ALTERATIONS IN MUSCLETENSION WITHOUTSIMILARCHANGESIN ELECTRICAL ACTIVITY IN PATIENTS

WITH MYASTHENIAGRAVIS'

By STELLA Y. BOTELHO

(From the Department of Physiology and Pharmacology, Graduate School of Medicine, Unti-versity of Pennsylvania, Philadelphia, Pa.)

(Submitted for publication February 25, 1955; accepted May 25, 1955)

Certain discrepancies between the electrical andmechanical activity of skeletal muscle in patientswith myasthenia gravis have been pointed outpreviously ( 1 ). However, simultaneously re-corded action potentials and tension have been re-ported only once in patients with this disorder (2)and no quantitative analysis has been made of therelationship between electrical and mechanical ac-tivity of skeletal muscle of patients with myas-thenia gravis. The results of the present study ofsimultaneously recorded electrical and mechani-cal activity of intact muscle in six patients withmyasthenia gravis indicate that changes in tensionreflect more accurately the degree of muscle weak-ness than do changes in electrical activity andchanges in contractile ability of the muscle can oc-cur, both before and after neostigmine, withoutsimilar changes detectable in neuromuscular trans-mission.

APPARATUSAND METHODS

The method and apparatus have been described in de-tail elsewhere (3). In brief, we recorded simultaneouslyaction potentials from and isometric tension developedby the adductor pollicis muscle, which was induced tocontract by supramaximal, percutaneous stimulation ofthe ulnar nerve at the elbow. We are cognizant of thefact that movements of the thumb are complex and there-fore we might not be recording mechanical and electri-cal activity from the same group of muscles. However,an electrode on the belly of the adductor pollicis wascertainly picking up electrical activity of muscles whichare innervated by the ulnar nerve. When supramaximalstimuli were used potentials of duplicate single twitches,successive twitches with 3/sec. and successive responseswith 30/sec. did not vary more than 10 per cent in 48normal subjects. The same was true of tension. Whenthe ulnar nerve is stimulated at the elbow, the hypothenar,

1 Aided by a research grant (USPHS B-128) fromthe National Institute of Neurological Disease and Blind-ness of the National Institutes of Health, U. S. PublicHealth Service. A portion of this paper was presentedat the International Myasthenia Gravis Conference inPhiladelphia, December 8, 1954.

forearm flexor, interosseus and palmaris brevis contract.However, if ulnar deviation, wrist flexion and finger flex-ion are prevented, as they were in the present study,these muscles contribute very little to the tension de-veloped by the adductor pollicis when it is initiallystretched. In fact, when these other movements occur,the bar tends to be pushed up and the recorded tensionis less. Any opposition of the thumb also tends to de-crease tension since an oblique rather than perpendicularforce is applied to the bar. In any case with supra-maximal stimuli the tension developed by successivetwitches or duplicate tetanus in 48 normal Isubj ects didnot vary more than 10 per cent. In this study, we haveconsidered only changes greater than 10 per cent to besignificant (three times standard deviation of the meanof individual differences).

The ulnar nerve was stimulated either with singletwitches or at the rate of 3/sec. for 2 seconds (at whichrate there was no fusion of successive twitches), 2 min-utes before and at intervals varying from 1 to 10 min-utes after a tetanus produced by 30/sec. stimulation for 1second. Records were taken in six patients both whenthey were weak but not necessarily in a basal state (atleast 3 to 4 hours after their usual oral dose of neo-stigmine, Table 1) and again when they felt stronger(20 to 30 minutes after intramuscular injection of 1.5mgm. neostigmine with 0.6 mgm. atropine). In addi-tion conduction velocities along the ulnar nerve weredetermined in three additional patients by measuring thedistance between two points of stimulation (elbow andwrist) and the difference in latencies between stimulusand artefact when the nerve was stimulated at these twopoints. Before tetanus the ulnar nerve was stimulatedsuccessively at elbow and wrist with single shocks todetermine the difference in latency. The differences inlatency of the first and fifth responses were determinedwhen tetanus was obtained with 30/sec. for 1 second bystimulation at both wrist and elbow. By this techniqueone determines the conduction velocity of the fastestfibers as they conduct only from elbow to wrist.

We are fully aware of the fact that different musclegroups are variously affected in patients with myastheniagravis and that voluntary muscle contraction does notnecessarily measure the same thing that is measuredwhen one measures tension developed with supramaximalindirect stimulation. However, the patients in this studyshowed a decrease in ptosis and increased strength ofadduction of the thumb (as judged by their ability to

1403

STELLA Y. BOTELHO

TABLE I

Clinical data

Age atonset of

Age symptomsyrs. yrs.

18 183 3

37 2144 2132 2524 2119 1320 1428 22

Daily oraldose of

neostigminemgm.

000

30 q 4h.0

7.5 q 3h.15 q 4h.15 q 4h.15 q 4h.

Intervalbetween

last dose ofneostigmineand study

hrs.

24*24*24*4-524*3-48-97-85-6

Calibrationof EMGmV/mm.

.3

.4

.5

.3

.3

.3

.3

.4

.4

* 1.5 mgm. neostigmine with 0.6 mgm. atropine had been given intramuscularly for diagnostic purposes 24 hoursbefore the studies were done.

hold a sheet of paper between the thumb and first meta-carpal by adduction of the thumb), 30 minutes after in-tramuscular neostigmine. Other pertinent clinical dataare given in Table 1.

Although it is known that tension and amplitude ofaction potentials vary with age and initial muscle length,insufficient data are available from normal human ad-ductor pollicis muscle to permit comparison with pa-tients with myasthenia gravis. For this reason absolutevalues for amplitude of muscle action potentials are notgiven but they can be obtained from the calibration fac-tors in Table 1 since the amplification system was linearand the sensitivity remained the same throughout any one

study.

RESULTS

3/sec. stimulation (Table 2): Before neostig-mine, there was a decrease in the tension devel-oped by successive twitches in each patient; thetension developed by the fifth twitch was, on theaverage, 70 per cent that developed by the firsttwitch. Such a decrease in the tension of suc-

TASE 2a Simultaneoue tension and action potentials developed ty the

cxor pollicie brevis muscle with 3/see. stimulation of the usbrnerve at the elbow. In this and subsequent tables, change lese than10% we not signifleant and the rumbers 1 and 5 refer to the responsesto the lot end 5th stirulus.

Prse-rostidne Post.n.ostigaine

Patent Teonion Action Pot. Tension Actlon Pot.p. Amplitude g. Ampltude

m. em,1 5 1 5 1 5 1 5

ER. 580 4.50 19 17.5 550 470 25 26

L.P. 680 430 31 22 770 680 32 31

WLW. 760 640 32 32 775 650 33 32.536.8W. 510 2140 36 35 470 390 31 34.

sean 633 440 29.5 26.6 64.1 518 31 30.9

cessive twitches at 3/sec. does not occur in nor-

mal subjects. There were no consistent changesin amplitudes of the accompanying action poten-tials, i.e., there was no change in amplitude inE. R., M. W. (Figure 1) and M. S. W., and a

moderate decrease in L. P.; thus per cent de-crease in tension did not parallel per cent decreasein amplitude of the action potential. After neo-

stigmine there was no significant change in thefirst twitch tension in E. R., M. W. (Figure 1)and M. S. W. and a slight increase in tension de-veloped by the first twitch in L. P. Again therewas no correlation between electrical and mechani-cal activity, e.g., there was an increase in amplitudeof the first action potential unaccompanied by an

increase in tension in E. R. whereas in L. P. therewas an increase in tension developed by the firsttwitch unaccompanied by any significant changein the action potential. The lower action poten-tial in E. R. was not the result of a change in am-

plification since the calibration factor remained thesame throughout this, as well as all other studies.After neostigmine, the fifth twitch developedgreater tension than before in E. R., L. P. andM. S. W. so that, on the average, the fifth twitchtension was 85 per cent that of the first. Therewas good correlation between electrical and me-

chanical activity in E. R., L. P. and M. W.; how-ever, in M. S. W., despite the greater tension de-veloped by the fifth twitch after neostigmine, therewas no increase in amplitude of the accompanyingaction potential. There were no significantchanges in either the delay (interval between be-ginning of stimulus artifact and beginning of the

1404

Patient

D. D.B. G.Na-1Na-2L. P.E. R.M. W.M. S. W.M. A.

Sex

FFFFMFFFF

SIMULTANEOUSTENSION AND POTENTIALS IN MYASTHENICS

action potential) or latent period (interval betweenbeginning of action potential and beginning ofthe rise in tension) of successive twitches at 3/sec.either before or after neostigmine.

30/sec. stimulation (Table 3): Before neostig-mine. The electromyographic picture of a decre-ment of successive action potentials described byHarvey and Masland in 1941 (1) occurred inthree patients (E. R., Figure 2, L. P. andM. S. W.) but there was no significant change inthe amplitude of successive potentials in the otherpatient (M. W.). The conduction velocities ofthe ulnar nerve were normal both before and dur-

TAKEZ 3: 30/sec. atimlation before and 20-30 tmtes afterintramuscular neostimie'.

Pre-Neostipine Post-Nsostpsine

Patient Tenoion Action Pot. Tension Action Pot.P. amplitude ga amplitude

1 5 1 5E.R. 3000 22 15 3150 26 30

L.P. 3500 30 16 3200 29 25

e.ws 2950 33 32 3950 29 26

Msews, 3460 35 20 1200 43 37

Mean 3228 30 20.9 3625 31.8 29.5

I

L_LFIG. 1. PATIENT M. W. FIRST (ON READER'S LEFT)

AND FIFTH (ON READER'S RIGHT) MECHANICAL (A AND

C) AND ELECTRICAL (B AND D) RESPONSESOF THE AD-DUCTORPOLLICIS TO 3/SEC. STIMULATION OF THE ULNARNERVE8 TO 9 HOURSAFTER 15 MGM. ORAL NEOSTIGMINE(A ANDB) AND20 MIN. AFTER 1.5 MGM. INTRAMUSCULARNEOSTIGMINE (C AND D)

In this and all subsequent figures, lines A and Crecord tension, maximum developed tension being meas-

ured as indicated by the upper arrow; lines B and Drecord electrical activity, the amplitude of the negativephase being measured as indicated by the lower arrow;

the small upward deflection preceding the negative phaseof the action potential when present is the stimulus arti-fact; and each of the widest time markers indicates 0.03sec. In this figure calibration of tension is in termsof gm./extremes of the bracket and of electrical activityin terms of mV/extremes of the bracket.

ing the tetanus at a time when the action potentialshad decreased in three other patients (Table 4).After neostigmine, there was an increase in tetanustension in two patients (M. W. and M. S. W.)and no change in tetanus tension in the other twopatients (E. R., Figure 2, and L. P.). Thus onlyin M. S. W. was there a correlation between themechanical and electrical activity since only herewas the increased tension accompanied by an in-

crease in amplitude of the first potential and a de-crease in decrement of successive potentials. Inthe other three patients there was no correlationbetween electrical and mechanical activity.

Post-tetanic potentiation of twitch tension: Be-fore neostigmine (Table 5): Single twitch ten-sion, as judged by the first twitch with 3/sec.stimulation, was greater after tetanus than thecorresponding twitch before tetanus. As in nor-

mal subjects (3), this potentiation of twitch ten-sion was unaccompanied by a change in the actionpotential. In two other myasthenia gravis patients(N. A.-2 and M. A., Table 6) we found that al-

TA3Z 4s Conduction velocity of ulnar nerve (CM/sec.). and

30 are ihe responses to the lbt and 30th stimuli: Tetanusproduced by 25/sec. stimulation for 1.2 sec.

Patient Before Tetanus R!ring Tetanus

C.o. C.V. Act. Pot.aspl. (u)1 30

N.A.-1 55.2 55.0 16 10

BeG. 75.0 714.2 29 25

D.D. 50.7 51.1 19 9

a a

IN r 14 nimmI

1405

I

STELLA Y. BOTELHO

TABLE 6: Single twitches before and after tetanus.

I I -I -IiAI i Ii I .i i

Pre-Tetanus Post-Tetaris

1 minute 2 minutesPatient Tension Act. Tension Act. Tension Act.

gm. Pot. gmn. Pot. gin. Pbt.(-) (-) (i)

Ne.A-2 250 26 310 25 250 25.5

M.A. 860 35 990 36 84sO 35

twitches after tetanus at this stimulation rate ismaintained at the potentiated single twitch level(4). This decrement in successive twitch ten-sions was greater after tetanus in patients L. P.and M. S. W. and about the same in patients E. R.

3.0f

FIG. 2. PATIENT E. R. SIMULTANEOUSMECHANICALAND ELECTRICAI RESPONSESOF THE ADDUCTORPOLLICISTO 30/SEC. STIMULATION OF THE ULNAR NERVE 3 TO 4HOURSAFTER 7.5 MGM. ORAL. NEOSTIGMINE (UPPER TwoLINES) AND 20 TO 30 MIN. AFTER 1.5 MGM. INTRAIMUS-CULAR NEOSTIGMINE (LOWER 2 LINES)

In this figure, calibration of tension is in terms ofgm./extremes of the bracket and of electrical activity interms of mV/extremes of the bracket.

though the twitch tension was potentiated at 1 min.it was not potentiated 2 min. after the tetanus;this is unlike normal human subjects, in whompost-tetanic potentiation of twitch tension lasts at least10 min. (3). After tetanus, there was a progres-sive decrement in tension of successive twitcheswith 3/sec. stimulation in all four myastheniagravis patients; again this is unlike normal humansubjects, in whom the tension of successive

TAMLE 5: 3/sec. stimulation after tetanus before and 20-30 minutesafter intramuscular neostigni_.

. .._4

PRE-lTETANUJS

IlI

P ST-T vTANU -r

Pre-neostigmine Post-neostignin.

Patients Tension Action Pot. Tension Action Pot.gm. Amplitude gm. Amplitude

Mo.1 5 1 5 1 5 1 5

E.R. 800 675 17.5 15 770 695 22.5 16.5

L.P. 800 230 33 13 1100 800 31 26

M.W. 1025 910 32.5 32 1150 975 32 32

V.S.C. 1200 190 35 27 IsOO 500 34 33

lan 956 509 29.5 21.8 1105 713 29.9 26.9

FIG. 3. PATIENT M. S. W. FIRST (ON READER'S LEFT)AND FIFTH (ON READER'S RIGHT) MECHANICAL ANDELECTRICAL RESPONSESOF ADDUCTORPOLLICIS MUSCLETO3/SEC. STIMULATION OF THE ULNAR NERVE BEFOREANDAFTER TETANUS OF 30/SEC. FOR 1 SEC. 7 TO 8 HOURSAFTER 15 MGM. ORAL NEOSTIGMINE

In this figure the number on the tension line indicatesthe maximum tension developed by the muscle in grams.

1406

i I !-I Li i:

SIMULTANEOUSTENSION AND POTENTIALS IN MYASTHENICS

and M. W. when compared to the pre-tetanusdecrement. There were no consistent changes inthe action potential after tetanus, e.g., the actionpotential decrement increased after tetanus inE. R. whereas the tension decrement did not andthe action potentials did not decline in M. W. butthe tension did. In L. P. and M. S. W. (Figure3) there was an increase in decrement of actionpotentials as well as in decrement of tension aftertetanus. After neostigmine, there was 1) an in-crease in the potentiation of the initial twitch ten-sion in three patients without an alteration in theaction potential and 2) either a decrease or in-crease in decrement of successive twitch tensionswith 3/sec. stimulation with no consistent changein the action potentials.

DISCUSSION

Weinterpret the results of the present study ofthe simultaneous electrical and mechanical ac-tivity of the adductor pollicis muscle during ac-tivity induced by supramaximal stimulation of thetulnar nerve in patients with myasthenia gravis asfollows: Tension changes, especially at slow rates(3/sec.) which are much more comfortable thantetanus rates (faster than 25/sec.) are more ac-curate guides to the degree of muscle weaknessthan the electrical changes. The data lend fur-ther support to previous observations of the diffi-culties encountered when using the action potentialpattern alone to diagnose myasthenia gravis (5).

Studies of either mechanical or electrical activ-ity but not of both simultaneously have led to ap-parently conflicting conclusions. Thus Lindsley(6) reported a decrease in single twitch tensionwhich increased with neostigmine but Pritchard(7) did not find that the single twitch tension ofone myasthenia gravis patient was less than nor-mal and Harvey and Masland (1) were unableto detect any decrease in the action potential ofsingle induced twitches in patients with myastheniagravis. All authors (7-9) agree that slow (about3/sec.) repetitive contraction produces a succes-sive decrease in tension in patients with myas-thenia gravis. However, it has been demonstratedthat some patients with this disorder do not have adecrement in amplitude of successive twitch po-tentials with induced 3/sec. stimulation (5).Furthermore, various workers (6, 7, 9) have been

unable to demonstrate any change in tetanus ten-sion until stimulation rates exceeding 80/sec. wereused. But some workers have found that pa-tients with myasthenia gravis show a progressivedecrease in amplitude of successive potentials dur-ing tetanus with stimulation rates less than 40/sec. (1) and other workers have shown that al-though some patients with myasthenia gravis doshow this electromyographic picture, others do not(5). These discrepancies cannot be explainedentirely by the suggestion that various technicaldifferences were present such as, 1) submaximalstimuli were used in some studies, 2) the intervalbetween single stimuli and duration of tetanusvaried from study to study, and 3) the severityof the disease varied from study to study. Webelieve that other explanations must be soughtand suggest the following: 1) Myasthenia gravis,at least in some muscles of some patients, is as-sociated with notable muscle weakness which isnot accompanied by any discernible degree ofneuromuscular transmission block. The findingsof the present study which support this are: (a)successive twitch tensions at 3/sec. stimulationdecreased significantly without any significantchange in the action potentials (E. R., M. W.,and M. S. W.), and (b) after an induced tetanus,successive twitch tensions at 3/sec. are initiallypotentiated but not sustained at the potentiatedlevel; this abnormal finding is not accompaniedby a consistent alteration in the action potential(E. R. and M. W.); 2) It is possible that post-tetanic potentiation of twitch tension, which wehave shown to occur in the present study, sustainsthe tension during the tetanus despite a decreasein the action potentials. Brown and Burns (10)have suggested that when mammalian muscle isfatigued, there is alternation of motor unit activityduring the tetanus; this explains the successive de-cline of the action potentials. The inactive fibersare now in the post-tetanus state even though thetetanus is continuing so that when they again con-tract they can contribute a greater increment oftension than they could have, had they been ac-tive throughout the tetanus. Since there is al-ternation of activity and inactivity of the variousfibers, the tension during tetanus is sustained atthe initial level because of this post-tetanic poten-tiation. Post-tetanic potentiation of twitch ten-sion unassociated with facilitation of the action

1407

STELLA Y. BOTELHO

potential has been shown to occur in normal hu-man muscle and the factors which may be re-sponsible for this phenomenon are discussed else-where (3).

Since the changes in muscle tension which oc-curred after neostigmine were not necessarily theresult of changes in neuromuscular block as judgedby the associated action potentials, we believe thatneostigmine can have a direct effect upon the con-tractile ability of muscle apart from or in additionto its effect upon neuromuscular transmission.Although their data do not necessarily indicatethat neostigmine has an effect on the contractilemechanism, Riker and Wescoe (11) have at leastshown that neostigmine can stimulate denervatedmammalian skeletal muscle. In the present studywe have not excluded the possibility that the con-traction was altered after neostigmine as a resultof changes in the circulation to the muscle.

Although it has been shown that neostigmineadministered intraarterially produces repetitivefiring of normal human muscle in response to asingle suprainaximal stimulus (12), we were un-able to demonstrate this in the present study,probably because we were using surface electrodes.It is therefore possible that repetitive firing of indi-vidual muscle fibers was present and could ac-count for an increase in tension without alteringthe summed potential. If this were so, the de-crease in successive twitch tensions without a de-crease in action potential with 3/sec. stimulationmight result from the fact that the small amountsof circulating neostigmine present from 3 to 9hours after oral neostigmine caused repetitive fir-ing when the muscle was stimulated through itsnerve; this repetitive response would decreasewith each such successive stimulus and thereforethe tension of each successive twitch would de-crease. If this were so, we would expect that in-creasing the amount of neostigmine by intra-muscular injection would cause an increase inrepetitive discharge of the muscle. We wouldthen expect an increase in initial twitch tension aswell as an increase in the tension of successivetwitches but this occurred in only one of the fourpatients.

It has been shown that evoked action potentialsrecorded from various parts of a muscle have dif-ferent forms and this might account for the dif-ferences in electrical and mechanical activity which

were found in this study. It is quite true that dif-ferent shaped action potentials are recorded fromdifferent parts of large muscles when surface elec-trodes are used (13) and from different parts ofsmall muscles when needle electrodes are used(14). However, we do not believe this to be animportant factor in the present study since thesurface electrodes we used practically covered theentire belly of the adductor pollicis muscle.

It is possible that the amplitude of the actionpotentials may not truly reflect changes in thenumber of active fibers since more synchronousfiring of fewer fibers could produce a potential ofgreater amplitude; however, in this case thereshould then be an alteration in the area or dura-tion of the potential (15). Since there were nochanges in potential duration or area, we believethat variations in potential amplitude did reflectvariations in the number of active fibers and thusalterations in the degree of neuromuscular block.The findings of increased, decreased, or unalteredmuscle tension in the absence of similar changes inneuromuscular block indicated that neostigminecan have a direct effect upon skeletal muscle ofpatients with myasthenia gravis. Webelieve thatthe efficacy of neostigmine must, at least in part,be the result of such a direct action.

We cannot overlook the possibility that myas-thenia gravis as diagnosed by clinical signs andsymptoms and frequently confirmed by the admin-istration of neostigmine or curare (which has alsobeen shown to have a direct effect both on mam-malian [16] and human [4] skeletal muscle) mayinclude not only patients who are weak because ofa neuromuscular transmission defect but also pa-tients who are weak because of the inability ofthe muscle to contract. It is apparent that themuscle as well as neuromuscular junction must bestudied in order to elucidate the pathophysiologyof myasthenia gravis and to obtain more effectivetherapeutic agents.

SUMMARY

The results of the present study show thatchanges in the degree of mechanical activity canoccur without similar electrical changes in the ad-ductor pollicis muscle of patients with myastheniagravis upon supra-maximal stimulation of theulnar nerve with single twitches, 3/sec. stimulation

1408

SIMULTANEOUSTENSION AND POTENTIALS IN MYASTHENICS

and 30/sec. stimulation both before and after in-tramuscular neostigmine. The findings suggest1) that changes in tension reflect more accuratelythe degree of muscle weakness than changes inelectrical activity, 2) in some muscles of somemyasthenia gravis patients the muscle weaknessresults from a defect in contractile ability apartfrom or in addition to a defect in neuromusculartransmission, 3) neostigmine is useful in some pa-tients with myasthenia gravis because of its abilityto alter the contractile processes of the muscleapart from or in addition to decreasing neuro-muscular block and 4) the muscle must not be ne-glected in studies designed (a) to elucidate thepathophysiology of myasthenia gravis and (b) toobtain better therapeutic agents in this disorder.

ACKNOWLEDGMENT

The author wishes to thank Dr. George Gammonforthe opportunity to study the majority of these patientsand Dr. Julius H. Comroe, Jr. for his valuable critiqueof the manuscript.

REFERENCES

1. Harvey, A. M., and Masland, R. L., The electromyo-gram in myasthenia gravis. Bull. Johns HopkinsHosp., 1941, 69, 1.

2. Grob, D., and Harvey, A. M., Abnormalities in neuro-muscular transmission with special reference tomyasthenia gravis. Am. J. Med., 1953, 15, 695.

3. Botelho, S. Y., and Cander, L., Post-tetanic potentia-tion before and during ischemia in intact humanskeletal muscle. J. Applied Physiol., 1953, 6, 221.

4. Botelho, S. Y., Simultaneous electrical and mechani-cal responses to indirect stimulation in curarizedsubjects. In preparation.

5. Botelho, S. Y., Deaterly, C. F., Austin, S., and Com-roe, J. H., Jr., Evaluation of the electromyogram

of patients with myasthenia gravis. Arch. Neurol.& Psychiat., 1952, 67, 441.

6. Lindsley, D. B., Myographic and electromyographicstudies of myasthenia gravis. Brain, 1935, 58, 470.

7. Pritchard, E. A. B., The occurrence of Wedensky in-hibition in myasthenia gravis. Proceedings ofthe Physiological Society, February 18, 1933. J.Physiol., 1933, 78, 3P.

8. Jolly, F., Ueber myasthenia gravis pseudoparalytica.Part I., Berl. Klin. Wchnschr., 1895, 32, 1.

9. Odom, G., Russell, S. K., and McEachern, D., Stud-ies of neuromuscular disorders: The myogram,blood cholinesterase and effect of prostigmine inmyasthenia gravis and progressive muscle dys-trophy. Brain, 1943, 66, 1.

10. Brown, G. L., and Burns, B. D., Fatigue and neuro-muscular block in mammalian skeletal muscle.Proc. Roy. Soc., London, s.B., 1949, 136, 182.

11. Riker, W. F., Jr., and Wescoe, W. C., The directaction of prostigmine on skeletal muscle; its re-lationship to the choline esters. J. Pharmacol. &Exper. Therap., 1946, 88, 58.

12. Harvey, A. M., Lilienthal, J. L., Jr., and Talbot, S.A., On the effects of the intra-arterial injection ofacetylcholine and prostigmine in normal man.Bull. Johns Hopkins Hosp., 1941, 69, 529.

13. Hodes, R., Larrabee, M. G., and German, W., Thehuman electromyogram in response to nerve stim-ulation and the conduction velocity of motor axons.Studies on normal and on injured peripheralnerves. Arch. Neurol. & Psychiat., 1948, 60, 340.

14. Jarcho, L. W., Eyzaguirre, C., Berman, B., andLilienthal, J. L., Jr., Spread of excitation inskeletal muscle: Some factors contributing to theform of the electromyogram. Am. J. Physiol.,1952, 168, 446.

15. Harvey, A. M., and Masland, R. L., A method forthe study of neuromuscular transmission in hu-man subjects. Bull. Johns Hopkins Hosp., 1941,68, 81.

16. Bean, J. W., and Elwell, L. H., Influence of d-tubo-curarine on contractile response of mammalianskeletal muscle in situ. Am. J. Physiol., 1951,165, 716.

1409