 0 0 ~~~~~0 0 0 C) 0 ) C) Uz C J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

Proprioceptive pathways of the spinal cord 425

0 5 4) --a -o 0 )0 -o 0 o o . 0o _ to_ 0 E to~_ Q~~~~~~~~~ 0 c0 c = .- co .

to , -; E 0 L. +>4 + +* -f+t + Q-f _7

m v0 m +Q +1 +

ot o - 0 0 v 0 -~~~~~~~~~~~~~~- + +

> I I + t _4 +- ± 1 s°0 +> 4

o 0 4> - + 1- + t-" Protected by copyright. s 1 +~~~~~~~~~4 co <-.

.00 L2 0 '1- + + O + o 0 0 O 0 0 0 0C {-0 O

4 o 0 4> 1- -) cz

o CC 0 0 0 0 0 0 0w

O 0 0 0 0 http://jnnp.bmj.com/ ~0 -4 0X - 0l 0 0 0 o 0 0 0 0, 0Z 7 0 O + + -0 , + 03 _ 4 In *tc o 0 I 0 0 on September 24, 2021 by guest. +- t 0 0 + 0c > 2-4- - - I-~ -I -+ + I- +1 -t±+ + + H--CO H t Qxo Q , 0 U Hu .0

-4 -,, "t _ _ _1 _2 X >I-L Cd C) 0 c QXa x x x U u U U U U 0 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

426 Richard J. Schneider, A. T. Kulics, and Thomas B. Ducker Thus, a fourth cervical lesion (DCX: 11), but not tion of exteroceptive information combined with an eighth cervical lesion (DCX: 5, DCX: 7), the proprioceptive redistribution which concen- abolishes forelimb instinctive grasping, and a sixth trates, within the posterior white columns, afferent cervical hemisection (DCX: 1) leads to abolition, fibres related to hair displacement and those re- while a partial lesion at the fifth or sixth cervical lated to light touch on the glabrous skin surfaces segment (DCX: 8, DCX: 9) may or may not spare of the distal limbs. We believe our results support instinctive grasping in the forelimb. In the hind- this scheme in the following ways. limb, a pure dorsal funicular lesion at high thoracic The contact (proprioceptive) placing reflex in to cervical levels will suffice to eliminate instinc- the hindlimb is not abolished in macaques whose tive grasping, as does expanding the lesion to dorsal funiculus is interrupted at levels above include additional pathology. But a seventh cer- T5 segment (DCX: 4, 5, 6, 7, 8, 9, 10, 11, 16), but vical dorsolateral lesion alone does not affect is absent in macaques with dorsal funiculus inter- instinctive grasp in any limb (DCX: 17, 18). ruptions at or below T5 (DCX: 13, 15), and in Tactile, but not contact, placing of the hind- rnacaques with lateral funiculus or hemicord in- limb is abolished by pure posterior white column volvement above T4 segment (DCX: 1, 2, 3, 4, lesions, except in the cases of DCX: 13 which was 10). However, lateral funiculus tractotomies alone an unusually low lesion, and DCX: 15 which had (DCX: 17, 18) do not abolish this reflex. The some postoperative complications affecting the contact placing reflex depends upon proprioceptive legs. Both hindlimb tactile and contact placing information being conveyed to the frontomotor disappeared with combination lesions, but were cortex (Woolsey and Bard, 1963). Since transect- intact with pure dorsolateral lesions. Hair dis- ing the dorsal funiculus does not necessarily placement reactions as well as light touch or tap abolish this reflex, the implication is that the reactions disappear pari passu with interruption necessary afferent proprioceptive information does of posterior white column fibres related to recep- Inot project in an uninterrupted manner exclu- Protected by copyright. tive fields on the body surface in question. With- sively through the posterior white column. In drawal to pinching is exhibited only in the contrast, hindlimb tactile (exteroceptive) placing, contralateral limb whose fibres enter caudal to a reflex depending on exteroceptive information hemisection. After a month, muscle tone is seldom being relayed to the parietal lobes (Woolsey and affected in pure posterior white column transected Bard, 1963), is abolished by dorsal funiculus preparations, but hypotonia ensues with hemi- lesions regardless of the level, and is not abolished sections. Proprioception is completely absent in by isolated lateral funiculus lesions. This implies the forelimb after either posterior white column that the necessary afferent exteroceptive informa- or a combination lesion above the brachial plexus. tion for the tactile placing reflex does project in an Proprioception is absent in the hindlimbs only uninterrupted fashion exclusively through the when the posterior white column lesion is just posterior white column. The fact that focal lateral above the , or when the lesion funiculus lesions which are dorsally placed do not includes the lateral funiculus at higher segmental abolish proprioceptive placing may imply a shared levels. projection of these afferent fibres between the two

tracts. But a more parsimonious conclusion would http://jnnp.bmj.com/ Discussion be that, since tactile placing is still intact in macaques with isolated lateral tractotomnies, skin We interpret our results to imply that the re- contact before the proprioceptive stimUlus thres- arrangement of sensory afferent fibres, which has hold is reached elicits a tactile placing reflex which been shown to occur anatomically and electro- obscures the fact that proprioceptive placing is physiologically (Glees and Soler, 1951; Petit and absent. Burgess, 1968; Burgess and Clark, 1969; Whitsel Placing reactions have been examined after et al., 1969), produces predictable behavioural and transections of the dorsal funiculus of macaques on September 24, 2021 by guest. neurological changes in the effects of tractotomies. (Ferraro and Barrera, 1934; Gilman and Denny- The main outlines of this rearrangement, which Brown, 1966; Liebman and Levitt, 1973), and in we call 'fibre sorting', are (1) the redistribution domestic cats (Lundberg and Norrsell, 1960; of proprioceptive information from the hindlimb Dobry and Casey, 1972). In cats it was found to pathways other than the dorsal funiculus as that tactile placing disappeared after ipsilateral this information is transmitted rostrally, (2) the transection of the medio-dorsal lateral funiculus appearance of this proprioceptive information in rostral to the level of the hindlimb afferent entry the lateral funiculus (dorsal aspect) at rostral of the spino-cervico-lemniscal tract (Morin's spinal cord levels, and (3) the rostral reorganisa- pathway), and caudal to the lateral cervical J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

Proprioceptive pathways of the spinal cord 427 nucleus (Lundberg and Norrsell, 1960). However, funicular tractotomies at low cervical levels lesions of Morin's pathway in its contralateral (DCX: 17, 18). Animals occasionally showed ventral funicular region, cranial to the lateral proprioceptive dysfunctions which did not fit cervical nucleus, do not abolish tactile placing in this systematic pattern (DCX: 5, 11), but we cats (Norrsell and Voorhoeve, 1962). It was con- believe these are the exception rather than the cluded that Morin's pathway is not the exclusive rule. afferent link in this cortical reflex for cats Electrophysiological results in the isolated (Norrsell and Voorhoeve, 1962). Lesions of the dorsolateral funiculus tractotomy experiments dorsal funiculus at the first cervical level which lend support to the occurrence of proprioceptive included more than 9000 of its afferent fibres, (deep-submodality) afferent fibres in the rostral were also reported to abolish tactile placing in dorsolateral funiculus. These data show that cats (Dobry and Casey, 1972). damage involving Morin's pathway does not affect In macaques, Gilman and Denny-Brown (1966) the distal, deep-submodality, single cell representa- reported that to demonstrate tactile placing in tion in Brodmann's area 2 of macaques as severely the hindlimbs after fourth cervical dorsal funi- as when the spinal cord position of the dorsal culus transection requires brisk cutaneous contact spinocerebellar tract is encroached upon to proximal parts of the leg, and in four of their (Schneider, 1972; Dreyer et al., 1974). This may animals (DC2, DC4, DC15, and DC46) they con- imply simply that there is a direct relationship sidered tactile placing to be absent. It was also between the extent of a dorsolateral funiculus found that the more extensive combination (dorsal lesion, and the loss of afferent information pro- and dorsolateral funiculus) lesion at the third jecting to cytoarchitectural area 2 of the brain. cervical level was required to abolish contact This evidence tends to allocate a role, if not the placing (since these authors did not use contact exclusive role, in relaying deep afferent impulses Protected by copyright. placing, we make this assessment from their to the primate dorsal spinocerebellar tract. How- statement, 'no limb placed with even broad, ever, the evidence is that the contact placing heavy, prolonged contact') along with tactile reflex is mediated through the frontomotor areas placing in the hindlimb, and that dorsolateral in macaques, so that change in occurrence of funicular lesions alone left no residual defect in proprioceptive units in Brodmann's area 2, while tactile placing (Gilman and Denny-Brown, 1966). significant for somatic sensation, would not ac- In support of the notion that both dorsal funiculus count for the abolition of this response. Recent and dorsolateral funiculus share fibres involved in results (Asanuma and Rosen, 1972a and b) show the projection of proprioceptive information neces- that, within the precentral motor area, there are sary to the contact placing reflex, other research cells which receive afferent proprioceptive input. indicates that this is the case for position sense When stimulated electrically these same cells pro- (Vierck, 1966). The same studies infer that the duce isolated peripheral muscle contractions. anterolateral funiculus also pays a role in position These cells may receive information identical to sense. that transmitted to the primary somatosensory We believe that the proprioceptive disturbances cortex which may subserve positive feedback or we observe are linked to afferent fibres which precentral cortical reflexes such as contact plac- http://jnnp.bmj.com/ travel both in the dorsal funiculus and in pathways ing. Elimination of such feedback has been shown in the dorsolateral funiculus at more rostral to affect the output of cells in the motor cortex levels. This may account for the occurrence of (Lewis et al., 1971). these disturbances in the hindlimbs of animals Single cell studies of the cortex (Schneider, with low cervical and high thoracic, dorsal funi- 1972; Dreyer et al., 1974) reveal the presence of culus plus lateral funiculus tractotomies (DCX: 1, deep-submodality receiving cells from the hind- 2. 3, 4, 10, 14. 15); in the forelimbs of animals limb in Brodmann's areas 1 and 2 (postcentral with high cervical (DCX: 6, 10, 11) and the hind- cytoarchitectural areas known to receive such a on September 24, 2021 by guest. limbs of animals with low thoracic (DCX: 13) proprioceptive population). However, there is a dorsal funicular tractotomies (in these, proprio- reduction of the numbers of proprioceptive ceptive fibres have not yet undergone fibre sorting, neurones from the forelimbs of animals with high although, a decrease in hindlimb position sense cervical dorsal funicular tractotomies, and from following a dorsal funiculus lesion as high as the the hindlimbs of animals with damage laterally in seventh thoracic level has been reported (Ferraro the dorsal spinocerebellar tract or with unilateral and Barrera, 1934) and we observed one (DCX: 15 hemisections. These data support the interpreta- at the fifth thoracic level) but not in the fore- tion that the segregation of exteroceptive and limbs or hindlimbs of animals with dorsolateral proprioceptive afferent fibres within the spinal J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

428 Richard J. Schneider, A. T. Kulics, and Thomas B. Ducker cord (Whitsel et al., 1969, 1970) can account for lack of sensation, we do believe that inattention the selective diminution of proprioceptive or or diminished sensitivity is indicated. This con- exteroceptive cells from the cortical population viction is supported by the radical diminution in when tractotomies transect these afferent fibres hair and touch related single cells encountered after they have been separated. Thus, spinal cord in microelectrode penetrations of the somato- fibres sorting can be shown to play an important sensory cortex (Schneider, 1972; Dreyer et al., role in the determination of the body representa- 1974). This diminution is significantly more ex- tion in the primary somatosensory cortex (Whitsel tensive when the population of tactile receptors et al., 1972). Such fibre sorting contributes to the of the glabrous skin of the fore and hindlimb is evaluation of the rostro-caudal level and extent of considered (Schneider, 1972; Dreyer et al., 1974). spinal cord lesions. In the macaque, this refers to the palm and sole. It should be noted that the assessment of pro- The population of tactile receptors on the palm prioceptive losses may involve a number of re- also supplies the largest input to those cells of the lated, but different qualities. For example, hand represented in the motor cortex which pro- Ferraro and Barrera (1934) used the failure of duce flexion of the fingers (Asanuma and Rosen, slow movements of any limb to evoke eye move- 1972b). These data suggest critical roles for such ments in their fourth cervical preparation to receptors in tactile exploration, manipulation, and attribute loss of proprioception to their dorsal grasping. This is not to say that these functions funiculus interruption. We find this phenomenon must be exclusively subserved by this receptor also, but are not willing to say it implies complete population. In the macaque, there is evidence absence of proprioceptive sense. These authors that other pathways can be used in tasks involving also noted the prominent dorsal placing of the these areas (Kuhn, 1949; Schwartz et al., 1972; distal limbs of such animals, and in one case used Vierck, 1973). Even less can we conclude that it to assign proprioceptive loss to the hindlimb. other body areas are the sole domain of one Protected by copyright. Our electrophysiological results indicate that the afferent system (Mettler and Liss, 1959; Christian- marked impairment of cutaneous sensation on the sen, 1966; Norrsell, 1966; Kitai and Weinberg, volar surface of both distal limbs, observed after 1968; Schwartzman and Bogdonoff, 1968, 1969; dorsal funiculus transection above the entry of Dobry and Casey, 1972; Eidelberg and Woodbury, afferent fibres from those limbs, could probably 1972; Mann et al., 1972; Schwartz et al., 1972; result in this unusual posture, the feedback Liebman and Levitt, 1973). However, it has been necessary for spatial position sense at rest being demonstrated (Ferraro and Barrera, 1934; Gilman obtainable only from the remaining innervation of and Denny-Brown, 1966; Wall, 1970; Dubrovsky the foot and hand. Thus, an exteroceptive rather et al., 1971; Melzack and Bridges, 1971; Schneider, than a proprioceptive loss may be the basis for 1972) that the quality of behaviour which uses this observation. Schiff, in 1857, made a similar these receptors changes after the elimination of claim (cf. Ferraro and Barrera, 1934). Gilman dorsal funiculus pathways, and even when be- and Denny-Brown (1966) cite catatonic postures havioural testing is included the disturbances are as evidence of loss of proprioception in animals nonetheless present. with a high cervical dorsal funiculus lesion. While With regard to the possibility that inattention is http://jnnp.bmj.com/ we have also observed this on occasion, the the result of a dorsal funiculus lesion, Wall (1970) ability to use the hindlimbs without visual guid- presented such a hypothesis of dorsal funiculiis ance in such animals both in walking and climb- function, especially in relation to the lack of ing, and in directing movements in space, indicate touch deficits in human subjects with high cervical to us at least some preservation of proprioceptive and thoracic spinal cord lesions (Boshes and Pad- function. Agreement on this seems to come from berg, 1953). In a retest of one of these patients, other dorsal funiculus tractotomy studies Wall was able to demonstrate a change in the (Christiansen, 1966) in which neurological testing quality of the remaining touch in the direction of on September 24, 2021 by guest. was also used. being less frequently felt during concomitant We believe that the concentration of extero- distraction of the patient by vibratory sensation, ceptive afferent fibres within the dorsal funiculus or by having him read aloud (Wall, 1970). Electro- confers upon it quantitatively the largest role physiological results (Schneider, 1972; Dreyer et in conveying information useful in tactile explora- al., 1974) indicate that many of the single neurones tion and manipulation through the macaque spinal which are encountered in the somatosensory cord. While we cannot draw the conclusion that cortex after posterior white column transection our hair movement and tap tests of animals with show a rapid habituation, and require higher transection of the dorsal funiculus demonstrated stimulus intensities to be excited. Such units, if J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

Proprioceptive pathways of the spinal cord 429 they are characteristic of the remaining form of Nevertheless, we interpret our data to indicate sensation in the patients, could account for their that the true grasp reflex depends on the dorsal lack of sensitivity to one form of stimulation funiculus. during simultaneous application of vibratory Dorsal funiculus lesions were also sufficient stimulation. Oscillatory stimulation could be like to replace forelimb grasp reflexes with active white noise, masking sporadic transmission from avoidance responses. This occurred even when another somatic source. the lesions were below the brachial plexus. Hind- Another question is whether the size of the cage limb tactile avoidance was seen early in the in which the animal is confined after surgery postoperative course with many animals, but determines the amount of recovery of function. generally disappeared or was greatly diminished We believe that residual postural and motor by the end of the first postoperative month. deficiencies at one month after surgery may be Gilman and Denny-Brown (1966) found such affected by cage size, but only to the extent that avoiding on the hindlimb but not the forelimb in free roaming or colony caged animals have a one animal (DCI) and found coarse avoiding in greater opportunity to develop strategies to cope the forelimbs of two (DC5 and DC15). A hyper- with their deficits. Such a strategy was reported aesthesia following primary somatosensory cortex (Goldberger, 1972) when tactile evasion was en- ablations in macaques is reported (Schwartzman hanced to enable a macaque with a lesion of and Semmes, 1971), which we believe is identical Brodnunn's area 6 of the cerebral cortex to over- to the phenomenon referred to by others (Denny- come the pathological grasp reflex. These Brown and Chambers, 1958) as release of tactile strategies may make the animal appear grossly avoiding, and which occurs with more extensive to be without deficits which subtler testing would postcentral cortical excisions in macaques. Thus, reveal. the limited cortical ablations will produce results A discussion of grasp responses must heed the similar to those which dorsal funiculus lesions Protected by copyright. differences in type and in nomenclature which create in forelimbs and hindlimbs at different post- have been used by various authors. The important operative times. It would seem that a tonic distinction is that in ascending order of com- input from the dorsal funiculus to the postcentral plexity lie the coarse grasp or traction reaction, cortex is necessary to maintain normal tactile the true grasp reflex, and instinctive grasping reactivity. (Denny-Brown and Chambers, 1958; Twitchell, Instinctive grasping, encompassing a higher 1965). This last named reaction is a contactual level of complexity, is not preserved in the hind- orienting reaction whose voluntary performance limb by partial sparing of the dorsal funiculus was called spontaneous grip or spontaneous grasp (DCX: 8, 9, 15), but a greater amount of sparing by Ferraro and Barrera (1934). caused by the same lesion (DCX: 9) being just The concentration of distal afferent fibres re- rostral to mid-plexus level, fails to abolish this lated to grasping in the dorsal funiculus causes response in the forelimb. Lesions above the both devastating loss of grasp behaviour with brachial plexus do abolish this response in the total tractotomies and resilience of it with partial forelimb (DCX: 10, 11). Ferraro and Barrera sparing. Our complete dorsal funiculus lesions (1934) were only able to get crude spontaneous http://jnnp.bmj.com/ were sufficient to abolish the hindlimb true grasp grasping after one month in any limb of an reflex. Gilman and Denny-Brown (1966), who animal with a midcervical, dorsal funiculus lesion tested this reflex, found a brisk grasp reflex on (CER80) or in the hindlimb of a midthoracic, the hindlimb in only one of their experimental dorsal funiculus lesioned animal (CER 135). The animals (DCI) which had a bilateral high cervical dorsolateral funiculus lesions alone yielded no dorsal funiculus section, but not on the other four effect on this response in either limb in our studies. animals with similar lesions (DC4, DC14, DC15, Identical results were obtained by Gilman and and DC46). Only coarse grasp reactions were Denny-Brown (1966) with a similar lesion. Ferraro on September 24, 2021 by guest. elicited in these latter animals. The occurrence of and Barrera (1934) described the deficits in spon- the true grasp reflex in the lone animal may have taneous grasping common to animals with a dorsal resulted from the sparing of small bundles of funiculus lesion. In an interesting observation dorsal funiculus fibres by the lesion. Such sparing they relate that animals with such lesions try to appears to have resulted in the retention of this grip things with their entire limb rather than just reflex even though diminished by our animals its distal part. We believe this perceptive observa- DCX: 9 and DCX: 15. However, the lesion in the tion can be explained by the relative preservation one animal mentioned above does seem com- of the proximal limb sensory input which we see plete; thus, we are at a loss to explain this result. in our electrophysiological work (Schneider, 1972; J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

430 Richard J. Schneider, A. T. Kulics, and Thomas B. Ducker Dreyer et al., 1974). These results may also ex- 3, 1, and 2 (Randolph and Semmes, 1974), the plain the necessity for using broad areas of region which is the primary central terminus of cutaneous contact with the limbs in producing impulses travelling along dorsal funiculus fibres, tactile placing and traction responses as described there are differentiable, discriminative, behavioural by Gilman and Denny-Brown (1966), and the deficits which last for months. Again, the receptor findings of Vierck (personal communication) that information which permits discrimination by pal- distal forelimb deficits in proprioception are more pation seems to be disturbed, and in addition, the marked than proximal deficits in macaques per- disturbance varies with the cytoarchitectural area forming behavioural tasks after posterior whitc involved. column tractotomies. The picture that emerges is that submodalities The existence of a sorting process and its conse- of mechanoreceptive somatosensation, and perhaps quences at levels rostral to the spinal cord intro- types of receptor information within submodali- duce a new dimension into the analysis of dorsal ties, are segregated as they ascend from caudal funiculus studies. For example, Melzack and to cephalic levels of the so as to Bridges (1971) found that, with motor performance form somatosensory subunits in the primary corti- in cats, there was less drastic interference by cal receiving area (Mountcastle and Powell, 1959a, lesions solely in the nuclei of the dorsal funiculus b; Powell and Mountcastle, 1959a, b; Whitsel et al., than with those which were made only in the 1972). The sorting process that takes place in the high cervical dorsal funiculus. They regarded this dorsal funiculus contributes to this segregation as a paradoxical result. Since their lesions did not (Whitsel et al., 1969; Whitsel et al., 1970). This extend into an area of the nuclei which has been sorting, and the existence of alternative somato- shown to receive a large number of proprioceptive sensory pathways, however different in the manner afferent fibres (Kuhn, 1949; Gordon and Paine, in which they handle the stimulus information, 1960; Gordon and Seed, 1961; Perl et al., 1962; allow for the development of strategies to assist Protected by copyright. Gordon and Jukes, 1964; Biedenbach, 1972), that the recovery of function assessed by many experi- is, rostral to the obex, they may have spared an im- ments after dorsal funiculus lesions in animals portant fraction of proprioceptive fibres. It would (Devito and Ruch, 1956; Mettler and Liss, 1959; thus become extremely important to delimit the Christiansen, 1966; Levitt and Schwartzman, 1966; precise caudal extent of each of their lesions since Vierck, 1966; Kitai and Weinberg, 1968; the less caudal a lesion is, the more fibres can be Schwartzman and Bogdonoff, 1968, 1969; Dobry spared by sorting. Since dorsal funiculus nucleus and Casey, 1972; Eidelberg and Woodbury, 1972; lesions require a more rostral central nervous Mann et al., 1972; Schneider, 1972; Schwartz et system exposure, the chances are that more spar- al., 1972; Liebman and Levitt, 1973). Neurological ing would have occurred in these animals than in observations do not permit the same statement to those with dorsal funiculus lesions alone, and the be made about humans, because those studies fibre sorting process could explain their paradoxical which report absence of deficits after dorsal funi- results. culus tractotomy lack the necessary histological The effect of dorsal funiculus lesions on the confirmation (Rabiner and Browder, 1948; Boshes

behaviour of the animal derives from the limita- and Padberg, 1953; Cook and Browder, 1965). http://jnnp.bmj.com/ tion of his ability to receive mechanoreceptive Also, neurological examinations in humans con- information, especially from the distal, volar re- tinue to implicate the dorsal funiculus in postural ceptors most involved in palpatory input. Tactual sensation (Rose and Mountcastle, 1959; Poggio guidance is necessary in fine manipulative tasks and Mountcastle, 1960). Whether differences re- where vision is excluded (Brinkman and Kuypers, lated to assessment procedures may account for 1972), and its sudden loss cannot be immediately these results in humans is not yet known. How- compensated by visual input alone. Manipulations ever attractive the new hypotheses of dorsal funi- where spared sensory input may be used-for culus function may be in offering explanations of on September 24, 2021 by guest. example, vision, facial somatosensation, crude the physiological role of the posterior white hand and foot dorsum somatosensation-like those columns, we suggest that further physiological and involving feeding and grooming, tend to show behavioural evidence is required before conclusions quicker recovery, but a return to the status before can be drawn about the function of the dorsal the transection does not occur within a month. funiculus. While we believe that our re-sorting Recent behavioural data have shown that when hypothesis is correct in monkey, and may apply selective ablations are made of the anterior and to man, we admit that it goes beyond existing posterior postcentral gyrus (Semmes and Porter, evidence. We are currently recording the appro- 1972), or of Brodmann's cytoarchitectonic areas priate clinical information in patients. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

Proprioceptive pathways of the spinal corcd 431 References monkeys. Experimental Neurology, 37, 573-581. Ferraro, A.. and Barrera, S. E. (1934). Effects of Albe-Fessard, Denise (1967). Organization of somatic experimental lesions of the posterior columns in central projections. In Contributions to Sensory Macacus rhesus monkeys. Brain, 57, 307-332. Physiology. Vol. 2. Edited by W. D. Neff. Academic Gilman, S., and Denny-Brown. D. (1966). Disorders Press: New York. of movement and behaviour following dorsal Andersson, S. A., Landgren, S., and Wolsk, D. (1965). column lesions. Brain. 89, 397-418. The thalamic relay and cortical projection of group Glees, P., and Soler, J. (1951). Fibre content of the I muscle afferents from the forelimb of the cat. posterior column and synaptic connections of Journal of Physiology. 183, 576-591. nucleus gracilis. Zeitschrift fur Zellforschung. 36, Asanuma, H., and Rosen, I. (1972a). Functional role 381-400. of afferent inputs to the monkey motor cortex. Goldberger, M. E. (1972). Restitution of function in Brain Research. 40, 3-5. the CNS: the pathologic grasp in Macaca mulatta. Asanuma. H.. and Rosen. I. (1972b). Topographical Experimental Brain Research. 15, 79-96. organization of cortical efferent zones projecting Gordon. G., and Jukes, M. G. M. (1964). Dual to distal forelimb muscles in the monkey. Experi- organization of the exteroceptive components of the mental Brain Research, 14, 243-256. cat's gracile nucleus. Journal of Physiology, 173, Biedenbach, Maria A. (1972). Cell density and regional 263-290. distribution of cell types in the cuneate nucleus of Gordon, G., and Paine. C. H. (1960). Functional the Rhesus monkey. Brain Research, 45, 1-14. organization in the nucleus gracilis of the cat. Boshes, B., and Padberg. F. (1953). Studies on the Journal of Physiology, 15, 331-349. cervical spinal cord of man. Neurological Society of Gordon, G.. and Seed. W. A. (1961). An investigation Minneapolis. 3, 90-101. of nucleus gracilis of the cat by antidromic Bowsher, D. (1965). The anatomophysiological basis stimulation. Journal of Physiology, 155, 589-601. of somatosensory discrimination. International Ha, H. (1971). Cervicothalamic tract in the Rhesus Review of Neutrobiology. 8, 35-75. monkey. Experimental Neurology, 33, 205-212. Protected by copyright. Brinkman. J.. and Kuypers, H. G. J. M. (1972). Split- Ha. H.. and Morin, F. (1964). Comparative anatomical brain monkeys: cerebral control of ipsilateral and observations of the cervical nucleus, N. cervicalis contralateral arm, hand, and finger movements. lateralis. of some primates. A natomical Record, Science, 176, 536-539. 1489 374-375. Burgess, P. R., and Clark. F. J. (1969). Dorsal column Hand, P. J. (1966). Lumbosacral dorsal root termina- projection of fibres from the cat knee joint. Journal tions in the nucleus gracilis of the cat. Some ob- of Physiology. 203, 301-315. servations on terminal degeneration in other Carlson, Kristen R. (1972). A temporary restraint medullarv sensory nuclei. Journal of Comparative chair for monkeys. Physiology and Behavior, 9, Neurology, 126, 137-156. 493-494. Iggo. A. (1963). An electrophysiological analysis of Christiansen, J. (1966). Neurological observations of afferent fibres in primate skin. A cta Neurovegetativa, macaques with spinal cord lesions. Anatomical 24. 226-240. Record, 154, 330. Jasper, H., and Bertrand, G. (1964). Stereotaxic Cook. A. W.. and Browder, E. J. (1965). Functions microelectrede studies of single thalamic cells and of posterior columns in man. Archives of Neuro- fibres in patients with dyskinesia. Transactions of logy (Chicago), 12, 72-79. the A merican Neurological A ssociation. 89, 79-82. Denny-Brown. D.. and Chambers, R. A. (1958). The Kitai, S. T., and Weinberg, J. (1968). Tactile discrimi- http://jnnp.bmj.com/ parietal lobe and behavior. In The Brain and Human nation study of the dorsal column-medial B&Aavior. Edited by H. C. Solomon, S. Cobb. and lemniscal and spinocervico-thalamic tract in cats. W. Penfield. Williams and Wilkins: Baltimore. Experimental Brain Research, 6, 234-246. Devito, J. L., and Ruch. T. C. (1956). Central path- Kuhn, R. A. (1949). Topographical pattern of cutane- wa's subserving weight discrimination in monkeys. ous sensibility in the dorsal column nuclei of the Federation Proceedings, 15, 48-49. cat. Transactions of the A merican Neurological Dobrv. P. J. K.. and Casey, K. L. (1972). Roughness A ssociation, 74, 227-230. discrimination in cats with dorsal column lesions. Kuypers, H. G. J. M.. and Tuerk, J. D. (1964). The Brain Research, 44, 385-397. distribution of the cortical fibres within the nuclei on September 24, 2021 by guest. Dreyer, D. A.. Schneider, R. J., Metz. C. B., and cuneatus and gracilis in the cat. Journal of Whitsel, B. L. (1974). Differential contributions of Anatomy, 98, 143-162. spinal pathways to body representation on the Landgren, A.. Nordwall, A.. and Wengstrom, C. postcentral gyrus. Journal of Neurophysiology, 37, (1965). The location of the thalamic relay in the 119-145. spino-cervico-lemniscal path. A cta Physiologica Dubrovsky. B.. Davelaar, E.. and Garcia-Rill, R. Scandinavica, 65, 164-175. (1971). The role of dorsal columns in serial order Levitt, M., and Schwartzman. R. (1966). Spinal sen- acts. Experimental Neurology, 33, 93-102. sory tracts and two-point tactile sensitivity. A na- Eidelberg, E.. and Woodbury, C. M. (1972). Apparent tomical Record, 154, 476. redUndancy in the somatosensory system in Lewis, M. McD., Porter, P.. and Horne, M. (1971). J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

432 Richard J. Schneider, A. T. Kulics, and Thomas B. Ducker The effects of impairment of afferent feedback The cytoarchitecture of the postcentral gyrus of from the moving limb on the natural activities of the monkey Macaca mulatta. Bulletin of the Johns neurones in the precentral gyrus of conscious Hopkins Hospital, 105, 108-132. monkeys: A preliminary investigation. Brain Powell, T. P. S., and Mountcastle, V. B. (1959b). Research, 32, 467-473. Some aspects of the functional organization of the Liebman, R. S., Levitt, M. (1973). Position sense cortex of the postcentral gyrus of the monkey: a after combined spinal tractotomies and cere- correlation of findings obtained in a single unit bellectomies in macaques. Experimental Neurology, analysis with cytoarchitecture. Bulletin of the 40, 170-182. Johns Hopkins Hospital, 105, 133-162. Lundberg, A., and Norrsell, U. (1960). Spinal afferent Rabiner, A. M., and Browder, E. J. (1948). Concern- pathways of the tactile placing reaction. ing conduction of touch and deep sensibilities Experientia, 16, 123. through the spinal cord. Transactions of the Mallart, A. (1964). Projection des afferences muscu- American Neurological Association, 73, 137. laires de la patte anterieure au niveau du thalamus Randolph, M., and Semmes, J. (1974). Behavioral chez le chat. Comptes Rendus Hebdomadaires des consequences of selective subtotal ablations in the Se'ances (Paris), 259, 1215-1218. postcentral gyrus of Macaca mulatta. Brain Mann, M. D., Kasprzak, H., Hiltz, F. L., and Tapper, Research, 70, 55-70. D. N. (1972). Activity in single cutaneous afferents: Rose, J., and Mountcastle, V. B. (1959). Touch and spinal pathways and cortical evoked potentials. kinesthesis. In Handbook of Physiology, Section 1, Brain Research, 39, 61-70. Neurophysiology, Vol. 1. Edited by J. Field, H. W. Melzack, R., and Bridges, J. A. (1971). Dorsal column Magoun, and V. E. Hall. American Physiological contributions to motor behavior. Experimental Society: Washington, DC. Neurology, 33, 53-68. Schneider, R. J. (1972). The effects of lesions of the Mettler, R. A., and Liss, H. (1959). Functional re- posterior funiculi of Macaca mulatta. PhD Disserta- covery in primates after large subtotal spinal cord tion, University of Pittsburgh.

lesions. Journal of Neuropathology and Experi- Schwartz, A. S., Eidelberg, E., Marchok, P., and Protected by copyright. mental Neurology, 18, 509-516. Azulay, A. (1972). Tactile discrimination in the Mountcastle, V. B., and Powell, T. P. S. (1 959a). monkey after section of the dorsal funiculus and Central nervous mechanisms subserving position lateral lemniscus. Experimental Neurology, 20, 43- sense and kinesthesis. Bulletin of the Johns Hopkins 51. Hospital, 105, 163-200. Schwartzman, R. J., and Bogdonoff, M. D. (1968). Mountcastle, V. B., and Powell, T. P. S. (1959b). Behavioral and anatomical analysis of vibration Neural mechanisms subserving cutaneous sensi- sensibility. Experimental Neurology, 20, 43-51. bility with special reference to the role of afferent Schwartzman, R. J., and Bogdonoff, M. D. (1969). inhibition in sensory perception and discrimination. Proprioception and vibration sensibility discrimina- Bulletin of the Johns Hopkins Hospital, 105, 201- tion in the absence of the posterior columns. 232. Archives of Neurology (Chicago), 20, 349-353. Mountcastle, V. B., and Smith, I. D. (1968). Central Schwartzman, R. J., and Semmes, J. (1971). The nervous mechanisms in sensation. In Medical sensory cortex and tactile sensitivity. Experimental Physiology. Vol.2. Edited by V. Mountcastle. C. V. Neurology, 33, 147-158. Mosby: St. Louis. Semmes, J. (1969). Protopathic and epicritic sensa- Norrsell, U. (1966). The spinal afferent pathways of tion: a re-appraisal. In Contributions to Clinical

conditional reflexes to cutaneous stimuli in the Neuropsychology. Edited by A. L. Benton. Aldine http://jnnp.bmj.com/ dog. Experimental Brain Research, 2, 269-282. Publishing: Chicago. Norrsell, U., and Voorhoeve, P. (1962). Tactile path- Semmes, J., and Porter, L. (1972). A comparison of ways from the hindlimb to the cerebral cortex in pre-central and postcentral cortical lesions on cat. Acta Physiologica Scandinavica, 54, 9-17. somatosensory discrimination in the monkey. Cor- Perl, E. R., Whitlock, D. G., and Gentry, J. R. tex, 8, 249-264. (1962). Cutaneous projection to second order Twitchell, T. E. (1965). The automatic grasping re- neurones of the dorsal column system. Journal of sponses of infants. Neuropsychologia, 3, 247-259. Neurophysiology, 25, 337-358. Vierck, C. J. (1966). Spinal pathways mediating limb

Petit, D., and Burgess, P. R. (1968). Dorsal column position sense. Anatomical Record, 154, 437. on September 24, 2021 by guest. projection of receptors in cat hairy skin supplied Vierck, C. J. (1973). Alterations of spatio-tactile dis- by myelinated fibers. Journal of Neurophysiology, crimination after lesions of primate spinal cord. 31, 849-855. Brain Research, 58, 69-79. Poggio, G. F., and Mountcastle, V. B. (1960). A Wall. P. D. (1970). The sensory and motor role of study of the functional contributions of the impulses travelling in the dorsal columns towards lemniscal and spinothalamic systems to somatic cerebral cortex. Brain, 93, 505-524. sensibility. Central nervous mechanisms in pain. Walshe, Sir Francis. (1970). Diseases of the Nervous Bulletin of the Johns Hopkins Hospital, 106, 266- System. Williams and Wilkins: Baltimore. 316. Whitsel, B. L., Petrucelli, L. M., and Sapiro, G. Powell, T. P. S., and Mountcastle, V. B. (1959a). (1969). Modality representation in the lumbar and J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.40.5.417 on 1 May 1977. Downloaded from

Proprioceptive pathways of the spinal cord 433 cervical fasciculus gracilis of squirrel monkeys. D. A. (1972). The resorting of spinal afferents as Brain Research, 15, 67-78. antecedent to the body representation in the post- Whitsel, B. L., Petrucelli, L. M., Sapiro, G., and Ha, central gyrus. Brain Behavior and Evolution, 5, H. (1970). Fiber sorting in the fasciculus gracilis of 303-341. squirrel monkeys. Experimental Neurology, 29, 227- Woolsey, C. N., and Bard, P. (1963). Cortical control 242. of hopping and placing reactions in Macaca mulatta. Whitsel, B. L., Petrucelli, L. M., Ha. H., and Dreyer, American Journal of Physiology, 116, 165. Protected by copyright. http://jnnp.bmj.com/ on September 24, 2021 by guest.