THE UNIVERSITY OF ILLINOIS LIBRARY NATURAL HISTORY SURVEY 5705 ILL v4cop.4 wii^'mmmmmmmmmmimimm Return this book on or before the Latest Date stamped below. A charge is made on all overdue books. U. of I. Library MAR 2 7 m29 19' ?9 ILLINOIS BIOLOGICAL MONOGRAPHS Vol. IV October, 19 17 No. 2 Editorial Committee Stephen Alfred Forbes William Trelease Henry Baldwin Ward Published under the Auspices of the Graduate School by THE University of Illinois Copyright, 1918 by the University of Illinois Distributed July 28, 1918 THE GOLDFISH (Carassius carassius) AS A TEST ANIMAL IN THE STUDY OF TOXICITY WITH GRAPHS AND TABLES EDWIN B. POWERS'^ Contributions from the Zoological Laboratory of the University of Illinois under the direction of Henry B. Ward, No. 109 TABLE OF CONTENTS V Page Introduction 7 Materials and Methods of Study ' 7 Acknowledgments 8 Experimental Data 9 Toxity of Cupric Chloride, Cadmiimi Chloride, Ferric Chloride, Colloidal Copper, and Distilled Water 35 The Efficiency of the Goldfish as a Test Animal 40 The Effect of Temperature upon the Resistance of the Goldfish ........!.... 42 Toxicity and the Expression of Relative Toxicity .....' 42 Is the Toxicity of a Substance Due to Osmotic Pressure 46 Theory of the Poisoning Effect of Toxic Substances .....: 46 A Comparison of Curves Obtained by other Workers 52 Discussion of Data „... 54 Smnmary of Conclusions j^. 66 Bibliography 69 127] THE GOLDFISH AS A TEST ANIMAL— POWERS INTRODUCTION The need of uniformity of remedial agents has long been recognized among therapeutists and pharmacologists. When the physiological activity of an agent is associated with or dependent upon its chemical constitution which can be determined by analysis the problem becomes a simple one. On the other hand when either the chemical constitution can not be determined or the activity of the substance is dependent upon some other factor independent of its constitution the problem becomes more complex. This difficulty has been overcome with more or less success by certain physiological assay methods, the history of which it is not deemed necessary to present as Baker (1913) has given an excellent review of the Uterature up to that date. Of the methods employed at present the most favored is Fagge and Stevenson's (1866) frog method as proposed by Houghton (1898) or some modification of it. Others have been suggested: — the blood pressure method, the guinea-pig method introduced by Labor de (1884) and further worked out by Reed and Vander- kleed (1908), the cat method of Hatcher and Brody (1910), the cockscomb method of Haskell (1914), the goldfish method of Pittenger and Vanderkleed (1915), and other methods. The purpose of this investigation was to test the validity of the Pittenger and Vanderkleed goldfish method or to work out a usable modification of it. The problem then involved the development of a method and technic by which drugs or remedial agents, not readily assayed chemically, could be assayed or standardized physiologically by means of goldfish. In order to use the goldfish as test animals they must meet the following requirements: 1. They must be relatively constant in their reactions or resistance to the drug or agent to be standardized or at least there must be a constant seasonal rhythm. 2. They should be capable of being standardized with some standard of a conveniently assayable substance if seasonal rhythm is present. 3. They must be relatively sensitive to small variations in the concentration of the substance to be tested. MATERIALS AND METHODS OF STUDY A series of experiments was run with the goldfish to determine how far they would comply with the three requirements just mentioned. In order to do this the goldfish were tested with a number of substances varying in toxic activities. A number of solutions of each substance tested were pre- pared of which the concentration of each differed only 10% or less from that of the one next to it in the series. Two goldfish of known weights were then placed in two Uters of each of the solutions contained in a three liter wide- mouth bottle. The survival time of each fish in each solution was noted and recorded. The bottles were kept stoppered when volatile substances 8 ILUNOIS BIOLOGICAL MONOGRAPHS [128 were being tested. The stock of fish was kept at as nearly constant tempera- ture as possible by the addition of hot water to the aerated tap-water. The experiments were run at the same temperature as that of the stock by placing the experimental bottles in the stock tank. This eliminated any ill effect due to the sudden change of temperature (Wells 1914). By these experiments it was hoped to determine two things: (1) the fitness of the goldfish as a test animal and (2) the most adaptable range of survival time of the goldfish for pharmacodynamic assay work. The goldfish used in these experiments were Carassius carassius L. (Meek and Hilderbrand 1910), the Crucian carp or goldfish. This is the goldfish commonly sold for aquaria. C. auratus is less common and was not obtainable for this work. ACKNOAVLEDGMENTS The author desires to acknowledge his indebtedness to Dr. J. H. Beal, Director of Pharmaceutical Research, who suggested the following research and supplied the necessary funds for its conduct, as well as many helpful suggestions throughout the course of the work. The author also extends his thanks to Dr. Victor E. Shelford, in whose laboratory the experiments were made, for much valuable advice and assistance and to Dr. C. H. Sisam for suggestions on mathematical points, for the applica- tion of which, however, the author assumes sole responsibility. The author's thanks are also due to Drs. Geo. D. Beal, H. B, Lewis, T. R. Ball, and H. E. Eastlack for frequent help in chemical problems. 129] THE GOLDFISH AS A TEST ANIMAL— POWERS EXPERIMENTAL DATA In the study of the goldfish as a test animal it was deemed advisable to determine the miiformity or lack of uniformity of the resistance or suscepti- bility of the goldfish first to certain of the less toxic simple substances; then later to take up a study of the more toxic and the more complex substances. First the chlorides and nitrates of the alkali and alkaline earth metals were employed. Later the heavier metals, cupric chloride, cadmium chloride, and ferric chloride, and finally hydrochloric acid, potassium cyanide, methyl, ethyl, and isobutyl alcohols, phenol, caffeine, and p5Tidine were used. It was found that all substances tested, with the exception of CuCl2, CdCl2, and to a certain extent FeCls, have certain points in common when the toxic activity of varying amounts of the substance tested is considered. This can be best illustrated by a detailed study of one or two representative substances and by comparing these with the other substances employed. The experiments with Hthium chloride will first be considered. In the Hthium chloride experiments the resistance of the goldfish to con- centrations varying from 0.089 N. to 0.466 N. was tested. By referring to Table I it will be found that the goldfish died fairly uniformly in any given concentration of the Jithium chloride solution. For other chlorides and nitrates see Tables II to XV. The 2.5 g. goldfish (Table III) having a sur- vival time of 135 minutes and marked with an asterisk (*) was taken out of the sodium chloride solution for dead, but when placed in an HCl solution TABLE I Lithium Chloride. Temperature 21° C. December 1, 1916 Weight Survival Weight Survival Normal offish time of Normal of fish time of Remarks in grams fish in minutes in grams fish in minutes 0.466 2.8 36 0.200t 2.8 141 Daggers (t) indicate the » 3.2 41 " 3.05 179 range in which the 0.388 3.0 46 0.166 2.8 234 greatest difference in » 4.2 44 )) 3.6 310 survival times occured 0.322t 2.9 54 0.133 2.9 525 in any two solutions >j 3.2 ' 58 )> 3.4 925 with the least differ- 0.266t 2.9 ' '^2 - 0.111 3.0 1204 ence in concentrations " 3.2 96 ■ i» 3.2 1214 in which the fish died 0.222t 3.3 116 0.089 2.7 ±1320 uniformly. 3.35 112 ' 3.0 1620 10 JLUNOIS BIOLOGICAL MONOGRAPHS [130 showed very slight signs of life. All goldfish when taken out of a solution for dead were tested by placing them in a hydrochloric acid solution and if any signs of life were evidenced the goldfish was designated by an asterisk in experimental data as not dead. No corrections have been made either in this or subsequent tables or graphs since it was deemed as being within experimental error as well as within individual variation of the goldfish. For further discussion of the uniformity of the survival time of the goldfish see page 40. By a close examination of data of killing goldfish in different con- centrations of Hthium chloride (Table I) it is found that the survival time of Minutes. O.OS. 0.1. "^ 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5. 0.5SN. Figure 1. Lithium chloride. LIJM is the svu-vival time curve. The survival time of the goldfish is plotted as ordinate and the concentration of the LiCl as abscissa. The crosses (-}-) are located from actual experimental data. C.\BG is the velocity of fatality curve. The concentration of the LiCl is plotted as abscissa and the reciprocal of the survival time in minutes or the velocity of fatality is plotted as ordinate. Instead of 1 over the survival time which gives the reciprocals, 100 over the survival time is used to avoid the use of fractions. The circles (•) are located by calciilations from actual experimental data. The numbers at the left hand of the graphs represent minutes of siuvival time and those at right hand side of graphs represent imits of velocity of fatalit}\ The line P.\BF is the theoretical x'clocity of fatality curve and is a straight line. HIJK is the thedtetical survival time curve. 131] THE GOLDFISH AS A TEST ANIMAL— POWERS 11 Minutes 2040 1920 1800 1680 1S60 1440 1320 1200 1080 960 840 720 600 480 360 240 120 Minutes 1200 1080 960 840 720 600 480 360 240 120 • • H- --^ / / /' • + \ • / \^ ! • •+ + V' / • +X/ . _J ^i^ ^t- -*— — *- 100 t 10 0.05. 0.1. 0.15. 0.2. 0.25. 0.3 0.35 0.4. 0.45. 0.5. O.SSN. Figure 3. Potassium chloride 100 t 0.5. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5. 0.55. 0.6. 0.65. 0.7. 0.75. 0.8N. Figure 2. Sodium chloride the goldfish does not increase in the same ratio as the concentration of the lithium chloride decreases, i. e., the survival time of the the gold- fish is not in inverse proportion to the concentration of the lithium chlo- ride employed. At higher concentrations (0.466, 0.388, and 0.322 N.) the increase in survival time of the goldfish was less rapid than the decrease in concentration of the solution. With concentrations from 0.322 to 0.133 N. there was a more rapid increase in survival time of the goldfish in proportion to the decrease in concentration of the solutions. And finally with concen- 12 ILLINOIS BIOLOGICAL MONOGRAPHS \m trations from 0.133 to 0.089 N. there was again a less rapid increase in survival time in proportion to the decrease in concentration of the lithium chloride solutions. These points can be better illustrated by the graphic method, (Fig. 1). Let one block abscissa represent 0.05 N. LiCl and one block ordinate represent 120 minute survival time of the goldfish and plot data as represented in Table I. When the points (+) have been located on the graphs the curve LIJM can be drawn by interpolation. The crosses of this and all subsequent graphs are located from actual experimental data. The curve LIJM thus Sodium Chloride. TABLE II Temperature 21.5° C. November 19 to 20, 1916 Weight Survival Weight Survival Normal of fish time of Normal of fish time of Remarks in grams fish in minutes in grams fish in minutes 0.600 2.5 28 0 277t 2.6 ±178 >j 2.7 30 » 2.7 260 Daggers (f) as in Table 0.500t 2.7 33 0.266 2.6 ±270 I. Fish were all alive » 2.7 34 >> 2.7 296 in. 0.18, 0.15 and 0.124 0.417t 2.85 61 0.253 2.7 457 N.solutions after 10200 >» 2.9 55 »> 2.8 304 minutes when experi- 0.347t 2.6 97 0.241 2.65 478 ments were discon- » 2.7 93 )> 2.7 1290 tinued. 0.289t 2.75 114 0.201 ? 2040 )> .2.8 155 >> 3.0 9240 SoDipM Chloride. TABLE III Temperature 21.5° C. April 10, 1917 Weight Survival Weight Survival Normal offish time of Normal of fish time of Remarks in grams fish in minutes in grams fish in minutes •.• 0.600 2.7 35 0.278t 3.3 134 » 3.8 38 " 3.4 129 0.500t 2.5 35 0.263 2.4 130 » 2.9 37 » 2.5 141 Daggef S' (t)' as iii Table 0.417t 2.5 40 0.253 2.4 255 I. *Fish was not dead » 3.2 42 »> 2.5 135* when taken out of solu- 0.347t 2.3 56 0.241 2.5 295 tion. »» 2.5 68 >) 2.9 400 0.289t 2.7 115 0.200 2.4 ±1020 2.9 125 » 3.0 ±1020 133] THE GOLDFISH AS A TEST ANIMAL— POWERS 13 TABLE IV POTASSItBI CHLORroE. TEMPERATURE 20.5° C. NOVEMBER 17, 1916 Weight Survival Weight Survival Normal of fish time of Normal of fish time of Remarks in grams fish in in grams fish in - .',..-■-. , minutes minutes 0.434 2.9 14 0.192 2.85 151 )) 3.2 17 )! 2.9 282 0.378 2.9 11 0.172 3.0 207 )) 3.0 16 )> 3.1 396 0.328t 2.8 22 0.134 2.9 286 Daggers (f) as in Table )j 3.0 26 " 2.9 ±900 I. *Fish was not dead 0.286t 3.2 25* 0.122 2.8 347 when taken out of solu- » 3.3 44 5J 3.1 527 tion. 0.243t 2.9 55* 0.112 3.1 318 )) 2.8 57 )> 3.2 292 0.214t 2.9 69 0.102 3.55 279 )> 2.8 150 2.8 ±900 drawn resembles that of an hyperbola and has been designated as the survival time curve the nature of which can be better illustrated by drawing the re- ciprocal curve. In the reciprocal curve the reciprocal of the survival time of the goldfish is plotted as ordinate. The normality is plotted as in the survival time curve. To avoid the use of fractions 100 over the survival time is taken as the reciprocal. Oiie block ordinate represents one unit reciprocal. The circles (•) represent the location of the reciprocals. The curve CABG is drawn by in- terpolation and has been designated as the velocity of fatality curve. By an ex- amination of this curve it is seen that it rises very slowly from the point C to A, i.e., in low concentrations, more rapidly from A to B, i.e., in higher concentra- tions, and again there is a falling off from B to G at still higher concentrations. By a closer examination of the curve it is seen that the portion from A to B approaches a straight line and for all practical purposes this portion can be considered as such. The point C where this curve cuts the X-axis has been designated as the actual threshold of toxicity concentration, i.e., the concen- tration below which the substance mil not kill the goldfish. The portion of the curve from A to B is not a straight line but approaches a straight line in that the direction of curvature is being reversed from that of the curvature from C to A to that of the curvature from B to G (See velocity of fatality curves, Figures 1 to 21.) For a further discussion of the velocity of fatality curve see pages 48, 52. The portion from A to B of the curve CABG if considered as a straight Hne and prolonged cuts the X-axis at the point P. The curve PABF thus drawn has been designated as the theoretical velocity of fataHty curve and the point P has been designated as the theoretical threshold of toxicity 14 ILUNOIS BIOLOGICAL MONOGRAPHS [134 concentration. The curve HIJK is an equilateral hyperbola drawn from calculations and has been designated as the theoretical survival time curve. The two curves HIJK and LIJM show the variation of the actual survival Ammontom Chloride. TABLE V Temperattieue 21° C. November 24 to 25, 1916 Weight Survival Weight Survival Normal offish time of Normal of fish time of Remarks in grams fish in minutes in grams fish in minutes 0.352 3.5 29 0.141 3.25 357 » 3.7 20 j> 3.4 ±600 0.319 3.6 30 0.115 3.15 320 >i 3.8 31 )> 3.9 297 0.268t 3.45 38 0.096 3.2 547 » 4.0 37 " 3.2 ±1020 Daggers (f) as in Table 0.249t 2.8 48 0.064 3.1 ±1020 I. » 3.2 64 » 3.3 ±1020 0.224t 2.5 99 0.032 3.1 376 » 3.7 50 » 3.4 1080 0.166 2.9 3.2 283 285 Ammonium Chloride. TABLE VI Temperature 21.5° C. April 12, 1917 Weight Survival Weight Survival Normal offish time of Normal offish time of Remarks in grams fish in minutes in grams fish in minutes 0.351 2.1 17 0.160 2.8 166 » 4.3 28* " 3.7 261 0.320 2.8 26 0.141 2.3 111 Daggers (t) as in Table >f 3.0 22 " ? I. *Fish was not dead 0.268t 3.0 48 0.115 3.0 376 when taken out of solu- » 3.2 40 " 3.5 106 tion. • vi^r4*J"/*V. 0.257t 2.7 51 0.096 2.7 123 **FishJ;was alive after " 2.9 43 » 2.7 663 5040 ? minutes when 0.224t 2.8 53 0.064 2.5 ±1080 experiment was discon- » 2.8 65 " 2.8 ±1080 tinued. 0.186 2.5 141 0.041 2.5 ±1080* »j 2.8 95 ? ** 135] THE GOLDFISH AS A TEST ANIMAL— POWERS 15 time curve from that of an equilateral hyperbola. To summarize: in the lithium chloride and all toxic substances tested with the exception of CuCl2, CdCl2, and to a certain extent FeCla the following conditions were met: — 1. A concentration (the point C, about 0.073 N. Li CI), the threshold of toxicity concentration, was found below which the goldfish were not killed. 2. Just above the threshold of toxicity concentration the rate of fatality, as expressed by the reciprocal of the survival time, increased very slowly with increase in concentration of the toxic substance employed. This is represented by the portion C to A, about 0.073 to 0.19 N. LiCl, of the velocity of fataUty curve CABG (Fig. 1). 3. With higher concentrations, concentrations 100 Minutes t 1200 \ lU 1080 + \ 9 960 840 720 600 *\ \ 8 1 6 S 4 480 360 240 120 / 1 4 + ! ' '--- Tr 1 i 1 2 1 \ X 7 * • J^t-^i-^' ^-f,j„„UL,j„J_J... 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4 0.45. 0.5. 0.55. 0.6N. Figure 4. Ammonium chloride sufficient to kill the goldfish in not less than 54 to 58 minutes and not more than 234 to 310 minutes in hthium chloride solutions, the velocity of fatality increased more rapidly in proportion to the increase in concentration of the toxic substance, and this portion of the velocity of fatahty curve approaches a straight line. (The portion A to B, curve CABG, Fig. 1.) Table XXIV shows minimum and maximum survival time of goldfish and concentration of toxic substances tested where its velocity of fatality curve approaches a straight line. The daggers (f) in Tables I to XXIII also indicate the con- centrations of the substances and the survival time of the goldfish where their velocity of fatality curves approach a straight line. 4. At stUl higher concentrations (0.375 to 0.45 N. LiCl) the increase in the velocity of fatality was again less rapid in proportion to the increase in concentration of the 16 ILUNOIS BIOLOGICAL MONOGRAPHS [136 TABLE Vn Sodium Nitrate. Temperature 21° C. December 9, 1916 Weight Survival Weight Survival Normal offish time of Normal offish time of Remarks in grams fish in minutes in grams fish in minutes 0.500 2.7 61 0.260t 3.0 134 Daggers (t) as in Table )j 2.9 164 j> 3.4 274 I. 0.460 2.7 116 0.220t 2.6 171* *Fish was not dead » 2.7 118 »> 2.7 196 when taken out of solu- 0.440 2.65 148 0.180 2.6 433 tion. » 3.2 68 » 3.0 656 Fish in 0.12 N. and 0.10 0.380t 2.9 122* 0.150 2.5 593* N. solutions were alive )> 3.1 109 i> 2.9 ±840 after 1680 minutes 0.310t 3.05 133 when experiments were » 3.4 180 discontinued. Minutes 1080 960 840 720 600 480 360 240 120 \ i \ \ Vi ! A \ + 1 +1 ^5> >^»± -r' • —l- -'-'r''^\ -— ^ "T7 ~ 100 t 9 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5. 0.55. 0.6N. Figure 5. Sodium nitrate solution. See the portion B to Q (Fig. 1). 5. At the highest concentrations tested the velocity of fatality curve a second time approaches a straight line but with a less rapid increase of velocity of fatality in proportion to the con- centration of the solution than the portion of the velocity of fatality curve repre- sented by the portion A to B, curve CABG (Fig. 1). (The portion Q to G, 137] THE GOLDFISH AS A TEST ANIMAL— POWERS 17 Fig. 1.) This last point is better illustrated by the curves represented in Figures 2 to 20, These curves are logarithmic in nature. See pages 48, 51 for discussion of the theoretical equation for the velocity of fatality curve. Ammonium Nitrate. TABLE Vm Temperature 21° C. December 18, 1916 Weight Survival Weight Survival Normal of fish time of Normal offish time of Remarks in grams fish in minutes in grams fish in minutes Daggers (f) as in Table I. *Fish was not dead 0.309 1.95 29* 0.120 2.0 ±720 )) 2.4 32 >' 2.3 ±2160 when taken out of solu- 0.253t 2.3 37 0.100 2.3 ±1080 tion. » 2.7 52 " 2.4 ±1800 *iA close approximate 0.213t 2.4 78 0.080 1.7 1620 survival time. » 2.4 82* >) 1.9 1800 *^ Fish was alive after 0.173 2.1 228*1 0.066 1.7 5625 4 days when taken out >) 2.5 228*1 » 1.7 *2 of solution and experi- 0.140 2.3 371 0.053 2.2 5400*1 ment discontinued. >) ? 404 2.0 ♦s Fish were alive after 4 days in 0.033 N. Solu- tion when experiment was discontinued. 18 ILUNOIS BIOLOGICAL MONOGRAPHS [138 100 inutes t 1 20 2280 ^ 19 2160 18 2040 17 1920 1 16 [. 1800 15 1680 -t 14 1560 13 1440 12 1320 11 1200 10 1080 1 9 960 8 840 1 7 720 \ 6 600 / 5 4 3 2 480 \ + / / • 360 \ \ , / / 240 / 1 120 . IrrJ -«-• —L P ^ 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35N. Figure 6. Ammonium nitrate 139] TEE GOLDFISH AS A TEST ANIMAL— POWERS 19 Minutes 840 720 600 480 360 240 120 ^ ' \ \ • '^ /^ \ \,\ ./ w 9 ^^. ti>4 r1^ -H- 4-^ ■■ 100 t 7 6 5 4 3 2 1 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5N. Figure 7. Magnesium chloride Magnesium Chloride. TABLE IX Temperature 20.5°C. November 8 to 14, 1916 Weight Survival Weight Survival Normal offish time of Normal of fish time of Remarks in grams fish in minutes in grams fish in minutes 2.158 4.4 10.5 0.432t 3.3 23 0.298 3.2 82 » 4.1 34 jj 3.4 91 )> 5.0 30* » 3.6 86 " 0.411t 4.2 33 0.283 3.6 117 » 5.4 35 )> 5.1 174 >) 6.2 58 » 6.3 270 Daggers (f) as in Table 0.373t 3.4 40 0.270 3.7 183 I. j> 3.7 45 » 5.5 132 *Fish was not dead » 5.8 70 ;> 6.0 239 when taken out of solu- 0.360t 3.9 42 0.257 3.3 306 tion. >j 5.7 57 »> 3.7 291 » 6.1 61 » 4.3 301 0.328t 3.4 62 0.216 5.4 780 » 3.9 118 » 7.6 99 )> 4.3 1200 , 0.313t 4.2 121 0.142 ? 30420 )> 5.0 88 » ? 15180 . )j 5.5 80 )> ? 4680 20 ILUNOIS BIOLOGICAL MONOGRAPHS [140 TABLE X Calcium Chloride. Temperature 21°C. Noxtmber 21 and December 2, 1916 Weight Survival Weight Surxnval Normal offish time of Normal offish time of Remarks in grams fish in minutes in grams fish in minutes 0.583 2.6 43 0.213 2.3 136 i> 2.6 43 II 2.5 ±900 0.356t 1.9 48* 0.202 ? ±1080 >> 2.05 59 II 3.1 ±1920 Daggers (t) as in Table 0.347t 2.6 57* 0.171 2.3 2640 I. » 2.7 66* II 2.5 4080 *Fish was not dead 0.299 2.0 65 0.168 2J& 6627 when taken out of solu- » 2.2 90 i> 3.1 ±2760 tion. 0.290t 2.8 91 0.142 2.5 1391 *»Experiment discon- II 3.0 139 II 3.0 4331 tinued after 4320 min- 0.250t 2.7 182 0.140 2.9 1920 utes. Fish were all alive. II 2.9 111 II 3.4 4320 0.249t 2.3 174 0.128 *i )> 2.25 190 1) *i Strontium Chloride. TABLE XI Temperature 20°C. November 25, 1916 Weight Survival Weight Survival Normal offish time of Normal offish time of Remarks in grams fish in minutes in grams fish in minutes 0.458 3.4 50 0.289t 3.3 80 II 2.6 59 II 4.1 88 0.380 2.8 46 0.237t 3.6 168 Daggers (f) as in Table II 3.75 51 II 3.6 347 I. 0.318t 3.4 59 0.193 2.7 ±1020 II 3.5 76 II 3.4 ±1860 141] THE GOLDFISH AS A TEST ANJMAl^POWERS 21 Minutes 2760 2640 2520 2400 2280 2160 2040 1920 1800 1680 1560 1440 1320 1200 1080 960 840 720 600 480 360 240 120 Minutes 1920 1800 1680 1560 1440 1320 1200 1080 960 840 720 600 480 360 340 120 f + 1 \ ,"'. \ \^^ ^ • ^-*'' X^'r- ^*-- iczi 100 t 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0 35. 0.4. 0.45. 0.5N. Figure 9. Strontium chloride 100 t 4 ».05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5. 0.55. 0.6N. Figure 8. Calcium chloride 22 ILUNOIS BIOLOGICAL MONOGRAPHS [142 Minutes 1200 1080 960 840 720 600 480 360 240 120 + -+ ♦ 1 100 t 10 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5N. Figure 10. Barium chloride TABLE XII Barium Chloride. Temperature 20.5°C. November 22, 1916 Weight Survival Weight Survival Normal offish time of Normal of fish time of Remarks in grams fish in minutes in grams fish in minutes 0.368 ■ 3.0 38 0.143 3.35 216 >f 3.2 38 1 > 3.7 314 0.302 3.2 51* 0.119 3.3 ±420 Daggers (t) as in Table » 3.2 56 »i 3.7 ±272 I. 0.249t 3.5 43 0.107 3.5 ±720 *Fish was not dead " 3.5 80 j> 3.6 ±720 when taken out of solu- 0.208t 3.35 113 0.090 3.2 ±720 tion. J) 3.5 144 » 3.8 1020 0.172t 2.8 169 » 3.7 316 143] THE GOLDFISH AS A TEST ANIMAL— POWERS 23 Minutes 1200 1080 960 840 720 600 480 360 240 120 1 1 1- t 1 i . [ ^ li-"^ » \ \. ^ -^^ ^%^- t^r^r^ 100 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5N. Figure 12. Magnesium nitrate TABLE XIII Magnesium Nitrate. Temperature 21°C. December 15 and 16, 1916 Weight Survival Weight Survival Normal offish time of Normal offish time of Remarks in grams fish in minutes in grams fish in minutes Daggers (t) as in Table I. *Fish was not dead 0.405 2.7 33* 0.229t 2.2 107* " 2.7 46*1 » 2.9 104* when taken out of solu- 0.370t 2.6 37* 0.185 2.4 ±240 tion. " 2.7 56* )) 2.6 ±240 *iFish had been dead 0.334t 2.4 33* 0.158 1.9 ±840 possibly a few minutes. " 2.5 35* » 2.5 ±960 Fish were aUve in 0.132 0.273t 2.9 70* N. 0.106 N. 0.088 N. jj 3.4 73* and 0.07 N. solutions after 312 0 minutes when e.xperiments were terminated. 24 ILUNOIS BIOLOGICAL MONOGRAPHS [144 ^5 100 t 7 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. O.SN Figure 13. Calcium nitrate TABLE XIV Calcium Nitrate. Temperature 21°C. December 16, 1916 Weight Survival Weight Survival Normal offish time of Normal of fish time of Remarks in grams fish in minutes in grams fish in minutes 0.311 2.6 42 0.133 2.0 ±840 » 3.0 44 )) 2.1 ±840 0.281t 2.2 45 0.111 2.4 ±1200 )> 2.4 50 >> 2.4 ±1200 Daggers (f) as in Table 0.229t 2.4 75 0.088 2.4 2116 I. ») 2.4 85 " 2.2 4936 0.192t 2.6 186 0.074 2.5 2580 » 2.6 186 >> 2.0 2920 145] THE GOLDFISH AS A TEST ANIMAL— POWERS 25 100 t 4 3 2 1 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5N. Figure 14. Strontium nitrate I I / « / / • m -7XY\ — ' Strontium Nitrate. TABLE XV Temperature 21°C. December 14, 1916 Weight Survival Weight Survival Normal of fish time of Normal of fish time of Remarks in grams fish in minutes in grams fish in minutes 0.394 2,7 63 0.165t 3.5 ±300 )> 2.5 75 " 3.5 600 0.347 2.7 108 0.137 2.9 435 Daggers (t) as in » 2.9 50 " 3.4 ±990 Table I. 0.283 3.0 60 0.110 2.2 816 Fish were alive in j> 3.2 111 »> 3.5 6240 0.073 N. solution after 0.238 3.0 70 0.091 3.2 1923 5760 minutes when ex- »j 2.8 85 )> 2.8 8760 periment was termin- 0.192t 3.0 92 ated. 3.2 87 26 ILUNOIS BIOLOGICAL MONOGRAPHS [146 TABLE XVI PoTASsroM Cyantde. Temperature 21.5° C. Weight Survival Weight Survival Normal of fish time of Normal of fish time of in grams fish in minutes in grams fish in minutes 0.2X10-1 2.9 35* 0.5X10-' 2.7 118 ■) 4.5 59 » 3.0 94 0.12X10-1 3.1 61 0.375 X 10-* 3.0 138 » 3.4 96 0.25tXlO-* 3.1 146* 0.6X10-= 3.7 96 " 3.5 204 » 4.3 64 0.235t X10-* 2.9 192 0.375X10-2 3.3 67 » 3.6 163 " 3.3 68 0.22tXl0-* 2.3 128 0.2X10-= 2.8 69 >> 4.2 179 )» 4.1 97 0.217t X10-* 2.5 138* 0.1X10-= 3.6 43* >> 3.7 253 » 4.3 71 0.215t X10-* 3.8 296 0.5 X 10-' 3.4 123 » 3.8 1658 » 3.7 90 0.211t X10-* 2.8 178 0.25X10-8 3.4 98 u 3.9 928 >> 4.2 98 0.205t X10-* 2.9 2135 0.162 XlO-s 2.7 84 j> 3.8 1945 )> 3.0 84 0.195 X 10-* 2.8 505 0.135 XlO-» 3.2 84 » 3.2 2785 >> 3.5 84 0.185X10* 3.1 715 0.1 xio-!" 3.3 70* » 4.2 655 " 3.6 67* 0.175 X 10-* 2.4 1519 0.85 X 10-* 2.9 79 » 4.2 ±1020 >> 2.95 79 0.12 X 10-* 2.9 2615 0.7 X 10-* >> 2.6 2.9 82 74* » 3.0 7080 Daggers (f) as in Table I *Fish was not dead when taken out of solution Fig. 14 (at right) . Potassium cjanide. Velocity of f atalit>' curve only 0.0003N. 0.00029N. 0.00028N. 0.00027N. 0.00026N. 0.00025N. 0.00024N. 0.00O23N. 0.00022N. 0.0002 IN. 0.0002N. 0.00019N. 0.00018N. 0.00017N. 0.00016N. 0.0001 5N. 0.00014N. 0.00013N. 0.00012N. O.OOOllN. 0.000 IN. 0.00009N. 0.00008N. 0.00007N. 0.00006N. 0.00005N. 0.00004N. 0.00003N. 0.00002N. O.OOOOIN. 147] THE GOLDFISH AS A TEST ANIMAL— POWERS 27 TABLE XVII Hyrdrochloric Acid. Temperature 21.5° C. Weight Survival Weight Survival Normal of fish time of Normal offish time of in grams fish in minutes in grams fish in minutes 8.89X10-3 2.8 27* 0.31tXlO-3 3.2 353 " 3.4 34 4.0 246 7.32X10-3 3.0 22* 0.209X10-3 4.5 ±1000 )) 5.7 33* »> 5.5 ±1000 4.19 X 10-' 2.8 30 0.125X10-3 3.4 ±1200 >» 5.0 36 " 4.0 ±2520 2.09X10-3 3.0 37 8.37X10-5 3.5 ±1200 " 5.1 40 7> 5.0 ±2520 1.05X10-3 3.0 50* 6.27X10-5 5.0 ±1200 J) 5.8 70 ■' 4.8 660 0.9tXlO-3 3.2 46* 4.18X10-5 4.7 ±1200 " 4.0 72* >j 4.1 697 0.774t 3.13X10-5 5.1 ±1200 XlO-3 2.2 78* " 5.2 ±2420 j> 3.3 87 2.09X10-5 3.3 553* 0.627t )> 4.1 ±1200 XlO-3 4.7 98 1.57X10-5 5.7 584 " 4.3 110 >) 3.9 794 0.523t XlO-3 3.9 4.1 134 161 0.418t XlO-3 4.2 339 " 5.2 480* Daggers (f) as in Table I *Fish was not dead when taken out of solution Fig. 15 (at right). Hydrochloric acid. Velocity of fatality curve only 0.0104N. O.OIN. 0.0096N. 0.0092N. 0.0088N. 0.0084N. 0.008 N. 0.0076N. 0.0072N. 0.0068N. 0.0064N. 0.006N. 0.0056N. 0.0052N. 0.0048N. 0.0044N. 0.004N. 0.0036N. 0.0032N. 0.0028N. 0.0024N. 0.0020N. 0.0016N. 0.0012N. 0.0008N. 0.0004N. 28 ILUNOIS BIOLOGICAL MONOGRAPHS [14S ,4 • ^>i ^ / X^ . / =^ y ^, '^» ; : ^ <.^ 100 t 6 10. 20. 30. 40. SO. 60. 70. 80. 90cc Figure 16. Methyl alcohol (M); Ethyl alcohi 4.0 258 52.5 2.3 40* 4.0 3.6 517 »> 3.2 45 » 4.2 502 42.5t 2.0 52 2.5 2.4 245 >> 3.6 57* ji 2.5 534 Daggers (t) as in Table 32.5t 3.6 ±109 1.5 3.2 651 I. » 4.2 ±109 » 3.4 906 *Fish was not dead 25.0 3.0 206 1.0 2.9 507 when taken out of solu- » 4.2 32 » 3.8 ±906 tion. 20.0 3.4 271 0.5 3.7 581 " 3.4 421 >i 5.0 901 16.0 2.8 151 0.25 2.4 791 " 3.8 299 ■ >i 3.2 646 12.5 3.0 163 n 2.4 325 149] THE GOLDFISH AS A TEST ANIMAL— POWERS 29 Minutes YLH 100 t 11 10 9 8 7 6 5 4 3 2 1 X 10. 20. 30. 40. 50. 60. 70. 80. 90cc. Figure 17. Ethyl alcohol TABLE XIX Ethyl Alcohol. Temperature 21.5° C. 1320 1200 1080 960 i 840 i' 720 -»'. y 600 I" + B^ X ^ 480 hi* / y 360 +\ + / 240 + \ V y^ ■^ 120 ^ ^*^i__ -+ -±j-. I 0 * i Weight Survival Weight Survival cc. per 1. offish time in cc. per 1. of fish time in Remarks in grams minutes in grams minutes 87.5 2.7 20 6.0 3.2 480 j> 3.3 22 )) 3.3 475 75.0 3.1 21* >> 2.6 462 )> 3.3 25 " 4.1 582 62.5 3.0 19* 4.0 3.0 469 )) i.i 24 >) 4.5 403 51.5t 2.8 26* >> 2.9 222 >> 3.5 28 >) 3.0 290 Daggers (f) as in Table 42.5t 3.5 31 2.5 3.5 ±900 I. " 3.9 Zi* )) 4.0 ±900 *Fish was not dead 32.5t 3.3 42 " 3.2 ±540 when taken out of " 3.5 44* )) 5.0 620 solution. 25.0t 2.8 56 " 3.4 561 )) 4.0 59* >) 5.0 771 20.0t 2.5 61 1.5 2.8 593 )> 3.9 85 » 3.4 538 16.0t 3.2 88* 1.0 2.4 738 » 2.5 98* j> 2.5 465 12.5 3.6 299 0.5 2.4 738 )> 4.3 579 " 3.3 583 » 2.2 397 0.25 3.0 396 " 3.4 472 )> 3.0 r.Q4 30 ILLINOIS BIOLOGICAL MONOGRAPHS [ISO TABLE XX Iso-BuTYL Alcohol. Temperature 21.5°C. cc. per 1. Weight offish in grams Survival time of fish in minutes cc. per 1. Weight of fish in grams Survival time of fish in minutes Remarks 14. 3.2 20 6.0 2.6 43* Boiling point 102° to }> 3.4 25 '> 2.8 14* 104° C. 12.5 2.9 24 5.0 3.3 64 » 3.2 30 » 3.9 78* Daggers (f) as in Table 11.0 2.9 28* 4.0 2.8 101* I. >> 3.2 £2 )> 5.5 130* *Fish was not dead 9.0 2.9 33 3.25 3.3 464 when taken out of solu- » 3.5 36* >> 3.4 654 tion. 7.5 2.9 41* 2.5 2.8 1680 » 5.5 46 " 4.2 900 20.0 3.0 17 4.4 3.0 57* Boiling point 104° to » 4.0 18 !J 4.9 63* 105° C. 16.0 2.4 23 4.0 3.8 . 70 » 3.3 29 " 3.9 80* 12.5 3.0 29 )) 3.6 70* » 4.0 29 »> 4.5 83 9.0 3.4 35 3.62 2.3 157* »> 4.2 39* >> 3.5 162 8.55 4.2 30* 3.3 2.6 302 »> 4.5 35* » 3.9 328* 7.77 4.3 31 2.97 3.2 628 >> 4.4 40 » 5.0 302 7.07 3.3 24 2.5 2.9 664 )) 5.9 48* j> 4.0 434 6.42t 3.6 33* 1.5 3.0 270 )> 5.4 51* j> 3.9 509 6.0t 3.3 46 1.0 2.8 1425 " 4.5 43 )> 4.0 ±1140 5.85t 3.5 61 0.5 2.3 510 J) 4.4 42* " 2.9 665 5.3 3.0 59 0.25 2.3 734 )) 4.3 61* )j 4.1 419 4.85 2.5 58* » 5.4 66* 9.0 2.8 40 1.5 3.1 389 Boiling point 105° to >» 3.3 . 40 J) 4.7 703 to 107° C. 6.0 3.3 42 1.0 3.8 783 j> 3.3 42 " 4.5 663 4.0 2.8 77 0.5 4.0 390 » 3.7 ±87 )) 4.2 664 2.52 3.1 282 0.25 3.8 784 )> 4.2 262* )> 4.0 ±1200 151] THE GOLDFISH AS A TEST ANIMAL— POWERS 31 TABLE XXI Phenol. Temperature 21.5°C. Weight Survival Weight Survival g, per liter of fish time of g. per liter offish time of Remarks in grams fish in minutes in grams fish in minutes 3.45 3.2 9 0.345t 2.4 59 >> 3.7 9 J) 2.6 103 2.58 Z.3, 10 0.31t 2,9 81 )) 3.9 10 »» 3.4 53 • 1.72 2.4 13 0.26t 2.55 104 )> 3.1 13 )) 2.8 132 1.036 2.5 17 0.17 3.2 125 Daggers (f) as in Table >) 3.7 24 )> 3.5 136 I. 0.69 2.8 18 0.10 3.2 91 » 3.8 30 ») 2.9 77 0.517 2.3 32 0.07 3.3 292 )) 2.5 38 " 3.7 217 0.414t 2.0 37 0.051 ? 92 >j 3.4 51 3.4 140 TABLE XXII Caffein'e. Temperature 17°C. Weight Survival Weight Survival g. per liter of fish time of g. per liter of fish time of Remarks in grams fish in minutes in grams fish in minutes 2.22 2.4 13 0.32t 2.8 22 )> 2.9 10 »j 3.5 83 1.58 3.0 14 0.2841 3.4 94 *■ )j 3.0 IS )> 3.8 256 0.95 3.0 22 0.19 2.4 319 " 3.2 23 !) 2.4 359 Daggers (f) as in Table 0.712 2.6 23 0.095 2.8 322 I. » 2.9 25 >! 3.0 1140 0.47 2.8 23 0.047 2.8 1200 )) 3.0 50 " 3.1 1652 0.40t 2.2 42 0.019 3.0 1320 » 3.2 45 » 3.5 2760 32 ILLINOIS BIOLOGICAL MONOGRAPHS [152 ^ ,-' ^ ^ 1 1 • / 1 1 1 / • • 1 ^ ^ 1 1 y 1 / ^ 1/ ' 1 • ->» 4. 1 0.2. 0.4. 0.6. 0.8. 1.0. 1.2. 1.4. 1.6. 1.8. 2.0. 2.2. lAg. Figure 18. Caffeine. Velocity of fatality curve only. Abscissa represents g. per liter \\^-" ^^^^ ^ y^ y^' x 1 *x '/ * m W 1 • • /*, >^l 100 t 12 11 Ij 9 8 7 6 5 4 2 1 0.25. 0.5. 0.75. 1.0. 1.25. l.S«. 1.75. 2.0. 2.25. 2.5. 2.75. 3.0. 3.25. 3.5. 3.75. 4.0g. Figure 19. Phenol. Velocity of fatality curve only. Abscissa represents g. per liter. 153] THE GOLDFISH AS A TEST ANIMAL— POWERS 33 '/ V ;\/ A • .1 ^- 100 t 4 3 2 10. 12cc. Figure 20. Pyridine. Velocity of fatality curve only. Abscissa represents cc. per liter TABLE XXIII Pyridine. Temperature 17.5°C. Weight Survival Weight Survival cc. per liter of fish time of cc. per liter of fish time of Remarks in grams fish in minutes in grams fish in minutes 7.87 2.0 26* 3.0 2.7 ±720 )> 2.6 38 >) 3.1 ±720 6.75 2.4 28 2.81 3.5 180 " 3.6 58 )) 3.7 ±900 5.81 2>.-i 63 2.62 3.3 420 Daggers (f) as in Table " 3.8 61 " 4.0 ±1080 I. 4.87 2.1 68 1.87 2.4 715 *Fish was not dead when )) 2.9 78 " 2 9 610 taken out of solution. 4.5t 2.2 82 10.0*> 3.1 13* *iTemperature 21.5 C. " 4.1 83 »*i 3.1 15* Fish were suffering with 4.12t 2.5 87 5.0*1 2.2 40 gas disease. " in 140 »*i 2.5 40 3.75t 2.2 175 2.0*1 3.3 ±1020 )) 3.1 187 " *i 3.3 660* 3.19t 2.2 2.2 180 190 34 ILUNOIS BIOLOGICAL MONOGRAPHS [154 TABLE XXIV The maximum and minimum concentration of substance and maximum and minimum survival time of goldfish within the range in which the velocity of fatality curve approaches a straight line. Survival Siirvival Substance Normal time of fish in minutes Substance Normal time of fish in minutes NaCl >> 0.500 » 34 Ca(N03). 0.281 45 50 » 0.277 178 260 0.192 186 186 LiCl 0.322 54 58 Sr(N0,)2 0.192 92 87 0.166 234 310 » 0.165 300 600 KCl 0.328 22 26 KCN » 0.25 X 10-* 146 204 0.214 » 69 150 0.2 X 10-* 2135 1945 NH«C1 0.:68 » 38 37 HCl 8.99X10^ J) 46 72 » 0.224 >» 0.380 » 99 90 122 109 171 196 » 0.313X10-* 1200 2520 NaNOs » g- per L. 0.220 Phenol 0.414 37 51 NH,N03 0.253 37 52 0.259 104 132 >> 0.213 » 78 82 Caffeine 0.396 42 45 MgCl, » 0.432 » 0.313 23 30 34 61 » 0.285 >> 256 94 » » cc. per L. >» 88 80 Methyl Alcohol 42.5 52 58 CaClj » » 0.356 » 0.249 48 59 174 190 Ethyl Alcohol 25.0 51.5 206 326 26 28 SrCU » 0.318 59 76 16.0 » 98 88 0.237 J) 168 347 Iso-butyl Alcohol 6.42 33 51 BaCl, 0.249 >> 43 80 » >> 5.85 » 61 42 » » Mg(N03)2 » 0.172 » 0.370 169 316 37 56 Pyridine 4.5 » 3.187 » 82 83 180 190 0.229 107 104 155] TEE GOLDFISH AS A TEST ANIMAL— POWERS 35 TOXICITY OF CUPRIC CHXORIDE, CADMIUM CHLORIDE, FERRIC CHLORIDE, COLLOIDAL COPPER, AND DISTILLED WATER It has already been pointed out that the toxic activity of various concen- trations of CuClz, CdCk and FeCls did not follow the same general law as did all other substances tested. The toxicity of the CuCl2 at very high concen- TABLE XXV Ctjpric Chloride. Temperatuee 21''C. Weight Survival Weight Survival Normal offish time of Normal offish time of Remarks in grams fish id minutes " in grams fish in minutes 0.66 2.0 78 S.7X10-' 3.2 62* >) 2.7 85 >j 3.2 137 0.58 2.6 83 4.5X10-3 2.5 58 )) 2.7 84 >> 2.6 53 0.43 3.0 85 2. 6 X 10-3 3.5 117 " 3.1 85 » 5.0 128* 0.33 2.0 45* 2.8X10-" 2.8 59 " 2.2 39 >i 2.8 69 0.25 2.2 38 2.3X10-3 2.8 83 " 3.0 30 " 3.3 80 0.19 1.8 44* 1.8X10-3 3.3 117 >! 1.95 49* »> 3.5 100 0.15 2.8 92 1.2X10-3 3.2 90 " 3.2 52* » 4.3 97 0.11 2.7 96 9.1X10-* 3.2 100 *Fish was aUve^when " 3.0 70 » 3.5 140 taken out of^solution. 0.068 1.9 85 4.5 X 10-* 2.9 120* )> 1.9 87 » 2.9 136 0.034 1.7 133 2.2 X 10-* 3.8 112 " 1.9 118 » 4,1 116 0.023 3.5 44* 1.1X10-* 3.7 158* " 4.0 56* " 3.8 112 0.017 2.1 60* 0.56 X 10-* 4.1 125 » 2.3 66 » 4.2 260 » 2.5 80 0.19 X 10-* 3.0 214 » 3.8 63 » 3.3 ±300 0.011 2.8 52* 0.13 X 10-* 3.5 ±315 )) 3.2 51* j> 3.9 367 9.1X10-* 2.5 146 0.34X10-* 3.9 117* » 2.5 164 M 3.9 208 7.9X10-* 2.9 115* O.llXlO-s 2.8 ±300 j> 3.2 59* >» 3.9 403 5.7 X 10-' 3.4 70 0.28X10-* 3.1 216 » 3.5 70 >> 4.4 430 36 ILUNOIS BIOLOGICAL MONOGRAPHS [156 TABLE XXVI Cadmium Chloride. Temperature 21.5°C. Weight Survival Weight Survival Normal offish time of Normal offish time of Remarks in grams fish in minutes in grams fish in minutes 0.157 1.95 68 0.92X10-3 3.2 167 )) 2.0 70 >j 4.5 177 0.1191 1.6 59* 0.42XlO-» 3.0 169 » 2.0 61 " 4.3 204 0.1036 1.7 61 0.24X10-' 3.8 272 » 2.3 62 » 4.0 208 0.0903 1.65 59 0.12X10-5 3.0 163 » 1.75 39 >) 4.0 165 0.078 1.9 43 6.6XlO-» 2.8 148 )> 2.0 42 » 2.9 189 0.065 1.9 54 3.7X10-» 2.7 215 " 2.0 57 » 4.1 292* 0.057 1.9 59 1.8X10-* 3.0 325 >» 1.9 60 J) 4.1 126 0.048 1.7 75 1.1 X 10-* 4.4 349 >) 2.1 77 )) 4.5 400 0.043 1.7 73 0.74X10-' 3.0 248 » 1.9 67 »> 3.0 335 *Fish was not dead 0.037 1.5 41 0.37X10-' 3.9 360 when taken out of solu- j> 1.9 82 » 4.0 349 tion. 0.032 1.65 78 jf 3.6 199 » 1.7 81 )j 3.6 224 0.024 1.8 120 0.18X10-' 4.0 311 j> 2.0 121 »» 4.7 261 0.021 1.6 93 >> 2.9 237 " 1.75 100 » 3.8 349 0.018 1.9 102 9.2X10-7 3.8 335 » 2.7 118 )> 4.3 373 0.015 1.6 164 5.2X10-7 4.1 286 J) 2.15 136 » 4.6 316 6.6X10-' 2.0 132 2.6X10-7 3.6 445 j> 2.9 130 » 4.6 353 » 3.7 206 1.3X10-7 3.4 635 » 3.7 216 » 3.7 635 3.7 X 10-' 1.9 115 0.74X10-7 4.7 861 » 2.0 135* >j 5.5 ±960 f> 3.4 233 0.37X10-7 3.7 442 i> 3.6 295 » 3.8 479 1.8X10-* 3.4 193* 0.18X10-7 3.9 ±1080 » 3.7 206 » 4.3 523 157] THE GOLDFISH AS A TEST AN IMAL— POWERS 37 trations (0.66 to 0.25 N. See Table XXV) increased with decrease in concen- tration. The shortest survival time of any goldfish was 30 minutes. From this point (0.25 N.) the toxicity of the CuCl2 decreased with decrease in con- centration but the decrease was very slight as compared to the dilution until the maximum survival time of the goldfish was 430 minutes. With the CdCl2 there was also an increase in toxicity with a decrease in concentration with the highest concentrations tested (0.157 to 0.078 N. See Table XXVI). The two goldfish died in 42 and 43 minutes in 0.078 N. CdCU. From this point TABLE XXVII Ferric CHLORmE. Temperatxjke 21.5° C. Weight Survival Weight Survival Normal of fish time of Normal of fish time of Remarks in grams fish in minutes in grams fish in minutes 0.284 2.1 28* 8.32X10-' 1.8 89 >) 2.3 29 » 1.9 109 0.213 1.5 26* )) 3.3 247 j> 1.7 28 >j 3.6 115 0.159 1.7 26* 3.88X10-" 2.8 101 )) 2.2 31 " 3.2 127 0.119 1.5 38* » 2.1 121 )> 1.8 32 )> 2.1 129 *Fish was not dead 0.0693 1.7 33* 1.94X10-3 3.0 131 when taken out of solu- )) 3.0 43* )> 3.4 113 tion. 0.0554 1.9 46* LlXlO-i" 3.4 164 )> 2.1 38 " 3.6 180 0.0332 1.8 56 5.5 X 10-* 3.3 450 >» 2.2 59 » 3.9 723 0.0166 1.7 84 2.8X10-* 2.7 790 )> 2.1 74 " 3.4 ±1080 )) 3.3 89 1. 66X10-* 3.0 ±1200 » 3.7 83 )> 3.2 ±1200 the toxicity decreased very slowly in proportion to the dilution of the solution until the maximum survival time of the goldfish was 1080 minutes inO.lSX 10"^ N. The toxicity of the FeCls, unlike CuCl2 and CdCl2, decreased pro- gressively with the dilution of the solution from the highest concentration (0.284 N. See Table XXVII) to the lowest concentration (0.166X10-3 N.) tested when the survival time of the goldfish reached a maximum of 1200 minutes. The variations of the toxicities of these three salts from that of other substances tested can possibly be explained by the fact that they form colloidal 38 ILUNOIS BIOLOGICAL MONOGRAPHS [158 solutions and that colloids are very toxic and tiiat the amount of coUoid in the solution is not in proportion to the concentration of the salt in solution. Miyake (1916) foiind 1/75000 N. AICI3 was toxic to the growth of rice seed- lings which he attributed to colloidal AICI3. Robin (1914) showed that the toxic activity of colloidal sulphur was much more intense than that of other forms of sulphur. The maximum survival time of the goldfish in ordinary distilled water was about the same as that in the most dilute CUCI2 tested. Mengarini and Scala have shown that nimierous metals, like copper and alumi- niun, form colloidal solutions in distilled water at room temperature. The toxicity of the distilled water, since it was distilled in copper stills with block- tin leads, was due possibly to the colloidal CuO X H2O and Sn02 X H2O. Ringer (1883) foimd that goldfish would five from 46 to 54 hours in distilled water. Bullot (1904) foimd that the survival time of Gammarus was shorter in water TABLE XXVin Water Distilued in a TABLE XXIX TABLE XXX COPPEK SinX WITH WaTF.R DTRTn,T.F.P IN Colloidal Copper Block-Tin Leads Glass Temperatxtee 21.5° C. Tfmfekatuke 21.5° C. Weight Survival time Weight Survival Weight Survival of fish offish offish time Exp. offish time offish in gram.s in minutes in grams in days m grams in minutes 3.4 352 2.7 41 1 4.3 900 3.8 512 ? 51 1 4.6 900 3.9 597 ? 30 2 4.0 2880 4.3 367 3.2 37 2 4.9 2820 distilled in copper than in ordinary glass, Jena glass, quartz glass, and plati- num. The latter were all toxic to about the same extent. Three sets of experiments were nm to test the supposition that a colloidal solution was responsible for the short survival time of the goldfish in ordinary distilled water. 1. The goldfish Uved 36 days in distilled water after an electrolyte (0.025 N. NaCl) had been added to precipitate the colloid. 2. The goldfish Uved 30 to 52 days in water distilled in ordinary glass. 3. A colloidal solution of copper was prepared in water distilled in ordinary glass by arcing copper wires under the surface of the water. To one quantity of water a smaller amount of colloidal copper was added and the goldfish lived about 2820 minutes. To another a larger amoimt of colloidal copper was added and the goldfish lived only about 900 minutes. Thus the greater toxic 159] THE GOLDFISH AS A TEST ANIMAL— POWERS 39 activity of water distilled in copper stills with block-tin leads and possibly also the CuCl2, CdCl2, and FeCls solutions is probably due to the presence of a colloid. This is in keeping with Locke's (1895) results in which he showed that distilled water may lose its toxic activity by long boUing, and especially when brought in contact with sulphur, carbon, manganic oxide, cotton, wool, silk, and other substances, and those of Wells (1915) who found distilled water no more toxic than tap-water so long as the distilled water was sUghtly acid. 40 ILUNOIS BIOLOGICAL MONOGRAPHS [160 THE EFFICIENCY OF THE GOLDFISH AS A TEST ANIMAL In all substances tested the goldfish died fairly uniformly in any given concentration where the survival time did not exceed five hours. See Tables I to XXIV for data of experiments in which two goldfish were killed in each solution tested. The variation of the survival time is better shown by data of experiments given in Tables XXXI and XXXII in which a number of fish were killed in each solution. The broken line in Figure 21 is a graphic repre- sentation of the survival times of the 8 goldfish killed in 0.272 N. LiCl solu- TABLE XXXI The Variation in the Survtval Time of the Goldfish. Temperatttre 20° to 21° C. 0.272 N. 0.229 N. 0.222 N. 0.231 N. Lithium Ammonium Ammonium Potassium Chloride Chloride Chloride Chloride Weight Survival Weight Survival Weight Survival Weight Siurvival offish time of fish offish time of fish of fish time of fish of fish time of fish in grams in minutes in grams in minutes m grams in minutes in grams in minutes 1.9 79 2.3 49.5 2.5 51 1.5 17 2.0 78 2.4 49.5 2.6 62 2.3 51 2.2 88 2.6 52 2.65 71 2.5 49 2.2 89* 2.6 74 2.7 51 2.5 43 2.25 87 2.8 53 2.9 78 2.55 69 2.3 81 2.9 54 2.9 85 2.6 69 2.4 70 2.0 88 2.9 126 2.8 47* 2.4 86 3.0 3.1 69 104 3.0 69 2.8 48 *Flsh was not dead when taken out of solution tion, Table XXXI. One block of abscissa represents an individual goldfish and one block ordinate represents 100 minutes survival time of the goldfish in the 0.272 N. LiCl. The goldfish with the shortest survival time is repre- sented first and the second next and so on to the goldfish with the longest survival time. Thus the deviation of the fine from the horizontal represents the variation of the survival time of the goldfish in the given solution. Each of the other lines of the graphs in Figures 21 and 22 represents the survival times of individual goldfish in a definite concentration of a particular sub- stance tested. In all graphs one block ordinate represents 100 minutes sur- vival time and one block abscissa represents an individual goldfish. In these 161] THE GOLDFISH AS A TEST ANIMAL— POWERS 41 graphs the weight of the goldfish is not taken into consideration. The gold- fish with the shortest survival time is always considered first, the second next, and so on. Small variations in the weight of the goldfish had very httle effect on their survival time, though in general the smaller died first. Of all goldfish killed the smaller died first in the ratio of 2.1 to 1. Where there was a greater variation in size the smaller died first in a much greater proportion. For variations of the survival time of the goldfish as compared to their relative weights see Tables I to XXXVI. In all the tables the smallest goldfish is placed first, the next smallest second, and so on. Minutes 200 100 ^, aa, SrTI — —^^r^ B.a Tin '"^ _.,. -J ■•^— _:/J:>^'" r"*^^^' 11111111 Figure 21. Graph showing the efl&ciency of the goldfish as a test animal, represents the survival time in minutes and abscissa represents individual fish Dot and dash line =0.284 N. CaCU Solid line = 0.274 N.SrCk Dotted line = 0.2 14 BaClz Broken line =0.272 N. LiCl Ordinate Minutes 200 100 • • Nua • NHa Mga pff^AW- — -_-l rr.:~:.. ^^-^"^ 11111111 Figure 22. Graph showing the efficiency of the goldfish as a test animal Broken line = 0.222 N. NH4CI ^ Dot and dash line = 0.229 N. NH4CI Solid line = 0.3 18 N. MgClj Dotted line =0.231 N. KCl 42 ILUNOIS BIOLOGICAL MONOGRAPHS [162 THE EFFECT OF TEMPERATURE UPOX THE RESISTANCE OF GOLDFISH A few experiments were run to determine the effect of temperature on the toxic activity of a substance on the goldfish. It is shown by Tables XXXIII, XXXIV, and XXXV that the survival time of the goldfish in a constant concentration of a substance is lowered by a rise in temperature. This agrees with other workers on toxicity. See bibliography for citations. No attempt was made to find the relation between temperature and the toxic activity of a substance. See discussion of Warren's (1900) temperature toxicity curve, page 52. TOXICITY AND THE EXPRESSION OF RELATIVE TOXICITY The relative toxicity of the respective elements has long interested the chemist. The views of workers on the toxic activity of the elements have varied from time to time to fit any newly discovered physico-chemical con- ception of the properties of the elements. James Blake (1883, 1887) associated the physiological action of the elements of the isomorphous groups with their atomic weights. Botkin (1885) suggested a relation between the toxic acti\'ity of the elements and their position in the periodic sj'stem. Pauh (1903), Kahlenberg and True (1896), Kahlenberg and Austin (1900), Loeb (1902), and Mathews (1903, 1907) held that the toxic activity of a substance is dependent either upwn the free electrical charge or the atom itself while in the atomic state. Richet (1881) and Rabuteau have suggested that toxicity is a function of the solubiUty of the substance. Mathews (1904) has related toxicity to the solution tension of the ion. While, finally, Bert (1871), Garrey (1905), Kfizeneck^ (1916) and others claim that the toxic effect of at least certain elements to fresh water animals is due primarily to osmotic pressure. After all these and other suggestions such as coagulation of the protoplasm, ionic combination, either physical or chemical, with the protein of the proto- plasm, and the change of permeability of the cell membrane, the cause of toxicity is still an open question. Aside from the divergence of opinions as to the cause of the active properties of the elements there has been no absolute agreement on relative toxicity itself. In this work some attempt has been made to determine the relative toxic values of certain of the alkah and alkaline earth metals when in combination with CI and NO3. First it is obvious that it is necessary to have some standard of measure of the elements themselves. The elements will be arranged differently from the same data when considered by actual weight of the element, actual weight of the salt, molecular concen- tration, and normahty. This of itself explains certain of the disagreements among workers, though after all the elements are reduced to a common standard of measure there is still a wide divergence in relative toxicity as reported by different workers. Osterhout (1915) has pointed out that " the relative toxicity of two substances may depend very largely on the stage of the reaction at which the measiirement is made"; i.e., the criterion employed. Osterhout objects 163] THE GOLDFISH AS A TEST ANIMAL— POWERS 43 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5. 0.55. 0.6N. Figure 23. Superimposition of survival time curves of the alkali and alkaline earth metals as chlorides showing crossing over of curves, thus showing varying relative toxicities of the group when different definite survival periods are taken as a criterion. to the utilization of the death point as a criterion and points out the fact that "it is impossible to determine the precise moment of death." He has shown that the death curve "approaches the axis asymptotically." He suggests that it may be assmned that as the reaction proceeds certain phenomena appear at definite points on the curve. He took as his criterion a point on the curve at which he said that the organism was "half dead." The writer has found the same objections pointed out by Osterhout but has come to the conclusion that the death point with certain precautions is a more exact cri- terion for an end point in the case of the goldfish than the loss of equilibrium or irritabihty. Aside from the objections raised by Osterhout there is a still more serious objection to any attempt to determine relative toxicities by comparing the concentrations of the solutions or the amount of the substance necessary to produce death or the appearance of any physiological phenom- enon in any arbitrary fixed time or the concentration necessary to just cause 44 ILUNOIS BIOLOGICAL MONOGRAPHS [164 death. The relative toxic activities of the substances will vary according to the criterion used. Only a glance at figure 23 is required to convince one that there will be a marked rearrangement in the relative toxic values of the elements when different survival times are used as criteria. For example, if a 120 minute survival time is chosen as a criterion the relative toxicities of the elements as chlorides with normahty as the standard of measure will be arranged from the most toxic to the least toxic in the following order: — Ba, K, Li, Sr, Ca, Mg, and Na, and with a 60 minute survival time they will be arranged as follows: — ^K, Ba, Ca, Li, Sr, Mg, and Na. Thus there is a marked rearrangement of the relative positions of the elements with a change from a 120 minute siu-vival time as a criterion to that of a 60 minute survival time. It is the opinion of the writer that it is the employment of different criteria that is responsible more than any other one thing for the differences of opinions of workers on the toxic values of the elements. The fact is not denied that the elements may have different relative activities toward different organs of the same animal as well as toward different animals (Kahlenberg and Mehl 1901). By a study of the survival time curves shown in Figure 23 it is seen that the activity of each element is a law unto itself, and thus there is no one expression that will include all specificities of the salts or toxic substances. But a standard criterion may be derived. From a study of the velocity of fatality curves it is seen that there are two variables and that these vary independently of each other. 1. The concentration necessary to just kill the goldfish varies with the element. 2. The acceleration of the change of direction of the survival time curve, i.e., the increase in the velocity of fatahty is different in each of the salts and does not necessarily have any relation to the amount of the salt neces- sary to just kill the goldfish. The value of the first factor is shown by the position of the point C and the second by the slant of the velocity of fatality curve. But since the slant of the velocity of fataUty curve is not imiform for all concentrations of a substance the theoretical velocity of fatahty curve will be considered. This curve bears the same relation to the physiological activity of the substance as the true velocity of fatahty curve. Thus the relative toxicities of substances due to the first factor can be measured by the reciprocal of a, the distance P from the origin O. That due to the second can be meas- lu-ed by the tangent of the angle XPB, 0. (See Figiure 1.) Both these factors must be taken into consideration in a criterion or an expression representing the relative toxicity of a substance. The tangent 6 increases as the toxicity increases with respect to the slope of the theoretical velocity of fatahty curve, i.e., the rapidity with which the activity of the substance increases with in- crease in concentration and the distance of the point P from the origin O or a decreases as the toxicity increases with respect to the theoretical threshold of toxicity concentration, i.e., as the theoretical threshold of toxicity concen- tration decreases. Thus it is seen that these two factors are the reciprocal of each other. This relation can be expressed mathematically by the equation^ 165] TEE GOLDFISH AS A TEST ANIMAL— POWERS 45 T (toxicity) =V tan. e This equation cannot represent the absolute or the exact relative toxic value of a substance since it is based only on a portion of the velocity of fatality curve, the portion A to B, Figure 1, and the assumption that this portion and the two extremes, i.e., the two extremes C to A and B to G, follow the reciprocal of an equilateral hyperbola which is not in keeping with experimental data or a curve drawn from the hypothetical equation repre- senting the true velocity of fatality curve. See pages 48, 52 for a discussion of this hypothetical equation. However this expression has been chosen to represent the relative toxic value of a substance rather than the curve itself as suggested by Osterhout (1915), as the latter leaves the formation of a criterion to the reader. It has the advantage in that the relative toxicity can be given a numerical value and is a natural criterion and not an arbitrary one. This equation, as has already been pointed out, does not express the specificity of the toxic activity of the salt, which can be shown only by the curve itself, as Osterhout has suggested, or by taking into consideration the factors which go to make up an equation which represents the ciuve itself (See hypothetical equation page 49). Taking the above equation as repre- senting the relative toxicities of the alkah and alkaUne earth metals, as chlorides and nitrates, the toxic values are found as given in Table XXXVT. These values are comparable in general, not so much in comparative numerical values but in position, to values obtained by other workers after the measurements of the elements are reduced to a common standard. See bibliography for citations. However, among all workers there is more or less difference of opinion. TABLE XXXVI Relative ToxiaiiES of the Alkali and Alkaline Earth Metals when in Combination with CI and NO3 Substance Relative toxicity NaCl 1.24 LiCl 1.53 KCl 2.3 NH4CI 2.87 CaCU 1.56 SrCli 1.57 MgCU 1.87 BaCU 1.94 NaNOa 1.00 NH4NO3 1.95 Mg(N03)2 1.75 CaCNOs). 1.92 Sr(N03)2 2.58 46 ILUNOIS BIOLOGICAL MONOGRAPHS [166 IS THE TOXICITY OF A SUBSTANCE DUE TO OSMOTIC PRESSURE? A series of experiments was run with d-glucose (Table XXXVII) to deter- mine the effect of osmotic pressure on the goldfish. When the molecular concentration of the glucose is divided by two to make its osmotic pressure comparable to that of a normal salt solution, it is seen that the toxicity of the glucose is still below that of the particular salt to which it is com- pared. This difference is even more striking when the fact that the salts are not completely ionized is taken into consideration and that this is more than sufficient to allow for complete ionization of salts of bi and tri valent acids and bases. The improbability that the toxicity of a substance is due to osmotic pressiu-e is further emphasized by the fact that the toxic activi- TABLE XXXVII d-Glucose. Temperature 21.5°C. Weight Survival Weight Survival Mole offish time of Mole of fish time of • in grams fish in minutes in grams fish in minutes 1. 2.3 158 0.15 2.4 5700 » 2.7 . 77 )> 2.3 7140 0.6 2.5 311 0.075 3.0 5640 » 3.8 356 " 3.3 7860 0.4 2.7 ±1200 0.05 2.8 2880 " 3.2 ±2640 )) 4.7 4560 0.2 2.2 ±8400 0.025 2.8 2880 » 2.9 ±8400 ties of substances like potassium cyanide, hydrochloric acid, the alcohols, caffeine, phenol, and pyridine are not in proportion to their molecular concen- trations. THEORY OF THE POISONING EFFECT ON TOXIC SUBSTANCES Osterhout (1915) has shown in a very interesting paper that death in the Laminaria, when placed in NaCl solution of the same conductivity as that of sea-water, as measured by the fall of electrical resistance, does not follow a straight line when electrical resistance is plotted as ordinate and time as abscissa, but foUows a curve which, he points out, "follows the course of a monomolecular reaction. " The effect of poison on protoplasm might be compared to a system of chemical reactions which finally result in the death of the organism. It has been shown outside the living organism that KCN retards oxidation. Kastle 167] THE GOLDFISH AS A TEST ANIMAL-POWERS 47 and Loevenhart (1901) pointed out the fact that the activity of oxidase of the potato was retarded by KCN. Shafer (1915) found that hydrocyanic acid affects catalases, reductases, and oxidases in insect tissues. Mathews and Walker (1909) showed that very small amounts of KCN is sufficient to check or prevent the spontaneous oxidation of cysteine to cystin both in neutral and alkaUne solutions. Loevenhart and Kastle (1903) have shown that hydrocyanic acid inhibits the catalytic activity of solutions of metallic colloids, Claud Bernard (1857) noted that venous blood of an animal poisoned with hydrocyanic acid takes on an arterial hue. Geppert (1889) showed venous blood to have a higher oxygen content than normal in mammals poisoned with hydrocyanic acid. Richards and Wallace (1908) found that protein metabo- Usm is increased in a dog by poisoning with KCN due in part to the retarding effect on oxidation by the KCN. Loeb and others have shown that oxygen requirement is decreased and that this decrease is greater the greater amount of KCN used. All have shown that KCN in very small amounts acts as a stimulus to oxidation both in and out of the animal organism. Moore and Moore (1917) have shown that the maturity of fruit was two weeks earher and that the total yield of fruit was larger on tomato plants fumigated with HCN than normal plants. Woodworth (1915) showed that scale insects eggs when fumigated with hydrocyanic acid in amounts not sufficient to kiU them hatched earlier than the normal eggs. Townsend (1901) has found that there is an increase in germination in seed fumigated with hydrocyanic acid. And Zieger (1915) in turn has shown that the activity of the catalase is in proportion to the general metabolic activities in animals. While Child (1915, p. 66 and citations) and others have shown that very small amounts of KCN increase the rate of metaboHsm while larger amounts decrease metaboHc processes, Mathews (1916, p. 813) suggests that since cysteine oxidizes itself and has many points in its oxidation which resemble respiration there is more than a superficial connection between the oxidation of cysteine and the respiration of the cell. The most favorable hydrogen ion concentration for oxidation is the same as that of protoplasm and both cysteine and protoplasm are poi- soned by many of the same substances, as nitriles, the cyanides, acids, and heavy metals. Their oxidations are catalyzed or hastened in the same manner by iron, arsenic, and certain other agents. Burge and Burge (1914, 1915) have shown that digestion of a dead round worm in activated pancreatic juice does not take place so long as its body wall is constantly permeated with nascent oxygen. They conclude that the oxidative processes of hving parasites enable them to withstand the enzymes by oxidizing the enzymes in contact with them. LiUie (1902) advances the theory that the tissues are protected from autolysis by oxidizing the enzymes in contact with them. Hofmeister (Mathews 1916, p. 83), Verworn (1909, p. 53), Hopkins (1914), and others have held that there is no vital matter in the cell and that the chemical trans- formations do not involve any large biogene molecule but only relatively simple compounds in solution. But Pfliiger, Du Bois-Rejonond (Mathews, 48 ILLINOIS BIOLOGICAL MONOGRAPHS ^ [168 1916, pp. 842-844), Driesch (1909), and Sir Oliver Lodge (1913) suggest that there is an organized force, entelechy, or vital force and that so long as the cell is ahve this force regulates the metabolism but as soon as this force is destroyed or inhibited the metabolic processes cease, i.e., the machinery no longer works and the cell dies. Robertson (1908) suggests that the normal rate of growth is an autocatalytic process. And finally Troland (1917) defends the thesis that all biological enigmas as he calls them are explainable on the basis of enzymes either autocatalytic or heterocatalytic. Thus it appears from either the mechanistic or the vitaUstic view and experimental data that metaboUsm can be conceived of as a self-perpetuating mechanism and that when metabolism is interfered with beyond a certain Umit, this mechanism or metabolism is progressively depressed or inhibited. That is, metabolism continues within a certain hmit at a normal rate; enzymes, excitors, or anti- bodies inhibiting autolysis are Uberated or generated which stimulate meta- boUsm or allow metabolic activities to continue at a normal rate. But as rapidly as the rate of metabolism is reduced below this normal rate, beyond a certain limit, the rate of Hberating or generating enzymes or antibodies is reduced in proportion to the reduction of the rate of Hberation or generation of the cell enzymes or antibodies. In other words, we are dealing here with the rate of inhibition of a process, metabolism, the rate of which is uniform imder any fixed set of normal conditions. One of the factors of a normal con- dition is a normal rate of metabolism. Thus when the rate of metabolism is reduced below the normal the conditions become abnormal. This in turn becomes an inhibiting factor which increases progressively with the decrease in the rate of metabolism, i.e., it becomes self -inhibiting. This portion of the effect on the speed of inhibition of metaboHsm is thus proportional to the product of the amount of reduction in the rate of metabolism at any time t, and the reduced rate of metaboUsm at that time. This can be compared in the speed of the reduction of the rate of metaboUsm to the velocity of an autocatalytic reaction and can thus be expressed mathematicaUy dz — = K2z(M-z). That is, at any given time, /, the speed of inhibition of meta- dt boUsm is in proportion to the product of the reduced rate of metaboUsm and the rate of metaboUsm at the given time. M= normal rate of metaboUsm under any fixed set of conditions, z= amount of reduction of rate of metaboUsm at any given time, K2=a constant representing the efficiency of the reduced rate of metaboUsm in inhibiting metaboUc processes, i.e., in inhibiting the activation or action of the metaboUc enzymes or antibodies or representing the liberation of autolytic enzymes, and /= time. The amount of reduction z in the rate of metaboUsm after any time t is due not only to the auto- inhibitory process but also to the continuous action of the protoplasmic poison introduced. But the equation above takes into consideration only the reduction due to the autoinhibitory process, and does not take into 1691 THE GOLDFISH AS A TEST ANIMAL— POWERS 49 consideration the continuous action of the poison on the rate of reduction of metabolism. But at any given time the speed of reduction of the rate of metabolism due to the continuous action of the protoplasmic poison is in proportion to the amount of poison acting at the given time and the actual rate of metabolism, i.e., (M-z). Thus the speed of the inhibition of the rate of metabolism by the continuous action of the protoplasmic poison can be compared to the velocity of a simple mass action. And since the fish is in a solution the volume of which is large as compared to the volume of the fish and since the poison is continuously diffusing into the protoplasm to the point of equilibrium, the amount of protoplasmic poison can be considered as con- stant throughout the entire process. The equation representing the speed of reduction of rate of metabolism by the continuous action of the protoplasmic dz poison then becomes — =KiX(M-z). Ki = the efficiency of the protoplasmic poison in reducing the rate of metabolism and X=the amount of poison em- ployed and is a constant for any definite amount of poison employed and appears in the equation only for the sake of comparing the action of different amounts of protoplasmic poison employed in any two experiments. Since both of the processes indicated by the above equations are acting simultaneously to reduce the rate of metabolism, the actual rate of reduction is the sirni of the two left members. If we now denote this actual rate of dz dz reduction by ^-, then -r- = KiX(M-z)-f-K2z(M-z). at dt Integrating we have t+C (constant of integration) = 1 , K1X+K2Z MK2+K1X ° K2(M-z) 1 TC X But when t = 0, z = 0. Then C = logg MKa-fKiX K2M By substituting the value of C and transforming v/e have: 1 , K1X+K2Z 1 , KiX MK2+K1X K2(M-z) MK2+K1X K2M 1 - ,M , K2MZ 1 , MK2+K1X M-a Ki(M-z) X _ 1 ^ MK2+K1X Or -= Y = t , , M , K2MZ 1 . loge ( \- • ) M-Z Ki(M-z) X The reciprocal of the survival time of the goldfish — is represented as ordi- nate in the graphs of the velocity of fatality curves, thus - = Y. 50 ILLINOIS BIOLOGICAL MONOGRAPHS [170 Any mathematical expression of the velocity of fatality curve must comply both with actual experimental data and with theoretical demands. 1. The mathematical expression should show that the velocity of fatality increases very slowly with increase in concentration when very low concentrations of the poison are used. 2. It should show that the velocity of fatality increases very rapidly with an increase in concentration when higher concentrations of poison are used. 3. Finally with still higher concentrations of the poison it should show a less rapid increase in velocity of fatality with increase in concentration. 4. At very high concentrations of poison it should show that the velocity of fatality curve approaches a straight hne. See discussion page 43. --,, -. . . ^, ... 1 K2M+KxX When X vanes m the equation \ = — = t , / ^I , K2MZ 1 , loge ( • — ) M-2 Ki(M-z) X and all other factors on the right hand side of the equation remain constant, it is found that when X is very small so that KiX is very small as compared to KiM the numerator (K2M+K1X) approaches a constant. Thus the value of — or Y is controlled by the reciprocal of the logarithm of a number which is controlled by 1 r^^ • 1 /M , K2]Mz 1 , „„ the reciprocal of X, i.e., loge ( 1 — )• When X is neither very M-z Ki(jM-z) X large nor very small as compared to the other factors neither the numerator nor the denominator approaches a constant and Y is controlled by the increase of X in the numerator and the reciprocal of the logarithm of a number which is increased by the reciprocal of X. Finally, when X is very large the denomina- , , M , K2MZ 1 X , M ^ tor loge ( 1 ) approaches a constant, becomes very large M-z Ki(M-z) X M-z as compared to — - — '- Thus Y is controlled by the increase of X in Ki(Mz) X the numerator (K2M+K1X) since KiX is large as compared to K2M. From these three conditions it is seen that the value of Y at first increases very slowly, being controlled by the reciprocal of the logarithm of the reciprocal of a number. After this it increases more rapidly, being increased by the same factor as the first and in addition is increased by a multiple of the number itself, i.e., numerator (K2M+K1X). Finally, in the third case Y increases only by some multiple of the number since X is very large as compared to K2MZ - . K2MZ 1 , rru V • u the expression — approaches zero. Ihus it is seen when Ki(M-z) Ki(M-z) X — or Y is plotted as ordinate and X as abscissa a curve, the velocity of fatality curve, will be formed which at first has a very gradual rise followed by a rapid rise which again is followed by a less rapid rise depending on the value of KiX which finally approaches a straight line at very high values of X. This curve 171] THE GOLDFISH AS A TEST ANIMAL— POWERS 51 will approach a straight Hne where the direction of the curvature changes (See the portion A to B, curve CABG, Figure 1). The nearness with which this portion approaches a straight hne depends on the values of M, Ki, and K2 as compared to X. Thus the theoretical velocity of fatahty curve complies with the actual experimtis^a' velocity of fataUty curve. The above equation though conforming in a general way to the experimen- tal data, is doubtless incomplete. It is only an attempt at a mathematical expression which might be taken to represent the experimental data and should be corrected for other factors not taken into consideration here. The vahdity of the equation is further emphasized by the work of Burge (1917) in which he shows that both ether and chloroform not only destroy the catalase of the blood of an anesthetised animal but also inhibit the production or the hberation of the catalase by the liver. In other words there are two factors. One is the effect of the poison in inhibiting the Hberation or produc- tion of an enzyme and the other is the continuous action of the poison in des- troying the enzyme which has already been formed or hberated. Both these factors are expressed in the theoretical equation. A mathematical study of the equation shows that as X decreases in value, Y becomes zero. This value of X corresponds to the threshold of toxicity concentration. When X is decreased below this point, Y becomes imaginary. This is in keeping with experimental data, as very small amounts of certain active substances do not inhibit but stimulate metabohc processes. 52 ILLINOIS BIOLOGICAL MONOGRAPHS [172 COMPARISON OF CURVES OBTAINED BY OTHER WORKERS Warren (1900) in his work on Daphnia was first to call attention to the fact that when aquatic animals are killed by placing them in solutions of a toxic substance if the survival time is plotted as ordinate and the concentration of the solution as abscissa, the points when joined will form a curve closely resem- bling an equilateral hj^erbola. Warren considered all of his data as complpng with this curve. Either he experimented with solutions which would fall within the concentrations where the velocity of fatality curve approaches a straight Hne or he disregarded data obtained outside this range of concentra- tions. Warren saw the similarity between his curves and the curves represent- ing Boyle's law and explained his results by supposing that the toxicity of a substance above a definite amount (the writer's theoretical threshold of toxicity concentration) depended upon the number of molecules which beat on the body of the Daphnia per unit of time. Ostwald (1905, 1907) was next to call atten- tion to a similar curve formulated from data obtained by killing Gammarus in different concentrations of sea-water or the constituents of sea-water. Ostwald, disregarding the extremes of his data, claimed that the curve was not an equilateral hyperbola and applied the modified absorption formula, tC" =Ki or t(C-n)'" =Ki (Ostwald 1907) (see page iZ). Kfizenecky (1916) next noted the resemblance to that of an autocatalytic curve of a curve which he obtained by determining the time required for an anneUd worm, Enchy- traeus humictUtor, to recover in ordinary tap water after having been in a solution of a toxic substance for one minute, when time was plotted as ordinate and concentration of the solution as abscissa. The curve obtained by Kri- zenecky resembles the writer's velocity of fatality curve. Krizenecky ex- plained his results in terms of osmotic pressure. Clayberg (1917) called atten- tion to such a curve and called it an hyperbola but made no attempt to show how nearly it approached an hyperbola or its variations. None of these workers have seen the significance of the entire curve nor did they construct what is designated by the writer as the velocity of fatality curve. Neither did they suggest the possibility of its utilization in physiological testing nor attempt to connect their entire data with life processes. Gregersen (1916) in his investigation on the antiseptic value of gastric juice found that the disinfecting value of gastric juice was in direct proportion to the free acidity and that the product of the survival time of the bacteria and the titration number was almost constant. This in fact represents an equilateral hyperbola but such a relation was evidently not noted by Gregersen. Warren also showed that v.hen the survival time of Daphnia killed in a constant concentration of a toxic substance at different temperatures was plotted as ordinate and tempera- ture as abscissa a similar curve was formed. 173] THE GOLDFISH AS A TEST ANIMAL— POWERS 53 Krogh (1914, 1914a) and Sanderson and Peairs (1913) almost at the same time and independently of each other noted that a similar curve was formed when rate of development of eggs of frogs, insects and sea-urchins and also larvae and pupae of insects was plotted as ordinate and temperature as abscissa. Sanderson and Peairs determined the reciprocal curve which they designated as the rate of development and considered it a straight Hne crossing the X-axis at the actual zero of development. They disregarded the variations from a straight line at the extremes altogether and made all calculations on the assump- tion that the temperature above their theoretical zero times the time required for the organism to pass through certain stages of development is a constant. Krogh in his work called attention to the variations of the rate of development begun below the point at which the straight line cut the X-axis, i.e., the lowest temperature at which development will take place is below the theoretical temperature for the initiation of development. Reibisch (1902) called this the threshold of development. Finally, Osterhout has shown that the dying curve of Laminaria v/hen killed in an NaCl or a CaCl2 solution of the same conductivity as that of sea-water as shown by the fall of electrical resistance after having passed the point of stimulation of the latter salt follows the course of the same general curve when resistance is plotted as ordinate and time as abscissa. The close similarity of the curves found by different workers in entirely different fields suggests that the extremes of the reciprocal curves should be more thoroughly investigated. The possibiUty is that no portion of the curve is a straight line. This supposition is evidenced by the equation l/t= (MK2-t- KiX)/ loge ( -- — H ) • This is further emphasized by the fact that M-z Ki(M-z) X Ostwald's data fits almost equally well the formulae y(x-a) = k, i.e., l/t= (x-a) /K = Ki(x-a) and t(C-n) =Ki, i.e., loge t-j-mlogg (C-n) = loge Ki. These two equations represent hyperbolae of entirely different orders and are never similar except when m=l. Thus the evidence is that neither of the two equa- tions fits exactly. But for all practical purposes in pharmaceutical work and in insect pest prediction (Shelford 1918) the equation l/t=y=Ki(x-a) can be considered as holding in very narrow hmits i.e., when the temperature to which the insect pest is subjected is never below the temperature at which the velocity of development curve ceases to approach a straight line; otherwise, some modification of the equation used by entomologists must be employed. 54 ILLINOIS BIOLOGICAL MONOGRAPHS [174 DISCUSSION OF DATA All substances investigated show very striking similarities in their toxic activities on the goldfish. 1. All substances show a slow relative increase in rate of velocity of fatality with increase of substance used when time required for fatality is above 720 minutes. 2. The greatest rate of increase of velocity of fatality occiu^s when amounts of substances used wiU kill the goldfish in periods from about 45 minutes to about 210 minutes vA\h only a few exceptions. (Table XXIV). 3. The rate of increase of velocity of fatality is less rapid with increase in concentration of the substance used when the time required to kill the goldfish is less than about 45 minutes. 4. When the velocity of fataUty is plotted as ordinate and normality or the amount of substance used per Uter as abscissa, the curve thus formed has a very striking resemblance to a monomolecular autocatal)^c curve, i.e., it is a logarithmic curve. HCl and KCN show this character more strikingly than other substances tested as experiments were run over a large range of concentrations. (Figures 14 and 15.) 5. The velocity of fatality of the goldfish is more nearly constant for any given concentration when the concentration of toxic substances used is sufficient to kiU the goldfish in a period of not less than 45 minutes nor more than 210 minutes, with very few exceptions. The range for KCN and HCl is from about 90 minutes to about 1800 minutes. This range must be determined independently for each substance. (Table XXTV.) 6. The velocity of fataUty curve approaches a straight line for amounts of substances that will kiU the goldfish in periods of time just mentioned. This range must be deter- mined independently for each substance by the graphic method from experi- mental data. (Table XXTV and Figure 1 and explanatory notes.) From these general results it is seen that when the toxic value or active principle of any substance is to be determined from a pharmacodynamic assay point of view; i.e., to determine the strength of an unknown, only such portions as wiU kill the goldfish within a range of a certain minimum and maximum time should be used. This range must have been previously determined for each substance to be tested. (See 5 and 6 above and Table XXTV^.) An unkno^vMi can be determined by any one of five methods. 1. A definite survival time can be utilized, ^.e., determine the concentration of the substance that ^^•ill kill the goldfish in a selected period of time. The survival time of the goldfish for most exact determinations should be within the range of time where the velocity of fatahty curve approaches a straight line. This range of time must have been previously determined for each substance to be tested. The portion A to B of the velocity of fataUty curve, (Fig. 1 and Table XXTV^.) This is a modification of the method proposed by Pittenger and Vanderkleed (1915). 2. A number of goldfish can be killed in the unknown and the average survival time taken and appUed directly as ordinate to a survival time curve of the same 175] THE GOLDFISH AS A TEST ANIMAL— POWERS 55 substance as that tested which has previously been very carefully prepared by killing a number of goldfish and placed in the hands of workers as a standard curve. For example, if the average survival time of a number of goldfish killed in an unknown Li CI solution was 150 minutes it is seen that the ordinate representing 150 minutes survival time will cut the LiCl survival time curve, LIJM (Fig. 1) at the point R and the normality (0.207 N.) of the LiCl solution, can be read directly from the abscissa. For this method to be most exact the point R must fall within the portion I to J of the survival time curve, i.e., the survival time of the goldfish must be within the portion of the curve where it corresponds to the portion of the velocity of fatality curve that approaches a straight line. 3. A number of goldfish can be killed in an unknown and the average of the survival time determined and data applied to the equation of an equilateral hyperbola, y{x-a) = k, where }» = survival time of the goldfish, x= amount of substance used per liter, a = theoretical threshold of toxicity con- centration, and k = 2, constant. For example the average survival time of a number of goldfish killed in an unknown LiCl solution is 150 minutes. By substituting in the equation the value of the constants (a =0.1 25 N. and k = 12.37 for LiCl) we have 150(a;-0.125) = 12.37. Solving ic=0.207 N. which is the strength of the unknown, k approaches a constant only when the velocity of fatality curve approaches a straight line. See Figure 1. The deviation of the velocity of fatality curve from a straight line increases progressively as the distance preceeding and following the portion A to B increases, and at the same time the survival time curve deviates from that of an equilateral hyper- bola. Curve CABG, Fig. 25, is a graphic representation of this fact [See Shel- ford (1918) for detail of this curve]. Ordinate of curve represents k [k — y{x-a)\. If the survival time required to kill the goldfish is less than 45 minutes k is greater than 12.37 and if it requires longer than 210 minutes k is less than 12.37. Thus to apply this equation to LiCl the survival time of the goldfish must not be less than 45 minutes nor longer than 210 minutes. This range of sur- vival time must have been previously determined for each substance to be tested. 4. Curve CABG, Fig. 25, when time is interpolated on abscissa (see abscissa at top of graph), can be utilized directly to determine the strength of the unknown, both where k approaches a constant and where ^ is a variable. Thus to determine the strength of an unknown apply the average survival time of a number of goldfish which have been lulled in the unknown to the curve CABG, Fig 25, and read the normality of the unknown directly from the abscissa. For example, if the average survival time of the goldfish killed in an unknown LiCl solution is 150 minutes it will be found that this corresponds to the point R of the curve CABG. Then read directly from the abscissa 0.207 N. which is the strength of the unknown LiCl solution. Again for most exact determinations, concentrations represented from A to B must be used. This curve is of special interest and value in that it emphasizes the variability of k. 5. A few experiments can be run with different concentrations of the substance to be tested and the survival time of the goldfish in each experiment 56 ILLINOIS BIOLOGICAL MONOGRAPHS [176 determined. All concentrations used must fall within the range of concen- trations where the velocity of fataUty curve approaches a straight line. (Con- centrations represented from A to B, curve CABG, Fig. 1). A graph should then be dravvTi with reciprocal of survival time as ordinate and the ratio of the substance used as abscissa (i.e., the theoretical velocity of fatality curve) on some standard scale. This curve which is practically a straight line and should be drawn as a straight line should then be compared to a standard curve which has been previously prepared from experimental data of the same substance at the same temperature or a curve d^a^vn from data of experiments testing a known solution or standard of the same substance where a standard solution is available. The latter eliminates any error due to seasonal variation, variation in stock of goldfish, physiological state of the goldfish as well as variations in temperature at which the experiments can be rim. Then the strengths of the known and the luiknown solutions are inversely proportional to the dis- tances from the origin O at which their theoretical velocity of fatality curves cut the X-axis. That is, the strength of the unknown solution is to the strength of the known solution as the distance from the origin O to the point where the theoretical velocity of fatality curve of the known solution cuts the X-axis is to the distance from the origin O to the point where the theoretical velocity of fatality curve of the unknown solution cuts the X-axis. For example, let e = strength of the known solution represented by the theoretical velocity of fatality curve SMP, Fig. 24, t'= strength of an unknown solution No. 1, repre- sented by the theoretical velocity of fatahty curve SP'M', and M=the strength of an imknown solution No. 2, represented by the theoretical velocity of fatality 1 11 111 curve SP"M." Then v:e:u= — :-=-:-^ = . . OP' OP OP" b a c where a = distance 9 u^ M ^ / ^ [^ 0 vy' ^1=^ ^ ^ ~ X y'^^^^ ^^^ ^i^^ 100 t 40. 60. 80. 100. 120. 140. 160. 180. 200cc. Figure 24. Lithium chloride. Graphs showing method of determining the normality of an imknown LiCl solution. SPM = Theoretical velocity of fatality curve of a known solution of LiCl SP'M' = SP"M"= " " an xmknown » » » » " No. 1 » » 2 1771 -•? ^ % Z. 3 D- '^ P 05 oi C Q —■ &- < c <' o •"■ n g P c» o* era 3. T-JETE GOLDFISH AS A TEST ANIMAL— POWERS ^ o p >-» h-k i-« o "S ^o> „"-» „oo „oo ^to s -0< -O -oo -tnON -;-J D •-! to b\ o bv O, •-I C/2 "• C B -( OJ K) H- I-' ►-« H-k 3 ° ^• i-'OJ--Ji4i.i-'i-»vO>O00CnCn i"g^ £L Ohfi.'Oi-'NJONONtOtOOOrfi. « 1 "^ff- o g 5* ch 3 Q. oj 4*. In '•-» bo bo p o io ^ bo proc ival sh 1 S ff- ^ \ 3 o r* n !^ d n p t-L 1-. H-i ^ l^g^l "S _-^ ^VO ^O ^NJNi „>f^ 3Cn So -O -0_^- -t-n ^ o ■-« In *^ ^ t^ 3 0) "-1 CAl £-• C 3 -( 3 2, ;^- OJ K3 I-' ^- I-* ^ 5* t^ C3 ^-Ca)— j*>.H-H^'0\OOooic/i 04^0>-'WOCNK3tOC»*i p s i-- * 5- ^a n S 3 £- c (T c^ p* <-► n ex. Red surv offi C>j 4a. In ^ CO bo p O t-O ^ bo to Ca) CN H- O CN 1^ OO ^I SJ Vl proc ival shl S ff. EL ^^ 3 o r* ft ii. o c! p Oo •f'- tn O* ON ^» „;-J ^ Ui ^O _pO _Ni ^ ^ ^ '9 P 2. 3 n "Ko" - ~*o---»''to ,., ■ ° C/1 Oi C/l bO ^ rV "^ ft CA c 3 T OJ KJ h- t-» ►-» H-' 3 £.1- >-«Oi~J4».^-^-^'OOOCnCn al tim fish inutes 04»'>0>-'KJONCNNiN300if>. « n p ^-^ ft s p (3" CO CL ►»» 3 ft » i- Q. o> < -r) i>J '*.■ "cy> '■« bo be p o to ^ bo tOtMONi— 'OONi^OO-JtOCn 3* P '1 ■ ■ O 8 g-£i- "i^ "• £. 57 58 ILLINOIS BIOLOGICAL MONOGRAPHS [178 OP, b = OP', and c= OP". The theoretical velocity of fatality curves can be bet- ter drawn and compared when it is remembered that all theoretical velocity of fatality curves of any one substance when drawn on a definite scale representing the reciprocal of the survival time of the goldfish and variable scales repre- senting amounts of substance used per liter or normaUty of substance have a definite common point of intersection. That is, if the survival time of the goldfish be plotted as ordinate, and the number of cc. or g. per 1. of knowTi and unknown used be plotted as abscissa the curves thus formed would constitute a system of confocal conies of equilateral hyperbolae each being dragged out of position a distance OP, OP', and OP" respectively. Therefore the recipro- cal curves will aU intersect on the Y-axis at the point S. (Fig. 24). For example two hypothetical unknowni LiCl solutions could be determined in the following manner. Make up six solutions from a known LiCl solution as given in the following table and determine the survival time and the recipro- cal of the survival time of the goldfish in each solution as recorded. Test the unknown solutions No. 1 and No. 2 as shown in table XXXVIII. Plot a theoretical velocity of fatahty curve for each set of data with reciprocal of sur- vival time as ordinate and cc. of a solution used per 1. as abscissa l:100/t is used instead of 1/t to avoid the use of fractions. One block ordinate =1 velocity of fatality. One block abscissa = 10 cc. of solution per 1. ooo mo o o o !S Minutes 14 12 10 8 6 4 2 0 -2 -4 0.05. 0.1. 0.15. 0.2. 0.25. 0.3. 0.35. 0.4. 0.45. 0.5N. Figure 25. Lithium chloride. Graph showing curve when normality is plotted as abscissa and concentration minus the theoretical threshold of toxicity concentration, times survival time of the goldfish is plotted as ordinate, i.e., k of the equation y(x-a) =k. The graph shows the deviation of k from a true constant. The portion AB is equivalent in range to AB of the velocity of fataUty curve Figure 1 . y = survival time of goldfish in minutes, x = normal- it>' of the LiCl solution, a = theoretical threshold of toxicity concentration =0.125 N. and k = a. constant = 12.37 which holds only between the two points A and B. lY ■ ■ IR 1 A 1 i . ^^ >!?= 1 1 0 X / / i 179] THE GOLDFISH AS A TEST ANIMAL— POWERS 59 24 23 20 Minutes K 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 -1000 L -2000 -3000 10. 20. 30. 40. 50. 60. 70. 80. 90. lOOcc. Figure 26. Ethyl alcohol. Graph showing the variation of k. Abscissa represents cc. of alcohol per liter and ordinate represents cc. of alcohol per liter minus the theoretical threshold of toxicity concentration times survival time of the goldfish in minutes. The portion AB is equivalent in range of concentration to AB of velocity of fatality curve Figure 17. o=2.5 cc. per liter. ^=2625. The activity of the goldfish represented by the dotted line has not yet been explained. t 1 s II 1 3 / . / f I I 1 / J V B 1 ^ / 1 t J y^ 1 i q 1 i ! 1 1 i 1 ! ! j 1 J 1 1 In Figure 24 the curve SPM= theoretical velocity of fatality curve of known LiCl solution, SP'M' = unknown solution No. 1, and SP"M"=un- known solution No. 2. 2.22 N. = normality of known LiCl solution. Then let t)= normality of unknown solution No. 1 and m= unknown solution No. 2. 1 1 1 1,1,1 2.78 * Then v: 2.22: u = = 1.95 : 2:22 : 3:91. OP' OP OP" 2.78 2.45 1.39 Thus unknown solution No. 1 = 1.95 N. and No. 2 = 3:91. For all methods suggested the temperature must be kept constant through- out the experiment and at the same temperature as that at which the standard theoretical velocity of fatality curve was prepared. These methods will not hold when the temperature varies. 60 ILLINOIS BIOLOGICAL MONOGRAPHS [180 Results which will allow the above interpretation have been derived in- dependently by dififerent workers (but not thus interpreted by them) in ' different groups of the animal kingdom as well as the plant kingdom. All will stand the above tests. Ostwald (1905) in his work on toxicity of sea- water and the constituents of sea-water on a fresh water isopod {Gammarus pidex De Geer) plotted survival time of the isopod as ordinate and proportion of sea-water or its constituents as abscissa. The curve thus formed when examined casually resembled an hyperbola, however he did not designate the curve as such, but stated, *'I have not been able so far to figiure out a formula that holds for the whole curve." Ostwald did not construct the reciprocal or velocity of fatahty curve. When this curve is constructed from his data, (Table, page 72, showing survival times of male and female Gammarus in different proportions of sea-water, 1905) it resembles very closely the velocity of fataUty curves shown in Fig*. 1 to 20. Doubtless the resemblance would have been more striking had Ostwald tabulated his data which were omitted and had he run a few experiments with higher concentrations, as he states, *'it was only from concentrations of 20 parts of fresh water with 80 parts sea- water and higher, that visible toxic effects began to appear. But these figures thus obtained were still so large and varied so much that I excluded them from my experiments. " And later he says that the toxicity of all salts inves- tigated increased very slowly at very low concentrations with increased con- centration of solution depending on the nature of the salt and at stronger concentrations there was a sudden rise, while at still higher concentrations there was again a slow rise in toxicity. This is in complete accord with data obtained in this investigation. He then suggests that this difference in the rate of increase of the toxicity of a substance with increase in concentration at very low and at very high concentrations was due to the inability to measure very exactly the low concentrations and to inexactness in determining the death point in a very short survival period. Ostwald (1905, 1907) then disregards these extremes and applies to his mean data the absorption formula (Ostwald, 1906) a=KC" where a = amount of salt absorbed, C = concentration of the solution, and K and m are constants depending on the nature of the salt investigated. He then assumes that the survival time of the Gammarus is inversely proportional to the amount of salt absorbed, i.e., l/t=a, thus l/t=KC™ or tC = Ki. Later Ostwald (Ostwald and Dernoscheck, 1910) recognized the inabiUty of the for- mula to fit his experimental data, and revised his formula by substituting (C-«) for C where « = amount of the salt tested normally found in the blood of the experimental animal. But it is diflScult to see why such a substitution should be made wih substances not normally found in the blood of the animal. This as has been pointed out would be necessary -with all substances tested, since all undergo the same variation in toxic activity with variation in con- centration as that pointed out by Ostwald. The only exceptions are CxiCh, 181] THE GOLDFISH AS A TEST ANIMAL— POWERS 61 CdCU, and FeCls. See page 35. Kfizeneck^ (1916) in his work on Enchy- traeus humicuUor, a fresh water annelid, showed that the survival time of these worms in different concentrations of sea-water, when data is plotted as above, gives a similar curve. Knzeneck;^ explained these results as well as the greater part of the toxic effect of the alkali and alkaline earth metals tested by him as being due to osmotic pressure. He then formulates a curve by tabulating results obtained by determining the time required for the worm to recover in ordinary tap-water after being placed in different concentrations of sea-water for one minute. He then states that both curves have the charac- teristics of curves of autocatalytic processes in that they fall within the province of the theory of the temporal properties of Ufe processes proposed by Ostwald (1908). Loeb (1903) in summarizing his work on a marine Gammarus, states that, "if sea-water be diluted by the addition of distilled water the duration of life decreases at first only slightly in the decrease of the degree of dilution. But as soon as a dilution of ten times is reached an abrupt decrease in the dura- tion of life takes place with further dilution. Whether the curve of the dura- tion of life at this place is discontinued is not yet proven. " Data obtained from the work of Loeb and Wasteneys (1913) on the reduction of oxidation of fertilized eggs of sea-urchins by the addition of .01% KCN solution to sea- water gives a similar curve when per cent of loss of rate of oxidation is plotted as ordinate and relative amount of KCN added to the sea-water as abscissa, Fig. 27. The per cent of reduction of the rate of oxidation of an individual sponge as given by Hyman (1916) in her work showing the effect of KCN on the reduction of the rate of consumption of oxygen by a marine sponge also gives a similar curve when per cent of reduction of oxygen consumption is plotted as ordinate and normaUty of KCN as abscissa (only data on a single Per Cent. 100 80 60 40 20 Figure 27. Curve showing the effect of potassium cyanide on the rate of oxygen absorp- tion of fertihzed eggs of the sea urchin. Abscissa represents cc. of 0.01% KCN p>er 50 cc. sea-water. Ordinate represents the per cent of reduction of rate of ox>'gen absorption below the normal by the potassium cyanide. 62 ILLINOIS BIOLOGICAL MONOGRAPHS [182 Per Cent. 100 80 40 - I i 'ill 1 1 ! i i r'' i \ i 8 2 Figure 28. Curve showing the effect of potassium cyanide on the oxygen absorption of a marine sponge. Abscissa represents mols, per Uter in sea water. Ordinate represents the per cent of the rate of ox;>''gen absorption reduced by the potassium cyanide below the normal during a vmit time (150 minutes). Data taken from Hyman (1916). All data taken from experiments on an individual sponge. sponge were taken) Fig. 28. The last two calculations are not made on the same basis as the preceding curves but are sufficiently comparable to show that both probably follow the same general law as the toxic activity of salts. A curve formulated from data given in Experiment 1, of Pittenger and Vander- kleed's (1915) preliminary work on the goldfish as a test animal is shown in Fig. 29. This is the same type of curve. This work of Pittenger and Vander- Minutes 2160 1920 1680 1440 1200 960 720 480 240 / / ,- /^' / ■ / L / / / / \ / / V 'l / "r^ — - H- Oi Figure 29. Digitalis. Velocity of fatality and survival time curves. Abscissa represents per cent concentration. Data taken from Pittenger and Vanderkleed (1915). 183] THE GOLDFISH AS A TEST ANIMAL— POWERS 63 kleed has been very severely criticized. Many have questioned, as some have put it, " the smooth and elegant results. " These smooth and elegant results obtained by these workers are due at least in part to the fact that they were working with solutions which fall within the portion- of the velocity of fataUty curve that approaches a straight hne. (Compare to the portion A to B, curve CABG, Fig. 1.) While the concentrations with which their critics worked possibly fell outside this range of concentrations, it is probable that if their data be reviewed from this point of view it will become more inteUigible. From data obtained in this investigation it is clear that all experimentation for pharmacodynamic assay work should fall within the portion A to B of TABLE XXXII The Variation of the Survtcval Time of Goldfish 0.318 N. 0.284 N. 0.274 N. 0.214 N. Magnesiom Calctctm Strontium Bartdm CHLORrOE 21° C. Chloride 21° C. Chloride 20.5° C. Chloride 20° C. Weight Survival Weight Survival 1 Weight Survival Weight Survival offish time of fish of fish time of fish offish time of fish of fish time ot fish in grams in minutes in grams in minutes in grams in minutes m grams in minutes 2.2 73 2.05 110* 2.95 96 1.9 86 2.3 74 2.3 145 2.2 85 2.2 85 2.4 41* 2.4 87* 2.25 73 2.2 89 2.4 75 2.4 95* 2.4 112* 2.3 94* 2.4 76 2.5 96 2.7 89 2.4 91 2.4 77 2.6 93* 2.85 99* 2.4 94* 2.5 78* 2.7 84* 2.9 101* 2.4 101 2.6 77* Z.?, 100* 3.0 115* 2.5 100 *Fish was not dead when taken out of solution. the velocity of fatahty curve CABG, Fig. 1. The question then arises by what method can one determine with the least number of preUminary experi- ments the portion of the curve at which one is working. This can be done either by knowing the extremes of the survival time of the goldfish when the concentrations are within the designated range (A to B) or by running only three preUminary experiments at different concentrations. If the velocity of fataUty curve formed by plotting the data from these experiments is convex with respect to the X-axis, the substance should be tested by using solutions falling near the weakest or between the two weaker solutions. If the curve is concave, solutions nearer the strongest or between the two stronger solutions should be used. If the curve is nearly straight and approaches a parallel position to the X-axis the solutions are either too weak or too strong. (These 64 ILLINOIS BIOLOGICAL MONOGRAPHS [184 seem to be the concentrations used by most pharmaceutical workers who have tested the goldfish method of Pittenger and Vanderkleed.) The time reqmred to kill the goldfish will generally tell one whether the solutions are too strong or too weak. See discussion of survival time curve, page 48. If the curve approaches a straight line and is at an angle to the X-axis the solu- tions are of the most effective concentration. A second method consists in running only one experiment, determining the survival time of the goldfish, and applying this datum to the survival time curve (Curve LIJM, Fig. 1) or to the curve of the constant (Figs. 25 and 26) and reading directly the approximate concentration of the unknown. These curves also show the range of concentrations (concentrations represented by the portions of the curves A to B) most advantageous to use. TABLE XXXIII 0.246 N. Potassium Chloride Weight Survival time Temperature Temperature offish of fish Experiment Stock ki grams in minutes Centigrade Centigrade 3.0 16 19.5° 19.5 to 22° 4.3 61 » " " " 4.7 60 » " >) >) 4.75 58 » )! » " 4.9 32 >» »! " " 5.1 45 )> " >) >> 5.05 35 21° 20° ' 21.5° 5.4 37 » » . „ 4.4 49 22° 15.5° ' 16.5 4.4 79 >> " > » 4.7 42 >) >) ' " 6.4 69 " " ) » 3.9 46 16.5° 19.5° ' 22° 4.9 37** " >) > )) 4.85 37 » » > i> 5.1 83 » >> ' " 4.8 57* » 15.5° ' 16.5° 5.3 64 » >) > >» 5.75 57 » » ) >» 5.8 77 » » » n *Fish was not dead when taken out of the solution **Fish was inactive and on side for about 1/2 minute or less immediately after being placed in solution 185] THE GOLDFISH AS A TEST ANIMAL— POWERS 65 TABLE XXXIV 0.2223 Potassium Nitrate Weight Survival time Temperature Temperature of fish of fish Experiment Stock in grams in minutes Centigrade Centigrade 3.4 31 19.5° 19° to 21° 3.4 35 » >) » »> 6.1 70 >» >) » )> 3.9 45 if 15.5° " 16.5° 4.1 95 » j> )j »> 5.4 31 » )» » M 4.3 45 15.5° 19° » 21° 4.7 59 » j> )> > 4.7 62 » >» j> ) 4.2 56 » 15.5° " 16.5° 4.8 70 )» j> >» ) ' 5.1 46 >i >i » » TABLE XXXV The Variation of Survival Time of Goldfish Weight Survival time Temperature Weight Survival time Temperature offish of fish Experiment of fish of fish Experiment in grams in minutes m grams in minutes .250 N. Magnesium Nitrate .244 N. Magnesium Nitrate 4.4 87 20° 4.7 40 19° 5.3 47 » 5.2 85 5.6 46 » 5.5 85 6.0 62 19° 5.5 48 6.0 79 >i 5.6 45 6.1 42 ») 5.6 128 6.3 59 20° 6.7 42 7.3 102 19° 7.1 57 » 66 ILUNOIS BIOLOGICAL MONOGRAPHS [186 SUMMARY OF CONCLUSIONS 1. The survival time of the goldfish {Carassius carassius L.) has a very definite relation to the concentration of the solution of the toxic substance used. This relation of the survival time of a goldfish to the concentration of the solution of the toxic substance follows a common general law with a very few exceptions which can be expressed as follows. a. There is a concentration of each of the toxic substances tested which will just cause the death of a goldfish and concentrations below this will not cause death. This concentration has been designated as the threshold of toxicity concentration. b. In concentrations of a toxic substance just above its threshold of toxicity concentration the velocity of fataUty of the goldfish (as measured by the re- ciprocal of the survival time of the goldfish) is increased very slowly with in- crease in concentration of the solution of thetoxic substance. c. In stronger concentrations the velocity of fatahty is increased more rapidly with the increase in the concentration of the solution. d. And finally at very high concentrations the increase of velocity of fataUty is again less rapid in proportion to the increase in concentration of the solution. 2. The survival time curve which is plotted by letting ordinate represent survival time of the goldfish and abscissa represent normaUty or the amount of substance used per 1. of water is not an equilateral hyperbola but is logarith- mic in function. 3. A curve, the velocity of fatality curve, which is formed by plotting the reciprocal of the siu-vival time of the goldfish as ordinate and concentration of the solution as abscissa resembles a curve which can be expressed by Y= — = M=normalrateof metabolism of a goldfish, t , / M , KsMz 1 , M-z KKM-z) X 2= loss of rate of metaboHsm of the goldfish when in a solution a toxic sub- stance, Y= reciprocal of the survival time of the goldfish, and Ki and K2=two constants depending on the nature of the metabolic process or processes in- volved and the nature of the toxic substance used. 4. The velocity of fatality curve approaches a straight line when the nor- mahty or amounts of toxic substance used per 1. of water will not kill the goldfish in less than 45 minutes and do6s not require longer than 210 minutes. Data within these limits when survival time of the goldfish is plotted as ordinate and the concentration of the toxic substance as abscissa forms a curve which approaches an equilateral hyperbola and can be expressed for all practical purposes by the equation y(jx-a) = k, where a = distance from the origin where 187] THE GOLDFISH AS A TEST ANIMAL— POWERS 67 the portion of the velocity of fatality curve that approaches a straight Une when prolonged cuts the X-axis and k = a. constant. This straight hne which is thus drawn has been designated as the theoretical velocity of toxicity curve. The concentration of the toxic substance represented by the point on the X-axis cut by the theoretical velocity of fatality curve has been designated as the theoretical threshold of toxicity concentration. 5. It has been suggested that relative toxicities of substances be expressed by the formula T = 'Y when only data which fall within the portion a of the velocity of fatality curve which approaches a straight Une are used- G = angle made by the theoretical velocity of fatality curve cutting the X- axis and a = the theoretical threshold of toxicity concentration of the substance tested. This expression does not represent either the absolute or the exact relative toxicities of the substances since it is based only upon the portion of the velocity of fatahty curve which approaches a straight hne, but has been chosen since it is a natural criterion and not an arbitrary one. 6. Four modifications of a definite survival time of the goldfish method of pharmacodynamic assay work as suggested by Pittenger and Vanderkleed have been proposed. a. A definite survival time of the goldfish can be employed provided that the concentration of the substance to be tested is within the range of concen- trations in which the velocity of fataUty curve approaches a straight line. b. The average survival time of a number of goldfish in a solution of the substance to be tested can be applied as ordinate to a standard survival time curve and the strength of the solution can be read directly from the abscissa provided this data falls within the limits of the survival time curve which approaches an equilateral hyperbola, i.e., where the velocity of fataUty curve approaches a straight line. c. The average survival time of a number of goldfish killed in a solution of the toxic substance to be tested can be substituted in the equation y{x-a) = k and the value of x determined which will be the concentration of the solu- tion of the substance tested, provided the survival time of the goldfish is within certain maximum and minimum of survival time, the reciprocals of which when plotted as ordinate and the concentrations of the solutions used as abscissa will approach a straight line, y = survival time of the goldfish, x = concentration of solution of the toxic substance in which the goldfish are killed, '. Report Brit. Assn. Adv. Sci., Birmingham, 1913; 652-668. Houghton, E. M. 1898. Pharmacologic Assay of the Heart Tonics. Jour. Amer. Med. Assn., 31:959-961. Hyman, L. H. 1916. On the Action of Certain Substances on Ox>-gen Consumption. I. The Action of Potassium Cyanide. Amer. Jour. Physiol., 40:238-248. Joseph, D. R. and Meltzer, S. J. 1909. The Comparative Toxicity of the Chlorides of Magnesium, Calcium, Potassium and Sodium. Jour. Pharm. Exp. Ther., 1 :l-26. Kahlenberg, L, and Austin, R. M. 1900. On the Toxic Action of Acid Sodium Salts on Lupinus albus. Jour. Phys. Chem., 4:553-569. Kahlenberg, L. and Mehl, H. F. 1901. Toxic Action of Electolytes upon Fishes. Jour. Phys. Chem., 5:113-132. Kahlenberg, L. and True, R. H. 1896. On the Toxic Action of Dissolved Salts and their Electrolytic Dissociation. Bot. Gaz., 22:81-124. Kastle, J. H. and Loevenhart, A. S. 1901. On the Nature of Certain of the Oxidizing Ferments. Amer. Chem. Jour., 26:539- 566. KiSSKALT, K. 1915. iJber die Beziehungen der todlichen Dosis zur Oberflache. Biochem. Zeit., 71:468^78. Krizenecky, J. 1916. Ein Beitrag ziun Studium der Bedeutimg osmotischer Verhaltnisse des Medium fiir Organismen. Arch. ges. Physiol., 163:325-354. Krogh, a. 1914. On the Influence of the Temperature on the Rate of Embryonic Development. Zeit. allgem. Physiol., 16:163-177. 1914a. On the Rate of Development and CO2 Production of Chr>-salides of Tenebrio molilor at Different Temperatures. Zeit. allgem. Physiol., 16:178-190. 1911 THE GOLDFISH AS A TEST ANIMAL— POWERS 71 Labokde, J. V. 1884. Sur le m6canisme de la mort ou des accidents qui succMent a I'injection sous- cutande du chloroforme, £l propos d'xine Note de M. Bouchard sur le mSme sujet. Compt. rend. soc. biol., 36:271-278. LiLLIE, R. S. 1902. On the Oxidative Properties of the Cell-nucleus. Amer. Jour. Physiol., 7:412-321. Locke, F. S. 1895. On a Supposed Action of Distilled Water as Such on Certain Animal Organisms. Jour. Physiol., 18:319-331. Lodge, O. 1913. Continuity. Science, n. s., 38:379-395, 417-430. LOEB, J. 1902. Studies on the Physiological Effect of Valency and Possibly the Electrical Charges of the Ions. I. The Toxic and Anti-toxic Effects of Ions as a Function of the Valency and Possibly their Electrical Charge. Amer. Jour. Physiol., 6:411-433. 1903. On the Relative Toxicity of Distilled Water, Sugar Solutions and Solutions of the Various Constituents of Sea-water for Marine Animals. Univ. Cal. Publ., Physiol., 1:55-69. LoEB, J. and Wasteneys, H. 1913. Is Narcosis due to Asphyxiation? Jour. Biol. Chem., 14:517-523. LoEVENHART, A. S. and Kastle, J. H. 1903. On the Catalytic Decomposition of Hydrogen Peroxide and the Mechanism of Induced Oxidations. Together with a Note on the Nature and Function of Cata- lase. Pt. I. Amer. Chem. Jour., 29 :397-437. Mathews, A. P. 1903. The Nature of Nerve Irritability, and of Chemical and Electrical Stimulation. Pt. II. Science, n.s., 17:729-733. 1904. The Relation between Solution Tension, Atomic Volume and the physiological Action of the Elements. Amer. Jour. Physiol., 10:290-323. 1907. The Cause of the Pharmacological Action of Ammonium Salts. Amer. Jour. Physiol., 18:58-63. 1916. Physiological Chemistry. New York. Mathews, A. P. and Walker, S. 1909. The Action of Cyanides and Nitriles on the Spontaneous Oxidation of Cystein. Jour. Biol. Chem., 6:29-37. Meek, S. E. and Hilderbrand, S. F. 1910. A Synoptic List of the Fishes Known to Occur within 50 Miles of Chicago. Field Mus. Nat. Hist. Publ., Zoology, 7:223-338. Miyake, K. 1916. The Toxic Action of the Soluble Aluminum Salts upon the Growth of the Rice Plant. Jour. Biol. Chem., 25:23-28. Moore, W. 1917. Toxicity of Various Benzene Derivatives to Insects. Jour. Agr. Res., 9:371-381. Moore, W. and Moore, J. J. 1917. Studies in Greenhouse Fumigation with Hydrocyanic Add: Physiological Effects on the Plant. Jour. Agr. Res., 11 :319-338. OSTERHOUT, W. J. V. 1915. The Measurement of Toxicity. Jour. Biol. Chem., 23 :67-70. OSTWALD, W. 1906. Lehrbuch der aUgemeinen Chemie, 2 Bd. Leipzig. OSTWALD, W. 1905. Studies on the Toxicity of the Sea-water for Fresh Water Animals (Gammarus pulex De Geer), Univ. Cal. Publ. Physiol., 2 :163-191. 72 ILLINOIS BIOLOGICAL MONOGRAPHS [192 1907. tlber die Beziehungen zwischen Adsorption und Giftigkeit von Salzlosungen fiir Siisswassertiere (Gammarus). Arch. ges. Physiol., 120:19-30. 1908. Uber die zeitlichen Eigenschaften der Entwlcklungsvorgange. Vortr. Aufsat. Entwicklimgsm., Heft 5. OsTWALD, W. and Dernoscheck, A. 1910. Ueber die Beziehungen zwischen Adsorption und Giftigkeit. Zeit. Chem. Ind, KoUoide, 6:297-307. Pauxi, W. 1903. Untersuchiingen iiber physikalische Zustandsanderungen der Kolloide, Beitr. chem. Physiol. Path., 3:225-246. PiTTENGER, P. S. and Vanderkleed, C. E. 1915. Preliminary Notes on a new Pharmacodynamic Assay Method. Jour. Amer. Pharm. Assn., 4:427-433. Reed, E. D. and Vanderkleed, C. E. 1908. The Standardization of Preparations of Digitalis by Physiological and Chemical Means. Amer. Jour. Pharm., 80:110-120. Reibisch, J. 1902. Ueber den Einfluss der Temperatur auf die Entwicklung von Fisch-Eiem. Wiss. Meersuntersuch. Kiel., N. F., 6:215-231. Richards, A. N. and Wallace, G. B. 1908. The Influence of Potassium Cyanide upon Proteid Metabolism. Jour. Biol. Chem., 4:179-195. RiCHET, C. 1881. De latoxicit6 compar6e des diff6rents m6taux. Compt. rend. acad. sci., 93:649- 651. Ringer, S. 1882. An Investigation Regarding the Action of Rubidium and Caesium Salts Compared with the Action of Potash Salts on the Ventricle of the Frog's Heart. Jour. Physiol., 4:370-379. 1883. Concerning the Influence of Saline Media on Fish etc. Jour. Physiol., 5:98-115. 1897. The Action of Distilled Water on Tubifex. Jour. Physiol., 22 :xiv-xv (Proc) . Robertson, T. B. 1908. On the Normal Rate of Growth of an Individual, and its Biochemical Significance. Arch. Entwickl. Organ., 25:581-614. Robin, A. 1914. Colloidal Sulphur. R6pert. Pharm. 70:161-162. (Cited from Chem. Abst., p. 2920.) Sanderson, E. D. and Peairs, L. M. 1913. The Relation of Temperature to Insect Life. Tech. Bull. N. H. Coll. Ag. Exp. Sta., 7:1-125. Shafer, G. D. 1915. How Contact Insecticides Kill. HI. Tech. Bull. Mich. Agr. Coll. Sta., 21:1-67. Sheltord, V. E. 1917. An Experimental Study of the Effects of Gas Weiste upon Fishes, with Special Reference to Stream Pollution. Bull. 111. State Lab. Nat. Hist., 11:381-412. 1918. Physiological Problems in the Life Histories of Animals with particular Reference to their Seasonal Appearance. Amer. Nat., 52:129-154. Sollmann, T., Mendenhall, W. L. and Stingel, J. L. 1915. The Influence of Temperatiue and Concentration on the Quantitative Reaction of the Heart to Ouabain. Jour. Pharm., 6:533-560. TOWNSEND, C. O. 1901. The Effect of Hydrocyanic-acid Gas upon Grains and Other Seeds. Bull. Md. Agr. Exp. Sta., 75:183-198. 193] THE GOLDFISH AS A TEST ANIMAI^POWERS 73 Troland, L. T. 1917. Biological Enigmas and the Theory of Enzyme Action. Amer. Nat., 51:321-350. Verworn, M. 1909. Allgemeine Physiologic. Jena. Warren, E. 1900. On the Reaction of Daphnia magna (Straus) to Certain Changes in its Environment Quart. Jour. Micr. Sci., 43:199-224. Wells, M. M. 1914. Resistance and Reactions of Fishes to Temperature. Trans. 111. State Acad. Sd., 7:48-59. 1915. Reactions and Resistance of Fishes in their Natural Environment to Acidity, Alkalinity and Neutrality. Biol. Bull., 29 :221-257. WOODWORTH, C. W. 1915. The Toxicity of Insecticides. Science, n.s., 41:367-369. ZlEGER, R. 1915. Zur Kenntnis der Katalase der niederen Tiere. Biochem. Zeit. , 69 :39-l 10. UNIVERSITY OF ILLINOIS-URBANA 570.5ILL C004 JLL^NOIS^BIOLOGICAL MONOGRAPHS URBANA 3 0112 017753440 i;',:'V;:i';i.';f , .. )i.- .,'1 1.-' *--\; ,^:;::;';: ;:,f ;::';::?:; ;B:^ 'Mi ■'■';;i':r' mm ■ ■■-v^:-^'--::'-'--^;'- :^*^ 8 ' ^iiji