Bulletin BULLETIN of the ILLINOIS NATURAL HISTORY SURVEY HARLOW B. MILLS, Chief The Bass-Bluegill Combination in a Small Artificial Lake GEORGE W. BENNETT Printed by Authority of the STATE OF ILLINOIS DWIGHT H. GREEN, Governor DEPARTMENT OF REGISTRATION AND EDUCATION FRANK G. THOMPSON, Dirtclor STATE OF ILLINOIS DwiGHT H. Green, Governor DEPARTMENT OF REGISTRATION AND EDUCATION Frank G. Thompson, Director NATURAL HISTORY SURVEY DIVISION Harlow B. Mills, Chief lume 24 B l" 1. 1. K'l" I N Articled The Bass-Bluegill Combination in a Small Artificial Lake CJEORCit: W. HENNET'I" Printed by Authority of the Stale of Illinois URBANA, ILLINOIS December 1948 STATE OF ILLINOIS DwiGHT H. Green, Governor DEPARTMENT OF REGISTRATION AND EDUCATION Frank G. Thompson, Director BOARD OF NATURAL RESOURCES AND CONSERVATION Frank G. Thompson, Chairman A. E. Emerson, Ph.D., Biology George D. Stoddard, Ph.D., Litt.D., L.H.D., L. H. Tiffany, Ph.D., Forestry LL.D., President oj the University 0} Illinois L. R. HowsoN, B.S.C.E., C.E., Walter H. Newhouse, Ph.D., Geology Engineering Roger Adams, Ph.D., D.Sc, Chemistry NATURAL HISTORY SURVEY DIVISION Urbana, Illinois Scientific and Technical Staff Harlow B. Mills, Ph.D., Chief Bessie B. Henderson, M.S., Assistant to the ChieJ Section of Economic Entomology George C. Decker, Ph.D., Entomologist and Head J. H. Bigger, M.S., Entomologist L. L. English, Ph.D., Entomologist C. J. Weinman, Ph.D., Entomologist S. C. Chandler, B.S., Associate Entomolo- gist James W. Apple, M.S., Associate Ento- mologist Willis N. Bruce, M.A., Assistant Ento- mologist John M. Wright, B.A., Assistant F.nto- mologist H. B. Petty, M.A., Associate in Entomolo- gy Extension George F. Ludvik, M.A., Special 'Research Assistant John E. Porter, M.S., Laboratory Assistant Section of Faunistic Surveys and Insect Identification H. H. Ross, Ph.D., Systematic Entomologist and Head Milton W. Sanderson, Ph.D., Associate Taxonomist B. D. Burks, Ph.D., Associate Taxonomist Lewis J. Stannard, Jr., M.S., Assistant Taxonomist Leonora K. Gloyd, M.S., Laboratory Assist- ant Philip W. Smith, B.S., Laboratory Assistant Dorothy A. Moulton, Technical Assistant Section of Applied Botany and Plant Pa- thology Leo R. Tehon, Ph.D., Botanist and Head J. Cedric Carter, Ph.D., Plant Pathologist J. L. Forsberg, M.S., Associate Plant Pa- thologist G. H. Boewe, M.S., Assistant Plant Pa- thologist Robert A. Evers, M.S., Assistant Botanist Section of Forestry WiLLET N. Wandell, M.F., Forester and Head Lawson B. Culver, B.S., Associate in For- estry Extension Section of Aquatic Biology George W. Bennett, Ph.D., Ai/uatic Bi- ologist and Head William C. Starrett, Ph.D., Associate Aquat- ic Biologist D. F. Hansen, Ph.D., Assistant Aquatic Biologist R. Weldon Larimore, M.S., Research Assist- ant Jacob H. Lemm, Field Assistant Daniel Avery, Field Assistant Section of Game Research and Manage- ment Ralph E. Yeatter, Ph.D., Game Specialist Frank C. Bellrose, B.S., Associate Gamf Specialist Harold C. Hanson, M.S., Assistant Game'' Specialist Section of Publications and Public Re- • lations James S. Avars, B.S., Technical Editor and Head Blanche P. Young, B.A., Assistant Tech- nical Editor Charles L. Scott, B.S., Assistant Technical Photographer Technical Library Marguerite Simmons, M.A., M.S., Tech- nical Librarian Cooperative Wildlife Research Paul J. Moore, B.S., Project Leader George C. Arthur, B.S., Project Leader Lysle R. Pietsch, M.F., Project Leader A. B. Cowan, B.S.F., Assistant Project Leader Consultant in Herpetology: Hobart M. Smith, Ph.D., Assistant Professor oJ Zoology, University of Illinois. This paper is a contribution from the Section of Aquatic Biology. (65976—3,500—10-48) CONTENTS Acknowledgments -^yg PROPPING PrOCEDLRE ^yo "iSH ^'lEI.D ^jjl OaM FAII.fRl! ^g9 'OND Habitat ^gj 'ecetation vs. Fish >'ii;i.i) ^gy iROvvTH Rates ^g7 .'ONDITION AND (jROWTH 39| CAl-E Analysis ,?9? PAWNING AND YoUNU FiSH 395 exual Cycle 397 OODS 398 )iscussioN 407 UMMARY 411 ITERATURE ClTED 412 The frontispiece shows wing nets being set in Fork Lake, 1939. In the first year of crop- ng, only two or three win;; nets were required ami these were set with iniliviiliial leads thai d no relation to one another. In other years, six wing nets were required and were arranged such a way as to trap fish in all but the deepest part of the pond. The Bass-Bluegill Combination in a Small Artificial Lake TEN years ago, Fork Lake, a pond of 1.38 acres on the farm of Paul S. Smith near Mount Zion, Illinois, Has probably typical of many of the older man-made ponds in central Illinois. When this pond was assigned for study to aquatic biologists of the Illinois Natural Histor\ Survey in 1938, it was 18 years old and had been used extensively for fishing, waterfowl shooting, and general outdoor recreation. A brief history of the recrea- tional benefits derived from the pond has been published elsewhere (Thompson ^ Bennett 1939«). With the outlawing of duck baiting and use of live decoys in 1935, the success of waterfowl shooting on Fork Lake was limited. Fishing, which was considered good from 1''26 to 1930. had become poor through the dexelopment of large popula- tions of black bullheads, carp, and bufia- loes. These undesirable fishes apparently were limiting the success of reproduction and curtailing the growth rates of large- mouth bass, crappies, bluegills, and other unfish in the pond. Fish in the pond were poisoned with rotenone on June 7, 1938, and a census was made of them (Thompson .^ Bennett 1939a). At this time. Fork Lake con- tained 5,350 fish weighing 774 pounds and consisting of 16 species. The rough fish, mostly carp and redmouth buffalo, made up 47.5 per cent by weight of all fish, and bullhead (plus four channel catfish) 41.2 per cent. Largemouth bass and panfish totaled only 6.3 per cent of the weight of all fish. The weight of fish per acre was 539 pounds, which indicated a moderately high natural fertility for the pond. At the time of the census. Fork Lake contained only 145 fish of desirable species and of a size large enough to interest anglers. I he pond was restocked bctw een June 11 and 18 with 270 stunted adult bluegills, GEORGK W. BENNETT Lepornis iiiacrnchirus Rafinesque, weigh- ing about 40 pounds and 1,440 largemouth bass fry, Miiropltrus saliiioides Lacepede, weighing 0.15 pound. Both the bluegills ;uid the bass were taken from Homewood Lake, near Decatur, Illinois. The blue- gills began to spawn almost at once; on June 22, 27 bluegill nests were counted. The bass averaged 0.87 inch total length when stocked, and, during October, 74 taken on fl\ -rod lures a\ eraged 6.45 inches. Some of the young bluegills that were spawned in 1938 and escaped being eaten by bass were nearly 3.75 inches long by October; others were less than 0.75 inch. Early in 1939 the decision was made to crop the new fish population by using 1 -inch-mesh wing nets, 1 -inch-mesh (bar) seines, and hook and line, and to determine the effect of such cropping upon this bass- hluegill population. Cropping efforts resulted in a substantial \ield of fish in 1939 (Bennett, Thompson, &' Parr 1940). The annual yields of blue- gills were smaller in 1940 and 1941. The xields of bass, although numericallv smaller in 1940 and 1941 than in 1939, increased in total weight each \ear from 1939 to 1941, inclusive. In the case of the bluegills, the first brood spawned in the lake grew more rapidly than all the others. Each succes- sive brood of bluegills grew less rapidl\ than the preceding brood in spite of in- tensive cropping, which might have been expected to reduce the competition for food and result in improved rates of growth. Successive yearly yields in pounds of blue- gills also progressively decreased. I believe that counter forces that reduced the food a\ailable for bluegills were responsible for the decline in yield and the absence of growth compensation. Two possible causes for a reduction in bluegill foods, which ma\ be assumed to [377] 378 Illinois Natural History Survey Bulletin 1'jI. 24. Art. i liave been responsible for a reduced yield of bluegills, were (1) loss of fertility of the pond resulting from the removal of a large poundage of fish and (2) the spread of rooted aquatic plants to reduce greatly the area of open water. The Fork Lake experiment was termi- nated by run-off water from a 4-inch rain that washed out the dam on July 8, 1942. As no Natural History Survey personnel were present at the time the actual break occurred, fish collected were those that remained in the pond after about two- thirds of the water had flowed out through the break, carrying with it a part of the fish population. In spite of this unfortu- nate circumstance, which eliminated the possibility of a final complete census of the fish, the partial census and the collections of previous years gave interesting and sig- nificant information on the bass-bluegill combination. Acknowledgments Many persons assisted in the Fork Lake investigation. Dr. David H. Thompson initiated the study in 1938 and cooperated with the author on the field investigations in 1939 and on the preparation of the two preliminary Fork Lake reports. Mr. Sam A. Parr, formerly an Inspector with the Illinois Department of Conservation and now Superintendent of Fisheries with that Department, gave invaluable assistance in field observations and cropping. Assistance in collecting monthly quotas of fishes was given by many members of the Natural History Survey staff, among whom were ' Dr. Donald F. Hansen, Dr. Louis A. Krumholz, Dr. Lee E. Yeager, Mr. Bruno von Limbach, Mr. Francis X. Lueth, Dr. C. L. Schloemer, and Dr. Gernon P. Hesselschwerdt. I am indebted to Mrs. Mary Shanor and Dr. Marian F. James, employed by the Natural History Survey for making stomach analyses of bass and bluegills. Dr. Herbert H. Ross, System- atic Entomologist of the Survey, and his- staff gave valuable assistance to Mrs. Shanor and Dr. James in the identification - of aquatic insects. To Mr. Paul S. Smith, owner of Fork Lake, I am grateful for the use of the pond and for the many pertinent observa- tions that he made during the early years of the experiment. Cropping Procedure It was not feasible to take a large annual crop of fish by hook and line from Fork Lake ; the Natural History Survey staff was unequal to the task and local people who cared to fish and who would keep records were few. Moreover, since many fish were needed for laboratory study, and a uniform sequence of collections was de- sirable, it was decided to crop the pond with nets and to augment the net catches with hook-and-line fishing. Nearly all anglers condemn the man who uses a trap net or seine as a game-fish poacher (whether the use of nets for commercial fishing is legal or not). Because such gear is frequently considered responsible for the r Fig. 1.—Outline map of Fork Lake, showing the customary set position of wing nets and lead nets used in cropping the pond during 1940, 1941, and 1942. It was not always possible to lilnck oflf the pond completely as is shown, for the bottom near shore was often hard and frequently would not support the outer wing pole of a net. With the arrangements of nets shown here, no Ush could swim for any great distance without running into a lead or net. Dec, 194S Bennett: Bass-Bi.uecill Combination 379 Fig. 2.—Six wing nets were needed to crop the Hsh in Fork Lake in 1940, 19+1, and 1942. These nets were set in groups of three, with leads blocking the pond between them, as shown in fig. 1. depletion ut our game fishes, it seemed of considerable value to use this type of equip- ment in cropping the pond. For a number of years, members of the aquatic biology stai? of the Natural His- torii- Survey have used a wing net or fyke net for sampling fish populations in Illi- nois lakes and streams. This net, similar to the wing nets used by commercial fisher- men on the Illinois River, consists of a tapered cylinder of webbing supported b\ hoops, open at the large end and closed at the small end by a drawstring. Inside are two funnels compo.sed of webbing, the first located just inside the open end of the net and the second about two-thirds of the distance from front to back. Attached to the hoop at the large open end of the net are two pieces of webbing, the wings, that are spread when the net is set and that function in leading fish into the net much as the wing fences on a cattle chute lead in the cattle, fig. 1. A separate lead net (much like a gill net or seine) is often used with a wing net to form an under- water "drift fence" ; occasionally wing nets are set at both ends of such a lead net. Fish are believed to swim up to this lead, and follow it along until they find themselves within the mouth of the wing net. ^Ving nets are not ordinarily baited except for catfish ; fish wander into the nets and are unable to find a way out. In shallow water these nets and leads are supported by long poles forced into the bottom mud. The webbing of legal com- mercial nets in Illinois must measure a minimum of 1.5 inches between knots on each of the four corners of a square mesh. Wing nets used at Fork Lake were of somewhat smaller mesh (0.50 to 0.75 inch ) and would hold bass as small as 5 inches in length and bluegills of ,^.5 inches. Most of the cropping of Fork Lake was accomplished in 1939 by using two or three wing nets of 3.5 feet and 4.5 feet 380 Illinois Natural History Survey Huli.etin /«/. 24, Art. 3 in hoop diameter. In 1940 and later years, it became necessary to use six wing nets, four having front hoops of 3.5 feet in diameter and two having front hoops of 4.5 feet, tig. 2. In 1939, each wing net was supplied with a separate lead net ; in 1940, 1941, and 1942, the six nets were set in two groups of three nets each with lead nets blocking the pond between them, as shown in fig. 1. \Vith this arrangement, no fish could swim for any great distance without running into a lead or wing net. Usually at 24-hour to 48-hour intervals after being set, nets were raised and the fish removed. Nets were moved frequently if they failed to catch fish. An attempt was made to fish the wing nets for a period each month from March to November of each year, beginning in March, 1939, and ending with the washout in July, 1942. Most of the fish caught were taken to the Urbana laboratories alive or iced, where they were weighed, measured, scaled, and dissected for stomachs and gonads. A small number of fish from Fork Lake were used to stock several new ponds constructed on farms near Mount Zion. Some collections in early spring and late fall were made by seine hauls with a 100- \ard, 1-inch-mesh seine. These hauls were necessary because the fish did not move enough to be caught in sufficient numbers in wing nets. Although some fish were taken with the seine, a satisfactory haul was difficult because of brush and snags in the bottom of the pond. Table 1 lists the net-days of fishing in Fork Lake for the vears 1939, 1940, 1941, and 1942. The length of time that nets were fished each month usually depended upon the catch. If good catches (15 to 25 pounds of fish) were made, the nets were removed after two raises. If catches were poor, the nets were left in the lake as long as practi- cable. Table 1 indicates that more net- days of fishing were required with each succeeding year, and table 5 shows that in spite of an increase in net-days of fishing the total yield in pounds went down each \ear. This decrease in the fish yield pro- duced by nets may have been influenced by one or both of two possibilities: (1) a reduction in the fish production of the pond and (2) a gradual improvement in the ability of the fish to avoid the nets or find a way out of them. Bass were much more successful in avoiding nets than were bluegills. Even in 1939 so few bass entered the nets that angling was necessary to produce adequate samples. As the origi- nal bass fry grew larger, 1939-1942, tjiey became more and more difficult to catch in nets. Bass of later broods, also, were caught more readily when small (5 to 7 inches) than when larger. Bluegills were taken in nets most readily in the spring throughout April and May and more readily in fall than in sum- mer. They were sometimes induced to move by violent storms or the flowing of warm w^ater into tlie pond from the water- shed. For example, nets set on March 9, 1942, caught few fish until a warm rain during the night of March 16 drained fiom the surrounding lands into the pond. When the nets were raised on March 17, they contained so many bluegills that some were replaced in the pond. The amount of hook-and-line fishing done in Fork Lake to augment the catch of bass by nets was moderate. Nearl\ al- ways done from a boat, fishing was with fly rod or bait rod and usually with arti- ficial baits. Angling periods were usually short (one-half to 3 hours) and repre- sented such time as was available on days when net raising or observations necessi- tated a trip to the pond. Hook-and-line catches varied from 16 fish per man-hour on May 6, 1941, to zero catches on several occasions. Table 2 gives the hook-and- line catch summary for years of the stud> and shows man-hours of fishing catch per man-hour, and numbers of bass and blue- gills taken. It is difficult to explain why no fish \\ere caught on hook and line in 1942. Part of the failure to catch fish was due to a late, cold spring accompanied by excessive rain- fall, which increased the turbidity of thci pond and kept the water abnormally cold.' The larger bass present (between 12 and 15 inches long and 1.0 and 1.5 pounds in weight) had become increasingly difficult to catcli in 1941, and apparently had learned through observation (because bas^ were seldom returned to the water when caught) to ignore artificial lures. In 1942 the poorness of fishing gave the picture ot a pond "fished out" for bass. That the lake was not "fished out" was shown b\ the catch of bass in the washout described later. Actually, during the spring of 1942, Deiemher, 194S Hknnf.tt: U.\ss-15i.ui;(;ii.i. Combination- SSI Table 1.—Net-days of fishiriii with 1-incli- mesh wing nets in Fork Lake, March. 1939, to June, 1942. Month .?82 Illinois Natur.al History Survey Bulletin Vol. 24. An. 3 Table 3.—Largemouth bass removed from Fork Lake, March-November, 1939, 1940, 1941; and March-June, 1942. Deccmhrr. I'US Bennett: Bass-Hi iFdiii. Combination 383 witli the initial failure of the dam, as the corn stalks were Hatteiied in the field be- low over a much wider area than the flow covered at the time of our arrival. Prob- ably any fish that were in the pond area immediately adjacent to the break at the time it occurred were washed across the \\ith the outflowing water into the net. Bluegills seemed less able to adjust them- selves to the situation and most of them were stranded in the pond basin. When we attempted to collect the stranded fish by walking into the pond basin, fig. 4, each of us in turn became Fig. 3.—Lower end of Fork Lake, showing the dam. Kipi apping uf broken contrete was held by fencing of wire mesh. This riprapping prevented washing from wave action but was attractive to burrowing muskrats. cornfield and into the stream that winds through the \alley below. AVater con- tinued to How out of the basin until 2:20 P.M. At the time of our arri\al, we staked a short piece of 1 -inch-mesh netting across the break in the dam to trap the larger (ish still within the pond basin. Fifty-four bass and 10 large bluegills were taken in the net set across the break. All these fish made an acti\e attempt to leave the lake with the water. Most of the bass held back until only about a foot of water re- mained in the basin ; then many swam mired in hip-deep silt a few feet from the former shore line. Because of the impossi- bility of collecting the hsh, we made a careful count b\- walking around the shore line and listing the bass and bluegills as belonging to the various broods on the basis of size. As the pond was long and relati\eh- narrow, we found it possible to walk along each side and count fish h ing stranded in the strip of bottom from the shore line to the trickle of water running through the center of the basin. Probably some fish had become buried in the soft mud, and so escaped being counted. 384 Illinois Natural History Survicy Bulletin I„l. 24. Art. 3 The band of vegetation completely en- circling the shove line and filling; the upper one-fourth of the pond (see map, fig. 5) flattened down over the mud when the water flowed out of the lake and ma\' have As the last of the water was running out of the lake, several small niuskrats left their burrows and ran around in the mud of the pond basin as if lost. Si.x adult bull- frogs crawled around in the mud. IVIan>- Fig. 4.—The basin of Fork Lake at the time of the dam failure, July 8, 1942. The break in the dam occurred near the center, supposedly as a result of a heavy rain and muskrat tunneling above the normal water line. The bottom of the white exposed portion of the gage board (upper left) marks the former water level. hidden a few additional large fish. This vegetation was covered by hundreds of small bass and bluegills that had wriggled to the surface of the vegetation mat with their dying exertions. We estimated that there were at least 10,000 bluegills be- tween one-half and 2 inches long, and not less than 5,000 bass between 1 and 4 inches. After the count was completed, the bass and bluegills that were captured in the seine were taken to the laboratory and processed. Table 6 gives an estimate of the total number and weight of fish observed. The weight amounted to ap- proxiinately 260 pounds. Man\' bullfrog tadpoles, Rana cates- beiana Shaw, were also in the vegetation. large clams, Anodontia grandis Say, were exposed and died after moving a few feet. Experience in draining ponds for mak- ing fish censuses indicates that, when a body of water is rapidly drained, bass swim against the current and usually most of them remain in the pond until the water has been lowered to a certain level, when they appear to reverse their behavior and attempt to move through the outlet. Blue- gills are weaker swimmers and conse- quently many are washed out with the water before the bass begin to appear in numbers at the outlet. The break in the dam at Fork Lake was V-shaped. The initial flow of water was large and probably carried away many D, 1948 Bennett: B.ass-Bluegill Combination 385 Fifi. 5.—Outline map of Fork Lake, showing encroachment of the pondweed, Potamogeton joiiosus Rafinesque. on the pond shallows. In 1939, plants of this species occupied a narrow band around the shore of the upper one-third of the pond ; in 1940, it spread into water 3 to 4 feet in depth and encircled the pond shore line; in 1941 and 1942, it filled the shallows up to 5 or 6 feet in depth, leaving only a little more than one-half of the pond area in open water. Open water was alwa\'S present in the upper forks of the pond because overhanging trees com- pletely shaded these pockets. more bluegills than bass. What the total weight of the population was can be onl\ conjectured. The deep, open water in the region of the dam was of a type attractive to bluegills beyond the size range of bass food. Certainly the numbers of Hsh actu- ally observed indicate that the pond con- tained a large fish population in spite of 3l/i years of heavy cropping with nets, and the many small fish inhabiting the vegeta- tion indicate a very successful spawn in 1942. Pond Habitat No attempt was made to stud\ the plankton or the invertebrate or \ertebrate fauna of Fork Lake other than the fish, except when an abnonnal abundance of some species could not escape observation or when plants or animals other than fish appeared in the nets. The most obvious change in the pond habitat during the 4! j years it was under observation resulted from the spread of a Table 6.—Fish observed at Fork Lake at the time of the dam failure of July 8, 1942. .?86 Illinois Natural History Survi-v Hulli;tin lol. 24. Art. 3 species of rooted aquatic vegetation, namcl>' the fine-leaved pondweed, Potmnocjetoit foliosus Rafinesque. Previous to the com- plete removal of fish in 1938, the pond contained no aquatic vegetation because the activities of bottom-rooting fish kept the water very turbid. The replacement of these fish by largemouth bass and blue- gills allowed the silt to settle, and after June of 1938 the pond was turbid only following heavy rains. The source of silt was a clay fill for a road about 150 yards above the pond. Most of the rest of the drainage basin was in timber and grass. The reduced turbidity of the pond was not immediately followed by a growth of aquatic plants; a few bunches of fine- leaved pondweed appeared in the late sum- mer of 1938. By midsummer of 1939 a dense, narrow band of Potainogeton foliosus was grow- ing in water from the shore out to a depth of 2 feet along the shore line of the upper one-third of the pond. In 1940, tjie fine- leaved plant spread into deeper water at the upper end of the pond and extended its distribution to form a band along the entire shore line of the pond—the edge of this band farthest from shore extended into water between 3 and 4 feet in depth. A vegetation map of the pond was made at the period of maximum plant growth in early August, fig. 5. When the open water of the pond was measured with a planim- eter on this vegetation map, it was found to be slightly more than 0.95 acre, as com- pared with an open water area of 1.25 acres in 1939. The plants in this band of vegetation extended to the surface and were so thick that it was difficult to row a boat through them. The following year (1941) the pota- mogeton extended its growth into still deeper water, fig. 5. The maximum depth at which the plants reached the surface was 6 feet, and most of the shore band of plants extended to depths a little beyond 5 feet. A measurement of the open water as outlined on a vegetation map made July 28 gave an area of 0.64 acre. By the time the 1941 vegetation map was made, a circular area of approximately 10 feet in diameter, at a point near shore immediately between the forks of the upper end of the lake, had become completely denuded of plants. As a heavy growth of plants had been present in this location earlier in the summer, the barren area was recorded. The water in the barren area was clear and it was possible to see bottom there. The water depth of this area ranged from 1 foot near shore to 3 feet at the outer edge of the opening. The pond was not visited again until August 19, 23 days after the vegetation was mapped. On this date the Potainoge- ton foliosus was completely gone, except for a few scattered stalks at the water's edge, and the water was turbid with a "bloom" of algae, Aphanizometion flos- aquae (Linnaeus), that obscured a Secchi disk lowered to 1.2 feet. Whether the small open area observed the latter part of July represented the beginning of a prog- ressive die-off of the potamogeton is not known. A few small bunches of Pota- mogeton nodosus Poiret growing in shal- low water along the south shore seemed to be unchanged. This phenomenon of a sud- den die-oft of P. foliosus in midsummer had been observed in one other experi- mental pond. In both instances, however, it was a "before and after" observation ; neither pond was observed during the prog- ress of the die-off. The cause of this sudden die-off is un- known and it is believed to be of uncom- mon occurrence. Usually, dense mats of Potamogeton foliosus remain until late September, when they gradually break loose from their attachments and float free for some time before finally disintegrating. The phytoplankton "bloom" that followed the disappearance of the potamogeton is believed to have resulted from the release of plant nutrients into the pond water. In 1942, in spite of a late, wet season, the fine-leaved pondweed again appeared in abundance and by the latter part of June had grown to fill the same parts of the pond that had been filled the preceding year. It is probable that the application of the rotenone in June of 1938 killed not only the fishes but also the Entomostraca and, where excessive concentrations occurred, a part of the insect lar\ae of the littoral zone and the benthos (Smith 1940; Brown & Ball 1943). The rapid growth of bass fry and bluegills during the summer indi- cates that replenishment of small aquatic invertebrates must have taken place within a very short time. The result of a sudden catastrophe, such as rotenone treatment of a pond, is often a simplification of the in- D,, 1^48 Bennett: Bass-Bluegii.i, Combination 387 vertebrate fauna, some species disappearing entirel) and other ft)rms disappearing tem- porarily but reappearing later in eruptive numbers. This reaction is assumed to ha\e taken place in 1Q38 in Fork Lake, particu- larly among the smaller invertebrates. Fork Lake contained a number of adult bullfrogs, Raiui cdtcsbeuina. and their tad- poles were extremely numerous throughout the summer of I'^iS. The only other year in which tadpoles were present in large numbers was 1Q41, although a iew could be seen at almost any time among the aquatic vegetation. Cra\fish, Cdiiibanis ririlis Hagen and C. propinquus Girard, must have been fairly abundant throughout the period of cropping of the pond, as large adults were taken often in wing nets. Both snapping, Chflylira sfrpentiiui (Linnaeus), and painted turtles, Chrysi-iiiys pitta maryiiiaia (Agassiz), found a way into nets and fre- quently were drowned before the nets were raised. In some raises the poundage of turtles greath' exceeded that of fish. Snap- ping turtles removed from the pond in 1*530 were '^ weighing 28.7 pounds; in 1940, 11 weighing 23.2 pounds; in 1041, 5 weighing 14.5 pounds; and in 1942, 3 weighing 7.1 pounds. The painted turtles removed in 1930 were il weighing 20.3 pounds; in 1040. 8 weighing 7.6 pounds; in 1941, 6 weighing 6.8 pounds; and in 1942, 4 weighing 5.3 pounds. Turtles found dead in the nets were remoxed from the pond. Some of the live turtles caught in nets were reiiio\ ed for laboratory study or for table use. Others caught were re- leased. \'egetation vs. Fish Yield .At the same time that the annual pound- age of fish from Fork Lake went steadih downward from 1939 through 1941, in spite of an increased intensity of fishing with nets and an increase in man-hours of angling, the mat-forming aquatic vegeta- tion continued to spread. While other factors undoubtedly reduced the yield of Hsh from the pond, I believe that the in- creased abundance of plants was the most important factor. A comparison of the yields of fish with the areas of open water is shown in table 7. The yields of Hsh are, of course, not exacth proportional to the areas of open water in the pond. But even though net- ting and angling pressures were increased during tlie 3 years, and many other factors probably influenced the fish yield, a re- niarkably close parallel existed between the fish \ields and the open pond acreages, table 7. Swingle (1045) in\estigated a pond that became filled with a heavy growth of naiad, Najas yuadaliipensis ( Sprengel ) , during \ears when this plant was not shaded out by muddy water. He concluded that the rank plant growths did not reduce the production of fish but did materially reduce the hook-and-line yield. Although information on the total weight of the fish population of Fork Lake, before and after the plants became abun- dant, is not available, the evidence pre- sented in table 7 and the fact that the growth rate of bluegills became slower in spite of heavy cropping (see section follow- ing) suggest that the production of food available to fish in the pond was actually reduced by Potaiii(i//i'triii OAiasus. Growth Rates The growing season for Fork Lake fish (water temperature abo\e 55 degrees F. ) was determined to be about 6 months long in 1030 (Bennett. Thompson, & Parr Table 7.—Yield of lish, fishing effort, and approximate area of open water in Fork Lake, 1939, 1940, and 1941. Illinois Natural History Survey ]3ulletin lot. 24, Art. 3 1940), and probably the growing season length \aried from this time no more than a few weeks in other years. These relatively small differences in the growing seasons had less influence upon fish growth than such factors as available food and related competition for food. Frequent collections and examinations of fish from Fork Lake gave total length measurements that were useful in plotting lected and preserved between late June and mid-September, 1938. The 270 bluegills that were moved to Fork Lake from Homewood Lake between June 11 and 18 were very thin and were between 5 and 7 inches total length. On the basis of the first spring collection of the following year, these fish may be said to have grown rapidly in 1938; they aver- aged 7.6 inches in March, 1939. Many T T ears; and 16 of the 1941 brood, 10 inches and 0.79 pound at 1 year. The large size of the bass of the 1940 and 1941 broods was due largely to growth made during the plant die-off period of 1941. Four broods of bluegills were repre- sented in the \'^\2 collections: 7 bluegills of the 1938 brood averaged 8.0 inches at 4 years; 35 of the 1939 brood, about 6.6 inches at 3 years; 193 of the 1940 brood, about 5.5 inches at 2 years; and 18 of the 1941 brood, about 3.5 inches at 1 year. Bluegill growth was consistently poorer with each successive year, fig. 7, in spite of heavy cropping. This slow growth rate is believed to have resulted from a reduc- tion in the available food supply associated with the spread of the plant, Putaiiiot/etdii folinsiis. Condition and Growth Condition, or relative plumpness, of fish is a measurement of some value in pond management. A high average condition usually indicates an abundance of available iood in relation to the number of fish present, and a low average condition de- notes slow growth and undue food com- petition. The index figure indicating condition of bass and bluegills, as calculated by any of several recognized formulas, increases with an increase in the length of the fish even for fish apparently of the same relative plumpness. The form of both bass and bluegills changes somewhat throughout their length range. In this manuscript the Index of Condi- tion formula (Thompson & Bennett 1 939i ) has been used : , , ,,,,.. VV 10,000 Index of Condition = ^-7, • L''W represents weight to the nearest hun- dredth pound, and L represents total length to the nearest tenth inch. ^Vhen this formula is used on lengths and weights of bass within the length range of 5 to 15 inches, an Index of Condition figure of 3.5 to 4.5 denotes a fish in poor Hesh; 4.6 to 5.5, one of about average or normal plumpness; and 5.6 to 6.5, a very fat fish. In bluegills, the increase in index figure with increasing size is more pronounced than in bass, but in fish of 5 to 8 inches an Index of Condition figure of 7.0 or below denotes a fish in poor flesh; 7.1 to 8.0, one of normal or average plumpness ; and above 8.0, one of unusual plumpness. The condition curve for 1938 bass in 1939, plotted in fig. 8, shows that fish of this brood were within the range of nor- mal condition during April, May, and June, then dropped into the thin classifica- tion, and remained there until July, 1940. This period coincides with the period of very small length increase (March, 1939- June, 1940) cited above in the discussion of growth. After the bass and bluegills had spawned in 1940, bass of the 1938 brood began to grow, and their condition curve rose gradually to 5.0 and slightly above. These bass remained within the range of normal plumpness until the period of vegetation die-off and "bloom," when they became very fat and their condi- tion curve rose rapidly to 6.15. A drop to normal followed, but in 1942 the condi- tion curve of the 1938 brood again rose above 6.0. The spectacular rise in the con- dition curve of bass of the 1938 brood dur- 392 Illinois Natural History Survey Bulletin Fol. 24, Art. 3 Decemhrr, 19fS Bennett: B.\ss-Bi.uEr;Ti.i. Combination 393 high. The early spring collections of 1942 showed a marked drop in Index of Condi- tion of the 1941 brood, but improvement to 5.77 in early July. As bluegills showed no increase in length or improvement in condition coin- ciding with the 1941 plant die-off, condi- tion curves for bluegills are plotted to show comparable conditions of the several broods of bluegills at comparable ages, fig. 9. Throughout the period of study (except for the first few months when the fish were ver\ small) the axerage condi- tion for the 1938 brood bluegills was high- est in early summer, but the curve re- mained always above 8.0, indicating that the fish were unusuall\ plump. Other broods showed a rise in condition during the spring months to a point above 8.0, followed by a severe drop in or after July to a low in November. The drop was much more pronounced in the 1939 and 1940 broods than in that of 1938. A reg- ular cycle of condition, probably influenced by spawning, seems to be characteristic of this species. The cycle reaches a high in early summer and a low in late fall. The fluctuations in the bluegill cycle must be considered when judging the condition of bluegills from a selected body of water on the basis of a single collection. Scale Analysis Annual rings, or annuli, that appear on fish scales are used frequent!}' by aquatic technicians to determine age of fish. The validity of this practice has been tested for only a few species, although fisheries biolo- gists, including the author, have applied the practice of "scale reading" to many species of fresh-water fishes. This study of bass and bluegills in Fork Lake gave a good opportunity to determine the validity of the scale method of age determination in these species, particularl\ in the original bass stocked as fry in 1938 and in the 1938 brood bluegills. As the original bass fry did not become sexually mature in 1939, no new brood of bass ap- peared until June, 1940, so that through- out the period of study the original bass could be separated from the 1940 and 1941 broods on the basis of size. The 1938 brood bluegills grew rapidlv during their first 2 years ( 1938-1939)" in the pond; this rapid growth, and the fact that body length, weight increases, and scale growth were followed from month to month throughout the growing seasons of 1939, 1940, 1941, and a part of 1942, gave as- surance as to the correct identification of this brood in all collections. In a previous study of the scales of Fork Lake bass and bluegills collected in 1939 (Bennett, Thompson, & Parr 1940), the annuli were found to appear on the scales at about the time growth was resumed fol- lowing a period of dormancy (winter). The length of the period during which individual fish of a given brood were in the process of forming annuli seemed to depend upon the amount of food available for that species of fish within the brood size range at the beginning of the growing season. If acceptable food was abundant, the beginning of growth was controlled by the temperature of the warming water, and most of the fish began to grow at about the same time. The appearance of the annulus, being dependent upon the addi- tion of concentric ridges of new material (circuli) on the scale margins, was closely associated with a length increase of the fish body. If acceptable food was relatively scarce at a time when temperature condi- tions were favorable for rapid growth, the beginning of body growth within a brood of fish was delayed until the individual members were able to ingest a quantity of food in excess of bod\ maintenance require- ments, and annulus formation was pro- longed over a much greater period than when food was abundant. There is no evidence to indicate that the annuli on the scales of warm-water fishes are more than visible marks produced by alternate periods of scale growth and growth stoppage. While the cessation of body growth and scale growth of fishes in winter is the result of low water temperatures which reduce the rate of metabolism, conditions might occur during the growing season (summer) which would stop growth, and, if growth were resumed later, produce false annuli. False annuli, which appeared on the scales of both the 1938 brood bass (original stock) and 1938 brood bluegills in mid- simimer of 1939, were quite common (Bennett, Thompson, & Parr 1940). While these marks were indistinct on the scales of many fish, on others they were very definite and were indistinguishable 394 Illinois Natural History Survey Bulletin J'ol. 24, Art. 3 from true annuli formed earlier in the season. The identification of these clear marks as false annuli aroused questions as to the validity of the scale method of age determination. Therefore, it was impor- tant to determine the percentage of clear false annuli among the hroods of bass and bluegills in Fork Lake. In the 1938 brood bluegills taken in 1939, 10.0 per cent had clear false annuli, formed in midsummer, that were in no way distinguishable from a true annulus. The false mark was in each case laid down outside of the first annulus (in the second summer of life). In the collections of 1940, 11.3 per cent of 329 fish of this brood (1938) showed a clear false annulus in the same position as those found in 1939, although in fish collected after May, when the 1940 annu- lus had formed, the false annulus lay be- tween the first and second true annuli. In the collections of 1941, 5 of 35 fish (14.3 per cent) of the 1938 brood showed a clear false annulus in the same position. Thus, throughout the period of study, from 10.0 to 14.3 per cent of the 1938 brood blue- gills collected each year showed a clear false annulus that formed during the sum- mer of 1939 (outside the first true annu- lus). This false annulus was in no way distinguishable from the true annuli. The scales of the 1938 brood bluegills were carefully checked for the presence of clear false annuli, other than those formed during the summer of 1939. Only 12 were found on fish collected in 1940 and later. These 12, located outside the second true annulus, appeared on the scales during the summer of 1940. The 1939 and 1940 brood bluegills grew less rapidly than the 1938 brood and, while the exact identification of members of these broods is less certain than of mem- bers of the 1938 brood, monthly collections failed to show clear false annuli with any degree of frequency. Table 8 gives the percentages of clear false annuli on the scales of 1939 and 1940 brood bluegills. Clear false annuli were common on largemouth bass scales. About 6 per cent of the 1938 brood bass caught in 1939 showed a false annulus that might be con- fused with true annuli. In the 1940 col- lections of 1938 brood bass, 15.3 per cent of the fish showed a false annulus outside the first true annulus (false annuli formed in the summer of 1939). In the 1941 and 1942 collections of the same brood bass, false annuli were found on the scales of 6.9 per cent and 4.5 per cent, respec- tively. No distinct false annuli were found in the 1938 brood bass other than those that were formed during the summer of 1939, located between the first and second true annuli. No clear false annuli were present on the scales of either the 1940 or 1941 brood bass. Other abnormalities associated with the scales of 1938 brood bass were found in this scale study. Some 1938 brood bass did not grow at all in 1939 and on the scales of these fish (three in number) the 1940 annulus replaced the annulus that should have appeared in the summer of 1939. Thus, these fish showed one less annulus than should have been present. In nine other bass of the 1938 brood, the increase in length was very small in i 1939. This small growth was reflected on i the scales in an addition of only three or four circuli outside the 1939 annulus and these in only the anterior field of the scale. When the 1940 annulus formed, the inter- space between it and the 1939 annulus consisted of the few circuli in the anterior field, and the two rings coincided in the lateral and posterior fields of the scale. Thus, a confusing partly double ring was formed, which probably would ordinarily be interpreted by a scale reader as a single Table 8.—Percentage of clear false annuli observed on the scales of 1939 and 1940 brood bluegills in 1940, 1941, and 1942. Di itmhi 194S Bi-XNun : 1? \ss-Hi.Li;(.ii.i. Combination 395 annulus. In the scales of two other Hsh of this brood, tlie 1939 and 1940 annuli were entirely distinct, but so close together as to throw suspicion on their validity. None of these abnormalities was accom- panied by serious scale erosion and none was found on the scales of 1940 and 1941 brood bass. Most of the annulus abnormalities given above—false annuli, skipped annuli. oxer- lapping annuli. and close spacing of annuli —were associated with the 1938 broods of bass and bluegills, and most of the abnor- malities were laid down on the scales dur- ing the growing season of 1939. During this period (1939), the 1938 brood bluegills were growing rapidly and the 1938 brood bass very slowl\- (see pre- ceding section on growth). Later broods of bluegills were subjected to more com- petition for food (growth was less rapid) and later broods of bass were subjected to less food competition. Therefore. I am inclined to follow the theory that scale abnormalities as related to annulus forma- tion are more common in fish growing at an abnormalK' rapid rate, or at an ab- normally slow rate, than in those subjected to moderate food competition, resulting in "average" growth. Some information on the way false an- nuli may arise on the .scales of "wild" fish was gained through experiments in feed- ing fishes confined in aquariums. Bruno von Limbach (unpublished experiments at Urbana) attempted to feed individual bluegills (one fish to an aquarium) on a heavy diet of earthworms (8 to 10 per cent of body weight per day). These fish fed well and gained rapidly for a few weeks. Then for no apparent reason they went "off feed" and refused to eat their quota of worms. In some individuals, this condition persisted for several months, dur- ing which they continued to eat onl\ enough to maintain their body weight. In others, heavy feeding was resumed after a "rest" of 1 or several weeks. In the latter fish, the periods of self-imposed starvation, followed by a resumption of feeding, produced false annuli on the scales. It cannot, of course, be pro\ed that "wild" fish go "off feed" when food is abundant, but the possibility is worth con- sidering. In other experiments in which bluegills were forced to alternate between periods of feeding and starvation, clear false an- nuli were produced on the scales if the starvation period between two 4-week periods of feeding was 3 or more weeks in length. Lesser periods of starvation pro- duced inconspicuous false rings. It is con- ceivable that, under conditions of crowd- ing in natural or artificial waters, indi- vidual fish might go practically without food for as long as several weeks, later to be supplied with comparati\el\ large quantities of food from a hatch of aquatic insects or a spawn of young fish. Skipped annuli, parth' double annuli, and close spacing of annuli are easily ex- plained as resulting from degrees of starva- tion extending throughout an entire grow- ing season. Spawning and ^ oung Fish No nests of largemouth bass were ob- served in Fork Lake during the period covered by this report. At the time of the dam failure, after all the water had drained from the pond basin, several craters were noted that may have been made by nesting bass in the spring of 1942. The nests of bluegills could be observed at almost any time during any summer and were most numerous in shallow water along the north shore of the pond near the spillway and on a submerged dome of earth at the east end between the forks. Nest-guarding males were nearly always present in these areas, where the nests of 5 to 10 inches in diameter were only a few inches apait, but more males were counted at the onset of the spawning season in May or early June than at any other period. Nesting males were least numerous on the spawning grounds from mid-June until the latter part of July. Later, the number of nest-guarding males increased. In 1939, young bluegills were scarce, although bluegill nests were in use throughout the summer. The 1938 brood bass (stocked as fr> ) were then 6 to 10 inches long and on a number of occasions were observed to enter bluegill nests and feed upon bluegill fry in the yolk sac stage. No voung bass were observed in 1939. On IVL-iy 27, 1940, 15 bluegill nests were counted in the spillwa\ spawning grounds, fig. 10, and several thousand fry were schooling near these nests. The bass 396 Illinois Natural History Survey Bulletin fol. 24. Art. 3 fry stocked in 1938 became sexually mature in 1940 and, on June 3 of that year, 11 scliools of fry were counted. Throughout the summer both young bluegills and young 20, five large schools of bass fry were counted, each containing several thousand fish. On July 19, schools of small Hsh — both bass and bluegills—were everywhere Fig. 10.—A group of bluegill nests near the spillway of Fork Lake. These nests were exposed by low water levels during August, 1940. bass could be seen amidst the submerged vegetation. The 1940 broods of bass and bluegills made poor growth in their first year, and many small bluegills were in evidence in the early spring of 1941. When nets were lifted on March 22, 1941, "showers" of small bluegills "rained" through the meshes and fell back into the pond. On April 15, 76 small bluegills were trapped, as a net was lifted, by a mat of Spirogyra. These fish averaged 1.48 inches long; the largest were 2.0 inches. Nineteen bluegill nests containing eggs were seen on May 6, 1941, and, on May along the edge of the pond in the fine- leaved potamogeton. When nests were raised on September 24, small bluegills and bass dropped through the meshes; most of the bass were less than 3 inches long and the bluegills less than 2 inches. Their numbers seemed scarceh' less than in July, although all rooted submerged vegetation had disappeared in early August. In 1942, the spawning season was un- usually late, due to a cold, wet spring, and neither young bass nor young bluegills were seen as late as May 26. The first schools of bass were observed on June 2y Dictmber, I'US Bexxett: Bass-Bluhcill Co.mbixatiox 397 on this date 17 liluegill nests were found to contain eggs, but no bluegill try were then in evidence. Throughout June the \oung of both species were again very numerous. Each year of the Fork Lake study, with the exception of \'~>i'S and 1939 when no mature bass were present, large numbers of young bass and bluegills were to be found in the pond. A minnow seine haul along the shore any time after June would have taken large numbers of both species. Swingle (1945) recommended minnow seining in ponds as a method of testing the "balance" of a bass-bluegill population. The presence of the young of both species in such seine hauls was said to indicate that the tish population of the pond was in "balance" and the pond should produce good fishing. While the adult bass and bluegills in Fork Lake produced a success- ful spawn each year (1940-1942), this spawn production bore little relationship to the annual fish \ield, average sizes of adult fish in the pond, or the catch of fish per man-hour. It indicated, however, that the adult population of fish was not crowded to the e.xtent that reproduction in either species was curtailed. Sexual Cycle In the process of determining by dissec- tion the sex of largemouth bass and blue- gills from Fork Lake, it was noted that changes in the appearance of the ovaries and testes occurred that could be readily identified. In immature fish of both spe- cies, the sex was easily told, although the sex organs were often very small. In these lish, the ovaries were spindle shaped and granular, while the testes were almost threadlike. In late fall and early spring the gonads of the larger adult fish re- sembled those of the smaller immature fish, except that they were larger. As the spring advanced, the gonads of mature fish began to swell and change in shape and color, until in May these organs reached a maximum size, and soon after the reproductive products were ready for deposition. At this stage the ripe males would give off milt when gently pressed in the lower abdomen and females would pass eggs that were translucent, yellow, and sticky. After the spawning period, the inaries and testes appeared smaller and liabby, and pink or red with blood. In the fall the gonads again would assume an immature appearance. In order to identify these changes with time in both species, a brief description of tlif stages was formulated. The classifica- tion given below was published in a previ- ous Fork Lake report ( Bennett; Thomp- son. c small oocytes, indicating that they were sexually imma- ture. The time schedule of the sexual c^cle of bluegills in 1940 was essentially the same as in 1939. Gonads of about 90 per cent of the bluegills examined were "poorly developed" in the period March 14-25; some were classified thus as late as August 19-22. "Enlarged" gonads were collected from April 12 to July 23; the higher percentages were in April, May, and June. Gonads in "spawning condi- tion" first appeared May 20 and were pres- ent in the August 19-22 ccllections but not later. "Partly spent" testes first ap- peared June 13-19, and "partly spent" ovaries, July 17-23; no partly spent ovaries or testes appeared after September. Four per cent of the testes were "com- pletely spent" June 13-19; only 58 per cent of the testes and 80 per cent of the ovaries were completely spent September 17-24. Gonads of all bluegills collected during the latter part of October and early November were completely spent and were in advanced reorganization stages (spermatogonia and small oocytes). Each year during the reorganization period, which followed after the completion of spawning, the spermatogonial cells and oocytes that became the spermatozoa and eggs for the next season were differenti- ated. The timing of the sexual c\cle in bass was a little in advance of that of bluegills. In the March 14-20 collecting period, 100 per cent of the gonads of 2-year-old bass were in the "enlarged" stage. In the May 6-26 collections, 27 per cent of the testes were in "spawning condition," and 73 per cent were "partly spent," while 67 per cent of the ovaries were "enlarged," 25 per cent in "spaviMiing condition," and 8 per cent "partly spent." No sexually mature ba^s were taken in June or July, but those taken in August or later contained reorganized gonads. These studies indicate an intermittent spawning season for bluegills, in 1940 beginning the latter part of May and last- ing through September. The bass spawn- ing season was confined to May, beginning probably a little earlier than that of the bluegills. Field observers often recorded schools of young bass in Fork Lake b\' the time the bluegills were guarding eggs. In both species the males appeared to come into spawning condition a little before the females. The histological study of James (1946) offers conclusive evidence that in Illinois bluegills mature at 1 year and bass at 2 years ; some of the larger bass males pro- duced a few sperms as yearlings. In the South, according to Swingle & Smith (1943), bluegills may reproduce at 4 months under unusually favorable condi- tions, but normally not until 1 year of age ; bass commonly reproduce at 10 to 12 months unless stunted. These differences in the ages at which the two species reach sexual maturity in the North and in the South must be considered in stocking new ponds with bass and bluegill fry. Foods Stomachs for food anal\sis were col- lected from Fork Lake bass and bluegills during 1939, 1940, and 1941. In 1941, the collections of bluegill stomachs were discontinued after May, but those of bass were continued throughout that season. The numbers -of fish stomachs containing food taken through the months of collect- ing during these vears were as follows: bluegills, 671 in 1939, 504 in 1940, 108 in March, April, and Mav, 1941 ; bass, 299 in 1939, 175 in 1940, 121 in 1941. As the stomachs were removed from the fish, each stomach was given an accession number and preserved in alcohol. Cor- responding accession numbers on the scale envelopes made it possible to relate the stomachs to fish of known ages, lengths, and weights. Methods of determining December, 194S Bennett: Hass-Bi.ui-gii.l Combination 399 volume of stomach contents were those described in a pre\ ious Fork Lake report (Bennett. Thompson, 5c Parr 1940). Althouj;h. in the years of this study the most pronounced change in the pond en- vironment was caused by the gradual spread of Potainogeton foliosus, other less obvious changes may have affected the relative abundance of fish foods during the period of this stud\'. No quantitative samples of aquatic in\ertebrates were made, and therefore changes in abundance of these animals were unrecorded except as indicated by field observations and stomach analyses. Fluctuations in the numerical abundance in individual broods of bass and bluegills in Fork Lake probably affected the degree of food competition at various times during the 3' J >ears of study. The degree of competition for a "staple" food often de- termines whether an individual fish is able to select this in preference to some less satisfactory substitute. Two species of hsh that compete but little for food in a favor- able aquatic environment may, under crowded conditions, be forced to change their normal feeding habits and become highl\' competiti\e. In l'^,?S, after the pond was restocked with adult bluegills and bass fry, there was apparently little competition for food, as indicated by the growth of the stocked fish and the 1^38 spawn of the bluegills. It is likely that in March, April, and May of 1939 some of the smaller bluegills !! the 193S brood were still available for l>a^s food, although bass stomachs collected ill March did not contain any fish (Ben- nett, Thompson, iS; Parr 1940). Small Muegills were observed in the fall of 1938. Also, the increase from ALirch to April, r'39, in the average lengths of 1938 brood Muegills collected, fig. 7, hardly can be ' \plained as growth, because of low water temperatures at that period; rather this Itngtli increase in the ccjllections suggests elimination of the smaller indi\iduals ihe brood. In .March, bass stomach con- 'cnts consisted of 80 per cent water boat- men and back swimmers, 4 per cent aquatic beetle larvae, 15 per cent terres- trial insects, and 1 per cent insect frag- ments. In April, fish and crayfish made up a total of 37 per cent of the diet, Ento- inostraca—largelv Dtipliiiiii—25 per cent, and water boatmen, back swimmers, aqua- tic beetle larvae, and insect fragments the remaining 38 per cent. In Ma\, fish and cra\/ish constituted 18.2 per cent of the diet. Other items, in addition to a trace of snails and coarse aquatic plants, con- sisted of insects, of which more than 25 per cent were adults—mostly dragonfiies, damselflies, and midges. The extensive use of insects during this period indicates a scarcitv of fish and crayfish of sizes that could be handled by 6-inch bass. During the spring period of 1939, the bluegills of the original stock and the 1938 brood fed heavily on Cladocera, midge larvae, and snails ; other aquatic insects were somewhat less important in the diet. Competition between bass and bluegills for any single group of insects was not ob\ious, but bass made wide use of the varieties of insects available. The nest robbing activities of bass on the bluegill spawning ground, mentioned previoush', indicate the extent of the shortage of bass foods. In the collections of July, 1939, very small bluegills made up 51.4 per cent of the bass diet, but the figure dropped to less than 30 per cent in August and Sep- tember and less than 10 per cent in Octo- lier. Both bass and bluegills depended upon aquatic insects (mostly larval) dur- ing the latter part of the 1939 collecting season—feeding most heavily on Diptera larvae and water boatmen. In 1939 the competition for insects be- t\veen these two species of fish had no appreciable effect upon the growth rate of either the original or 1938 brood bluegills, fig. 7. Cladocera and midge larvae were more or less staple foods for the small- mouthed bluegills, but apparently Cladoc- era and aquatic insects were not conducive to rapid growth in bass. In 1940. when aquatic vegetation be- gan to appear in Fork Lake in abundance, stomach collections taken from March through June were largely from bass and bluegills of the 1938 broods, table 9. In .March and April, both species fed largelv' on midge larvae, and tiie quantity of fish and crayfish in bass stomachs was insignifi- cant. In May only one bass stomach was obtained and it contained 95 per cent crayfish. Contents of the bluegill stomachs collected in May consisted of Diiphii'ui (50 per cent of the total), insects, and a few miscellaneous items; Diptera larvae consti- tuted nearh' 21 per cent of the total. 400 Illinois Natural History Survey Bulletin Fol. 24. Art. o ibcr, I'US Bennett: Bass-Bt.uecii.i. Combination 401 402 Illinois Natural History Survey Bulletin J-ol. 24. Art. December, 19-fS Ben-nett: Bass-Hi.i f.cii.i. CoMniXATiox 403 The large spawn of both bass and blue- J gills in late May and early June, 1940, ' probabl) improved the food situation for : bass, although the hea\-.\- stand of aquatic • vegetation offered excellent protection for young fish. The growth curve for IQJS brood bass indicates that most of the length increase occurred after June, rig. 6. In contrast to the spring diet of bass, which was largel>- insects, the late summer and fall diet consisted of more than 70 per cent fish, many of which were young bass. Other items of importance were crayfish, t snails, and miscellaneous aquatic insect larv.ae. Diptera larvae were commonly found in the stomachs of bluegills of all sizes throughout 1Q40, and entoniostra- cans were important from March through May and in October and November. Ex- cept that smaller numbers of Cnrixa and Xotonecta were taken in 1940 than in 1939, bluegili foods were essentiallv the same in the 2 years. The 1939 brood'blue- gills grew less rapidly in 1940 than the 1938 brood when of comparable age (dur- ing 1939), f^g. 7. At the beginning of 1941, Vo\\ Lake contained two broods of bass (1938 and 1940) and three broods of bluegills (1938 1939, and 1940), as well as the few re- maining original adult bluegills. No bass stomachs were taken in March, but, in April, 9 of the total of 15 original stock bass collected (1938 brood) contained fish and 7 contained crayfish, table 10. These two items together made up 54 per cent of the weight of all food taken. Other im- portant items were Diptera larvae, dragon- fly nymphs, and miscellaneous insect frag- ments. As indicated previously, growth of the broods of bass was slow until the plant die-oi^' in August. In August and September, after the rooted plants dis- appeared, stomach contents of bass of all broods (1938, 1940, and 1941) showed a smaller variety of foods and a higher percentage of fish than in any other 2- month period, and growth was very rapid. It is of some interest to note here that in August and September a few of the small 1941 brood bass ate small leopard frogs ind their metamorphosing tadpoles. Al- though bullfrog tadpoles were always lumerous in Fork Lake, they were found rarely in bass stomachs. In 1941, as in previous years, bluegills ted largel\- on Diptera larvae, damsel fly n\niphs, dragonfly nymphs, I),i/>liiiiri, and snails. 1 able 1 1 represents a sununary of food Items for all years shown as percentages of the total weights of all foods taken. Any figure of less than 4 per cent has been onu'tted. This table also lists the average lengths of the broods of bass and bluegills collected at the beginning and end of each collecting season, so that a comparison may be made between length increment and kind of food consumed. Although Diptera larvae appeared to be very important in the diet of bluegills, as did Entomostraca and snails, no dift'er- ence in growth of 1938 and 1939 brood bluegills can be shown to be associated with certain foods. The importance of specific t.\ pes of food in the diet of bass is obvious. In the 1938 brood bass, the onl.\ brood of this species present throughout all years of this study, there seemed to be a direct relationship between rate of growth and the percentage b\- weight of fish and crayfish in the diet. In 1939, collections of bass indicated that this brood increased 1 inch in length and made a relatively small weight increase; the percentages of fish and cra\fish in the diet were 13.8 and 9.2, respectively, table 11. In 1940, the average length of 1938 brood bass in the collections increased from 7 to 10 inches and the weight more than doubled; the percentages of fish and cray- fish in the diet were 21.2 and 13.0, re- spectively. In 1941, in spite of the fact that bass in the collections averaged 10 inches and about 0.5 pound in weight at the beginning of the collecting season, they mcreased an average of 3 inches in length and nearly tripled their weight in that \ ear ; the percentages of fish and cra> fish in their diet were 34.5 and 34.4, re.spectively. In both the 1940 and the 1941 broods of bass, growth was rapid on a diet of fish, small frogs, and tadpoles. The more common forms of Diptera lar\ae found in bass and bluegili stomachs were the Chironomidae (midges) — Chi- rniKiiiius and Tunypus: Chaoboridae — Chaoharus ; Ceratopogonidae (biting midges) — Pcdfioiiiyici, Bezzia. and I'riibez- zifi; Simuliidae (black flies) — Simtilium; Stratiomyidae (soldier flies) — Stratininxs, OdDiilrjmyia: and Culicidae (mosquitoes)—Cuh'x. No attempt was made to identify families or genera of Mayfl\ n\niphs. The 404 Illinois Natural History Survey Buli.etij Vol. 24. Art. 3 1— « £ o o er, W4H Bennett: Bass-Bluegill Combination 407 houses of caddis worms were largely those of Oecetis inconspicua (^Valker), Mol- aniia. Oxyelliira, and Orthntrichia. The damsel flies were largely of the genus Eiial- la