THE JOURNAL OF COMPARATIVE NEUROLOGY 269:425-430 (1988)

Medial Gastrocnemius Motor Nucleus in the Rat: Age-Related Changes in the Number and Size of Motoneurons KEN HASHIZUME, KENRO KANDA, AND ROBERT E. BURKE Department of Kinesiology (K.H.) and Department of Physiology (K.K., R.E.B.), Tokyo Metropolitan Institute of Gerontology, Tokyo 173, Japan

ABSTRACT The age-related alterations in the number and size of alpha- and gammamotoneurons were studied in the medial gastrocnemius (MG) motor nuclei in rats a t four ages: young (5 months), middle aged (10-13 months), old (26 months), and very old (31 months). Small volumes (0.1-0.5 ~ 1of) 40% horseradish peroxidase (HRP) solution were injected into the cut MG nerve bilaterally by using glass micropipettes and a pressure injection system. The number, position, and soma size (average soma diameter) of MG motoneurons were determined by using photographic maps of each TMB-stained section. The total number of myelinated axons was counted in seven MG nerves from the same animals. The average soma diameters in each MG nucleus were distributed bimodally; cells with average diameter greater than 21.0-24.0 pm were presumed to be alpha-motoneurons and those with smaller diameters were presumed to be gamma. The mean number of presumed alpha-motoneurons was significantly less in the old and very old groups as compared with the young and middle-aged. In contrast, the number of presumed gamma-motoneurons was the same across age groups. The mean average soma diameter of both alpha- and gamma-motoneuronswas smaller in the old animals. The apparent decrease in the total number of labeled motoneurons in old animals was also reflected in a decrease in myelinated axon counts. We conclude that there is a significant decrease in the absolute numbers of motoneurons in rats aged 26 months and older, with most of the decrease occurring among the larger alpha-motoneurons. Key words: alpha-motoneuron,gamma-motoneuron, spinal cord, cell survival, aging

It has been reported that spinal motoneurons decrease in number with increasing age in man (Kawamura et al., '77; Tomlinson and Irving, '77; Tsukagoshi et al., '79) and in animals (Wright and Spink, '59), although most of these studies were restricted to counting total numbers of ventral horn neurons in particular spinal segments without definitive identification of motoneurons or motor nuclei. It has also been shown that the total number of myelinated fibers in peripheral nerves decreases significantly with increasing age (Samorajski, '74; Mittal and Logmani, '87). On the other hand, age-related changes in the neuromuscular system appear to differ in different muscles (Caccia et al., '79; Eddinger et al., '85), which suggests that the effects of aging on motoneurons may not be equal among different cell groups, including the motoneurons that innervate slow-

0 1988 ALAN R. LISS, INC.

vs. fast-twitch muscle units. The relative proportions of fast-twitch (type F) and slow-twitch (type S) motor units are known to differ in different muscles (see Burke, '81). Our previous results showed a differential effect of aging on the physiological properties of type F and type S motor units in the rat medial gastrocnemius (MG) muscle (Kanda et al., '86). In order to further clarify such effects, quantitaAccepted October 12, 1987. Address reprint requests to Dr. K. Kanda, Department of Physiology, Tokyo Metropolitan Institute of Gerontology, 35-2, Sakaecho Itabashi-ku, Tokyo 173, Japan. R.E. Burke's present address is Laboratory of Neural Control, National Institute of Neurological and Communication Disorders and Stroke, National Institutes of Health, Bethesda, MD 20892.

K. HASHIZUME ET AL.

426 tive estimates of motoneuron numbers in individual motor nuclei are essential. The motor nuclei also contain gammamotoneurons, which differ from alpha-motoneurons in both morphology and function (Burke, '81; Burke et al., '82; Strick et al., '76; Westbury, '82). There is little information available about possible differential effects of aging on these two classes of motoneurons. The purpose of this experiment was to study age-related alterations in the number and size of alpha- and gamma-motoneurons in a particular motor nucleus, the MG motor nucleus in rats.

MATERIALS AND METHODS Experiments were performed on 12 male Fischer rats (F344/DuCrj),including two young (5 months; body weight, 340-350 g), four middle-aged (10-13 months; 410-500 g), three old (26 months; 425-500 g), and three very old (31 months; 340-410 g) animals. All rats were raised in a specific pathogen-free colony in which the 75, 50, and 25% survival times of the rats were 778, 855, and 957 days, respectively. They were kept in conventional housing, less than four per cage, and without restriction of feeding or any special exercise requirements. All animals were anesthetized with pentobarbital (35 mgl kg; i.p.) preceding surgery. Under aseptic conditions, the MG nerve was freed of surrounding tissues by microdissection and was cut and tied with suture material at a point near the entry to the MG muscle. Glass micropipettes, broken to a tip diameter of 20-40 pm and filled with a fresh solution of 40% horseradish peroxidase (HPR, Sigma type VI) in distilled water, were used to impale the proximal stump of the MG nerve under a dissecting microscope. Small volumes (0.1-0.5 p1) of HRP solution were injected by using 20-50 brief (20-30 msec) pulses of pressure (about 25 psi), produced by a pulse-operated valve device (Picospritzer 11). After HRP injection, the wound was washed with sterile saline and then closed in layers. In all rats, both legs were treated identically. After a survival period of 48-52 hours, the animal was reanesthetized with pentobarbital and perfused transcardially with 500 ml of warmed saline, followed by 700 ml of cold fixative mixture (1.25% glutaraldehyde and 1.0% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, 4°C) over 30 minutes. The fixative solution was then washed out by perfusion with 500 ml phosphate buffer containing 20% sucrose at 4°C. After the perfusion, the spinal cord was exposed and the exit levels of the various spinal roots were identified. Then the cord was removed and placed in 20% sucrose buffer at 4°C for 15-20 hours. The spinal segment boundaries were marked with small holes made by inserting a pin into the dorsal columns at points between the entry zone of adjacent, identified dorsal roots. The cord was then cut into suitable blocks, with the L4-L6 segments included in a single block. Serial sections in the horizontal plane were cut at 80-pm thickness on a freezing microtome. All sections were saved in serial order and collected individually into 0.1 M phosphate buffer. Sections were processed for demonstration of HRP with the chromogen 3,3',5,5'-tetramethy1 benzidine (TMB) by using Mesulam's ('78) protocol. After processing, all sections were mounted on gelatinized slides in serial order without counterstain. Sections containing HRP-labeled cells were photographed on high-contrast film (Kodak Technical Pan), and photomontage maps for cell identification were made of each section by printing the negatives on Kodalith Ortho film. Spinal cord sections were examined with brightfield mi-

Fig. 1. Photomicrograph of a single, 80-pm-thick horizontal section from the spinal cord of a young (5 months old) rat, showing retrogradely labeled motoneurons in the MG motor nucleus. Calibration bar = 0.5 mm.

croscopy and a projection microscope. Labeled neurons were considered "cell bodies" when the entire nucleus was included in the section or when a nucleolus could be visualized. Cell bodies, as defined above, were individually numbered for reference on the photomontage maps. In each motor nucleus, lateral distances from the central canal and caudal distances from the most rostra1 cell to each center of a labeled cell were measured on the photographic maps. The maximum and minimum diameters of the somatic profile of each mapped celI body were measured at high magnification (780 X ) with a projection microscope directly from the original sections. Average soma diameters (one-half of the sum of the measured maximum and minimum diameters) were then calculated (see Burke et al., '77, '82). No correction for shrinkage during fixation was attempted in this study. In seven animals, a small piece of the MG nerve was excised from the distal cut end just before HRP injection and fixed in 2.5% glutaraldehyde. This was later osmicated in 1%Os04. Semithin cross sections were cut at about 1pm thickness from the specimen embedded in the Epon and stained with toluidine blue. The total numbers of myelinated axons were counted from photographic prints made of these sections at 400 x magnification. We used the one-way analysis of variance to test for differences between the mean numbers and the mean average

AGING OF RAT MOTOR NUCLEUS

427

I 40

30

L4

.T

.

s. .*

r

20

10

0

15.0

22.5

30.0

37.5

Average Soma Diameter

I

T

Fig. 2. Reconstruction of both MG motor nuclei from serial horizontal sections in a middle-aged (12 months old) rat, showing a dorsal view (white matter-pia boundary in heavy lines and gray-white matter boundary in light lines). Superimposed positions (dots) of all labeled MG motoneuron cell bodies are shown. Boundaries between L4,L5 and L6 segments are indicated by dashed lines.

diameter of cells in the four different age groups. When the F ratio indicated a significant difference among the groups ( P < .001), the Tukey multiple comparison test was used to test for significant differences between individual group means (Zar, '84).

RESULTS Retrogradely labeled motoneurons innervating the MG muscle were located in the dorsolateral part of the ventral horn in spinal segments L4-L6 (mainly L5 segment, Fig. 1).Both large and small motoneurons were intermingled and arranged in a discrete longitudinal column. In a few cases, cells outside the darkly labeled cell column were very lightly labeled and these were not included in the data. Such occurrences were assumed to be due to leakage of HRP into nerve bundles other than MG (see, e.g., Burke et al., '77). Figure 2 shows a reconstruction of the left and right MG motor nuclei in a middle-aged rat. In the dorsal view, each MG motor nucleus had a lens-shaped outline. The motor nuclei on both sides in each animal were quite symmetrical with respect to the rostrocaudal location and the number of included motoneurons. There were some interanimal differences in the rostrocaudal location of the cell columns. The histogram in Figure 3 shows the distribution of average soma diameters for the entire population of labeled motoneurons belonging to both left and right MG motor nuclei in a middle-aged rat. The average soma diameters of

45.0

(rm)

Fig. 3. Histogram of the average soma diameters for the entire population of labeled motoneurons belonging to both left and right MG motor nuclei shown in Figure 2. Note the bimodal distribution, with a distinct boundary a t about 22.5 pm. The smaller cells (black) are presumed to be gamma-motoneurons and the larger ones are presumed to he alpha (see text).

motoneurons were distributed bimodally, with a n apparent transition at about 22.5 pm between small and large groups. The cumulative frequency diagram showed an inflection at the same point. Similar bimodal distributions of cell size were found in all motor nuclei examined, with size transition values ranging between 21.0 and 24.0 pm in different nuclei. The values for the three very old animals ranged from 21.0 to 22.5 pm and were not different from those for the Younger groups. Since it has been shown that gamma-motoneurons are systematica~~y smaller in average diameter than alpha cells (Bryan et al., '72; Cullheim and Ulfhake, '79; Burke et al., '82; Westbury, '82), we assumed that cells with average diameter G22.5 pm (Fig. 3, black) were gammamotoneurons and that those with larger diameters were alpha-motoneurons. Therefore, it seems likely that both alpha- and gamma-motoneurons are intermingled within the rat MG nucleus (Fig. I),as is also true in the cat (e.g., Burke et al., '77). Most of the presumed gamma cells were labeled more heavily than the larger, presumed alpha cells, as also found in cats (Strick et al., '76; Burke et al., '77; however, see Peyronnard and Charron, '83). This was also true for the old and very old animals, although in general all cells in these animals were less heavily labeled than those in younger rats. Table 1 shows the numbers of motoneurons in four age groups. There was no significant difference in the total numbers of motoneurons between the young (5 months) and middle-aged (10-13 months) groups. The MG nuclei in the old (26 months) and the very old (31 months) groups, however, contained significantly fewer motoneurons compared with the young and middle-aged groups (01 < 0.001 by oneway analysis of variance and Tukey multiple comparison test; see Materials and Methods). The numbers of alphaand gamma-motoneurons were estimated from the size distribution in each animal (see Figs. 3, 5). Motoneurons in the gamma size range constituted between 25 and 33% of each labeled cell population in the young and middle-aged groups (means 27.4% and 28.8%, respectively). The numbers of presumed alpha-motoneurons in the old and very old groups were significantly smaller than those in the

K. HASHIZUME ET AL.

428

TABLE 1.Mean Numbers (2SD) of Total, Alpha- and Gamma-Motoneurons in the MG Motor Nuclei, and Myelinated Axons of MG Nerves Age (months)

MG nuclei

Total cells

Alpha MNs

Gamma MNs

Myelinated axons

4 6 5 6

134.5 3.2 137.0 k 4.0 121.4 3.9* 107.2 6.9*

* **

97.8 f 2.4 97.5 f 4.2 85.0 f 2.9* 68.3 f 5.4*

36.8 f 3.2 39.5 k 3.3 36.4 2.2 38.8 k 2.9

297.0 f 10.0 (3) 263.0 (1) 241.7 16.6** (3)

5 10 - 13 26 31

*

*

*Significant difference (01 < 0.001) from the means obtained for 5-month and 10 - 13-month animals by the Tukey multiple comparison test (see Materials and Methods). **Difference between 10 - 13 month and 31 month animals was significant (P< .01) by the t-test.

120 0 L

151

100

80

60 15.0

22.5

30.0

37.5

Average Soma Diameter 0

10

20

Age

30

(months)

Fig. 4. Age-related alterations in the mean percentage of presumed alpha (open circles). and gamma-motoneurons (filled triangles) making up the MG notoneuron populations, normalized to the values found in the young (5 months) group (dashed line, 100%). Vertical bars indicate 1 standard deviation about the mean.

young and middle-aged groups. In contrast, however, the numbers of presumed gamma-motoneuronswere not significantly different among the age groups. The decrease in total motoneuron numbers in the old and very old groups was due entirely to an apparent loss of motoneurons in the alpha size range. The number of presumed alpha-motoneurons decreased by 13%for the very old group and by 30% for the old group (Fig. 41, compared to younger animals. The histograms in Figure 5 show the distribution of average soma diameters for motoneurons pooled from the middle-aged (heavy outline, open bars) and the very old (lighter line, shaded) rats plotted as percentage of the respective populations. These histograms include all motoneurons obtained from six motor nuclei in each of these two age groups. The average soma diameters of MG motoneurons in both age groups were clearly distributed in a bimodal manner. Although the total ranges of cell size and the transition point between alpha- and gamma-motoneurons were almost the same at both ages, the data suggest a general decrease in soma diameters in the very old rats. The mean average soma diameters of gamma-motoneurons and alpha-motoneurons were smaller in the very old than

45.0

(w)

Fig. 5. Comparison of average soma sizes of MG motoneurons in middleaged (10-13 months, heavy lines, open bars) and very old (31 months, lighter lines, shaded) rats. These histograms include all motoneurons obtained from six motor nuclei of these two groups (n=822 and 643, respectively, see Table 1). Note the clear bimodal distribution a t both ages. The data suggest a general decrease in soma diameters in the very old animals but no definite shift in the transition diameter between alpha- and gamma-motoneurons,

in middle-aged rats (17.4 f 2.3 vs. 18.2 f 2.1 pm for gammamotoneurons, 29.1 f 3.7 vs. 30.4 f 3.5 pm for alpha-motoneurons in the respective age groups; a < 0.005 by one-way analysis of variance and Tukey multiple comparison test). We also counted the total number of myelinated axons in the MG muscle nerve taken from seven rats (see Materials and Methods) as an independent check of our estimates on the size of the MG motoneuron pool. The mean total number of myelinated axons (i.e., both af'ferent and efferent fibers) present in MG nerves in the middle-aged rats was 297 (three nerves, see Table 1).In the very old rats, the mean axon count was 242 (three nerves), indicating a 19% decrease compared with the middle-aged rats. This difference was statistically significant (P < .01; t-test)despite the small sample sizes. The number in a single example from a 26-month-old rats was 263, intermediate between these extremes.

DISCUSSION The main finding of this study was a significant decrease in the total number of HRP-labeled cells in rats aged 26 months and older. The most specific method currently available to examine the number and distribution of motoneurons in a particular motor nucleus is the use of retro-

AGING OF RAT MOTOR NUCLEUS grade axonal transport of HRP (Burke et al., '77; McHanwell and Biscoe, '81; Nicolopoulos-Stournaras and Iles, '83; Peyronnard and Charron, '83; Swett et al., '86; Weeks and English, '85). However, the method has several obvious problems. Spurious labeling of cells in different motor nuclei has been observed, particularly when muscles rather than nerves are injected with HRP (e.g., Burke et al., '77; Richmond et al., '78; Haase and Hrycyshyn, '85). If unrecognized, this would lead to overestimation of motoneuron numbers. In our experience, spurious labeling appeared much less likely with direct injection of HRP into muscle nerves, although on occasion it was seen in these experiments despite efforts to localize the injected material. The converse problem is underestimation of cell number. It is difficult to ensure that all of the target motoneuron axons are exposed to adequate amounts of the label with intraneural injection, or that all motor axons adequately exposed to tracer actually take up and transport the material. The smaller number of labeled motoneurons in old rats might have been due to decreased uptake or axonal transport of HRP, which has in fact been observed with other tracer materials in aging rats (McMartin and O'Conner, '79). We did observe that the staining intensity in motoneurons of old rats was generally less than in younger animals, although all cells were still well labeled and the number of presumed gamma cells remained constant. The counts of myelinated axons in the muscle nerve were undertaken in order to check on this possibility. The counts of total axons in the MG muscle nerve (Table 1)in 31-month-old animals were about 80% of those in 10-13 months-old rats (means 297.0 vs. 241.6, respectively). This decrease is the same order of magnitude as the decrease in total labeled motoneurons (mean 137 vs. 107,. respectively; Table 1).Since the total axon counts include an unknown proportion of &erent fibers, which may also change in number with age, the data are not definitive but do support our conclusion that the decline in labeled motoneurons in rats 26 months of age and older is in fact due to cell death among motoneurons. By using the HRP method, bimodal distributions of cell size, as shown in Figure 3, have also been found in various motor nuclei innervating hindlimb muscles in cats (Strick et al., '76; Burke et al., '77; Weeks and English, '85), rats (Peyronnard and Charron, '83; Swett et al., '86), and mice (McHanwell and Biscoe, '81). The proportions of presumed gamma cells in the young and the middle-aged groups in the present experiment (i.e., 2 5 3 3 % of total motoneurons) agree very well with those reported for cat MG (Burke et al., '77). The most interesting and significant finding in the present experiment was the 30% decrease of presumed alpha-motoneurons in very old rats, while the mean number of presumed gamma-motoneurons remained constant in all age groups. As a result of studying the age-related changes in the caliber spectra of the dorsal and ventral root fibers and in the conduction velocity of the peripheral nerves, it has been suggested that the aging process preferentially involves the largest motoneurons (cf. Larsson, '78; see also Table 1of Tsukagoshi et al., '79). However, this interpretation is somewhat difficult because there is no precise relation between average soma diameter and axonal conduction velocity in cat alpha-motoneurons (Burke et al., '82) and because of a general apparent shrinkage of axonal diameters in the aged (Mittal and Logmani, '87). Although there are no prior reports concerning differential effects of aging on alpha- and gamma-motoneurons, the present results provide strong evidence for a preferential involvement of large alpha-motoneurons in the aging process.

429 By using intracellular injection of HRP in cat MG motoneurons, Burke and co-workers ('82) showed that the alphamotoneurons with the largest soma diameters exclusively innervated type F (fast-twitch) muscle units. Our previous results showed a relatively low proportion and a decrease in axonal conduction velocity of type F motor units in old rats, indicating preferential atrophic changes in type F motoneurons with increasing age (Kanda et al., '86). Campbell et al. ('73) have also reported a progressive fall in the number of functioning motor units in aged human subjects, with most of the surviving motor units of the slow-twitch type. Moreover, it has been reported that atrophic changes with increasing age are seen predominantly in type I1 muscle fibers (Tomonaga, '77; Larsson, '78; Caccia et al., '79), which are innervated by the generally larger type F motoneurons (Burke et al., '73). Since the bimodal transition point in average soma diameter between alpha- and gammamotoneurons did not vary with age (see Fig. 5), a preferential loss of larger alpha-motoneurons may also contribute to the decrease in the mean average soma diameter of alpha-motoneurons found in the very old rats. At minimum, the present evidence suggests that a possible link between motoneuron type and survival in old age deserves further study. A major difficulty in the study of aging animals and humans is distinguishing between specific disease conditions and the aging process per se. It has been reported that aged Fischer rats, such as used here, have various diseases even when they are raised in a specific pathogen-free condition. Two of the three very old rats in the present experiment had apparent tumors, including one probable fibroadenoma of the mammary gland and one interstitialcell tumor of the testis. We cannot exclude that the animals also may have had nephrosclerosis and/or pituitary adenoma because of the high incidence of these diseases in this strain. However, all of the rats used in this experiment were in generally good condition, as far as can be judged from their activity patterns, body weight, and food intake. We can conclude that there is a specific decrease in motoneuron number in the old and very old animals, with most of the decrease occurring among the alpha-motoneurons. A number of species of animals exhibit an approximately exponential increase in cellular mortality rate with age. One of the mathematical expressions for this relationship is the nonlinear Gompertz function. The apparent motoneuron death in the rat spinal cord was also a nonlinear function of time (Fig. 4; see also Johnson and Erner, '72; Tomlinson and Irving, '77; however, see Fig. 3 in Kawamura et al., '77). Because of the small number of age groups in present experiment, we did not try to fit our data to any of the functions proposed so far.

ACKNOWLEDGMENTS Dr. Burke was supported by the scientist exchange program between NIA and "MIG.

LITERATURE CITED Bryan, R.N., D.L. Trevino, and W.D. Willis (1972) Evidence for a common location of alpha and gamma motoneurons. Brain Res. 38:193-196. Burke, R.E. (1981)Motor units: Anatomy, physiology, and functional organization. I n V.B. Brooks (ed): Handbook of Physiology. Section 1. The Nervous System, Vol. 11. Motor Control, Part 1. Washington, DC: American Physiological Society, pp. 345-422. Burke, R.E., D.N. Levine, P. Tsairis, and F.E. Zajac (1973) Physiological types and histochemical profiles in motor units of the cat gastrocnemius. J. Physiol. (Lond.)234:723-748.

430 Burke, R.E., P.L. Strick, K. Kanda, C.C. Kim, and B. Walmsley (1977) Anatomy of medial gastrocnemius and soleus motor nuclei in cat spinal cord. J. Neurophysiol. 40:667-680. Burke, R.E., R.P. Dum, J.W. Fleshman, L.L. Glenn, A. Lev-Tov, M.J. O’Donovan, and M.J. Pinter (1982) An HRP study of the relation between cell size and motor unit type in cat ankle extensor motoneurons. J. Comp. Neural. 209:17-28. Caccia, M.R., J.B. Harris, and M.A. Johnson (1979) Morphology and physiology of skeletal muscle in aging rodents. Muscle Nerve 2202-212. Campbell, M.J., A.J. McComas, and F. Petito (1973) Physiological changes in ageing muscles. J. Neurol. Nenrosurg. Psychiatry 36:174-182. Cullheim, S. and B. Ulfhake (1979) Observations on the morphology of intracellularly stained y-motoneurons in relation to their axon conduction velocity. Neurosci. Lett. 13:47-50. Eddinger, T.J., R.L. Moss, and R.G. Cassens (1985) Fiber number and type composition in extensor digitorum longus, soleus, and diaphragm muscles with aging in fisher 344 rats. J. Histochem. Cytochem. 33r10331041. Haase, P., and A.W. Hrycyshyn (1985) Labeling of motoneurons supplying the cutaneous maximus muscle in the rat, following injection of the triceps brachii muscle with horseradish peroxidase. Neurosci Lett. 60r313-318. Johnson, H.A., and S. Erner (1972) Neuron survival in the aging mouse. Exp. Gerontol. 7:111-117. Kanda, K., K. Hashizume, E. Nomoto, and S. Asaki (1986) The effects of aging on physiological properties of fast and slow twitch motor units in the rat gastrocnemius. Neurosci. Res. 3242-246. Kawamura, Y., P. O’Brien, H. Okazaki, and P.J. Dyck (1977) Lumbar motoneurons of man I1 The number and diameter distribution of large- and intermediate-diameter cytons in “motoneuron columns” of spinal cord of man. J. Neuropathol. Exp. Neurol. 36t861-870. Larsson, L. (1978) Morphological and functional characteristics of the ageing skeletal muscle in man. A cross-sectional study. Acta Physiol. Scand. [Suppl.] 457:l-36. McHanwell, S., and T.J. Biscoe (1981)The sizes of motoneurons supplying hindlimb muscles in the mouse. Proc. R. Soc. Lond. [Biol.] 213201-216. McMartin, D.N., and J.A. OConner, Jr. (1979) Effect of age on axoplasmic transport of cholinesterase in rat sciatic nerves. Mech. Ageing Dev. 10:241-248.

K. HASHIZUME ET AL. Mesulam, M.-M. (1978) Tetramethyl benzidine for horseradish peroxidase neurohistochemistry: A non-carcinogenic blue reaction-product with superior sensitivity for visualizing neural afferents and efferents. J. Histochem. Cytochem. 26:106-117. Mittal, K.R., and F.H. Logmani (1987) Age-related reduction in 8th cervical ventral nerve root myelinated fiber diameters and numbers in man. J. Gerontol. 422-10. Nicolopoulos-Stournaras, S., and J.F. Iles (1983) Motor neuron columns in the lumbar spinal cord of the rat. J. Comp. Neurol. 217r75-85. Peyronnard, J.M., and L. Charron (1983) Motoneuronal and motor axonal innervation in the rat hindlimb: A comparative study using horseradish peroxidase. Exp. Brain Res. 50t125-132. Richmond, F.J.R., D.A. Scott, and V.C. Abrahams (1978) Distribution of motoneurones to the neck muscles, biventer cervicis, splenius and complexus in the cat. J. Comp. Neural. 181:451-464. Samorajski, T. (1974) Age differences in the morphology of posterior tibia1 nerves of mice. J. Comp. Neural. 157t439-452. Strick, P.L., R.E. Burke, K. Kanda, C.C. Kim, and B. Walmsley (1976) Differences between alpha and gamma motoneurons labeled with horseradish peroxidase by retrograde transport. Brain Res. 113,582-588. Swett, J.E., R.P. Wikholm, R.H.I. Blanks, A.L. Swett, and L.C. Conley (1986) Motoneurons of the rat sciatic nerve. Exp. Neurol. 93:227-252. Tomlinson, B.E., and D. Irving (1977) The numbers of limb motor neurons in the human lumbosacral cord throughout life. J. Neurol. Sci. 34213219. Tomonaga, M. (1977) Histochemical and ultrastructural changes in senile human skeletal muscle. J. Am. Geriatr. Soc. 25:125-131. Tsukagoshi, H., N. Yanagisawa, K. Oguchi, K. Nagashima, and T. Murakami (1979) Morphometric quantification of the cervical limb motor cells in controls and in amyotrophic lateral sclerosis. J. Neurol. Sci. 41:287297. Weeks, O.I., and A.W. English (1985) Compartmentalization of the cat lateral gastrocnemius motor nucleus. J. Comp. Neurol. 235,255-267. Westbury, D.R. (1982) A comparison of the structures of a- and y-spinal motoneurones of the cat. J. Physiol. (Land.) 325.79-91. Wright, E.A., and J.M. Spink (1959) A study of the loss of nerve cells in the central nervous system in relation to age. Gerontologia 3:277-287. Zar, J.H. (1984) Biostatistical Analysis. Prentice-Hall: Englewood Cliffs, NJ.