Reprinted from THE JOURNAL OF CLINICAL INVESTIGATION, Vol. 37, No. 6, pp, 838-845, June, 1958 Printed in U. S. A. THE EFFECTS OF A VARIETY OF HEMODYNAMIC CHANGES ON THE RAPID AND SLOW COMPONENTS OF THE CIRCULATION IN THE HUMAN FOREARM1 BY EDWARD D. FREIS AND HAROLD W. SCHNAPER (From the Cardiovascular Research Laboratory, Georgetown University School of Medicine, and the Veterans Administration Hospital, Washington, D. C.) (Submitted for publication July 23, When a tracer substance which does not diffuse through the capillary walls is introduced into the brachial artery and the effluent blood is sampled continuously thereafter, a time-concentration curve of the labeled substance may be plotted which de- scribes the velocity-flow pattern of the blood in the forearm ( 1) . It was found that the descending limb of the time-concentration curve usually was biphasic. Evidence was presented which indi- cated that these biphasic curves were not due to differences in path lengths or to velocity flow through different tissues such as skin and muscle. It was suggested that this biphasic pattern was due to two basically different rates of flow in the highly anastomotic small vessel circulation (l-3). Since pressures are very nearly equal at either end of an anastomotic intercapillary or intervenular chan- nel, flow may be sluggish in that channel. Direct observation of the circulation in the hamster cheek pouch disclosed that varying rates and direction of Aow occur in the anastomotic channels-at times rapid, at times slow, at times surging, stop- ping or reversing. It seemed of interest to determine what effect changes in both the local and general circulation would have upon the characteristics of this biphasic velocity-flow pattern in the forearm. METHODS The dye T-1824 was used as the tracer substance. The source of subjects, experimental technique, methods of analysis and calculation of results were identical to those previously described (1). The ratio of blood flow in the rapid component (fj) to that in the slow (f,) was esti- mated from the mean circulation times of the total curve (T), the rapid component (t,), and the slow component (tl) as described in a previous paper (1). The param- eters f,/f, and t,/t, were used to estimate the ratio of the 1 Supported in part by United States Public Health Service Grants H-720 and H-1903 (National Heart Institute). 1957; accepted February 20, 1958) blood volumes (v,/v,) in the rapid and slow portions. The ratio of flows rather than absolute blood flows were estimated because the distribution of dye in the various branches of the brachial artery may not be uniform, and hence, the peripheral vasculature of the forearm cannot be regarded as a uniform mixing chamber (1). Hyperemia restricted to the forearm under study was produced by the following techniques: 1) by the local intra-arterial injection of 5 mg. tolazoline (Priscoline@). After flushing of the skin had become plainly evident (usually after one to three minutes) the dye T-1824 was injected into the brachial artery; 2) by local exercise in which the subject squeezed a hand ergometer repeatedly at one-second intervals as forcibly as he was able until he could carry on no longer because of fatigue and aching of the forearm muscles. This varied from 65 to 124 squeezes in the 3 subjects. At this point the dye was in- jected through the previously placed intra-arterial needle without delay; 3) by reactive hyperemia in which the cuff of a blood pressure apparatus was wrapped on the upper arm and inflated to pressures 100 mm. Hg above systolic for a period of 10 minutes. Dye was injected im- mediately following release of the cuff. In all of the hyperemia experiments it was necessary to double the dose of T-1824 in the experimental as com- pared to the control period. The reason for this was that the greatly increased volume flow reduced the dye con- centrations. In all of the other procedures the dosage of the dye was the same in both the experimental and control periods. Local injection of epinephrine and norepinephrine was carried out as follows : Epinephrine was diluted to a con- centration of 0.4 pg. per ml. and levarterenol (norepineph- rine) to 2.0 pg. per ml. in isotonic saline. These dilute solutions were injected intra-arterially through a three- way stopcock at a rate of 2 ml. per minute. When pal- lor of the skin occurred (usually after one to two min- utes) an assistant attached the syringe containing the dye to the free arm of the three-way stopcock and in- jected it without delay. Following this the stopcock was immediately turned and the infusion of the pressor amine continued at the same even rate until the end of the col- lection period. The pressor amine also was added to the dye in a concentration similar to that present in the infu- sion. The blood pressure was recorded in the opposite arm. In no instance was a significant change in systemic arterial pressure observed. When epinephrine and norepinephrine were given sys- 838 839 EDWARD D. FREIS AND HAROLD W. SCHNAPER temically the infusion contained either 1 mg. of epinephrine or 2 mg. of norepinephrine per liter of isotonic saline. The infusion was connected via a three-way stopcock to a needle in a vein of the arm opposite to the forearm un- der study. A flask containing 1 liter of isotonic saline without added pressor amine was connected to the free limb of the three-way stopcock and infused slowly dur- ing the control period. Following the control experi- ment the stopcock was turned and the infusion of pres- sor amine was adjusted until a significant change in blood pressure and usually of the heart rate was obtained. The dye was then injected into the brachial artery of the forearm under study and the infusion rate of the pressor drug maintained at a level which produced a significant blood pressure change throughout the sample collection period. In the case of epinephrine the elective infusion rates averaged 15 * 5 pg. per minute; for norepinephrine they averaged 36 * 8 pg. per minute. In the venous congestion experiments a narrow blood pressure cuff was wrapped proximally on the arm under study and covered with gauze bandage to prevent baSloon- ing. The cuff was inflated to pressures of 40 to 60 mm. Hg for a period of five minutes before as well as through- out the injection and collection periods. RESULTS Control determinations were carried out im- mediately prior to the experimental period in 25 of the 47 subjects. Controls, being technically laborious, were not obtained in the remaining cases. Therefore, in analyzing the data, some with controls, some without, the coefficients of variation were used to obtain a combined estimate of the standard deviation. This seemed valid since the ratio of the standard deviation to the mean was nearly constant for the various groups (Table I). Using this combined estimate the t test for signifi- cance then was applied in the usual way. In the group of subjects given norepinephrine sufficient controls were carried out among these subjects to permit an intragroup comparison. The latter method of analysis was carried out in this case since the combined estimate provided suggestive but not conclusive indications of a significant difference. This was produced in several ways as described above : by local exercise (three cases), intra-ar- terial tolazoline (eight cases), and reactive hypere- mia (four cases). Overall results indicated a highly significant shortening of T and t, (p > 0.01) with an increase in the ratios f,/f, (p > 0.01) and v,/v, (p > 0.05), indicating a marked relative shift of blood flow and volume to the rapid component (Tables I and II). These changes were most marked in the cases given in- tra-arterial tolazoline (Table I and Figure 1). Control determinations were carried out in three cases with local exercise, in two of the sub- jects given intra-arterial tolazoline, and in one case subjected to reactive hyperemia. In each of these there were decreases in mean circulation times and increases in f,/f2 and v,/v, as compared to the controls (Table I and Figure 1) . T-w24 - OPTICAL DENSITY OPTICAL DENSITY s.* `. - CONTROL - CONTROL It o --* AFTER TOLAZOLINE o --* AFTER TOLAZOLINE 1 a MINUTES MINUTES a r I T-1824 - OPTICAL DENSITY ~CONTROL *--AFTER LOCAL EXERCISE MINUTES b FIG. 1. TYPICAL EFFECTS OF HYPEREMIA ON THE BI- PHASIC TIME-CONCENTRATION CURVES IN THE HUMAN FOREARM Figure la (above) illustrates the effects of intra- arterial tolazoline. Figure lb (below) shows the effects of local exercise. 840 VARIATIONS IN BLOOD FLOW COMPONENTS OF FOREARM TABLE I Mean circulation times and ratios of blood j&nu and vascular volume in the rapid and slow components under various experimental conditions Control Experimental Subject Age Diagnosis T t1 t_ fllfi VI/W T t1 tn h/f: VllV.2 Hyperem; Local exercise Peptic ulcer Urticaria Urethritis Intra-arterial tolazoline Ft. :: Urinary tract infect. Lung abscess 26 fc.). 43 Psychoneurosis Acute infect. arthritis . . 38 A. M. 38 E ii: Pulmonary histo- plasmosis Peptic ulcer Boeck's sarcoid Peptic ulcer Reactive hyperemia H. D. ii Multiple sclerosis Infectious arthritis ii Pneumonia Mean S. D. Venous congestion :: Ei. ;t Psychoneurosis W. H. 35 Go11 t Normal JR.. if Rheum. spond. . . Gastroenteritis Mean S. D. Norepinephrine (intra-arterial) kti 2": Cirrhosis R. N: 46 Rheum. arthritis Diab. mellitus L"-2 48 No disease c: hi. ii Peptic ulcer Convalescent rheum. fever Es". 33 Peptic ulcer w. Ii. :: Penicillin reaction Rheum. arthritis Mean S. D. Epinephrine (intra-arterial) Ei Peptic ulcer G: S: 2 2 Psychoneurosis Irritable colon tt: i!' 23 Peptic ulcer Convalescent F. H. meningitis E. D. :t Peptic ulcer Fatty liver Mean S. D. 102 5": 49 17 19 247 9": z 1:o 92 33 179 1.5 82 15 97 0.2 49 24 79 1.2 38 113 8.6 89 123 0.6 ;: 62 135 112 127 E 40 75 30 31 43 21 19 :4" 15 20 168 86 75 12.5 A:; 1.5 :.: 1:4 0.9 8:; 0.2 0.3 0.1 0.3 2.2 0.2 it; 0.2 8.; 0:3 0.1 64 a: 51 1: :x 1.5 46 150.4 75 127.1 46 16.2 16 :"z 14 60 33.3 15 42 5.2 12 65 4.3 E 16 27 13 17 9 25 f14 18 f12 ii 8.9 1.4 49 3.9 68 33.8 f60 f58.4 f19.4 149 307 210 209 215 1;: 1:; 48 223 6.52 E X:i 251 406 E 8:: 320 0:6 0.1 218 82 370 1.2 0.2 *57 540 ~172 ~0.9 f0.2 46 129 0.6 :t 203 132 hi ::i 2 z 1:7 8:: 48 190 3:: 8:: 57 100 127 95 126 :i 118 o.8 1.7 E 24 `E 1.1 0:2 34 1.50 1.3 0.3 &lo f4.5 SO.6 ho.14 18 139 1.7 0.2 :: 97 2.6 106 1.5 8:: 13 78 2.1 0.3 38 104 1.9 16 126 1.5 X:i 2.8 i:: 51.1 39.9 5.3 7.9 ii:: 2.7 8:`: 10.7 21 108 1.8 f9 f22 f0.5 *E 841 EDWARD D. FREIS AND HAROLD W. SCHNAPER TABLE r-Continued Control Experimental Subject Age Diagnosis T t1 tl fI/fP wlvr T t1 ts fI/f_ VI/V, VS. Epinephrine (systemic) s. w. 21 Convalescent pneumonia hf. IL 28 Psychoneurosis D. C. Peptic ulcer H. s. z': Pyeloneph. right Mean s. D. Norepinephrine (systemic) W. J. No disease 0. Y. :3" Convalescent spontaneous pneumothorax J. P. 30 Convalescent E pneumonia 2 Peptic ulcer G. ti. Boeck's sarcoid Traumatic wrist G. R. 26 drop No disease Mean S. D. 1:: 97 75 f38 Mean (all controls) 65.7 S. E. (all controls) ~22.7 47 5': 109 44 seconds seconds 18 117 27 180 :: 1:: 2.4 0.4 0.7 0.2 0.6 0.1 0.8 0.2 34 12 81 2.2 0.3 103 40 144 0.7 1:: :i 2:: 1.1 1.6 2 0:3 80 31 139 1.4 0.25 f43 f21 f80 ho.6 f0.06 :z 164 67 24 77 z 123 174 18 132 26 123 f8 f44 25 125.2 h10.8 f43.8 3.6 3.2 2.4 0.6 1.7 0.5 A:: ii6 0.4 0.1 1.6 0.8 zlz1.3 f1.2 1.6 0.5 lt1.8 f0.7 33 29 87 11.8 3.9 75 30 162 1;; 2 103 188 E 0.6 f:: 0:09 70 20 128 1.2 0.2 58 27 12.5 6.7 1.8 f24 zt7 f12 h5.5 Al.8 Venous congestion the three cases in which control determinations im- In the five subjects studied there were marked mediately preceded the experimental period the prolongations of mean circulation times (Tables only change consistent in all three was a slight to 1 and II, Figure 2). These were significant for moderate increase in T (7, 15 and 109 per cent, T, t, and t, (p > 0.05,0.05 and 0.01, respectively). respectively). Because of the nearly proportionate increase in-all of these parameters the changes in f,/f, and v,/v, were not significant. It should be noted that the number of cases studied was not sufficient . to detect small changes in these ratios. It would appear, however, that the predominant effects of congestion were to greatly retard blood flow and velocity in both components without much change in the relative flows and volumes. Local rpinephrine and norepinephrine The changes observed after the intra-arterial in- fusion of epinephrine and norepinephrine were not significant. Following norepinephrine the average values in nine cases disclosed slight and insignifi- cant increases in mean circulation times and re- ductions in f,/f, and v,/v2 (Tables I and II). In T-1624 - OPTICAL DENSITY 5. -CONTROL -* *--a AFTER VENOUS CONGESTION i I. 1 `0, i 0. I I i 0.1 OS .Ol \-A----&- --- \ \ \ \ \ \ \ \ \ 2 4 6 6 IO c2 MINUTES FIG. 2. EFFECTS OF VENOUS CONGESTION See text for details. VARIATIONS IN BLOOD FLOW COMPONENTS OF FOREARM 842 TABLE II Means, standard deviation of means, and statistical analysis of the T t1 Difference Difference between means* between means No. of Differ- SD. Differ- S.D. Procedure cases Meall S.D. ence of diff. t Mean S.D. ence of diff. t Control 25 66 23 2.5 11 Hyperemia 1.5 25 14 41 6 6.5 18 12 7 3 2.2 Congestion 5 218 57 1.52 37 4.1 82 40 57 19 3.0 Epinephrine. local 7 52 10 14 11 1.3 21 9 4 5 0.8 Norepinephrine. local 9 95 26 29 17 1.7 34 10 9 6 1.5 Epinephrine, systemic 4 80 43 14 12 1.2 31 21 6 8 0.7 Norepinephrine. systemic 7 58 14 8 10 0.8 27 7 2 5 0.4 * Coefficients of variation used to obtain a combined estimate of the standard deviation. See text. In the seven cases receiving intra-arterial epi- nephrine no significant changes were observed in the average values of T, t,, t,, f,/f, and v,/v, as compared to the combined series of 25 controls. In four of these subjects control determinations preceded the experimental period. Results in these cases failed to reveal consistent changes in the mean circulation times. In three of the four cases moderate increases in f,/f, and v,/v, oc- curred. This must be regarded as being of ques- tionable significance since the average values for the entire group of seven failed to confirm such a trend. The changes after norepinephrine and epineph- rine locally, therefore, were not striking. After norepinephrine there seemed to be a moderate de- lay in mean circulation time but this was not sta- .-_bFTER NOREPINERI FIG. 3. EFFECTS OF SYSTEMICALLY ADMINISTERED NOR- EPINEPHRINE IN THREE SUBJECTS Note increased dye concentrations and broadening of the curve around the peak. tistically significant. After epinephrine no con- sistent changes were observed. Systevnic norepinephrine and epinephrine Epinephrine was infused at a rate of 10 to 22 pg. per minute until there was a significant eleiration of systolic pressure. This was always accompanied by a detectable fall in diastolic blood pressure. The mean elevation of systolic pressure was 34 + 8 mm. Hg, and the fall of diastolic pressure aver- aged 11 + 7 mm. Hg. The average increase in mean [ (systolic + diastolic)/21 pressure was only 11 f 5 mm. Hg. The changes in mean cicula- tion times, relative flows, and relative volume in the two compartments failed to indicate any con- sistent pattern (Tables I and II). Norepinephrine was infused at a rate to pro- duce not only significant systolic but also diastolic hypertension and bradycardia. The average ele- vation of systolic pressure was 52 * 18 mm. Hg and of diastolic pressure 30 +- 5 mm. Hg, the in- crease in "mean" arterial pressure averagin: 42 -C 11 mm. Hg. Thus, the elevation of "mean" ar- terial pressure was approximately four times as great after norepinephrine as after epinephrine. The mean decrease in heart rate was 17 +7 beats per minute. In contrast to the insignificant effects of epi- nephrine there was a significant decrease in T and increase in f,/f, when the six cases witt controls were compared on an intragroup basis (Table III). The ratio f,/f, increased in all of the cases who had their own controls (Table I), the mean increase being 375 per cent over the wntrol and the range 94 to 755 per cent. Although t, and t, 843 EDWARD D. FREIS AND HAROLD W. SCHNAPER TABLE II-Continued differences between means under the various experimental conditions tn fllfl VI/V1 DifferlXKe Difference Difference between means between means between means Differ- SD. Differ- S.D. Differ- S.D. Mean SD. ence of diff. t Mean SD. ence of diff. t Mean S.D. ence of diff. t 125 44 1.6 1.7 0.5 0.7 68 60 57 15 3.7 34 58 32 9 3.6 10.7 19.5 10 4. 2.8 370 172 245 8 31.0 1.2 0.9 0.4 0.6 0.7 0.2 0.2 0.3 0.16 1.8 108 22 17 22 0.8 1.8 0.5 0.2 0.75 0.2 0.4 0.4 1.0 0.21 0.4 150 45 25 25 1.0 1.3 0.6 0.3 0.4 0.8 0.3 0.14 0.2 0.17 1.1 139 80 14 28 0.5 1.4 0.6 0.2 0.7 0.3 0.25 0.06 0.25 0.2 1.3 125 12 0.2 25 0.004 6.7 5.5 5 2.4 2.2 1.8 1.8 1.3 0.8 1.6 were not significantly changed from the controls they both increased in relation to the decrease in T. Inspection of the curves disclosed a uniform change characterized by a broadening of the curve with increased dye concentrations around the peak. The junction or transition zone between t, and t, was shifted to the right (Figure 3). Thus, nor- epinephrine given systemically in dosages suffi- cient to produce a significant elevation of "mean" arterial pressure, including diastolic pressure, seemed to produce a change in the distribution of blood flow so that a larger proportion was carried in tie rapid component. This change apparently did not occur when norepinephrine was given lo- cally nor when epinephrine was given locally or systemically in dosages to produce a systolic but not diastolic hypertension. DISCUSSION These studies demonstrate that the biphasic pat- tern of blood flow in the forearm is subject to dy- namic changes. When total blood flow was aug- mented by any one of several means (local exer- cise, intra-arterial tolazoline, reactive hyperemia) flow in the rapid component increased relative to that in the slow component. The increase in ilow of the rapid component was accompanied by an increase in the relative vascular volume of the rapid omponent. These data represent an ex- tension of the previous observations on a group of subjects studied under resting conditions. In that series, f,/f, and v,/v, were found to be in- versely related to T (1) . Whereas the ratios f,/f, and v,/v, could be in- creased by various maneuvers, no means was found to significmtly reduce them. Even venous con- gestion, which resulted in marked prolongation of the mean circulation times, probably reflecting a reduction in total blood flow (4), failed to produce a significant decrease in relative flow in the rapid as compared to the slow component. The local (intra-arterial) infusion of epinephrine and nor- epinephrine also failed to change the relationship between flows or volumes in the rapid as compared to the slow components. It must be admitted, however, that the number of observations was not sufficient to rule out the possibility of minor changes in the measured parameters. In addition, the method of statistical analysis was more apt to underestimate than over- estimate significant changes. Nevertheless, there were three cases with paired controls given nor- epinephrine intra-arterially and four given epi- nephrine. Inspection of these data fails to sug- gest the obvious and striking changes seen in other groups (Table I). Epinephrine continuously infused intra-arterially produces no change in forearm blood flow while sustained intravenous infusion is followed by a slight vasodilation (5). Because of these opposing TABLE III Statistical analysis of data on six subjects given norepinephrine systemically, compared with their own controls Mean difference between Parameter means zt standard error T 19.7 f 6.2 t1 t2 8.2 2 iF L/f2 5:9 f 2:o Vl/VZ 1.2 f 0.9 * Significant at the 5 per cent level. t iFi* 0:s 2.9* 1.3 VARIATIONS IN BLOOD FLOW COMPONENTS OF FOREARM a44 actions, the total forearm flow representing both skin and muscle may not be significantly changed. Norepinephrine, on the other hand, is said to pro- duce slight decreases in forearm or calf blood flow (6) which were somewhat more prominent after intra-arterial infusion (7). This constrictor effect may have accounted for the slight increase in T fol- lowing intra-arterial norepinephrine. However, the failure to reduce the flow and volume ratios in the rapid as compared to the slow component by any of the various experimental maneuvers, in- cluding intra-arterial norepinephrine and venous congestion, suggests that in the normal resting state the ratios f,/f, and v,/v, approach their mini- mal values ; and whereas these ratios can be made to increase considerably under various stresses, apparently they cannot easily be reduced. A surprising observation was that flow in the rapid component increased relative to that in the slow component following systemic norepinephrine infusion. The increase in dye concentration fol- lowing systemic norepinephrine indicates a re- duction in total forearm blood flow or vascular volume, or both. The broadening out of the early portion of the curves also suggests this. However, the earlier "disappearance" time observed in five out of six cases suggests that the predominant ef- fect was on vascular volume rather than on blood flow since a decrease in flow without a volume change would tend to prolong rather than shorten the disappearance time. Obviously, the elevated arterial pressure would also tend to counteract a possible decrease in blood flow. Plethysmographic measurements during norepinephrine infusions in- dicate only slight decreases in forearm blood flow (6). Thus, the characteristic change in the dye concentration curve following systemic norepi- nephrine seemed to be due to the combined effects of increased pressure and increased local vaso- constriction. These observations are in keeping with the previous studies, indicating a decrease in periph- eral vascular volume after norepinephrine (8). [The previous studies showed that epinephrine also had this activity in the dog, but it should be recalled that unlike man the hemodynamic effects of epinephrine in the dog resembled those of nor- epinephrine (9).] It was indicated that either sympathetic activity or norepinephrine reduces the volume of the small vessels, including postarterio- lar vessels, with a resulting increase in venous return. It is suggested that the present data may throw further light on this phenonmenon. A decrease in T and increase in the ratio f,/f, could not oc- cur as a result of arteriolar constriction alone, but must include postarteriolar narrowing as well. The failure of v,/v, to change significantly indi- cates that both the rapid and slow components shared in the vasoconstriction. The net effect of such a decrease in peripheral vascular capacity, aided and abetted by an increase in arterial pres- sure, would be to increase the velocity of flow through the generally constricted peripheral vas- cular bed. The result would be more fast-moving and less slow-moving blood flow; that is, an in- crease in the ratio f,/f,. SUMMARY AND CONCLUSIONS Various experimental procedures were uti- lized to study their effects on the relationship be- tween the rapid and slow components of blood flow in the human forearm ( 1). Local hyperemia pro- duced by intra-arterial injection of tolazoline, re- active hyperemia, or local exercise increased the flow and volume in the rapid component relative to that in the slow component. Venous congestion resulted in marked but nearly proportionate pro- longations of mean circulation times. The rela- tive flows and volumes of the two components were not significantly changed. Local (intra- arterial) infusion of either epinephrine or nor- epinephrine and systemic administration of epi- nephrine produced no obvious changes in these parameters although the number of cases, es- pecially those with paired controls, was insuffi- cient to rule out small changes. Systemic infu- sion of norepinephrine produced characteristic changes in the time-concentration curve of the dye, indicating a relative increase in blood flow of the rapid component. It is suggested that the latter was due to the hypertension plus vasoconstriction in the forearm occurring in response to significant elevation of mean pressure. These observations demonstrate that the bi- phasic system of forearm blood flow and volume is dynamic, the relative proportion of the rapid and slow components changing under appropriate stim- uli. The data also suggest that under normal 845 EDWARD D. FREIS AND HAROLD W. SCHNAPER resting conditions the ratios f,/f, and v1/v2 are near the lower levels of their possible range. ACKNOWLEDGMENTS The authors are indebted to Drs. W. J. Youden and W. S. Connor, Jr., of the National Bureau of Standards, for advice and counsel in carrying out the statistical analysis of the data. REFERENCES 1. Freis, E. D., Schnaper, H. W., and Lilienfield, L. S. Rapid and slow components of the circulation in the human forearm. J. clin. Invest. 1957, 36, 245. 2. Nicoll, P. A., and Webb, R. L. Vascular patterns and active vasomotion as determiners of flow through minute vessels. Angiology 1955, 6, 291. 3. Zweifach, B. W., and Mets, D. B. Selective distribu- tion of blood through the terminal vascular bed of mesenteric structures and skeletal muscle. Angi- ology 1955, 6, 282. 4. Halperin, M. H., Freidland, C. K., and Wilkins, R. W. The effect of local compression upon blood flow in the extremities of man. Amer. Heart J. 1948, 35, 221. 5. Duff, R. S., and Swan, H. J. C. Further observations on the effect of adrenaline on the blood flow through human skeletal muscle. J. Physiol. 1951, 114, 41. 6. Duncanson, D., Stewart, T., and Edholm, 0. G. The effect of I-arteranol on the peripheral circulation in man. Fed. Proc. 1949, 8, 37. 7. Barcroft, H., and Konzett, H. Die Wirkung von Noradrenalin, N-Isopropylnoradrenalin (Aleudrin) und Adrenalin auf Blutdruck, Herzschlagzahl und Muskeldurchblutung am Menschen. Helv. physiol. pharmacol. Acta 1949, 7, C4. 8. Freis, E. D., and Rose, J. C. The sympathetic nervous system, the vascular volume and the venous return in relation to cardiovascular integration. Amer. J. Med. 1957, 22, 175. 9. Rose, J. C., Freis, E. D., Hufnagel, C. A., and Mas- sullo, E. A. Effects of epinephrine and nor-epi- nephrine in dogs studied with a mechanical left ventricle. Amer. J. Physiol. 1955, 182, 197.