Neurosciem'c Letters, 145 (1992) 129 132
129
c'~ 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00
NSI. 08992
Walking of cats on a grid: performance of locomotor task in spinal intact and hemisected cats Masamichi Kato Department ~Jl Physiology, Hokkaido University School ol Medicine, S~q~poro ( J~q~a##s (Received 15 May 1992; Revised version received 6 July 1992; Accepted 6 July 1992) Ke~ words.'
Walking on grid: Pattern generator; lnterlimb coordination: Hemisected cat
We have investigated interlimb coordination during walking by observing locomotion of cats on a grid. Control cats made no error in forelimb placement on the grid, and landed their hindlimbs on the rung where each ipsilaleral t\~relimb left moments earlier. This suggests that locomotor c o m m a n d signals were modified by grid information including its size and direction and affected the pattern generators of bnlh forelimbs and hindlimbs. Hemisected cats, made both at high cervical and lower thoracic, showed deteriorated placement of the hindlimbs suggesting that interlimb coordination is carried out mainly by descending pathways from the brainstem such as ventrolateral fasciculus and dorsolateral fasciculus. Interlimb reflex pathways may play a limited role.
When a cat walks straight, looking ahead, the cat does not c o n f r m visually the footstep of its own hindlimbs. Nevertheless the footprints of their hindlimbs coincide with those of the ipsilateral forelimbs in many steps (Fig. 1 in ref. 5). This phenomenon suggests that the pattern generator t\)r the forelimbs sends synchronizing signals of spatial placement, if not temporal, to the pattern generator of the hindlimbs. In the present experiments we encouraged the cats to walk on a grid. By doing this we have tried to (1) observe how the cats place their torelimbs successively on the grid, (2) observe quantitatively how accuralely the hindlimbs step over the forelimb footsteps in normal spinal intact cats, and (3) observe how the placement of the hindlimbs is affected by chronic spinal cord hemisection. Nine cats were used for these experiments: 2 nornlal spinal intact cats (control) and 5 chronically hemisected cats at The-, or g~ and 2 chronically hemisected cats at C~ and C~. The cats were encouraged to walk on a grid (80 cm wide × 190 cm long, made of 50 m m x 50 m m steel mesh of 3 mm ~#) with food reward and affection, usually early in the morning before they were fed. No systematic or special training was given to any cats. The walking was taken on 8 mm movie film (18, 24 or 36 frames/s depending on the walking speed of the cat) and also was inspected by 3 observers from different angles. They Corre,g~o#lde#u'e: M. Kato, Department of Physiology, Hokkaido University School of Medicine, Sapporo, 060 Japan.
counted both successful and failed landings on the rung of each limb. The experiment was repeatedly carried out, usually at intervals of 3 10 days on each cat. Stick diagrams of fore- and hindlimb movement oil the grid were drawn by projecting the movie f l m on the screen. After completion of the experiments the operated cats were sacrificed by an overdose of Nembutal (100 mg/kg i.v.) following ether anesthesia. Histological investigation was carried out after the operated site was carefullx, inspected. The cats walked on the grid at velocities ot"40 70 cm/s on average. One control cat walked at a speed of 120 cm/s in one session. The cat walked on the grid quickly and successively oil one day, though in the next observation the same cat did not always walk smoothly and stopped walking at the middle of the grid. This tendency was rather c o m m o n among both control and operated chronic cats. Therefore the data obtained when cats walked successively and continuously fiister than 40 cm/s were analyzed in this report. Pictures of representative steppings of a control cat are shown in Fig. 1A. The control cat walked on the grid, looking straight at the t\~od reward, accurately stepping on the rung of the grid. As shown in these pictures and diagrams in Fig. 2 (control cat) the cat did not look at its legs during walking. Fig. I B illustrates how a hemisected cat walked on the grid. As seen in these photographs the spinal hemisected cat often failed to step on tile rung and dropped its hindlimb(s) through the mesh. The rate of success to land
131)
A
B
Fig. 1. Walking on a grid. A: normal spinal intact cat. B: spinal hemisected cat (right Th~). Control cat (A) can walk on the grid without failing to step on the rungs of the grid (from tipper to lower). Cat 55 shown in B was hemisected at the right Th~ on December 28, 1985. These pictures were taken on January 13, 1986, 16 days after the hemisection, t:or this session tile cat walked 36 steps on the grid. The right hindlimb failed to land on a rung, as is shown in the lower picture of B. in 12 out of the 36 steps (33.3%) while the other 3 limbs stepped on lhe mesh.
on rungs in the case of the tbrelimbs, and the rate of stepping on the rung where the ipsilateral forelimb stepped a m o m e n t earlier in the case of the hindlimbs, were calculated from 30-50 steps at each session. The control cats made no error in forelimb placement. Also the control cats made almost no error in hindlimb placement on the rung where the ipsilateral forelimb had been placed (mean error < 3% in 103 sessions). This value corresponds well to Bregman and Goldberger's [1] observation, although the size of their grid was larger than ours. Fig. 2A illustrates stick diagrams of a control cat. The cat walks straight, looking ahead, and the hindlimb exactly steps on the same rung where the ipsilateral forelimb stepped a m o m e n t earlier. This diagram also shows that the forward movement of the limbs is smooth and placement is accurate. Fig. 2B illustrates how the hindlimb of a spinal hemisected cat failed to reach the rung where the ipsilateral forelimb had just left. This diagram also shows that the forward movement of the hindlimb is clumsy. The postural control, including lateral stability of the body and distribution of floor reaction forces of
the limbs, and locomotion of the hemisected cats recovered within 1 week, and at this time they walked and ran over the ground like the control cats, confirming our earlier observation [6]. Phase relation between vastus lateralis muscle and ipsilateral tibialis anterior muscle was either almost normal or the latter muscle discharged earlier than the control values, indicating that dorsiflexion of the foot joint was already finished before the hindlimb landed on the rung. The hemisected cats failed, in some steps, to land the hindlimbs on the rung where the ipsilateral forelimbs stepped a moment earlier either by exceeding the "target' or by stepping down too early, as shown in Fig. 2, right (hemisected cat 55). There was no firm tendency towards a dominant result, but the latter case was more frequently observed. Fig. 3 illustrates representative scores of the 4 limbs of the operated cats during walking on the grid. Cat 47 accurately placed its forelimbs on the grid at any experimental sessions following hemisection at the left Thl3. For the hindlimbs, the left side showed great deterioration down to about 10% success rate for about 10 days,
131
6
A
Hemisected Cat No.55
Cat
R H
LH
RF
f-
2 1 3___j~
456
6
_
// grid
gr
10 Fig. 2. Stick diagrams of walking on a grid. The diagrams were constructed from movie pictures which were taken at 24 framcs's. L means left, R means right, F means forelimb and H means hindlimb. The steppings were numbered from the beginning of the lk)relimb in A and o1"the hindlimb in B and the following numbers indicate frame numbers. Numbers 1 and 8 in the heads in A indicate the head positions respectivel~ when the left forelimb left (1) and landed on the grid (8). In A the control cat walked from right to left, and in B the hemisected cat walked from left to righl. The hemisected cat is the same cat as is shown in Fig. 1B. but on a different day.
then rapidly recovered to about 70% during the following 10 day period. Then this percentage remained virtually unchanged until the 55th day of observation. The right hindlimb, which is contralateral to the hemisection, showed a decreased success rate of 70-85% for about 10 days, then recovered to the control values. For cat 45, which received a hemisection at the left C~, it took longer to recover from the operation and grid walking was inNo. 47
%
~.-
loo
v e s t i g a t e d f r o m t h e 3 5 t h p o s t o p e r a t i v e d a y o n w a r d s , as i l l u s t r a t e d in Fig. 3. B o t h f o r e - a n d h i n d l i m b s o f t h e o p e r a t e d side s h o w e d d e t e r i o r a t e d p l a c e m e n t f o r m o r e t h a n 1 month thereafter. The present experiments showed that once a normal c a t b e g a n w a l k i n g it w a l k e d s u c c e s s i v e l y a n d c o n t i n u ously at almost constant speed from one end to the other e n d o f t h e grid. D u r i n g t h e l o c o m o t i o n
No. 45
%
~ ~ ~
~
~
~
~ ~ ~,
~
o
o
o
oo
the cats walk
loo-
°
-
~
~ ~
8
o c~ t
o
o
t
*
t*
t
*
•
•
•
*
t a
t
*
•
t
I
50 •
*
50
. • *
IT13
15
~1'I T13
3 '13
4'5
days
LF RH
° o
LH
*
IC1
6 'O
3'5
4'0
5'0
6'0
day s
7'0
~tlC 1
Fig. 3. Success rate of stepping on rungs of 4 limbs following hemisection. Cat 47 was hemisected at the left Th~ on September 13, 1984. Cat 45 was hemisected at the left C~ on July 17, 1984. The abscissa indicates the number of days after the hemiscction and the ordinate indicatcs the success rate of stepping on the grid, ,~, right forelimb: e, left forelimb: ,=, right hindlimb: and *, left hindlimb. The scores were calculated l'rom 25 3(1 steppings at each point.
132 looking ahead at the food reward, and do not search and confirm even their forelimb positioning step by step. This suggests that grid information, including its size, location and direction entered the brain through visual pathways, affecting the pattern generator o f the forelimbs. Furtherm o r e the control cat can place the hindlimbs on the rungs where the ipsilateral forelimbs stepped a m o m e n t earlier during successive locomotion. This indicates that comm a n d signals, p r o g r a m m e d somewhere in the brain to perform limb movements, are supplemented by visual information o f the grid and that the pattern generators o f the forelimbs and the hindlimbs are controlled by the supplemented c o m m a n d signals to place the limbs precisely on the rung. The pattern generator for the forelimbs is located in the brainstem [11] and those o f the hindlimbs are located in the lumbar spinal cord [2, 3]. J o r d a n and his colleagues [4, 7] discussed the organization o f l o c o m o t o r pathways descending from the mesencephalic l o c o m o t o r region ( M L R ) [10] to the spinal cord via the ventrolateral fasciculus (VLF) and dorsolateral l;asciculus (DLF). However, not much attention has been paid to the anatomical and physiological relations between the forelimb pattern generator and the hindlimb pattern generator. The V L F and D L F seem to be g o o d candidates linking the two generators. In thoracic hemisected preparations, like cat 47 in Fig. 3A, the deteriorated accuracy o f the ipsilateral hindlimb could be explained by the severed V L F and/or D L F or by the interlimb reflex pathways being sectioned. The severance m a y be responsible for the deterioration o f accuracy o f the ipsilateral hindlimb for about 10 days. However, the performance then improved quickly to the level o f about 70%. This indicates that c o m m a n d signals descending contralateral V L F and/or D L F ' c o m p e n s a t e ' to some extent the pattern generator o f the hindlimb o f the hemisected side. As to the slight decrease o f performance score o f the contralateral hindlimb for a b o u t 1 week postoperatively, the a u t h o r has no plausible explanation, t h o u g h physical unbalance during walking might be one factor. In cervical hemisected cats like 45 ipsilateral V L F and D L F for cervical enlargement were sectioned chron-
ically. However, interlimb reflex pathways remained intact. In these preparations, performance scores o f both ipsilateral limbs decreased to about the same degree as those o f thoracic hemisected cats (Fig. 3B). These results suggest that descending signals from supraspinal centers are mainly responsible for interlimb coordination and the interlimb reflex pathways play a limited role. The a u t h o r would like to thank Mr. S. Doki and Miss Y. K o b a y a s h i for their excellent technical assistance and Ms. K. A m a n o tbr her editorial assistance. I Bregman, B.S. and Goldberger, M,E., Infant lesion effect: It Sparing and recovery of function after spinal cord damage in newborn and adult cats, Dev. Brain Res., 9 (1983) 119 135. 2 Deliagina, T.G., Orlovsky, G.N. and Pavlova, G.A., The capacity for generation o1' rhythmic oscillations is distributed in the lumbosacral spinal cord of the cat, Exp. Brain Res., 53 (1983) 81-90. 3 Grillner, S. and Zangger, P., On the central generation of locomotion in the low spinal cat, Exp. Brain Res., 34 (1979) 241 261. 4 Jordan, L.M., Brainstem and spinal cord mechanisms fbr the initiation of locomotion. In: M. Shimamura, S. Grillner and V.R. Edgerton (Eds.), Neurobiological Basis of Human Locomotion, Japan Scientific Society, Tokyo, 1991, pp. 3 20. 5 Kato, M.. Motoneuronal activity of cat lumbar spinal cord lbllowing separation from descending or contralateral impulses, Central Nerv. Syst. Trauma, 4 (1987) 239-248. 6 Kato, M., Murakami, S., Hirayama, H. and Hikino, K., Recovery of postural control following chronic bilateral hemisection at different spinal cord levels in adult cats, Exp. Neurol., 90 (1985) 350 364. 7 Noga, B.R., Kettler, J.J. and Jordan, L.M., Locomotion induced in mesencephalic cats by injections of putative transmitter substances and antagonists into the medial reticular formation and the pontomedullary locomotive strip, J. Neurosci.. 8 (1988) 2074.-2086. 8 Selionov, V.A. and Shik, M.L,, Medullary locomotor strip and column in the cat, Neuroscience, 13 (1984) 1267 1278. 9 Shefchyk, S., McCrea, D., Kriellaars, D., Forticr, E and Jordan, L., Activity o1 interneurons within the L4 spinal segment of the cat during brainstem-evoked ficitive locomotion, Exp. Brain Res., 80 (1990) 290 295. 10 Shik, M.L., Severin, F.V. and Orlovsky, G.N.. Control of walking and running by means of stimulation of the mid-brain, Biophysics, 11 (1966) 756 765. I 1 Shimamura, M.. Kogurc, 1. and Fuwa, T., The role of the paralemniscal pontine reticular tormation in forelimb stepping of thalarnic cats, Neurosci. Res., 1 (1984) 393 410.
129
c'~ 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00
NSI. 08992
Walking of cats on a grid: performance of locomotor task in spinal intact and hemisected cats Masamichi Kato Department ~Jl Physiology, Hokkaido University School ol Medicine, S~q~poro ( J~q~a##s (Received 15 May 1992; Revised version received 6 July 1992; Accepted 6 July 1992) Ke~ words.'
Walking on grid: Pattern generator; lnterlimb coordination: Hemisected cat
We have investigated interlimb coordination during walking by observing locomotion of cats on a grid. Control cats made no error in forelimb placement on the grid, and landed their hindlimbs on the rung where each ipsilaleral t\~relimb left moments earlier. This suggests that locomotor c o m m a n d signals were modified by grid information including its size and direction and affected the pattern generators of bnlh forelimbs and hindlimbs. Hemisected cats, made both at high cervical and lower thoracic, showed deteriorated placement of the hindlimbs suggesting that interlimb coordination is carried out mainly by descending pathways from the brainstem such as ventrolateral fasciculus and dorsolateral fasciculus. Interlimb reflex pathways may play a limited role.
When a cat walks straight, looking ahead, the cat does not c o n f r m visually the footstep of its own hindlimbs. Nevertheless the footprints of their hindlimbs coincide with those of the ipsilateral forelimbs in many steps (Fig. 1 in ref. 5). This phenomenon suggests that the pattern generator t\)r the forelimbs sends synchronizing signals of spatial placement, if not temporal, to the pattern generator of the hindlimbs. In the present experiments we encouraged the cats to walk on a grid. By doing this we have tried to (1) observe how the cats place their torelimbs successively on the grid, (2) observe quantitatively how accuralely the hindlimbs step over the forelimb footsteps in normal spinal intact cats, and (3) observe how the placement of the hindlimbs is affected by chronic spinal cord hemisection. Nine cats were used for these experiments: 2 nornlal spinal intact cats (control) and 5 chronically hemisected cats at The-, or g~ and 2 chronically hemisected cats at C~ and C~. The cats were encouraged to walk on a grid (80 cm wide × 190 cm long, made of 50 m m x 50 m m steel mesh of 3 mm ~#) with food reward and affection, usually early in the morning before they were fed. No systematic or special training was given to any cats. The walking was taken on 8 mm movie film (18, 24 or 36 frames/s depending on the walking speed of the cat) and also was inspected by 3 observers from different angles. They Corre,g~o#lde#u'e: M. Kato, Department of Physiology, Hokkaido University School of Medicine, Sapporo, 060 Japan.
counted both successful and failed landings on the rung of each limb. The experiment was repeatedly carried out, usually at intervals of 3 10 days on each cat. Stick diagrams of fore- and hindlimb movement oil the grid were drawn by projecting the movie f l m on the screen. After completion of the experiments the operated cats were sacrificed by an overdose of Nembutal (100 mg/kg i.v.) following ether anesthesia. Histological investigation was carried out after the operated site was carefullx, inspected. The cats walked on the grid at velocities ot"40 70 cm/s on average. One control cat walked at a speed of 120 cm/s in one session. The cat walked on the grid quickly and successively oil one day, though in the next observation the same cat did not always walk smoothly and stopped walking at the middle of the grid. This tendency was rather c o m m o n among both control and operated chronic cats. Therefore the data obtained when cats walked successively and continuously fiister than 40 cm/s were analyzed in this report. Pictures of representative steppings of a control cat are shown in Fig. 1A. The control cat walked on the grid, looking straight at the t\~od reward, accurately stepping on the rung of the grid. As shown in these pictures and diagrams in Fig. 2 (control cat) the cat did not look at its legs during walking. Fig. I B illustrates how a hemisected cat walked on the grid. As seen in these photographs the spinal hemisected cat often failed to step on tile rung and dropped its hindlimb(s) through the mesh. The rate of success to land
131)
A
B
Fig. 1. Walking on a grid. A: normal spinal intact cat. B: spinal hemisected cat (right Th~). Control cat (A) can walk on the grid without failing to step on the rungs of the grid (from tipper to lower). Cat 55 shown in B was hemisected at the right Th~ on December 28, 1985. These pictures were taken on January 13, 1986, 16 days after the hemisection, t:or this session tile cat walked 36 steps on the grid. The right hindlimb failed to land on a rung, as is shown in the lower picture of B. in 12 out of the 36 steps (33.3%) while the other 3 limbs stepped on lhe mesh.
on rungs in the case of the tbrelimbs, and the rate of stepping on the rung where the ipsilateral forelimb stepped a m o m e n t earlier in the case of the hindlimbs, were calculated from 30-50 steps at each session. The control cats made no error in forelimb placement. Also the control cats made almost no error in hindlimb placement on the rung where the ipsilateral forelimb had been placed (mean error < 3% in 103 sessions). This value corresponds well to Bregman and Goldberger's [1] observation, although the size of their grid was larger than ours. Fig. 2A illustrates stick diagrams of a control cat. The cat walks straight, looking ahead, and the hindlimb exactly steps on the same rung where the ipsilateral forelimb stepped a m o m e n t earlier. This diagram also shows that the forward movement of the limbs is smooth and placement is accurate. Fig. 2B illustrates how the hindlimb of a spinal hemisected cat failed to reach the rung where the ipsilateral forelimb had just left. This diagram also shows that the forward movement of the hindlimb is clumsy. The postural control, including lateral stability of the body and distribution of floor reaction forces of
the limbs, and locomotion of the hemisected cats recovered within 1 week, and at this time they walked and ran over the ground like the control cats, confirming our earlier observation [6]. Phase relation between vastus lateralis muscle and ipsilateral tibialis anterior muscle was either almost normal or the latter muscle discharged earlier than the control values, indicating that dorsiflexion of the foot joint was already finished before the hindlimb landed on the rung. The hemisected cats failed, in some steps, to land the hindlimbs on the rung where the ipsilateral forelimbs stepped a moment earlier either by exceeding the "target' or by stepping down too early, as shown in Fig. 2, right (hemisected cat 55). There was no firm tendency towards a dominant result, but the latter case was more frequently observed. Fig. 3 illustrates representative scores of the 4 limbs of the operated cats during walking on the grid. Cat 47 accurately placed its forelimbs on the grid at any experimental sessions following hemisection at the left Thl3. For the hindlimbs, the left side showed great deterioration down to about 10% success rate for about 10 days,
131
6
A
Hemisected Cat No.55
Cat
R H
LH
RF
f-
2 1 3___j~
456
6
_
// grid
gr
10 Fig. 2. Stick diagrams of walking on a grid. The diagrams were constructed from movie pictures which were taken at 24 framcs's. L means left, R means right, F means forelimb and H means hindlimb. The steppings were numbered from the beginning of the lk)relimb in A and o1"the hindlimb in B and the following numbers indicate frame numbers. Numbers 1 and 8 in the heads in A indicate the head positions respectivel~ when the left forelimb left (1) and landed on the grid (8). In A the control cat walked from right to left, and in B the hemisected cat walked from left to righl. The hemisected cat is the same cat as is shown in Fig. 1B. but on a different day.
then rapidly recovered to about 70% during the following 10 day period. Then this percentage remained virtually unchanged until the 55th day of observation. The right hindlimb, which is contralateral to the hemisection, showed a decreased success rate of 70-85% for about 10 days, then recovered to the control values. For cat 45, which received a hemisection at the left C~, it took longer to recover from the operation and grid walking was inNo. 47
%
~.-
loo
v e s t i g a t e d f r o m t h e 3 5 t h p o s t o p e r a t i v e d a y o n w a r d s , as i l l u s t r a t e d in Fig. 3. B o t h f o r e - a n d h i n d l i m b s o f t h e o p e r a t e d side s h o w e d d e t e r i o r a t e d p l a c e m e n t f o r m o r e t h a n 1 month thereafter. The present experiments showed that once a normal c a t b e g a n w a l k i n g it w a l k e d s u c c e s s i v e l y a n d c o n t i n u ously at almost constant speed from one end to the other e n d o f t h e grid. D u r i n g t h e l o c o m o t i o n
No. 45
%
~ ~ ~
~
~
~
~ ~ ~,
~
o
o
o
oo
the cats walk
loo-
°
-
~
~ ~
8
o c~ t
o
o
t
*
t*
t
*
•
•
•
*
t a
t
*
•
t
I
50 •
*
50
. • *
IT13
15
~1'I T13
3 '13
4'5
days
LF RH
° o
LH
*
IC1
6 'O
3'5
4'0
5'0
6'0
day s
7'0
~tlC 1
Fig. 3. Success rate of stepping on rungs of 4 limbs following hemisection. Cat 47 was hemisected at the left Th~ on September 13, 1984. Cat 45 was hemisected at the left C~ on July 17, 1984. The abscissa indicates the number of days after the hemiscction and the ordinate indicatcs the success rate of stepping on the grid, ,~, right forelimb: e, left forelimb: ,=, right hindlimb: and *, left hindlimb. The scores were calculated l'rom 25 3(1 steppings at each point.
132 looking ahead at the food reward, and do not search and confirm even their forelimb positioning step by step. This suggests that grid information, including its size, location and direction entered the brain through visual pathways, affecting the pattern generator o f the forelimbs. Furtherm o r e the control cat can place the hindlimbs on the rungs where the ipsilateral forelimbs stepped a m o m e n t earlier during successive locomotion. This indicates that comm a n d signals, p r o g r a m m e d somewhere in the brain to perform limb movements, are supplemented by visual information o f the grid and that the pattern generators o f the forelimbs and the hindlimbs are controlled by the supplemented c o m m a n d signals to place the limbs precisely on the rung. The pattern generator for the forelimbs is located in the brainstem [11] and those o f the hindlimbs are located in the lumbar spinal cord [2, 3]. J o r d a n and his colleagues [4, 7] discussed the organization o f l o c o m o t o r pathways descending from the mesencephalic l o c o m o t o r region ( M L R ) [10] to the spinal cord via the ventrolateral fasciculus (VLF) and dorsolateral l;asciculus (DLF). However, not much attention has been paid to the anatomical and physiological relations between the forelimb pattern generator and the hindlimb pattern generator. The V L F and D L F seem to be g o o d candidates linking the two generators. In thoracic hemisected preparations, like cat 47 in Fig. 3A, the deteriorated accuracy o f the ipsilateral hindlimb could be explained by the severed V L F and/or D L F or by the interlimb reflex pathways being sectioned. The severance m a y be responsible for the deterioration o f accuracy o f the ipsilateral hindlimb for about 10 days. However, the performance then improved quickly to the level o f about 70%. This indicates that c o m m a n d signals descending contralateral V L F and/or D L F ' c o m p e n s a t e ' to some extent the pattern generator o f the hindlimb o f the hemisected side. As to the slight decrease o f performance score o f the contralateral hindlimb for a b o u t 1 week postoperatively, the a u t h o r has no plausible explanation, t h o u g h physical unbalance during walking might be one factor. In cervical hemisected cats like 45 ipsilateral V L F and D L F for cervical enlargement were sectioned chron-
ically. However, interlimb reflex pathways remained intact. In these preparations, performance scores o f both ipsilateral limbs decreased to about the same degree as those o f thoracic hemisected cats (Fig. 3B). These results suggest that descending signals from supraspinal centers are mainly responsible for interlimb coordination and the interlimb reflex pathways play a limited role. The a u t h o r would like to thank Mr. S. Doki and Miss Y. K o b a y a s h i for their excellent technical assistance and Ms. K. A m a n o tbr her editorial assistance. I Bregman, B.S. and Goldberger, M,E., Infant lesion effect: It Sparing and recovery of function after spinal cord damage in newborn and adult cats, Dev. Brain Res., 9 (1983) 119 135. 2 Deliagina, T.G., Orlovsky, G.N. and Pavlova, G.A., The capacity for generation o1' rhythmic oscillations is distributed in the lumbosacral spinal cord of the cat, Exp. Brain Res., 53 (1983) 81-90. 3 Grillner, S. and Zangger, P., On the central generation of locomotion in the low spinal cat, Exp. Brain Res., 34 (1979) 241 261. 4 Jordan, L.M., Brainstem and spinal cord mechanisms fbr the initiation of locomotion. In: M. Shimamura, S. Grillner and V.R. Edgerton (Eds.), Neurobiological Basis of Human Locomotion, Japan Scientific Society, Tokyo, 1991, pp. 3 20. 5 Kato, M.. Motoneuronal activity of cat lumbar spinal cord lbllowing separation from descending or contralateral impulses, Central Nerv. Syst. Trauma, 4 (1987) 239-248. 6 Kato, M., Murakami, S., Hirayama, H. and Hikino, K., Recovery of postural control following chronic bilateral hemisection at different spinal cord levels in adult cats, Exp. Neurol., 90 (1985) 350 364. 7 Noga, B.R., Kettler, J.J. and Jordan, L.M., Locomotion induced in mesencephalic cats by injections of putative transmitter substances and antagonists into the medial reticular formation and the pontomedullary locomotive strip, J. Neurosci.. 8 (1988) 2074.-2086. 8 Selionov, V.A. and Shik, M.L,, Medullary locomotor strip and column in the cat, Neuroscience, 13 (1984) 1267 1278. 9 Shefchyk, S., McCrea, D., Kriellaars, D., Forticr, E and Jordan, L., Activity o1 interneurons within the L4 spinal segment of the cat during brainstem-evoked ficitive locomotion, Exp. Brain Res., 80 (1990) 290 295. 10 Shik, M.L., Severin, F.V. and Orlovsky, G.N.. Control of walking and running by means of stimulation of the mid-brain, Biophysics, 11 (1966) 756 765. I 1 Shimamura, M.. Kogurc, 1. and Fuwa, T., The role of the paralemniscal pontine reticular tormation in forelimb stepping of thalarnic cats, Neurosci. Res., 1 (1984) 393 410.
