- Email: [email protected]
The pretectum as a site for relaying dorsal column input to thalamic VL neurons
Brain Research, 476 (1989) 135-139 Elsevier
135
BRE 23252
The pretectum as a site for relaying dorsal column input to thalamic VL neurons Robert Mackel and Terumi Noda The Rockefeller University, New York, NY10021 (U.S.A.)
(Accepted 30 August 1988) Key words: Pretectum; Dorsal column; Ventrolateral thalamus; lntracellular recording
Imracellular recording techniques were used to study dorsal column input to 122 feline ventral thalamus (VL) relay neurons, before (61 cells) and after (61 cells) lesioning the pretectum. Prior to the lesions, 75% (46161) of the neurons responded with short and longer latency postsynaptic potentials to dorsal column stimulation. Latencies of the postsynaptic potentials ranged firom4 (short) to 20 ms (long). After the lesions, only long latency responses were encountered, and those responses were seen in only 16% (10161) of the cells. These data indicate that the pretectum may play an important role in mediating dorsal column information to VL, ultimately influencing cerebellar commands to the motor cortex.
It is generally believed that neurons in the ventrolateral nucleus (VL) of the thalamus lack significant input from the spinal cord and simply mediate cerebellar information onto motor cortical areas 2"9a5. However, some anatomical studies have demonstrated a distinct spinothalamic input to VL 5"7'8"~6. Although there is no anatomical evidence for lemniscal input to the bulk of VL, there is controversy about a lemniscal projection to a small border area between VL and VPL 5'16. Nevertheless, in view of the anatomical work, more spinal information would be expected to reach VL than has been reported in the earlier physiological studies. Indeed, in recent experiments using intracellular recording techniques, it was shown ~2that many VL neurons responded with shortand long-latency postsynaptic potentials to stimulation of spinothalamic and/or dorsal column afferents. The frequent input from the dorsal columns was a surprising finding. Since all dorsal column input to VL is indirect and hence mediated polysynaptically, it follows that it must be relayed via a structure (or structures) intercalated between the dorsal column nuclei and VL. A candidate structure could be the pretectum, since a number of anatomical studies
have recently shown that the pretectum receives lemniscal input 4"Is and projects to VL re°a3. The present report shows that the pretectum may indeed mediate dorsal column input, since pretectal lesions were found to abolish lemniscal information to VL neurons. In the present work, intracellular recording techniques were used to study dorsal column input to VL neurons, before and after lesioning the pret 'urn. The data were obtained from five cats, ,i, sere anesthetized with alpha-chloralose (50-6" mg/k~;), paralyzed, and artificially ventilate, Supplementary doses of the anesthetic (10-20 mg/kg) were given intravenously, if necessary. The blood pressure was monitored and maintained above 80 m m H g by infusion of a solution of 5% dextrose in saline 0.9% or a solution of metaraminol bitartrate 40 ~d/ml (Aramine); rectal temperature was maintained between 37 and 39 °C. The general preparation for stimulating the spinal cord and for recording from VL neurons has been described elsewhere m2. Briefly, a craniotomy and a cervical laminectomy were performed. The dorsal columns were separated from the spinal cord from seg-
Correspondence: R. Mackel, The Rockefeller University, 1230 York Avenue, New York, NY 10021, U.S.A.
0006-8993/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
136 ments C s to C3 and mounted, for stimulation, on bipolar hook electrodes. Single or double shocks of 0.2 ms duration were used. The stimulus strength was adjusted to be 4 times threshold for the cortically evoked potential. For studying the cerebellar input to VL, one stimulating electrode was inserted into the intermediate, and a second into the lateral, deep cerebellar nuclei, contralateral to the recording site. The stimulus strength was adjusted to produce maximum positivity over the contralateral pre-cruciate (motor) cortex. The silver ball electrodes used to record the evoked potential were left on the surface of the sensorimotor cortex and later used to stimulate axons of thalamocortical projection neurons. To gain access to VL neurons and to the pretectum, the dorsal surface of the thalamus and midbrain was exposed by aspiration of the overlying parietal lobe, fornix and part of the hippocampus. For recording from VL neurons, glass micropipettes filled with 2 M potassium acetate, or 2% pontamine sky blue in 0.5 M potassium acetate, were inserted perpendicularly into the surface of the thalamus (stereotaxic coordinates anterior 10, lateral 3-4 (ref. 11)). The region containing relay neurons in the cerebello-cerebral path was localized by monitoring the cerebello-thalamic field potentials. Neurons in this region were penetrated for intracellular recording. Neurons were identified as VL relay neurons by their monosynaptic excitation from the contralateral deep cerebellar nuclei and, when possible, by their antidromic activation from the motor cortex (see Fig. la). Only cells with stable resting membrane potentials of at least 40 mV were studied. After a relay cell was identified, its response to stimulation of the dorsal columns was examined. Responses were sampled in a number of VL neurons prior to lesioning of the pretectum (see Table 1). Pretectal lesions were made by means of a radiofrequency probe (Radionics), which was inserted according to stereotaxic coordinates (anterior 5, lateral 3, horizontal 12 (ref. 3)), perpendicular to the exposed surface of the midbrain. When the target area was reached, RF current was applied to raise the tip temperature to 90 °C for 3 min. Subsequently, the thalamic VL relay area was re-examined. (The region was quickly relocated using surface anatomical landmarks and stereotaxic coordinates ascertained in the first half of the experiments, as well as by monitor
TABLE I Number and percentages (in parentheses) of VL neurons receiving and tested for dorsal column input, before and after pretectal lesions Cats
Before lesion
After lesion
1 2 3 4 5
5/10(50) 15/18(83) 12/14(86) 7/11 (64) 7/8 (88)
0/5 (0) 2/15 (13) 1/7 (14) 6/22 (27) 1/12(8)
Total
46/61 (75%)
10/61(16%)
of cerebello-thalamic fields.) VL relay neurons were identified and tested for dorsal column input, as in the first half of the experiment. Selected recording sites were marked by ejecting dye from the tip of the micropipettes (cathodal DC current of 2-4/~A, passed 3-5 rain). An example is illustrated in Fig. 2B,C. At the end of the experiment, the animal was perfused with 0.9% saline followed by 10% formalin and the brain removed for storage in fixative. The thalamus and pretectum were cut into
I0-
Z u0 ~n w .,i
= 0
l
v o
,-. 5-
F1
I1,,, IhJi
4Y {461611
10 ms
L " I
~
"o-
ImV
Fig. 1. Distribution and latencies of postsynaptic potentials recorded in VL neurons to stimulation of the dorsal columns, before (upper) and after (reversed histograms) pretectal lesions. Percentages and numbers (in parentheses) refer to all cells tested (see also Table I). Typical intracellular records obtained prior to a lesion are shown for short, intermediate and long latency responses by the inserts b-d. The framed inset in a, illustrates an example of a fully identified VL neuron which was antidromically activated from the motor cortex (Mt ctx) and monosynaptieally excited from the contralateral deep cerebellar nuclei (CN). After a lesion, dorsal column input was abolished as illustrated by inset e. All postsynaptic potentials are averages of 4 sweeps. The time base for all records is under e. Calibration: 1 mV for all records except 20 mV for motor cortex. O, IPSP, inhibitory postsynaptic potential.
137
80-/~m-thick frontal sections,
stained with thionin or Neutral red, and examined for lesions or dye marks. Location of unmarked recording sites in the VL region was confirmed by histological reconstruction
with reference to the dye marks and stereotaxic coordinates. A total of 122 cells were studied: 61 cells before, and 61 cells after the lesions. All cells were monosyn-
¢
Fig. 2. Photomicrographs of the pretectum (A) and ventrolateral thalamus (B). A: the location of the lesion (arrow) on a frontal section at A 4.5. PT, pretectum; ML, medial lemniscus;MRF, midbrain reticular formation; RN, red nucleus; MG, medial geniculate nucleus; LG, lateral genieulate nucleus. B: the location of a dye mark (arrow) on a frontal section at A 10. VL, ventrolateral nucleus; VPL, ventroposterolateral nucleus; LP, lateral posterior nucleus; LD, lateral dorsal nucleus; CL, central lateral nucleus; MD, mediodorsal nucleus. C: at high magnification, the dye mark (arrows) located in the area of detail which is outlined in B.
138 aptically excited following stimulation of the contralateral deep cerebellar nuclei; and 32 of the cells were activated antidromically from the motor cortex (19 before and 13 after the lesions). Prior to the lesion, 46/61 (75%) cells responded to stimulation of the ~ r s a i columns. In different animals, the percentages showing dorsal column input ranged between 50 and 88%, as shown in Table L All responses, except one, were excitatory. Typical examples of postsynaptic responses are illustrated by the insets in Fig. l b - d . The latencies ranged between 3.8 and 20 ms, with responses clustered in 3 groups. These were a short (around 4 - 5 ms), an intermediate (around 10 ms), and a long (around 15 ms and longer) latency group, as illustrated in Fig. 1. The wide range of latencies probably reflects both synaptic transmission through a chain of intercalated neurons and the wide range of conduction velocities among dorsal column afferents 6a7. The general distribution of dorsal column input prior to a lesion is consistent with earlier data t2. After the lesions, only 10/61 (16%) cells responded to stimulation of the dorsal columns. The percentage ranged from 0 to 27% in different animals. This decrease in the percentage of cells showing dorsal column input is statistically significant (at a -- 0.03; one-tailed, sign test for matched samples, n = 5). All short and intermediate latency responses were abolished (in response to single or double volleys) and only a few long latency postsynaptic potentials remained, as illustrated in Fig. 1.
! Anderson, M.E. and DeVito, J.L., An analysis of potentially converging inputs to the rostral ventral thalamic nuclei of the cat, Exp. Brain Res., 68 (1987) 260-276. 2 Asanuma, H., Fernandez, J., Scheibel, M.E. and Scheibel, A.B., Characteristics of projections from the nucleus ven. tralis lateralis to the motor cortex in the cat: an anatomical and physiological study, Exp. Brain Res., 20 (1974) 315-330. 3 Avendano, C. and Juretschke, M.A., The pretectal region of the cat: a structural and topographical study with stereotaxic coordinates, J. Comp. Neurol., 193 (1980) 69-88. 4 Berkley, K.J. and Mash, D.C., Somatic sensory projections to the pretectum in the cat, Brain Research, 158 (1979) 445-449. 5 Berkley, K.J., Spatial relationships between the termination of somatic sensory and motor pathways in the rostral brain stem of cats and monkeys. If. Cerebellar projections compared with those of the ascending somatic sensory path-
The location and the size of the lesions were comparable for all cats. A typical example of a lesion is shown in Fig. 2A. It had a diameter of 2 mm and extended sagitally from anterior 4 to 6. Such a lesion did not abolish lemniscal input to VPL because short-latency excitatory postsynaptic potentials ( 3 - 4 ms) could still be recorded in VPL and the cortically evoked potential was unaffected (not illustrated). This suggests that the lesions removed effects of a specific suhpopulation of lemniscal projections onto pretectal neurons. At present it is not known whether the lesions destroyed pretectal relay neurons or lemniscal afferents to the pretectum, nor can the possibility of lesioning fibers en passage he excluded. Future studies will be directed towards the clarification of these aspects. These experiments altogether demonstrate dorsal column input to VL passes through the pretectum. This input could play an important role in setting the excitability of VL neurons during movement execution, and could act to modify cerehellar commands before the commands reach the motor cortex. The present data also shed new light on pretectal functions by indicating that the pretectum processes not only visual t4, but also somatosensory information.
The authors wish to thank Dr. E.E. Brink for reading the manuscript and Mrs. E. Oquendo for technical assistance. The study was supported by grant NS10705.
ways in lateral diencephalon, J. Comp. Neurol., 220 (1983) 229-251. 6 Brown, A.G., Cutaneous afferent fibre collaterals in the dorsal columns of the cat, Exp. Brain Res., 5 (1968) 293-305. 7 Craig, A.D. and Burton, H., The distribution and topographical organization in the thalamus of anterogradelytransported horseradish peroxidase after spinal injections in cat and raccoon, Exp. Brain Res., 58 (1985) 227-254. 8 Greenan, T.J. and Strick, P.L., Do thalamic regions which project to rostrai primate motor cortex receive spinothalamic input?, Brain Research, 362 (1986) 384-388. 9 Horne, M.K. and Porter, R., The discharges during movement of cells in the ventrolateral thalamus of the conscious monkey, J. Physiol. (Lond.), 304 (1980) 349-372. 10 Itoh, K., Efferent projections of the pretectum in the cat, Exp. Brain Res., 30 (1977) 89-105. 11 Jasper, H.H. and Ajmone-Marsan, C., A StereotaxicAtlas of the Diencephalon of the Cat, National Research Council
139 of Canada, Ottawa, 1954. 12 Ma~kel, R. and Noda, T., Sensory input to cer "bellocerebral relay neurons in the cat thalamus, Brain Research, 440 (1988) 348-351. 13 Nakano, K., Kohno, M., Hasegawa, Y. and Tokushige, A., Cortical and brain stem afferents to the ventral thalamic nuclei of the cat demonstrated by retrograde axonal transport of horseradish peroxidase, J. Comp. NeuroL, 231 (1985) 102-120. 14 Rodiek, R.W., Visual pathways, Annu. Rev. Neurosci., 2 (1979) 193-225. 15 Strick, P.L., Activity of ventrolateral thalamic neurons dur-
ing arm movement, J. NeurophysioL, 39 (1976) 1032-1044. 16 Tracey, D.J., Asanuma, C., Jones, E.G. and Porter, R., Thalamic relay to motor cortex: afferent pathways from brain stem, cerebellum and spinal cord in monkeys, J. Neurophysiol., 44 (1980) 532- 554. 17 Uddenberg~N., Differential localization in dorsal funiculus of fibres originating from different receptors, Exp. Brain Res., 4 (1968) 367-376. 18 Wiberg, M. and Blomqvist, A., The projection to the mesencephalon from the dorsal column nuclei: an anatomical study in the cat, Brain Research, 311 (1984) 225-244.
135
BRE 23252
The pretectum as a site for relaying dorsal column input to thalamic VL neurons Robert Mackel and Terumi Noda The Rockefeller University, New York, NY10021 (U.S.A.)
(Accepted 30 August 1988) Key words: Pretectum; Dorsal column; Ventrolateral thalamus; lntracellular recording
Imracellular recording techniques were used to study dorsal column input to 122 feline ventral thalamus (VL) relay neurons, before (61 cells) and after (61 cells) lesioning the pretectum. Prior to the lesions, 75% (46161) of the neurons responded with short and longer latency postsynaptic potentials to dorsal column stimulation. Latencies of the postsynaptic potentials ranged firom4 (short) to 20 ms (long). After the lesions, only long latency responses were encountered, and those responses were seen in only 16% (10161) of the cells. These data indicate that the pretectum may play an important role in mediating dorsal column information to VL, ultimately influencing cerebellar commands to the motor cortex.
It is generally believed that neurons in the ventrolateral nucleus (VL) of the thalamus lack significant input from the spinal cord and simply mediate cerebellar information onto motor cortical areas 2"9a5. However, some anatomical studies have demonstrated a distinct spinothalamic input to VL 5"7'8"~6. Although there is no anatomical evidence for lemniscal input to the bulk of VL, there is controversy about a lemniscal projection to a small border area between VL and VPL 5'16. Nevertheless, in view of the anatomical work, more spinal information would be expected to reach VL than has been reported in the earlier physiological studies. Indeed, in recent experiments using intracellular recording techniques, it was shown ~2that many VL neurons responded with shortand long-latency postsynaptic potentials to stimulation of spinothalamic and/or dorsal column afferents. The frequent input from the dorsal columns was a surprising finding. Since all dorsal column input to VL is indirect and hence mediated polysynaptically, it follows that it must be relayed via a structure (or structures) intercalated between the dorsal column nuclei and VL. A candidate structure could be the pretectum, since a number of anatomical studies
have recently shown that the pretectum receives lemniscal input 4"Is and projects to VL re°a3. The present report shows that the pretectum may indeed mediate dorsal column input, since pretectal lesions were found to abolish lemniscal information to VL neurons. In the present work, intracellular recording techniques were used to study dorsal column input to VL neurons, before and after lesioning the pret 'urn. The data were obtained from five cats, ,i, sere anesthetized with alpha-chloralose (50-6" mg/k~;), paralyzed, and artificially ventilate, Supplementary doses of the anesthetic (10-20 mg/kg) were given intravenously, if necessary. The blood pressure was monitored and maintained above 80 m m H g by infusion of a solution of 5% dextrose in saline 0.9% or a solution of metaraminol bitartrate 40 ~d/ml (Aramine); rectal temperature was maintained between 37 and 39 °C. The general preparation for stimulating the spinal cord and for recording from VL neurons has been described elsewhere m2. Briefly, a craniotomy and a cervical laminectomy were performed. The dorsal columns were separated from the spinal cord from seg-
Correspondence: R. Mackel, The Rockefeller University, 1230 York Avenue, New York, NY 10021, U.S.A.
0006-8993/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
136 ments C s to C3 and mounted, for stimulation, on bipolar hook electrodes. Single or double shocks of 0.2 ms duration were used. The stimulus strength was adjusted to be 4 times threshold for the cortically evoked potential. For studying the cerebellar input to VL, one stimulating electrode was inserted into the intermediate, and a second into the lateral, deep cerebellar nuclei, contralateral to the recording site. The stimulus strength was adjusted to produce maximum positivity over the contralateral pre-cruciate (motor) cortex. The silver ball electrodes used to record the evoked potential were left on the surface of the sensorimotor cortex and later used to stimulate axons of thalamocortical projection neurons. To gain access to VL neurons and to the pretectum, the dorsal surface of the thalamus and midbrain was exposed by aspiration of the overlying parietal lobe, fornix and part of the hippocampus. For recording from VL neurons, glass micropipettes filled with 2 M potassium acetate, or 2% pontamine sky blue in 0.5 M potassium acetate, were inserted perpendicularly into the surface of the thalamus (stereotaxic coordinates anterior 10, lateral 3-4 (ref. 11)). The region containing relay neurons in the cerebello-cerebral path was localized by monitoring the cerebello-thalamic field potentials. Neurons in this region were penetrated for intracellular recording. Neurons were identified as VL relay neurons by their monosynaptic excitation from the contralateral deep cerebellar nuclei and, when possible, by their antidromic activation from the motor cortex (see Fig. la). Only cells with stable resting membrane potentials of at least 40 mV were studied. After a relay cell was identified, its response to stimulation of the dorsal columns was examined. Responses were sampled in a number of VL neurons prior to lesioning of the pretectum (see Table 1). Pretectal lesions were made by means of a radiofrequency probe (Radionics), which was inserted according to stereotaxic coordinates (anterior 5, lateral 3, horizontal 12 (ref. 3)), perpendicular to the exposed surface of the midbrain. When the target area was reached, RF current was applied to raise the tip temperature to 90 °C for 3 min. Subsequently, the thalamic VL relay area was re-examined. (The region was quickly relocated using surface anatomical landmarks and stereotaxic coordinates ascertained in the first half of the experiments, as well as by monitor
TABLE I Number and percentages (in parentheses) of VL neurons receiving and tested for dorsal column input, before and after pretectal lesions Cats
Before lesion
After lesion
1 2 3 4 5
5/10(50) 15/18(83) 12/14(86) 7/11 (64) 7/8 (88)
0/5 (0) 2/15 (13) 1/7 (14) 6/22 (27) 1/12(8)
Total
46/61 (75%)
10/61(16%)
of cerebello-thalamic fields.) VL relay neurons were identified and tested for dorsal column input, as in the first half of the experiment. Selected recording sites were marked by ejecting dye from the tip of the micropipettes (cathodal DC current of 2-4/~A, passed 3-5 rain). An example is illustrated in Fig. 2B,C. At the end of the experiment, the animal was perfused with 0.9% saline followed by 10% formalin and the brain removed for storage in fixative. The thalamus and pretectum were cut into
I0-
Z u0 ~n w .,i
= 0
l
v o
,-. 5-
F1
I1,,, IhJi
4Y {461611
10 ms
L " I
~
"o-
ImV
Fig. 1. Distribution and latencies of postsynaptic potentials recorded in VL neurons to stimulation of the dorsal columns, before (upper) and after (reversed histograms) pretectal lesions. Percentages and numbers (in parentheses) refer to all cells tested (see also Table I). Typical intracellular records obtained prior to a lesion are shown for short, intermediate and long latency responses by the inserts b-d. The framed inset in a, illustrates an example of a fully identified VL neuron which was antidromically activated from the motor cortex (Mt ctx) and monosynaptieally excited from the contralateral deep cerebellar nuclei (CN). After a lesion, dorsal column input was abolished as illustrated by inset e. All postsynaptic potentials are averages of 4 sweeps. The time base for all records is under e. Calibration: 1 mV for all records except 20 mV for motor cortex. O, IPSP, inhibitory postsynaptic potential.
137
80-/~m-thick frontal sections,
stained with thionin or Neutral red, and examined for lesions or dye marks. Location of unmarked recording sites in the VL region was confirmed by histological reconstruction
with reference to the dye marks and stereotaxic coordinates. A total of 122 cells were studied: 61 cells before, and 61 cells after the lesions. All cells were monosyn-
¢
Fig. 2. Photomicrographs of the pretectum (A) and ventrolateral thalamus (B). A: the location of the lesion (arrow) on a frontal section at A 4.5. PT, pretectum; ML, medial lemniscus;MRF, midbrain reticular formation; RN, red nucleus; MG, medial geniculate nucleus; LG, lateral genieulate nucleus. B: the location of a dye mark (arrow) on a frontal section at A 10. VL, ventrolateral nucleus; VPL, ventroposterolateral nucleus; LP, lateral posterior nucleus; LD, lateral dorsal nucleus; CL, central lateral nucleus; MD, mediodorsal nucleus. C: at high magnification, the dye mark (arrows) located in the area of detail which is outlined in B.
138 aptically excited following stimulation of the contralateral deep cerebellar nuclei; and 32 of the cells were activated antidromically from the motor cortex (19 before and 13 after the lesions). Prior to the lesion, 46/61 (75%) cells responded to stimulation of the ~ r s a i columns. In different animals, the percentages showing dorsal column input ranged between 50 and 88%, as shown in Table L All responses, except one, were excitatory. Typical examples of postsynaptic responses are illustrated by the insets in Fig. l b - d . The latencies ranged between 3.8 and 20 ms, with responses clustered in 3 groups. These were a short (around 4 - 5 ms), an intermediate (around 10 ms), and a long (around 15 ms and longer) latency group, as illustrated in Fig. 1. The wide range of latencies probably reflects both synaptic transmission through a chain of intercalated neurons and the wide range of conduction velocities among dorsal column afferents 6a7. The general distribution of dorsal column input prior to a lesion is consistent with earlier data t2. After the lesions, only 10/61 (16%) cells responded to stimulation of the dorsal columns. The percentage ranged from 0 to 27% in different animals. This decrease in the percentage of cells showing dorsal column input is statistically significant (at a -- 0.03; one-tailed, sign test for matched samples, n = 5). All short and intermediate latency responses were abolished (in response to single or double volleys) and only a few long latency postsynaptic potentials remained, as illustrated in Fig. 1.
! Anderson, M.E. and DeVito, J.L., An analysis of potentially converging inputs to the rostral ventral thalamic nuclei of the cat, Exp. Brain Res., 68 (1987) 260-276. 2 Asanuma, H., Fernandez, J., Scheibel, M.E. and Scheibel, A.B., Characteristics of projections from the nucleus ven. tralis lateralis to the motor cortex in the cat: an anatomical and physiological study, Exp. Brain Res., 20 (1974) 315-330. 3 Avendano, C. and Juretschke, M.A., The pretectal region of the cat: a structural and topographical study with stereotaxic coordinates, J. Comp. Neurol., 193 (1980) 69-88. 4 Berkley, K.J. and Mash, D.C., Somatic sensory projections to the pretectum in the cat, Brain Research, 158 (1979) 445-449. 5 Berkley, K.J., Spatial relationships between the termination of somatic sensory and motor pathways in the rostral brain stem of cats and monkeys. If. Cerebellar projections compared with those of the ascending somatic sensory path-
The location and the size of the lesions were comparable for all cats. A typical example of a lesion is shown in Fig. 2A. It had a diameter of 2 mm and extended sagitally from anterior 4 to 6. Such a lesion did not abolish lemniscal input to VPL because short-latency excitatory postsynaptic potentials ( 3 - 4 ms) could still be recorded in VPL and the cortically evoked potential was unaffected (not illustrated). This suggests that the lesions removed effects of a specific suhpopulation of lemniscal projections onto pretectal neurons. At present it is not known whether the lesions destroyed pretectal relay neurons or lemniscal afferents to the pretectum, nor can the possibility of lesioning fibers en passage he excluded. Future studies will be directed towards the clarification of these aspects. These experiments altogether demonstrate dorsal column input to VL passes through the pretectum. This input could play an important role in setting the excitability of VL neurons during movement execution, and could act to modify cerehellar commands before the commands reach the motor cortex. The present data also shed new light on pretectal functions by indicating that the pretectum processes not only visual t4, but also somatosensory information.
The authors wish to thank Dr. E.E. Brink for reading the manuscript and Mrs. E. Oquendo for technical assistance. The study was supported by grant NS10705.
ways in lateral diencephalon, J. Comp. Neurol., 220 (1983) 229-251. 6 Brown, A.G., Cutaneous afferent fibre collaterals in the dorsal columns of the cat, Exp. Brain Res., 5 (1968) 293-305. 7 Craig, A.D. and Burton, H., The distribution and topographical organization in the thalamus of anterogradelytransported horseradish peroxidase after spinal injections in cat and raccoon, Exp. Brain Res., 58 (1985) 227-254. 8 Greenan, T.J. and Strick, P.L., Do thalamic regions which project to rostrai primate motor cortex receive spinothalamic input?, Brain Research, 362 (1986) 384-388. 9 Horne, M.K. and Porter, R., The discharges during movement of cells in the ventrolateral thalamus of the conscious monkey, J. Physiol. (Lond.), 304 (1980) 349-372. 10 Itoh, K., Efferent projections of the pretectum in the cat, Exp. Brain Res., 30 (1977) 89-105. 11 Jasper, H.H. and Ajmone-Marsan, C., A StereotaxicAtlas of the Diencephalon of the Cat, National Research Council
139 of Canada, Ottawa, 1954. 12 Ma~kel, R. and Noda, T., Sensory input to cer "bellocerebral relay neurons in the cat thalamus, Brain Research, 440 (1988) 348-351. 13 Nakano, K., Kohno, M., Hasegawa, Y. and Tokushige, A., Cortical and brain stem afferents to the ventral thalamic nuclei of the cat demonstrated by retrograde axonal transport of horseradish peroxidase, J. Comp. NeuroL, 231 (1985) 102-120. 14 Rodiek, R.W., Visual pathways, Annu. Rev. Neurosci., 2 (1979) 193-225. 15 Strick, P.L., Activity of ventrolateral thalamic neurons dur-
ing arm movement, J. NeurophysioL, 39 (1976) 1032-1044. 16 Tracey, D.J., Asanuma, C., Jones, E.G. and Porter, R., Thalamic relay to motor cortex: afferent pathways from brain stem, cerebellum and spinal cord in monkeys, J. Neurophysiol., 44 (1980) 532- 554. 17 Uddenberg~N., Differential localization in dorsal funiculus of fibres originating from different receptors, Exp. Brain Res., 4 (1968) 367-376. 18 Wiberg, M. and Blomqvist, A., The projection to the mesencephalon from the dorsal column nuclei: an anatomical study in the cat, Brain Research, 311 (1984) 225-244.