Bulletin OF THE Illinois State Laboratory OF Natural History Urbana, Illinois, U. S. A. STEPHEN A. FORBES, PhD., L. L. D., Director Vol. XI. March, 1017 Article VI. AN experimental, STUDY OF THE EFFECTS OF GAS WASTE UPON FISHES, WITH ESPECIAL, REFERENCE TO STREAM POLLUTION BY Victor E. Shelford, Ph. D. ERRATA AND ADDENDA. Page 50, second column, line 13 from bottom, for Danais arcMppus read Anosia plexippns ; line S from bottom, for melliflca read mellifera. Page 51, line 11 from bottom, for Danais read Anosia. Page 159, at right of diagram, for Bracon agrilli read Bracon agrili. Page 289, second column, last line but one, for Scalops real Scalopus. Page 294, line 3, for c.atesheana read catesiiana. Pages 327 and 330, line 12, for oreus read oreas. Page 347, line 4, for Cecidomyidae road Cecidomyiidae. Page 356, line 7, for Anthomyidse read Anthomyiidae. Page 368, line 18, dele second word. Page 373, after line 10 insert as follows: 53a, suipruinosa Casey, 1884, p. 38. Page 375, after suXimucula Le Conte, 48, insert subpruinosa Casey, 53a. Page 377, after line 7, insert as follows: — 1884. Casey, Thomas L. Contributions to the Descriptive and Systematic Coleopterology of North America. Part I. Page 379, line 11 from bottom, for sen.iu lata read sensu lato. Page 382, line 12, for VII read VIII. Page 408, line 2, for the next article in read Article VIII of. Page 410, line 6 from bottom, for = 4 read '11. Page 412, line 7, for 31 read 30. Page 421, line 17 from bottom, insert it before grows. CONTENTS PAGE I. IntrotUiction 381 II. Statement of Fish and Gas-waste Pollntion Problem 381 III. Material and Methods 383 1. Character of University of Illinois Water 383 2. Treatment for keeping Fishes Alive 384 3. Difficulties to be Guarded against in Fish Experiments 385 4. Fishes used 387 IV. Gas Waste—Its Character and Constituents 388 V. Toxicity of Waste ' 389 1. Methods of Experimenting ' 390 2. Toxicity of Waste and Tar 391 3. Toxicity of Illuminating Gas and Constituent Gas-mixtures 392 4. Eeactions of Fishes to Waste 392 VI. Toxicity of Illuminating Gas Waste Constituents 394 VII. General Discu-'sion 406 VIII. Summary 409 IX. Acknowledgments 410 X. Literature Consulted 410 Article VI. — An Experimental Study of the Effects of Gas Waste upon Fishes, zvith Especial Reference to Stream Pollution. By Victor E. Shelford. I. Introduction. The products of destructive distillation of coal include an innum- erable series of substances representing most of the important groups of organic compounds ranging from gases to solids. In the manu- facture of illuminating gas all these substances are thrown into streams in varying amounts, depending upon the manner of treating l)v-products. The gases and volatile products are in solution in water used in washing the gas, and are often introduced into waterways. By-products, except the heavy tars, are often thrown away. This is especially true in the case of the smaller plants where the quantity is insufficient to make the further treatment of it profitable. Thus in many plants only the heavy tars are saved, the gas liquor drip from the mains and holders being dumped into waterways without the re- moval of even ammonia. The immense commercial value of these wasted products has been more generally appreciated since the out- break of the European war, which cut off the large supply of foreign dyes and important organic compounds and increased the demand for such products as may.be used in the manufacture of explosives. The value of these wasted products should be sufficient to prevent their wastage, but their injurious efifect upon fishes and other life of streams generally is itself sufficient to justify the prohibition of pollution by this means. II. Statement of the Fish and Gas-waste Pollution Problem. The gas waste problem is concerned witli the effects upon fishes of the gas liquor untreated, the effect after the removal of the heavy tar, the effect of tar, the effect of gas-washing water, and that of lime, etc. from the purifiers. It is the purpose of this paper to show that essen- tially all the products of the distillation of coal are very toxic to fishes, some of the most toxic being those which are commonly regarded as "insoluble" in water. From the standpoint of fishes the waste problem is concerned with the reactions of fishes when encountering the pol- 382 luted waters. The reactions of fishes to the resuhs of contamination with natural organic matter, such as decomposing bodies of plants and animals, are generally advantageous, as the fish turn awav from the polluted area. The result of this investigation shows that in the case of gas wastes the reactions are usually disadvantageous,—the fishes swim into the polluting substances without recognizing them or turning back from them even when their toxicity is such as to cause death within a short period. The detrimental character of gas wastes is thus increased many fold. The toxicity of waste differs for different species of fish and is greatest for the more valuable fishes as indicated by Dr. Wells' work (Article VTI of this volume). It will be shown to be generally greater for the smaller and younger fishes. The writer's investigations will show that this rule holds good down to the youngest fry studied. Snllmann ('06) found some evidence that the eggs and newly hatched embryos of marine Fitndulus are more resistant to poisons than the adults. He however seems to question his results in this respect because of the long exposure in the poison solutions and small Cjuan- tity of the solution in proportion to the size and total oxygen demand and excretory output of the adult fish. Child's work with phenyl urethane which was done after long experience in the use of such poisons showed that in marine FuudtilKS tlie resistance declined rapid- \y from a maximum in the two-cell stage of the tgg, to the time of hatching. In Tautogolahris the resistance fell from a survival time of 675 minutes soon after fertilization to 1 5 minutes at the time the heart began to beat and rose to 20 minutes at the time of hatching, when the experiments were discontinued. The resistance of the eggs and embryos of fresh water fishes has not been studied and compared with that of the adults, but there is every evidence that the rule reported here will hold good throughout the age and size series beginning aliout the time of hatching. The most sensitive period must be determined before the minimum fatal quantity can be established with any cer- tainty. For this reason no attempt has herein been made to determine the minimum which will prove fatal to the fishes studied. Dr. Wells has found that the resistance of some fishes to various factors varies greatly with the time of year. The lowest point comes between the middle of June and the last of July when such fishes as the cyprinids can hardly be taken from the water before death sets in. From this time the resistance slowly rises until September. Then the rise be- comes more rapid and reaches its highest point in March and April, when all the fishes are exceedingly resistant. With the onset of the breeding season the resistance falls, though whether or not it con- 383 tinues to fall until the period is well passed has not been determined. The effects on the breeding operations while of paramount importance have not been touched in this investigation. In this work no experi- ments were performed between June 8 and Aug. i8. In the course of the investigation the working out of the toxicity of the different compounds has been rendered essential, first because of their general occurrence as by-products and secondly because va- rious methods of treatment remove some compounds and not others. This toxicity is further of interest in connection with the effects of these compounds as drugs and poisons. The recent use of gold fishes, frogs, etc. as means of standardizing drugs, such as digitalis, renders these data of interest to the pharmacist and legal toxicologist. The timed killing of upwards of 1,500 fishes has, it is hoped, made clear some facts and methods which may be useful in the study of these problems with domesticated species such as gold fish. III. Material and Methods. The character of the water used is of much importance in the study of toxicity of polluting substances. The loss of oxygen and accumula- tion of waste matter in standing water renders experiments conducted with it open to criticism and necessitates the use of a short period to death in comparatively high concentrations of the drugs as a criterion in determining relative toxicity. It further necessitates the running of control experiments in running water. Experiments in running water are usually necessary in the case of gases. Toxicity is frequent- ly different in distilled water and tap water. I. THE character of university OF ILLINOIS WATER, AND OTHER WATER PROBLEMS. The water supply of the University comes from deep wells and the salts are nearly all present in the form of carlionates instead of a mi.xture of carbonates, chlorides, and sulphates as is the case in waters where fish normally occur. It also contains about twice as much magnesium and calcium and eight times as much iron as is commonly present in such waters. As it comes from the tap the university water contains no oxygen and about iScc. per liter of carbon dioxide. The lack of oxygen alone makes it unsuitable for fishes, and the presence of so much carbon dio.xide renders it wholly unfit for them. Fishes die in it quickly. The mortality among fishes brought in from streams was very great when this water was used in aquaria in which they were kept. 384 2. TREATMENT FOR KEEPING FISHES ALIVE- Treatment A The water in this case was Ijoiled in an apparatus ( Fig. i ) which continuously boils and cools it, being run through at the rate of 500 cc. per minute. This removed all of the readily i)recipitahle iron and the excess of magnesium and calcium, thus reducing the total solids to about what one commonly finds in the average stream ; but the water so treated still differed from stream water in that the salts pres- ent were nearly all carbonates, instead of a mixture of carbonates, chlorides, and sulphates, and decidedly alkaline. The water was aerated after boiling. The mortality became markedly less among fishes when they were first brought in, but on the whole it was not less than in water which received Treatment B. Treatment B In this treatment the water was aerated in an aerating device, so as to give air saturation. This removed nearly all the free carbon dioxide and rendered the water alkaline. In this the fishes lived fairly well I)ut became very sluggish, so that they were not suitable for be- havior experiments. Treatment C Thinking that the above sluggishness might be due to the absence of sulphates and the presence of carbonates only, a small quantitv of sulphuric acid was added to the water. This rendered it acid by dis- placing some of the carbonic acid in the carbonates with the sulphate radical. This treatment proved beneficial, but the requisite manipu- lation was cumbersome. Treatment D For this treatment aerated water and direct tap water were run, half and half, into the aquaria. This rendered the fishes active and suitable for behavior experiments and the difficulties of manipulation were reduced. Later a less complete aeration in the aerating device shown in figure i was found to give equivalent results, and fishes lived unusually well for months without attention. In this the water was treated by running down twelve feet of incline at a rate of about two liters per minute. It then usually contained sufficient oxygen to support fishes and from 1-3 cc. of free CO2 per liter, and had lost much of its iron and a little of its excess magnesium and calcium. 385 3- DIFFICULTIES TO BE GUARDED AGAINST IN FISH EXPERIMENTS. a. Character of Water. At the beginning of the work Dr. Wells ('15 and '15a) undertook a careful study of the relation of fishes to salts, acids, and alkalies. In general he found that carbonates do not have detrimental effects upon fishes when the water is acid. He further found many minor complications in connection with diiiferent salts which occur in some waters but none of these occurred in the water used. His findings relative to acidity, alkalinity, etc. are of general application and may be summarized as follows : Water which is consistently slightly alkaline lessens the activity of fishes and the mortality is high. N/ioo alkahnity, KOH, (56 pts. per m. ) kills them in a few hours. Neutral water also seems to be toxic to the fishes, and they become less and less active until death may occur. An optimimi acidity is obvious. 2—6 cc. of COo per liter, (4-12 pts. per m. ) seems to be the proper acid concentration for many fresh water fishes. Higher concentrations prove fatal very soon, though fishes will live for some time in 10—20 cc. per liter (20—40 pts. per m. ) of carbon dioxide. N/io,ooo H2SO4, (4.9 pts. per m.) is fatal in a day or so, but N/20,000 HoSOi, (2.4 pts. per m.) seems to be near their optimum as they live in this concentration for a long time. Fishes react very definitely to exceedingly small concentrations of hydrogen and hydroxyl ions. Fresh-water fishes in a gradient which is slightly acid at one end and neutral near the middle and slightly alkaline at the other end will spend most of their time in the acid end, turning back from the alkaline end at a point just on the acid side of neutrality. The concentration here when tested shows that they turn back when the acid concentration falls below N/ 12,000 carbonic acid (3.5 pts. perm.). In a gradient where the fishes may select between alkalinity and neutrality they avoid the neutral water to some extent and spend the greater part of the time in slightly alkaline water. b. Quantity of Water. In the aquaria suckers, small-mouthed and large-mouthed black bass died frequently when the flow of water was small and the depth in the aquaria more than 6 inches. This was probably due to insuf- ficient oxygen. When the amount of water in the aquaria was small and the flow sluggish as was the case when the water was 2 or 3 inches 386 deep the greatest mortality was among the darters and the minnows (Notropis and Pimephales). These died in numbers in the aquaria, no darters at all being kept alive. After a number of trials a series of experiments was performed to demonstrate the cause of the death of the fishes (darters, Btheostoiiia cocndcum, and minnows, Pimephales notatus) . The procedure was as follows : Twenty-two 5 in. x 8 in. battery jars were set in a water bath,—the tank into which and out of which tap water flowed,—ready for filling with water modified variously, by boiling, aeration, and the addition of various substances as shown in Table I. Minnows and darters were given separate jars. The following table shows the results. Table I Average Life dp to Ten Days. Two individuals in each condition except where otherwise stated. 750 cc. H,0 387 c. The Transportation of Fishes. In collecting fishes for such experimental work they may be se- cured and brought to the laboratory in numbers if only a very small quantity of water is used. In general it is best to allow the dorsal fins of sunfish, basses, crappies, and suckers to protrude from the water. Minnows on the other hand, live best in about 3 inches of water. In this way many fishes may be safely brought in without the usual l"bor of carrying a quantity of water. 4. FISHES USED. The fishes used in this experiment belong to the species mentioned below. Common name Orange-spotted sunfish Blue-spotted sunfish Blue-gill Long-eared sunfish Rock bass Small-mouthed black bass Large-mouthed black bass Blunt-nosed minnow Steel-colored minnow Common shiner Golden shiner Common sucker Bullheads Brook silverside Rainbow darter Scientific name Abundance Lepomis humilis Gir. Abundant Lcpomis cyanellus Raf. Common Lepomis pallidus Mit. Common Lepomis tnegalotis Raf. Common Ambloplitcs rupestris Raf. Common Micropterus dolomieu Lac. Common Micropterus salnioides Lac. Common Pimeplmles notatiis Raf. Very common Notropis zuhipplii Gir. Abundant Notropiis cornntus Mit. Abundant Abramis crysoleuca Mit. Abundant Catostomits commersonii Lac. Common Ameiiirus nebulosits LeS. Common Labidcstlies siccultt-s Cope Occasional Btheostoma coeruleum St. Common The small sunfish, Lcpomis Iiiimilis, was used as a standard fish. It is only about 4" long when adult, is widely distributed in Illi- nois and without value as a food fish. A sufficient number of other fishes were studied to make its relative sensitiveness clear, and min- nows and one of the basses were nearly always used in reaction ex- periments. Minnows were used also to show toxicity. The condition of individual fishes is also a matter of importance. In a few cases fishes with obvious external protozoan parasites were killed in coal-tar products, and in every case they died sooner than the 388 normal fish. Thus in detailed work it is important to open and exam- ine all fishes dying sooner than other fish of the same size. When fishes are brought into the laboratory they do not ordi- narily take food and are often not well-fed or in a semi-starved state when the experiments are performed. Wells found that in the case of salts the resistance to adverse conditions is slightly increased by starvation. To test this, fishes were kept in the aquaria from May 15 to Aug. 23. All died but six; those which died being their only source of food. On Aug. 23 fishes recently caught were compared with tlie starved ones. The starved fishes were from 3 to 3^ inches long (7-9 cm.) and had an average weight of 7.6 gm. while fishes of this length collected from the streams weighed twice as much. In the fresh waste the starved fish died somewhat sooner on the average though the time of some individuals of about the same length as the well-fed individuals was almost the same as the latter. In aerated waste the starved fishes lived longest. On account of the small num- ber (six) of starved fishes available the experiment could not be car- ried out on a large enough scale to establish significant averages but there was nothing to indicate that ariy important differences existed. IV. Gas-Waste;—Its Character and Constituents. The waste of the Champaign gas plant consists of what is known as the "drip", which accumidates in the bottom of the holders and in the pipes leading to and from them, also in the mains throughout the town. It consists of water with illuminating gases and other coal products in solution. On the surface of this water a light tar floats, while some heavy tar may rest at the bottom. The waste is pumped from the inlet and outlet of the holder onto the ground beside the tank, and is alleged to flow into the Boneyard Creek in wet weather. The light tar is used by the gas-works people for paint, for which purpose it appears to have some value. It dries hard and rather quickly. The heavy tar is removed but as is the usual case with small plants, everything else is thrown away. Coal-tar is an excessively complex mixture of chemical compounds many of which occur in its distillation between naphthalene on the one hand and anthracene on the other. It contains nitrogenous com- pounds, chiefly of a basic nature. The usual constituents of the waste and tar varies with the coal used, the temperature and the method of washing and testing the gases etc. during the process of manufacture and the amount of water gas added. These constituents may be classified and described as follows (Lunge '00). 389 A. Nitrogcnizcd Compounds. Of this group ammonia and its salts are of constant occurrence. The volume of ammonia in the drip from tiie Champaign iiolder inlet is usually about 200% of the volume of liquid. The salts are abundant in all parts of an ordinary plant. Such well known compounds as ethylamine, aniline, pyridine, and quinoline belong to this group. B. Sulphuretted Compounds. To this group belong such well known compounds as hydrogen sulphide, sulphur dioxide and carbon bisulphide, and the less well known licjuid thiophene, which is common as an impurity in benzene. All are very poisonous. C. Oxygenized Compounds. In this group are included such well known substances as acetone, acetic and benzoic acids, and phenol and the cresols. D. Hydrocarbons. To this group belong the solids phenanthrene, anthracene, naphtha- lene, and the volatile liquids, xylene, toluene, benzene, etc. The gases are numerous, including acetylene, ethylene, and methane. E. Carbon Oxides. These are the two well known gases carbon dioxide and monoxide. Gas waste from plants which remove only the heavy tar may be regarded as containing all of these compounds. The dissolved gases of course escape into the air but are held in great quantity and given of? slowly from the tarry materials. V. Toxicity of Wastes from the Champaign Plant. The toxicity of different samples differs greatly, some samples be- ing ten or twelve times as toxic as others. This depends upon the interval since the main was pumped and whether it comes from the inlet or the outlet to the holder. Attempts were made to determine the toxicity of waste by means of indicators and acid. There appears to be no relation between the amount of normal acid required to pro- duce a red color with methyl orange and toxicity to fish. The same difficultv was encountered when normal alkali was used. Likewise BOO the amount of iodine absorbed appeared to bear no definite relation to toxicity. It is probably best to determine the toxicity of waste with fishes rather than by chemical means. I. METHODS OF EXPERIMENTING WITH WASTE. a. Standing water. Battery jars 5 inches in diameter and 8 inches deep are filled to a depth of 4 inches (10 cm.) with waste diluted for use. This gave 2,000 cc. of liquid with 113 sq. cm. of exposed surface and gave con- ditions under which one or two fishes would live for days. This method simulated in a general way the conditions in polluted standing water. The period of toxicity determination being one or two hours the method was free from serious objections. h. Running Water Mctlwd. A bottle with a very wide neck, holding a liter is fitted with a rub- ber stopper in which are three holes (M. Fig. i). A 12 liter aspira- tor bottle (Fig. I W) with stopper tubulature is closed at the bottom aperture antl filled with waste about ten times as strong as is required for the experiment at hand. One part of the diluted waste is run into the bottle through one opening in the three holed rubber stopper, while 9 parts of water are introduced through another. The water flows from this bottle into a larger bottle holding about three liters, in which the fishes were confined. The flows were set with pinch cocks on rubber tubing and adjusted from time to time. The flows used varied from time to time but usually were between 100 to 300 cc. per minute. The object was to secure definite concentrations rather than definite flows, as all that is necessary is to change the water often and simulate the conditions in running streams. The temperature of such experi- ments was usually ly'C c. Bottle method. For determining the exact toxicity of any sample of waste when unexposed to the air it is necessary to proceed in an entirely different way. A bottle with a wide mouth, holding a little more than four liters, is supplied with a close fitting rubber stopper. It is first filled with water to the four liter mark scratched on the outside. A definite amount of waste is then run in from a burette or Mohr's pipette. The bottle is then shaken until all of the substance is in solution. The free air space, which should not exceed 2% of the volume of the 391 water, serves when the bottle is laid on its side to show any undis- solved substance lighter than the water, thus making the method later applicable to tiie light slightl)- soluble constituents of waste. The temperature of such experiments was usually 20°C. 2. TOXICITY OF WASTE AND TAR. The toxicity of waste from the Champaign plant varies so that a general statement as to the toxicity can hardly be made. In general the greater the amount of tar the more toxic the waste. The most toxic sample contained niucli tar. Eight hundredths of a cc. of the waste was introduced from a Mohr pipette into four liters of water in a four liter bottle. The water was shaken until all the waste had gone into solution excepting a slight tarry film on the sides of the bot- tle near the surface of the water. It is impossible to say how much of the substance actually went into solution, but assuming that half of it did, it may be safely said that ten to twenty parts per million of this waste killed a 4-5 gram Lcpoiiiis liuiiiilis in an hour, while twice that amount killed such fishes in from ten to thirty-five minutes. An- other sample with less tar killed fishes of the same size in five hours when 1,000 parts per million were present. A small amount of tar was rubbed on the sides of several full grown Lepomis humilis and the fishes left in open aquaria ; all died in from one to nineteen hours. A small amount of tar was rubbed in the mouths and on the sides of several suckers and orange-spotted sunfishes in open aquaria. All died in from one to nineteen hours. Marsh ('07) found tar very toxic to perch and bass. Aerating and boiling removed toxic constituents of the waste. For example, a sample of waste was treated as follows : 1. Fresh waste was added to 99 times its volume of aerated water as quickly as possible, and the small space above the water in the large bottle was filled with illuminating gas. 2. Some of the same waste was aerated by pouring from one beaker to another for three minutes. This was added to 99 parts of water and corked. 3. Some of the same waste was boiled vigorously for several minutes, until all odor of ammonia was removed, and added to 99 parts of aerated water. The efifect of these treatments is illustrated by the following ex- periment which is one of many. A liter of each of the three kinds of waste was put into each of three battery jars and 4-5 gm. orange- spotted sunfishes placed in them. They survived as follows : } } 393 to the polluting substances. Fishes turn away from dangerous sub- stances which are normally found in their usual environment, but with strange and unusual substances such as are thrown into streams by gas-works and other industrial plants, they frequently enter and follow up to points where the concentrations are fatal, or fail to recog- nize the dangerous substance at all and often stay in it until they are intoxicated and finally die there. (Chart II, graphs 8-i i ; Chart V, graph 60. ) Conditions and Methods of Study. The experiments were performed in a gradient tank (N), figure I. The tank used in these experiments was 122.3 cm. long, 15 cm. wide, 13 cm. deep. The front wall was of plate glass and a plate glass top was used at times. Water of two kinds, normal and polluted, was used in the experiments. One kind was allowed to flow into one end at a definite rate and another kind into the other end at the same rate. It flowed out at the middle at the top and at tlie bottom so that the two kinds of water met at the center. The outflow at the center did not of course prevent the mixing of the two kinds of water and thus the middle section, equal to one half or one third of the tank was a gradient between two kinds of water. The water entered both ends at the same rate (usually 600 cc. per minute) through tees the cross-bars of which contained a number of small holes. The cross-bars of the tees were at the center of the ends of the tank behind screens. The drain openings were located at the center near the top and in the bottom. The outer openings of the drain tubes were at the level of the water in the tank. We found no evidence that fishes reacted to the slight current produced by the water flowing in at the ends and drifting toward the center and out through the drains. Since each half of the tank held about 9 liters, it required 15 minutes to fill it or to replace all of the water in one of the halves. The tank was enclosed under a dark hood. Two electric lights were fixed in the rear and above the center of the two halves, i.e., above a point midway between the screen partition and the center drain. The light was 15—20 cm. above the surface of the water which was 13 cm. deep. The experiments were observed through openings in the hood above the lights or through the glass side late at night. Fishes do not usual- ly note objects separated from them by a light. Water differing as little as possible from that in which the fishes usually live was used for control readings. Controls were observed and conditions in the two ends of these were the same either because the water introduced at the two ends was alike or because no water 394 was run into either end (standing water). In the controls (Chart I) the fishes usually swam from end to end in a rather symmetrical fashion, and thus comparing tliese movements with those occurring when the fishes encountered differences in water, we are able to de- termine the reactions of the fishes to the differences. When the differences between the solutes at the two ends of the tank were not great we found by chemical tests that the central portion of the tank was a gradient between the characteristic waters intro- duced at the two ends. Usually the end thirds were essentially like the inflowing water. When the difference in concentration was great the region of the gradient was proportionally longer and the ends with th.e inflowing concentrations correspondingly shorter. When the dif- ference in concentration was very great the entire tank was gradient. For an experiment a fish was placed in a dish containing enough water to barely cover it and set above the tank. When all was in readiness the fish was liberated in the center of the tank. Marks on the sides divided the tank into thirds. The fish nearly always swims back and forth, apparently exploring the tank. The movements of the fish were recorded graphically as shown in Chart I. For this purpose sheets of ruled paper were used. Four vertical double rulings corresponded to the thirds and two ends of the tank. Distance from right to left was taken to represent the length of the tank, vertical distance to represent time and the graphs drawn to scale. The width of the tank was ignored. The graphs on the following pages are copies of the originals. The experiments were conducted with water at about i7°C. Before or after the experiment, the headings of the sheets were filled with data regarding the kind, size, and previous history of the fish, the conditions in the tank, concentration of the solutes and other significant data. The fish was observed continuously for twenty or more minutes. Fishes are positive to waste in all concentrations tried. Fishes are positive or indefinite to illuminating gas, and to com-, binations of the most important illuminating .gas constituents in both acid water with 2-3 cc. of oxygen per liter and in alkaline water at oxygen saturation (Chart II, graphs 11 and 12; Chart \^, graphs 53 and 54). VI. The Toxicity of Ii o S ; 2. • . "^ !^ O fl 05 „. Oq I—I tr' rr Cfl c» ^ B" 5" S S ^§ra -^ -* P* &= h- ti S" " '~'^~- :a 2 S g- 03 Ei tn u CD d c" "^ (-1- . CO "^ O C O) tr" u I* tJ O ^ !-. ^C o 2. * G. 3 ? -^ S r;- »" : c^:: 5" to d o p ii m "-^ B B . J3 [^ o h^ ra « S 5" 2. 5^ S 2 ^--w 3- ^Td B/^ j3 e-- b-3 o"-w B- ^ =" ;i 53 B B - ^p^3 ^ |_H "^ 5 -- ^ '^ — to ;2j pj- 2- — . . ,—. o M " h3 P 5^ o ' '^ TJ n^ G i Cj -" ™ ^ ' ^ '^ a> <^ re o ^ B" !i jt ^ "^ ^ "^"5 W 2reft.t;rr re ^. ^ , *p (»_ S ^2 S- re < re B re cr ^ =5 "-h :? '''' (W o B" & Ei-re ^ " ",§ re ^ p < p --^ B 3 '4 S. JO C r^ - £3* p ft) ^Si-jp*' PLrpti o M. r* OJ C « sr:i p-^SgPB- S„ B-S ?!£=-£,"• B^P re^ B„. SnPS^ B giK! B re p B 2 o a 2 re 1 '^ ap^c-"^c-rt-„o-ireoBpre^Q -^^Cl.X(^u;BOr^ B 2, re " re B* B- ^ ^ B - p o I^re o „ B^ r/) ^- Cij g'2.re5S:Sprere B S„>a ^ g; g_ P P re ^crq S o p o S -. '-» ^ t-."-B-«- c-ra o B- - o p 5a E '^ re S- = re E re C^^^ i7Q <1 O ^ ^ O CO "-rf CO M- Oj ^.^^^ o re p '^ B-re B ^ .,^^° E. K'B- B- S! H. re re _. ° re a re Si" ° CO - EJ* fo re S-i -fc S.re Prg^ re a- 2 3 2.S " _ ^- CO 2 1° r 'S gr g g- S- s& »= B re B to • P_ P' 1^ ^ tiB-re 3 p p p n a ^o-^re ji i:d fo re S°E.B ., B S P ^ B re C" re B p ^"^ B Bcg; 3. 5; =° W-S2 . ffi -! B_ w 5 a Chart I. The graphs on this chart show the movements of fishes in the gradient tank %vhen no contaminating substance has been added at the end, and the water is there- fore of equal purity throughout. Graphs 1-5 have been previously published. The gradient tank is shown in Figure 1, N, on preceding page. This is a diagram of a longitudinal section of the lank. The left hand end was used for the introduction of water such as the fishes were taken from and the right hand end was used for the introduction of water to which the substance being tested had been added. The water was introduced through a number of small openings in pipes which extended crosswise at each end of the tank, midway between top and liottom. The water flowed out at the center from both top and bottom. This gave pure water at the left hand end and usually in the ease of dissolved solids and liquids, throughout about one third of the tank while the approximate full concentration of the polluting substance extended throughout the right hand third. The central third contained a mixture in which the concentration of the substance added at the right decreased from right to left. The central portion of the tank was accordingly a gradient between the two kinds of water introduced into the ends. The fishes introduced into the tank usuall}- swim from end to end. The record of the movements of the fish ^\'as made by tracing their longitudinal movements in the tank on paper with reference to a time scale. Thus in Graph 1 the fish passed from the center to the left end and back to the right end during the first minute. Graph 1 shows the longitudinal movements of a river chub made up as a com- posite of a number of such graphs to show that on the whole no more time was spent in one end of the tank than in the other. Graph 2 shows the movements of two individuals of the green sunfish. Where the broken line appears, the two were moving separately. It will be noted that fishes made long staj's in the ends but usually moved back and forth nearly always without turning back at the center. Graph 3 shows the movements of a golden shiner in a uniform tank. It will be noted that the fish rarely turns around except at the end. Graph 4 shows the movements of a specimen of Notropis. There was little activity and the fish turned back at the center once. Graph 5 shows the movements of a specimen of rock bass. Graph 6 shows the movcme-nts of a specimen of the orange-spotted sunfish. This species often turned before reaching the end of the tank but the number of turnings in the central third were the same from each direction. Graph 7 shows the movements of a long-eared sunfish. It sometimes turned near the center, but about the same number of times from each end. Chart I. Chart IT. The relations of tank length to time scale is the same as in Chart I. In the case of this and all the charts which follow, the iioUuting substance was introduced into the right hand end of the gradient tank and is accordingly shown at the right side of the graphs. The vertical broken lines are intended to indicate the location of thirds of the tank length. The solid black area at the right between the two lines at the head of each grai>h is intended to show the part of the tank in which the polluted water is full strength and the narrowing of this black area from right to left in the middle third is intended to indicate the region of principal gradient. The unpolluted water contained about 5 cc. CO, per liter. X indicates that the fish became intoxicated; the arrow that it was driven. Graph 8 shows the positive reaction of an orange-spotted sunfish to 1 part of weak waste to 25 parts of water. The tishes avoided the normal water and remained most of the time in the high cojicentration and gradient. Graph 9 shows the reaction of the golden shiner to waste, 1 part in 100 of water, which killed the standard fish in a little more than one hour. In this case the fish avoided the nupollutcd water and did not enter it at all until driven as indi- cated by the arrow. Graph 10 shows the reaction of an orange-spotted sunfish to the same solution as was used in the case of graph 8. In this case the fish was negative, showing that there is some variation in the reaction. Graph 11 shows the reaction of a minnow (Pimephales) , indicated by the broken line. The fish was negative for a time but became intoxicated after a little more than two minutes and then remained positive after being driven into the strongest solution. Graph 12 shows the reaction of an orange-spotted sunfish to illuminating gas. There was little activity and one fish remained in the clear water while the other remained in the polluted water during the period of observation. The amount of illuminating gas was not determined; much more was forced into the water than would go into solution in the pipe of the lower cooler (Fig. 1, preceding Chart I). Graph 13 shows the reaction of a largie-mouthed black bass to illuminating gas under the same conditions as in graph 12. The fish was driven into the stronger solution of gas and reacted positively thereafter. Graph 14 shows the reaction of the golden shiner to a solution of ammonia which would prove fatal in a short time. The fish on the whole remained most of the time in the part of the tank containing a somewhat diluted solution but gave no avoiding reactions. Graph 15 shows the jjositive reaction of a minnow {Notropis) to ammonia solution under the same conditions as graph 14. In this case the fish was clearly positive. Graph 16 shows the positive reaction of two individuals of Notropis to a solution of approximately one gram per liter of NH;(CO), which kills such fishes in less than an hour. Graph 17 shows a decidedly positive reaction of a roek bass to approximately 1 gram per liter of ammonium chloride. The experiment continued for 10 minutes after the portion shown without change of result. Graph 18 shows the reaction of a roek bass to 0.7 gram of ammonium sulphate ]>er liter. The fish rested in the polluted water throughout the greater part of the time and turned back when a decreased concentration was encountered. Graph 19 shows the reaction of a minnow {Pimephales) to approximately 0.25 gram of ammonium sulphocyanate (sulphocyanide) per liter. The fish moved back and forth actively and turned back regularly from a pure water. Graph 20 shows the reaction of a full-grown rock bass to the same concentration as in Graph 19. The fish was driven into the pure water at the end of seven minutes but soon returned to the polluted portion. Chart II. Chart III. Graph 21 shows the reaction of a minnow (Notropis) to approximately 0.25 fjram of ammonium ferrocyanide per liter. The fish was active and swam back and forth turning back from the pure water repeatedly. Graph 22 shows a slight preference on the part of an orange-spotted sunfish for water containing ethylamine. Graph 23 shows the indifference of two orange-spotted sunfishes to about one- tenth ec. of aniline per liter. Graph 24 shows the marked activity of two individuals of Noiropis in a pyridine gradient and their repeated avoidance of the pure water throughout the experiment. Graph 25 shows the marked negative reaction of the minnow {Pimephales) and long-eared sunfish to 0.16 cc. of quinoline per liter. Graph 26 shows a marked negative reaction of a minnow (Notropis) to a weak solution of isoquinoline. In the case of both quinoline and isoquinoline the fishes were active and turned back from the polluted water; the rule for the former, and the exception for the latter. Graph 27 shows the negative reaction of a minnow (Pimephales) to water con- taining 8 cc. per liter of hydrogen sulphide. The fish became intoxicated at the end of five minutes as indicated by the X in the graph. Graph 28 shows the positive reaction of a full-grown rock bass to two cc. per liter of h.vdrogen sulphide. The pure water was encountered repeatedly and re- peatedly avoided. Graph 29 shows the reaction of two orange-spotted sunfishes tO' approximately .500 cc. of sulphur dioxide per liter. The fishes spent the greater part of the time in the central part of the tank. Graph 30 shows the positive reaction of two orange-spotted sunfishes to 5 cc. of sulphur dioxide per liter. Graph 31 shows the positive reaction of a long-eared sunfish to water containing less than 1 cc. per liter of carbon disulphide. Graph 32 shows the reaction of a minnow (Pimephales) under the same con- ditions as graph 31. The fish was positive during the first seven minutes and nega- tive during the last three. Graph 33 shows the positive reaction of a rock bass to a fatal concentration of thiophene. Graph 34 shows an equally positive reaction of a long-eared sunfish. Chart J II. 29 . 31 Sulphur C«rbon Dioxi«B BlPJlplUda 33 Wl^ I I __ , ^ , , Hydrogen Sulpbur Carbon IthylMiln* _ Pyrldln* iBoquinolln* Sulphide DloiidM Sulp^lda 2l; —T? Tnioptaon* Chart IV. Graph 35 shows the positive reaction of two minnows (Pimephales) to water containing 2% cc. per liter of acetone. The fishes turned back repeatedly from the pure water. Graph 36 shows the decidedly negative reaction of a minnow {Pimephales) to water containing a fatal concentration of benzoic acid. Graph 37 shows the reaction of a green sunfish to water containing 0.5 cc. per liter of phenol. The fish was markedly positive to the phenol during the first ten minutes, when the activity increased, probably duo to its irritating effects. The greater part of the time was spent in the phenol however. Graph 38 shows the reaction of a full-grown rock bass to approximately 0.1 cc. per liter of orthocresol, which would prove fatal to the fish in an hour or more. When the fish entered the polluted water the first time it did not recognize it at all. It gave no avoiding reaction. Later it moved toward the weaker solution and turned back again into the stronger solution. After becoming partially intoxicated, it moved into the pure water but returned to the fatal solution again and was completely over- come there. Graph 39 shows the reaction of an orange-spotted sunfish to 0.3 cc. paracresol per liter—about three times as much as is required to kill one of the fishes in one hour. It is to be. noted in particular that the fish after trying the pure water twice, gradually avoided it more and more until it finally came to rest in the strongest solution of paracresol. Graph 40 shows the reaction of two orange-spotted sunfishes to 0.12 cc. of metacresol, sufficient to kill them in an hour. One fish was negative and the other positive. The fish which happened to enter the polluted water at first became intoxicated and remained positive thereafter. Fishes are often negative to meta- cresol. Graph 41 shows the reaction of an adult rock bass to a saturated solution of phenauthrene. Fishes are often indefinite to this substance. Graph 42 shows the negative reaction of an adult rock bass to a saturated solu- tion of naphthalene. Graph 43 shows the positive reaction of an individual orange-spotted sunfish to a saturated solution of naphthalene. Fishes are generally positive to this deadly substance. Graph 44 shows the reaction of an adult rock bass to a mixture of pure water 3 parts and water saturated with xylene 1 part. The fish was decidedly positive and was soon intoxicated. Graph 45 shows the reaction of a minnow (Notropis) to water containing ap- proximately 0.08 cc. per liter of toluene. The fish is decidedly positive, though this concentration would kill it in less than an hour. Graph 46 shows the reaction of two orange-spotted sunfishes to 0.04 cc. of ben- zene per liter—sufficient to kill them in an hour. In this experiment the fishes avoided the pure water and finally came to rest in the center. Graph 47 shows the reaction of a orange-spotted sunfish and a rock bass to a slightly weaker concentration of benzene than was used in the case of graph 46. In this case the fishes both finally avoided the polluted water. Chart IV. Chart V. Graph 48 shows the positive reaction of an orange-spotted sunfish to 0.22 ee. of amylene. Eeactions to this drug arc usually positive. Graph 49 shows the positive reaction of two minnows (Notropis) to amylene in which they appear to have selected an optimum concentration, near the center. Graph 50 shows the reaction of a largc-mouthod black Vjass to water containing 34.4 cc. of ethylene per liter. It is clearly positive, though this concentration would kill the fish in less than an hour. Graph .51 shows the reaction of a minnow (Notropis) to water containing about ten cc. per liter of acetylene. The reaction is clearly positive though the gas is not fatal. Graph 52 shows the reaction of an orange-spotted sunfish to about ten cc. of acetylene per liter; the reaction is clearly positive. Graph 53 shows the i-caction of three orange-spotted sunfishes to a mixture of carbon monoxide (1.4 cc. per liter) and ethylene (9.6 cc. of ethylene). Graph 54 shows the reaction- of two suckers to the same solution as in graph 53. Graph 55 shows the reaction of an individual (Airamis) to ammonia in alkaline \' atcr. The fish was positive, as in acid water. , Graph 56 shows the reaction of a large-mouthed black bass to paracresol in alkaline water. The general result is the same as in acid water. Graph 57 shows the reaction of an orange-spotted sunfish to orthocresol in alka- line water. The fish was positive, as in the acid water. Graph 57 shows the reaction of an orange-spotted sunfish to phenol in alkaline water. This and other fishes are positive, as in acid water. Graph 58 shows the positive reaction of an orange-spotted sunfish to naphthalene in alkaline water. Graph 59 shows the positive reaction of a large-mouthed black bass to toluene in alkaline water. Graph 60 shows the positive reaction of a blue-gill to gas waste in alkaline water. Chart V.