Methyl alcohol. The narcotic action of this alcohol on tadpoles. brackets, equilibrium was attained as a rule in about 10 to 15 minutes.

Transcription

1 THE ACTION OF HOMOLOGOUS ALCOHOLS AND ALDEHYDES ON THE TORTOISE HEART. BY H. M. VERNON. (From the Physiological Laboratory, Oxford.) IN a previous paper' I showed that when the tortoise heart is perfused with <saline containing alcohol, chloroform, ether or formaldehyde, its beats are depressed to a level proportionate to the concentration of the poison, and then remain at a constant height for as long as the poisoning is continued. I have now extended these observations to nine homologous alcohols of the aliphatic series, and have found that their depressant power on the tortoise heart corresponds closely with their narcotic action on tadpoles as determined by Overton2. The method of experiment was the same as before. The heart of Emys europcea was perfused with oxygenated Ringer's solution, and the beats, magnified 14-fold, recorded on a smoked surface. When the heart was perfused with the alcohol-containing saline, it was surrounded by a bath of the same solution. The temperature of the bath was kept approximately at 180 C. in all the experiments. The solutions were made up volumetrically with Ringer's solution, due allowance being made for the specific gravity of the alcohols. The alcohols and aldehydes were obtained from Merck and Kahlbaum. Methyl alcohol. The narcotic action of this alcohol on tadpoles is less than that of any other, and it likewise has the smallest effect upon the heart. The data in the table show the contraction heights of the heart, when equilibrium was attained, as percentages on the initial contraction height. The hearts were generally perfused for 15 to 30 minutes with the alcohol, and as is shown by the times given in brackets, equilibrium was attained as a rule in about 10 to 15 minutes. 1 Vernon. This Journal, XLI. p Overton. Studien ulber die Narkose, p Jena, 1901.

2 326 H. M. VERNON. The N/2 methyl alcohol depressed the beats on an average to 890/0 of their initial height, whilst 2N methyl alcohol, or a 6-4 0/o solution, produced them to 120/o on the initial height. In all four of the experiments made with this strong alcohol solution the spontaneous beats of the heart were stopped, whilst the N solution likewise stopped them in some instances. Whenever an alcohol stopped the heart in these and the other experiments, it was gently pinched with forceps about once a minute, and it generally continued to respond to suich mechanical stimulus throughout the poisoning. In one of the four experiments quoted in the table, however, the beats were reduced to zero after 13 minutes' poisoning. The effect of methyl alcohol and of other alcohols on the rate of heart beat is described in detail in a subsequent section. Concentration of methyl alcohol A A B B C Mean N/2 900/0 (19 m.) 800/0 (15 m.) 90 0/0 (10 m.) 94 0/0 (9 m.) 89 /0 Nll 18 0/0 (15 m.) 36 0/0 (26 m.) 75 0/O (6 m.) 970/0 (10 m.) 41 /o (10 in) 530/n 2N - (00/oin 13 m.) 20/0 (12i.) 340/0 (13 m.) 11 0/ (3) 12 0/0 A tvpical series of experimental results is plotted out in Fig. 1. Ordinates represent the contraction height of the heart in millimetres, and abscissae the time in minutes. In the experiments of this series equilibrium was attained in 6 to 12 minutes, so the poisoning was continued in each case for only 15 minutes. At the points marked R in the curves the alcohol saline was replaced by fresh oxygenated Ringer's solution both in the perfusion cannula and in the bath surrounding the heart, and it will be seen that the recovery of the heart began almost immediately. The rate of recovery of the heart, as judged by the slope of the curves, is almost the same whatever the concentration of alcohol used for poisoning. Also the heart, after the alcohol was washed out, generally beat for a short time more vigorously than before the poisoning began. As a rule two or three consecutive series of experiments were made on each heart with different alcohols, and in most cases the results obtained on already poisoned hearts were similar to those on absolutely fresh hearts. To minimise any error so arising, most alcohols were tested twice on fresh hearts, these series being designated in the tables by the letter A; twice on hearts previously subjected to one series of poisonings (B); and once on hearts previously subjected to two series of poisonings (C). In the data under discussion it will be seen that the two hearts most depressed by the methyl alcohol were fresh hearts, whilst that least depressed was a " B" beart. Though the depression

3 HIEART OF TORTOISE. 327 of contraction height by N/2 methyl alcohol was fairly constant, that by N alcohol varied from 18 0/0 to 97 0/0. These great differences of reaction were undoubtedly due in chief part to differences in the hearts themselves, for a heart, if more, or less, sensitive to the action of one alcohol, was similarly responsive to other alcohols. The heart least affected by methyl alcohol was likewise the least affected of the five poisoned with iso-propyl alcohol, whilst that most affected by methyl alcohol was the one most affected of the five poisoned by propyl alcohol. These differences in the reactivity of the hearts were not dependent on season, as all the experiments were made during the summer months, but the tortoises used differed greatly in size and condition. r1ime IN MINUTES Fig. 1. Methyl alcohol. Fig. 2. Iso-propyl alcohol. Iso-propyl alcohol. (CH3)2. CHOH. Next to methyl and ethyl alcohols, iso-propyl alcohol is the least depressant on the heart, and the least narcotic to tadpoles. The data in the table show great differences in the reaction of different hearts, for with N/4 alcohol the depression of contraction height varied from 8 0/0 to 73 0/0. Concentration of iso-propyl alcohol A N/ /0 (12 m.) N/8 77 0/ (10 m.) N/4 50 /0 (9 m.) N/2 13 0/0 (9 m.) A 100 /0 (15 m.) 89 /0 (10 m.) 73 0/0 (9 m.) 21 0/0 (12 m.) B 840/o (7m.) 50 0/0 (7 m.) 8 0/0 (9 m.) B C Mean _ 93 0/0 100 /o (10 m.) 81 0/o (9 m.) 79 /0 61 0/o (6 m.) 40 0/0 (14 m.) 46 0/0 2 /o (5 m.) 1 /o (l m.) 7 /0 In calculating the mean contraction heights, two additions have been made to the data. It will be seen that in one experiment N/8

4 328 H. M. VERNON. alcohol had no depressant effect on the heart in ten minutes, so the corresponding experiment with N/16 alcohol was not made, though there can be no doubt that this concentration would likewise have failed to exert an influence. Hence in calculating the mean depression produced by N/16 alcohol, a 100 0/0 value is assumed in this series. In another series it was found that N/4 alcohol depressed the heart to 8 0/0 its initial cotntraction height, and so there can be no doubt that N/2 alcohol, if tried, would have reduced the beat to zero. Consequently a zero value is assumed in calculating the general mean. In Fig. 2 is reproduced a typical series of experiments with isopropyl alcohol. The individual results happen to be in close agreement with the mean values given in the table. It will be seen that in two of the four curves the heart recovered some of its contraction height in the last few minutes of the poisoning. In such cases the equilibrium level is taken as the average contraction height from the time of maximum depression till the time the Ringer's solution was substituted. The slope of the curves shows that the rate of recovery of the heart was very similar throughout. Propyl alcohol. CH3. CH2. CH2OH. This alcohol is slightly more depressant on the heart than iso-propyl alcohol, and slightly more narcotic on tadpoles. The individual results do not show quite so much variation as those obtained with iso-propyl alcohol. In calculating the means no corrections have been applied, but no mean is given for Concentration of propyl alcohol A A B B C Mean N/ /0 (5 m.) - - _ N/ /0 (8 i.) 100 0/0 (15 m.) 93 0/0 (7 m.) 72 0/0 (15 m.) 88 0/0 N/8 49 0/ (6.) 98 0/0 (3 m.) 72 0/0 (8 i.) 51 0/0 (31 m.) 77 0/0 (11 m.) 690/0 N/4 22 0/0 (4 i.) 59 0/0 (10 i.) 31 0/o (13 i.) 8 0/0 (23 m.) 320/0 (12 i.) 300/0 N/2-8 /o (5 m.) (1 /0oin 15 m.) N/2 alcohol, as the data are insufficient. ID that N/4 alcohol lowered the beat to 30 0/0 its initial height, it was as a rule useless to investigate the action of N/2 alcohol. It would have reduced the beat to zero in a very few minutes, and so have. afforded a result useless for the calculation of an equilibrium level. In the N/2 experimental result given in brackets, equilibrium of contraction height was not reached in 15 minutes. Butyl alcohols. Overton found that of isomeric alcohols, those with the least branched carbon chains had the strongest narcotic action, and those with the most branched carbon chains, the weakest narcotic action. My results closely correspond to his, though secondary butyl alcohol,

5 HEART OF TORTOISE. which was not investigated by him, forms an exception to his rule. Thus I found that normal butyl alcohol (CH3. CH2. CH,. CH2OH) has the most depressant action on the heart; then iso-butyl alcohol ((CH2)2. CH. CH20H); then secondary butyl alcohol (CH3. CH2. CHOH. C011), and lastly tertiary butyl alcohol ((COH)3. CO). The weaker action of iso-propyl alcohol as compared with propyl alcohol has already been pointed out, whilst below it will be shown that tertiary amyl alcohol is much less depressant than iso-amyl alcohol. The data obtained with butyl alcohol are rather less variable than those for methyl and the propyl alcohols. The heart least affected was likewise the least affected of the five hearts poisoned with tertiary Concentration of butyl alcohol A A B C Mean N/ /0 (3 i) 77 0/0 (7 m.) 79 0/0 (10 m.) 84 /0 N/32 57 /0 (7 m) 55 /0 (11 m.) 89 I/0 (5 m.) 50 0/0 (10 m.) 63 0/0 N/ /0 (12 i.) 9 0/0 (4 i.) 42 /0 (11 i.) 4 0/c (4 i.) 18 /0 N/8 - (O0/0 in 2 m.) - butyl alcohol, whilst that most affected was one of the two most affected by iso-amyl alcohol, and one of the two most affected by tertiary butyl alcohol. The data obtained with iso-butyl alcohol vary about as much as those for butyl alcohol. The heart least affected was one of the two least affected by iso-propyl alcohol, whilst that most affected was one of Concentration of iso-butyl alcohol A A B C Mean N/ /0 (5 m.) 95 0/0 (4 in) 86 /0 (4 in) /n N/ /0 (7 m.) 87 /0 (4 i.) 61 0/0 (6 i.) /0 N/16 15 /0 (12 m 30 /0 (5 i.) 6 /0 (6 i.) 48 0/0 (9 m.) 25 0/0 N/8 ( 0/0in 13 m.) (O0/0 in 3 m.) - 5 0/0 (13 m.) the two most affected by propyl alcohol. A series of results obtained with iso-butyl alcohol is reproduced in Fig. 3. They correspond well with the previous records. After poisoning with N/16 alcohol the contraction height of the heart rose on recovery to 20 O/o above the initial height. With secondary butyl alcohol three experiments were made, as only a very small quantity of the alcohol was available. In these Concentration of secondary butyl alcohol A A B Mean N/ /0 (15 i) 93 0/0 (4 m.) 93 0/O N/ i/e (15 m.) 58 0/0 (18 m.) 81 0/ (9 m.) 65 0/0 N/ /0 (24 m.) 57 0/0 (18 m.) 72 0/0 (10 m.) 52 0/0 N/8 10 0/0 (18 m.) 45 0/0 (12 m.) (3 0/o in 30 m.) 19 0/3 N/4 - (00/0in5im.) 329

6 330 H. M. VERNON. experiments the heart was not surrounded with a bath of saline containing the alcohol, but was suspended in air, so it is possible that the action of the higher concentrations was not so marked as it would have been if the usual conditions of experiment had been observed. Fig. 3. Iso-butyl alcohol. TIME in MINUrES Fig. 4. Iso-amyl alcohol. Of the five series of experiments made with tertiary butyl alcohol, four show about the usual amount of correspondence. The last series in the table is ignored in taking the general mean, as it was made on a heart which had previously been poisoned with four concentrations Concentration of tertiary butyl alcohol N/32 N/16 N/8 N/4 N/2 A 920/0 (9 m.) 100 0/0 (17 m.) 79 0/0 (7 m.) 83 0/0 (9 m.) 75 0/0 (8 m.) 48 0/0 (16 m.) 370/0 (6 m.) 27 0/o (13 m.) 6 /o (17im.) A B 91 0/o (6 m.) 65 0/0 (10 m.) 88 0/0 (10 m.) 79 0/0 (11 m.) 43 /0 (10 m.) 59 0/o (10 m.) 8 0/0 (10 m.) 1-5 0/0 (11 m.) B B Mean 59 /0 (6 m.) 93 0/0 24 0/0 (17 m.) 77 /0 5 0/o (11 m.) 56 /o 22 0/0 of acetaldehyde. Poisoning with aldehydes, in contradistinction to that with alcohols, sometimes permanently depresses the vitality of a heart, and renders it more susceptible to the action of other poisons. With this one exception none of the alcohol experiments were made on aldehyde-poisoned hearts. Amyl alcohols. Of these alcohols, several are not available because of their insolubility in water, so only two were experimented with. Iso-amyl alcohol, or fermentation amyl alcohol, which is a mixture of

7 HEART OF TORTOISE. (CH8)2. HCH. CH2. CH2OH and (COH). CH (C2H,). CH2OH, was found by Overton to be one of the most powerfully narcotic of the alcohols investigated by him, and I found it to be more than twice as depressant on the heart as any of the other alcohols investigated. Most of the individual results in the five series of experiments made show fairly good agreement, whilst the curves in Fig. 4 correspond well with those obtained with the other alcohols. Concentration of iso-amyl alcohol A A A B C Mean N/ /0 (26 m.) 88 0/0 (10 m.) 85 0/0 (11 n.) 800/0 (12 m.) /o N/ /0 (20 m.) 470/0 (10 m.) 57 0/0 (10 m.) 14 0/0 (8 m.) 760/0 (3 i.) 54 0/0 N/ /o (18 m.) 2 0/0 (10 m.) 24 0/o (10 m.) - 1 0/o (3m.) 11 0/o The other amyl alcohol investigated was the tertiary alcohol, or amylene hydrate, C2H5. C (CH3)2. OH. It had only about a fourth as Concentration of tertiary amyl alcohol A A B C Mean N/ /0 (3 i.) 100 0/0 (10 m.) /0 N/ /0 (13 i.) 95 O/0 (4 i.) 73 0/0 (25 m.) 77 0/0 (6 i.) 84 /0 N/ /0 (17 i.) 72 0/0 (9 i.) 46 0/0 (17 m.) 49 0/0 (20 i.) 57 0/0 N/8 28 0/0 (7 i.) 4 /0 (5 i.) 8 /0 (23 i.) 24 0/0 (10 i.) 16 o/0 N/4 ( 0/0in 4 m.) - much depressant power on the heart as iso-amyl alcohol, for the data in the table show that N/16 amylene hydrate had about as much effect as N/64 iso-amyl alcohol. Comparison of the different alcohols. In order to bring out the differences in the activity of the various alcohols investigated, the mean percentage values given in the above tables of data have been plotted out in Fig. 5. The abscissa represents concentration in terms of normality. The experiments with ethyl alcohol described in the previous paper were made at, concentrations of 1, 2 and 4I/0, and so slight corrections were necessary to reduce them to N/4, N/2 and N values respectively. It will be seen that the slope of all the curves is very similar, or it appears that tile depressant action of each alcohol upon the heart increases at about the same proportionate rate with increase in its concentration. Considering the comparatively small number of observations made with each alcohol, and the considerable variations in the individual results a each concentration, the mean results as PH. X[LIII. 22 3:31

8 332 H. M. VERNON. reproduced in this figure show a very striking correspondence. One set of mean values has been omitted from the figure, viz. that obtained with secondary butyl alcohol. The curve has a smaller slope than that of the other alcohols, and cuts across the iso-butyl alcohol and tertiary amyl alcohol curves. It is very likely an erroneous result for the reason already described, and as it is founded on only three sets of experimental data, but little weight can be attached to it X80 Z A' ~~~~~~~~~~~~~- -CI Q U,~40. r30 20 tn N N N/4 N/8 a/16 CONCE NTRArJON N/3b2 N/S4 ipzs Fig. 5. In order to compare m-y results with those obtained by Overton for the narcosis of tadpoles, I have calculated the concentration of each alcohol required to depress the heart 50O/,. It ranges from 1 iii 29X4 for methyl alcohol to 1 in 676 for iso-amyl alcohol, or as 23 to 1. With one exception the order of activitv of the various alcohols on the heart is the same as that of their narcotic action on tadpoles. Overton found that tertiary butyl alcohol is less active than the propyl alcohols, whilst I find it is more so. On taking ratios between the pairs of con- Concentration Concentration depressing heart narcotising Ratio of beat 60/ tadpoles concentrations Methyl alcohol... 1 in 29'4 1 in to 1 Ethyl alcohol... 1 in 41 1 in 75 1'83 Iso-propyl alcohol... 1 in 71 1 in Propyl alcohol... 1 in 88 1 in 150 1'70 Tertiary butyl alcohol... 1 in 93 1 in 100 1'08 Tertiary amyl alcohol... 1 in in 200 1'26 Secondary butyl alcohol 1 in Iso.butyl alcohol.. 1 in in Butyl alcohol... 1 in in Iso-amyl aloohol... 1 in in

9 HEART OF TORTOISE. centrations obtained with each alcohol, it will be seen that the numerical relationship is by no means constant. The concentration of methyl alcohiol required to reduce the heart beat 500/o is nearly twice as great as that required to narcotise a tadpole, whilst the concentration of iso-amyl alcohol required for the same purpose is only three-fourths as great. The other alcohols fall between these extremes, or with the exception of the aberrant tertiary butyl alcohol one may say that as the narcotic action of homologous alcohols increases, a relatively smaller and smaller amount is needed to depress the heart, as compared with that required to depress the nervotus system. The explanation of this changing ratio is unknown, but it may be due to differences in the nature of the lipoids of the central nervous system and of the heart. MeyerI and Overton2 have shown that the activity of a narcotic varies roughly as its partition coefficient between fat-like substances and water, but the correspondence is not numerically exact3, and there is no information as to how far the coefficient is affected by the nature and composition of the fat. The results obtained in my previous paper show that the depressant effect of increasing concentrations of ether and chloroform upon the heart by no means corresponds to the effect of alcohol. It is calculated that on an average the quadrupling of the concentration of the alcohols lowered the cardiac contraction from 84 /oi of its initial height down to 160/o. To effect a similar lowering of contraction height with ether it is necessary to increase its concentration 16-fold, so the ether, though it comes into association with the lipoids of the heart, cannot act upon them in just the same way as the alcohols do. The toxic action of homologous alcohols has been determined by a number of other investigators besides Overton, though not so completely. Upon mammals the effects of intravenous interjection have been studied by Joffroy and Serveaux4, whilst' Baer5 adopted the simpler method of introducing the alcohols into the stomach of the animal by means of a sound. Picaud6 placed fish and amphibians in solutions of the alcohols, and subsequently a similar method was adopted by Bradbury and Marshall7, and by Cololian8. Finally Wirgin. Meyer. Arch. f. exp. Path. XLII. p Overton, I.c. 3 Cf. Baum. Arch. f. exp. Path. XLII. p Joffroy and Serveaux. Arch. de med. exp. et d'anat. path b Baer. Arch. f. Physiol. 1898, p Picaud. Comptes Rendus, cxxiv. p Bradbury and Marshall. Brit. Med. Journ. 1899, ii. p Cololian. Journ. de phys. et de path. 1901, p Wirgin. Ztsch. f. Hyg. xi.vi. p

10 334 3H. M. VERNON. determined the concentrations required to paralyse the development of M. pyogenes aureus, and found that they ranged from 8 /o for methyl alcohol to 10/0 for iso-amyl alcohol. He also investigated the capacity of alcohols for laking the corpuscles of rabbit's blood, and found that the concentrations required ranged from 11-60/o for methyl alcohol down to 60/o for iso-amyl alcohol. Taking the toxicity of ethyl alcohol in every case as unity, we get the following comparative toxicities of the other alcohols. Wirgin Vernon Joffroy Baer Picaud Bradbury Cololian Overton Wirgin red tortoise Alcohol mammals mammals fish fish fish tadpoles bacteria corpuscles heart Methyl *46 *8 * *73 *73 *84 *72 Ethyl Propyl Butyl Iso-amyl In some of the experiments amyl alcohol was said to have been used, but the statements as to boiling point and solubility make it clear that the alcohol was really the iso-body. It will be seen that my results on tortoise hearts are in best agreement with those of Joffroy and Serveaux on mammals, and those of Cololian on fish. The closest correspondence of all is not with any series of toxicity determinations, but with the results of Wirgin on the laking of red blood corpuscles. As this laking is due to the solution of the lipoids of the corpuscles by the alcohols, it follows that my results afford substantial support to Meyer and Overton's theory of the relationship between toxicity and lipoids. The action of alcohols on the rate of heart beat. In addition to rowering the height of contraction of the heart, alcohol poisoning as a rule slows the beat. Sometimes as a result of this slowing the heart is able to give more forcible beats, but there is no constancy either in the effect of the alcohol on the rate of beat or in the indirect result thereby produced on the contraction height, so for statistical purposes the two effects have to be considered independently. Following the plan adopted in tny previous paper, I have split up all the results obtained with each alcohol into four groups. The first group includes hearts of which the beat was reduiced by less than a fourth. The effect is then said to be nil. Another group includes hearts which were stopped completely for two or more minutes, whilst the hearts slowed to various degrees are split up into the two

11 HEART OF TORTOISE. remaining groups. The table shows the percentage of hearts falling into each of these grotups. The number of data available is too small to afford more than a rough basis for comparison, but we see that five out of the ten alcohols did not slow the heart at all in 41 to 540/0 of the experiments made. Excluding the results obtained with methyl and ethyl alcohols, the proportion of hearts stopped by the alcohols varied from 8 to 29 0/0, or in six out of the ten series it varied only from 18 to 29 0/0. The special tendency of methyl alcohol, and to a less extent of ethyl alcohol, to slow and stop the heart, may perhaps be due to the considerable concentrations of these alcohols necessary to affect the heart. Tiie 2N solutio.n of methyl alcohol contains 6A40/0 of alcohol, or has a very much greater osmotic pressure than the 1750/0 saline solution it replaces. Reduction Iso- Iso- Secondary Tertiary Isoin rate Tertiary Methyl Ethyl Propyl propyl Butyl butyl butyl butyl amyl of heart beat amyl alcohol alcohol alcohol alcohol alcohol alcohol alcohol alcohol alcohol alcohol Nil i to i i or less Stop No. of Exps The general conclusions to be drawn from these data are mostly negative ones. One can say that the actioin of the alcohols on the rate of individual heart beat is very irregular, but that most if not all of the alcohols-with the exceptions just noted-tend to act in more or less the same way. A source of error in these results lies in the fact that all the data obtained with each alcohol are thrown together, regardless of whether the concentrations of alcohols used were great or small. Still the much more complete series of experiments with ethyl alcohol described in my previous paper showed that even allowing for differences of alcohol concentration the irregularities are very great. The rate of recovery of the heart from alcohol poisoning. :33.5 It has been pointed out that in the sets of experimental data reproduced in Figs. 1 to 4 the slope of the curves of recovery of the heart produced by washing out the alcohol with Ringer's solution is more or less the same in each case. This slope presumably affords an index of the rate at which the alcohol breaks away from the heart, and in order to obtain comparable data I have calculated the time required by the heart beats to increase from 200/0 of the height finally

12 336 3H. M. VERNON. attained by them up to 800/0 of this final height. The experiments made with the smaller concentrations of the alcohols are not available for such a calculation, but when the lowering of contraction height produced by the alcohol was 2.5 0/0 or so, instead of the required 200/G, it was generally possible to make a small correction and determine approximately the /0 recovery time. The recovery times are given in the table, together with a part of those quoted in my previous paper for ethyl alcohol. The individual times are very variable, and the mean times range from 2-6 to 6'8 minlutes. Judging from the comparative constancy of the recovery times obtained with ethyl alcohol, ether and chloroform, I suggested in my previous paper that these three narcotics all broke away from the tissues at the same rate. The present data show that a similar conclusion cannot be accepted for these alcohols. Time of recovery of /o Mean Corresponding time of attainment Mean Alcohol final contraction height (mins.) of constant contraction height (mins.) Tertiary butyl 10, 9, 8, 4, , 22, 11, 17, Tertiary amyl 10, 9, 6, 3, , 10, 7, -, 5 11 Propyl 9, 9, 4, , 5, -, 4 11 Iso-amyl 9, 8, 5, 5, , 18, 10, 3, Ethyl 9, 7, 7, 6, 5, 5, 5, 5, 4, , 18, 11, 9, 12, 11, 12, 9, 15, Methyl 6, 5, 4, 3, , 13, 3, 12, - 11 Iso-propyl 5, 3, 3, 2, , 11, 9, 9, 5 9 Butyl 3, 3, 3, ,-, 4, 4 7 Iso-butyl 4, 3, 2, 2, , 5, 13, 6, - 8 In the right half of the table are given the times required for the attainment of a constant contraction height in the corresponding experiments, so far as they were determined, whilst the means of these times are given in the last column of the table. They range from 14 to 7 minutes, and it will be seen that they roughly run parallel to the mean /, recovery times. That is to say, hearts which are quickly depressed by a given alcohol are as a rule able to recover quickly, and if slowly, then slowly. Such differences of response are probably due in part to the nature of the alcohol, but they are also partly dependent on the peculiarities of individual hearts. A comparison of the pairs of data shows that a heart which was poisoned slowly by an alcohol generally recovered more slowly than usual. In a few instances this result was dependent on too slow perfusion of the heart, the result of vaso-constriction of the coronary vessels. Great care was taken to keep the perfusion rate at a fairly rapid and constant rate by varying

13 HTEART OF TORTOISE. the pressure suitably, but in a few hearts it was impossible to get a sufficiently rapid perfusion, even at a pressure of two metres (water pressure). The data obtained in such experiments are italicised in the table. T'he action of aldehydes on the heart. In my previous paper I described the action of formaldehyde on the heart, and I have now extended these observations to acetaldehyde, propyl aldehyde and iso-butyl aldehyde. These three aldehydes are twenty to forty times less toxic than formaldehyde, and their action differs in other ways. Fig. 6 shows the action of acetaldehyde, and we see that though the poisoning with N/128 and with N/64 aldehyde was continued for 50 minutes, the constancy level, reached after 15 and 33 mins. respectively, was maintained perfectly, and the heart rapidly 3s 33}7 30t0 -/ 6ICtEN\_;&tEXr3 1:~~~~~~~~~~~~~~1 00~~~~~~~~~~~~~~~~C$ so 60 7 I010 is 20 2s TIMPE / N 4INUTES TIME IN MINUTES Fig. 6. Acetaldehyde.3 Fig. 7. Iso-butylaldehyde. recovered when the poison was washed out. In the fo'rmaldehyde experiments the heart generally took about 80 mins. to attain constancy level, aiid on washing out the poison with Ringer's solution there was a latent period of 3 to 1- mins. before the heart began to recover at1all. The / recovery time was 24 to 104 mins. (50 mins. on an average), and the heart seldom recovered its initial contraction height. If a formaldehyde-poisoned heart were poisoned a second time, it recovered much more slowly than before, the latent period averaging 30 0ins., and the /0recovery time 89 mi0s.

14 338 H. M. VERNON. Concentration of acetaldehyde A A A B B Mean N/ /0 (11 m.) /0 (3M.) - - N/ /0 (12 m.) 67 0/0 (8 i.) 56 0/0 (15 i.) 640/0 (6 m.) 65 0/ (20 m.) 670/0 N/64 54 C/o (17 m.) 35 0/o (11 m.) 13 0/0 (33 m.) 36 0/0 (10 m.) 230/0 (12 m.) 320/0 N/32 23 /0 (19 m.) 5 0/0 (24 m.) (1 0/o in 15 m.) 4 0/0 (36 m.) 80/ On the other hand the data in the table seem to show that acetaldehyde had no greater depressant action on "B" hearts than on " A " hearts, or that it did no more permanent damage to the heart than poisoning with an alcohol. The results obtained with propyl and iso-butyl aldehydes show that the recovery is not quite so perfect as that from alcohols, and a repetition of a poisoning on a heart generally gave a greater depression of contraction height than the first poisoning. Still the injury is comparatively slight, and not to be compared with that induced by formaldehyde poisoning. Concentration of propyl aldehyde A A B B Mean N/ /0 (12 m.) - N/ /0 (8 m.) 72 /0 (15m.) 52 0/0 (7 m.) -64 /o N/ /0 (13 m.) 52 0/0 (8 m.) 38 0/0 (7 i.) 40 0/0 (13 i.) 44 /0 N/32 25 /o (11 m.) (23 0/0 in 25 m.) 19 0/0 (9 i.) 11 /0 (15 m.) 20 /o N/16 ( O/ in 3 m.) The data obtained with propyl aldehyde are even more consistent than those with acetaldehyde, but those with iso-butyl aldehyde are rather less so. Fig. 7 shows the effect of poisoning with four concentrations of iso-butyl aldehyde. The N/128 poisoning was continued for.50 mins., but for convenience in reproduction part of the record is omitted from the figure. The heart was not damaged at all, and in fact its contraction height rose steadily during the last 29 mins. of the poisoning. It will be noted that the rate of recovery of the heart was nearly the same after each poisoning, both in this series of experiments and in that obtained with acetaldehyde and reproduced in Fig. 6. Concentration of iso-butyl aldehyde A A B B Mean N/ /0 (10 m.) 68 O/0 (6 i.) 60 0/0 (13 i.) /0 N/ /o (21 i.) 40 0/0 (5 i.) 53 0/0 (10 m.) 28 0/o (15 m.) 47 O/0 N/64 35 / (17 m.) 12 0/0 (14 m.) 21 /n (13 m.) 11 /o (23 m.) 20 0/0 N/32 1 /o ( In Fig. 8 a comparison is made of the mean constancy levels obtained at different concentrations with the four aldehydes. The formaldehyde valuies are those yielded by fresh hearts only. The general

15 HEART OF TORTOISE. contour of the curves differs from that of the alcohol curves. The slope is less, and two of the curves seem to show that the depressant effect of an aldehyde is relatively rather smaller with great concentrations than with small concentrations. The other two curves show slightly the reverse effect, whilst a mean of all of them would indicate that the depressant effect of an aldehyde is approximately in direct proportion to its concentration. 'S.. The relative position of the four curves seems to show that there is no regular connection between molecular weight and toxicity. Perhaps one would have shown itself if butyl aldehyde had been employed instead of iso-butyl aldehyde, but in any case formaldehyde offers a complete exception. A source of error lies in the ~~~4O~ ~ ~ 4 '30 &. O,soV,s /lso a30 20 ~-10 z3t N/64 N/128 N//26 CONCENTRATION Fig difficulty of ascertaining the exact concentration of the acetaldehyde used. The sample employed was said to be of 800/0 strength, and fresh solutions were made up on this basis for each set of experiments. It contained a little free acid, but not enough to neutralise the *01 0/0 NaHCO3 present in the Ringer's solution. Extra NaHCO,3 to neutralise it was added in two series of experiments, but this made no difference in the result. The rate of recovery of a heart poisoned with ethyl, propyl or iso-butyl aldehydes is about the same as that of its recovery from an alcohol, for the mean values vary from 4-o to 6-2 miiins. The times of attainment of constancy in the experiments with ethyl and iso-butyl aldehydes quoted in the table are for the most part considerably

16 340 H. M. VERNON. Time of recovery Corresponding time of of o/0 final Mean attainment of constant Mean contraction height (mins.) contraction height (mins.) Acetaldehyde 12, 7, 5, 5, 4, 3, , 32, 19, 12, -, 36, Propyl aldebyde 9, 7, 7, 4, , 11, -, 15, 9 12 Iso-butyl aldehyde 9, 5, 2, , 13, 17, greater than the times obtained in the alcohol experiments, but the means of all the constancy times quoted in the previous tables are 16 mins. for acetaldehyde, 11 mins. for propyl aldehyde, and 14 mins. for iso-butyl aldehyde, or do not differ greatly from the alcohol values. Upon the rate of heart beat the action of the aldehydes differs considerably from that of alcohols. In only one out of the 44 experiments made with ethyl, propyl and iso-buityl aldehydes was the heart stopped completely, whilst it was slowed in only 320/o of all the experiments. The data in the table show that in 38 or 39 0/0 of the experiments made with each aldehyde the effect on the heart was nil, Influence on rate of heart beat Acetaldehyde Propyl aldehyde Iso-butyl aldehycde Increased 100/0 or more Nil Reduced to Reduced to j or lees Stopped Number of Exps whilst in a considerable number of experiments the beat was quickened. Only experiments with a quickening of 10 0/0, or more are included, and in three of the experiments with acetaldebyde the acceleration amounted to 500/o. Formaldehyde has a simnilar action, as it occasionally quickens the heart beat considerably, and it slowed or stopped it in only 45 0/D of all the experiments described in my previous paper. Judging from the very slow recovery of formaldehyde-poisoned hearts, especially if they have been diminished in vitality by a previous poisoning, it is probable that the action of formaldehyde on cardiac muscle is different from that of the higher aldehydes. In my previous paper I suggested that the formaldehyde, once bound up in the protoplasm of the heart, is unable to break away again, and is only got rid of after a preliminary oxidation. In that hearts recover from the other three aldehydes as rapidly as from alcohols, there can be no doubt that these aldehydes readily break away from their combination with the

17 HEA RT OF TORTOISE. 341 cardiac substance when the heart is washed out with Ringer's solution. As to the nature of this combination we know nothing. Though paraldehyde, the polymer of acetaldehyde, and chloral and acetal, its derivatives, are powerful narcotics, acetaldehyde itself has little or no narcotic power. Neither has formaldehyde, so it seems unlikely that any of the four aldehydes investigated enter into relationship with the tissue lipoids as the alcohols do. The nature of this relationship is itself uncertain. The results described in the present paper suggest that it is a purely physical one, dependent on the solution of the alcohols in the lipoids, buit the different ratios between toxicity and concentration obtained with them, with ether, and with chloroform, do not agree with so simple an explanation. An alternative hypothesis is referred to in my previous paper. SUMMARY. When the tortoise heart is perfused with oxygenated Ringer's solution containing methyl, ethyl, propyl, iso-propyl, butyl, iso-butyl, secondary butyl, tertiary butyl, iso-amyl or tertiary amyl alcohols, its beats are depressed in about ten minutes to a definite level, and then continue unchanged for the remainder of the poisoning. The depression increases with the concentration, but for all the alcobols is relatively greater at great concentrations than at small ones. On washing out the alcohol with fresh Ringer's solution, the heart quickly and completely recovers. The toxicity of the alcohols increases with molecular weight, that of iso-amyl alcohol being 23 times greater than that of methyl alcohol. Iso-propyl alcohol is less toxic than propyl, and iso-butyl alcohol less than butyl. Secondary butyl alcohol is less toxic than iso-butyl, and tertiary butyl less than secondary butyl. With one exception the order of toxicity is the same as that found by Overton for the narcotisation of tadpoles, but the numerical relationship of the two series of values changes steadily on passing from the lower to the higher alcohols. The action of the alcohols on the heart corresponds closely with their capacity for laking red corpuscles. Formaldehyde, acetaldehyde, propyl,aldehyde and iso-butyl aldehyde depress the heart beats to a constant level which is directly proportional to their concentration, and on washing out the poison with Ringer's solution the heart recovers from the latter three aldehydes as

18 342 H. M. VERNVON. quickly-but not quite so perfectly-as from alcohol poisoning. The relative toxicities of these aldehydes are: propyl aldehyde 10, acetaldehyde 1P2, iso-butyl aldehyde 1P8, and formaldehyde 40, or show no relationship to molecular weight. All the alcohols from propyl alcohol onwards have a similar action on the rate of heart beat, but methyl alcohol and to a less extent ethyl alcohol slow or stop it much more frequently than they do, probably because of the high osmotic pressure of their solutions. Aldehydes very seldom stop the heart, and often quicken its beats considerably. The expenses of this research were met in part by a grant from the Royal Society.