Bulletin BULLETIN ] of the ILLINOIS NATURAL HISTORY SURVEY HARLOW B. MILLS, Chief i Efficiency and Selectivity of Commercial Fishing Devices Used on the Mississippi River WILLIAM C. STARRETT PAUL G. BARNICKOL Printed by Authority of the STATE OF ILLINOIS WILLIAM G. STRATTON, Governor DEPARTMENT OF REGISTRATION AND EDUCATION VERA M. SINKS, Director . 'nfimm: STATE OF ILLINOIS William G. Stratton, Governor DEPARTMENT OF REGISTRATION AND EDUCATION Vera M. Binks, Director NATURAL HISTORY SURVEY DIVISION Harlow B. Mills, Chief Volume 26 BULLETIN Article 4 Efficiency and Selectivity of Commercial Fishing Devices Used on the Mississippi River WILLIAM C. STARRETT PAUL G. BARNICKOL Printed by Authority of the State of Illinois URBANA, ILLINOIS July 1955 STATE OF ILLINOIS William G. Stratton, Governor DEPARTMENT OF REGISTRATION AND EDUCATION Vera M. Binks, Director BOARD OF NATURAL RESOURCES AND CONSERVATION Vera M. Binks, Chairman A. E. Emerson. Ph.D., Biology Walter H. Newhouse, Ph.D., Geology L. H. Tiffany, Ph.D., Forestry Roger Adams, Ph.D., D.Sc, Chemistry Rouert H. Anderson, B.S.C.E., Engineering Lloyd Morey, B.A., B.Mus., C.P.A., LL.D., D.Sc, President of the University of Illinois Delyte W. Morris, Ph.D., President of Southern Illinois University NATURAL HISTORY SURVEY DIVISION Urbana, Illinois SCIENTIFIC AND TECHNICAL STAFF HARLOW B. MILLS, Ph.D., Chief Bessie B. East, M.S., Assistant to the Chief Section of Economic Entomolofey George C. Decker, Ph.D., Entomologist and Head J. H. Bigger, M.S., Entomologist L. L. English, Ph.D., Entomologist S. C. Chandler, B.S., Associate Entomologist Willis N. Bruce, Ph.D., Associate Entomologist Norman C. Gannon, Ph.D., Associate Ento- mologist John M. Wright, Ph.D., Associate Entomologist Paul Suranvi, Ph.D., Associate Entomologist W. H. Luckmann, M.S., Associate Entomologist Ronald H. Meyer, B.S., Assistant Entomologist John W. Matteson, B.A., Field Assistant Robert Snetsinger, M.S., Field Assistant Sue E. Watkins, Technical Assistant H. B. Petty, Ph.D., Extension Specialist in En- tomology* Stevenson Moore, III, Ph.D., Extension Specialist in Entomology* John Arthur Lowe, B.S., Research Assistant* Mohan Rao, M.S., Research Assistant* Clarence E. White, B.S., Research Assistant* Louise Zingrone, B.S., Research Assistant* Section of Faunlstlc Surveys and Insect Identification H. H. Ross, Ph.D., Systematic Entomologist and Head Milton W. Sanderson, Ph.D., Taxonomist Lewis J. Stannard, Jr., Ph.D., Associate Tax- onomist Philip W. Smith, Ph.D., Associate Taxonomist Leonora K. Gloyd, M.S., Assistant Taxonomist Thomas E. Moore, M.S., Technical Assistant Barbara Gutowsky, M.A., Technical Assistant Section of Aquatic Biology George W. Bennett, Ph.D., Aquatic Biologist and Head William C. Starrett, Ph.D., Aquatic Biologist R. W. Larimore, Ph.D., Associate Aquatic Bi- ologist Donald F. Hansen, Ph.D., Assistant Aquatic Biologist Robert D. Crompton, Field Assistant Leonard Durham, M.S., Research Assistant* WiLLiA.M F. Guilders, B.S., Technical Assistant* Section of Applied Botany and Plant Pathology J. Cedric Carter, Ph.D., Plant Pathologist and Head J. L. Forsberg, Ph.D., Plant Pathologist G. H. Boewe, M.S., Associate Botanist R. J. Campana, Ph.D., Assistant Plant Pathol- ogist I. R. Schneider, Ph.D., Assistant Plant Pa- thologist E. B. Himelick, M.S., Assistant Plant Pathologist Robert A. Evers, Ph.D., Assistant Botanist Rovenia F. Fitz-Gerald, B.A., Technical Assistant James D. Bilbruck, M.S., Research Assistant* Section of Game Research and Management T. G. Scott, Ph.D., Game Specialist and Head Ralph E. Yeatter, Ph.D., Game Specialist F. C. Bellrose, B.S., Game Specialist H. C. Hanson, M.S., Assistant Game Specialist J. S. Jordan, Ph.D., Assistant Game Technician Ross J. Miller, M.S., Field Ecologist Frances D. Robbins, B.A., Technical Assistant Virginia A. Whipple, Technical Assistant John M. Schilling, Field Assistant William B. Robertson, Jr., Ph.D., Research Assistant* James Opsahl, M.S., Field Assistant* Section of Publications and Public Relations James S. Avars, B.S., Technical Editor and Head Blanche P. Young, B.A., Assistant Technical Editor William E. Clark, Assistant Technical Photog- rapher Milan Dobrovic, B.A., Technical Assistant Technical Library Ruth R. Warrick, B.S., B.S.L.S., Technical Librarian Olga E. Griminger, B.A., Assistant Technical Librarian Consultants: Herpetology, Hobart M. Smith, Ph.D., Associate Professor of Zoology, University of Illinois; Parasitology, Norman D. Levine, Ph.D., Professor of Veterinary Parasitology and of Veterinary Re- search, University of Illinois. Employed on co-operative projects with one of several agencies: IlHnois Agricultural Extension Service, Illinois Department of Conservation, United States Army Surgeon General's Office, United States Depart- ment of Agriculture, United States Fish and Wildlife Service, United States Public Health Service, and others. This paper is a contributionfrom the Section of Aquatic Biology. (13S22—3.''>00—3-5.5) ^.tS'^SSSc;. CONTENTS Materials and Methods 325 Characteristics of Area 327 Commercial Fishing Activities 327 Terminology for Commercial Fishing Devices 329 Angling Devices 330 Jugs or Floats ( Single Hook ) 330 Trot Lines (Multiple Hooks) 330 Encompassment Devices 331 Entrapment Devices 333 Analyzing Entrapment Catch Data 337 Catches With Basket Traps 344 Catches With Wing Nets 345 Catches With Hoop Nets 351 Catches With Trap Nets 353 Entanglement Device 353 Catches With Set Trammel Nets 354 Experimental Sets at Grafton 359 Gilling Experiment With Set Trammel Net 361 Catches With Floated Trammel Net 361 Experimental Floats at Quincy 363 Discussion 364 Summary 364 Literature Cited 365 Cleaning fish in a commercial fish market at Dallas City, Illinois, on the Mississippi River. Efficiency and Selectivity of Commercial Fishing Devices Used on the Mississippi River* WILLIAM C. STARRETTt PAUL G. BARNICKOL+ IN 1944 and 1946, a fishery survey was made of the Mississippi River between Caruthersville, Missouri, and Du- buque, Iowa, under the auspices of the Technical Committee for Fisheries, a sub- group of the Upper Mississippi River Conservation Committee. Collections of fish were made with various types of fishing devices at 31 field stations. Based largely on these collections, a report rela- tive to the status of commercial and sport fishing in the Caruthersville-Dubuque section of the river was written and pub- lished (Barnickol & Starrett 1951). The present paper is a statistical analy- sis of the catch data relative to the effi- ciency and selectivity of the various com- mercial fishing devices used during the survey. It is believed that such an analysis could be of value to persons interested in the management of the river's commercial fishery. At the present time, law enforce- ment is the chief form of fish management practiced in the Caruthersville-Dubuque section of the river. The fishery survey was financed by the Illinois Department of Conservation, the Illinois Natural History Survey, the Iowa Conservation Commission, and the Mis- souri Conservation Commission. The Illi- nois Natural Histon,- Survey's laboratory boat, the Aiiax, was used as field head- quarters. The writers are indebted to Dr. George W. Bennett of the Illinois Natural His- tory Survey for suggestions and encour- * The investigation on \vhicli this paper is based was conducted under the auspices of the Technical Committee for Fisheries of the Upper Mississippi River Conservation Committee. t Aquatic Biologist, Illinois Natural History Survey. t Formerly Ichthyologist, Illinois Natural History Survey, novv Chief Biologist of the Fisheries Section of the Missouri Conservation Commission. agement in the preparation of this paper and to Dr. Robert Touchberry of the University of Illinois for technical advice in the statistical treatment of the data. The following persons, at the time asso- ciated with the Illinois Natural History Survey, aided in the collection of data: Mr. Daniel Averv, Mr. Leonard Dur- ham, Dr. B. Vincent Hall, Dr. Donald F. Hansen, Mr. Don W. Kelley, Mr. Jacob Lemm, Dr. Hurst H. Shoemaker, and Dr. David H. Thompson. Mr. James S. Avars of the Illinois Natural History Survey edited the manuscript. Many other persons contributed directly in mak- ing this investigation possible through their administrative assistance, notably Dr. Harlow B. Mills, present Chief, and the late Dr. Theodore H. Frison, for- mer Chief, of the Illinois Natural History Survey; Dr. G. B. Herndon of the Mis- souri Conservation Commission ; Mr. Sam A. Parr of the Illinois Department of Conservation ; and Mr. Everett B. Speaker and the late William E. Albert of the Iowa Conservation Commission. Photographs other than that for fig. 10 M'ere taken by Natural History Survey staff photographers William E. Clark and Charles L. Scott and by the senior author of this paper. Materials and Methods In 1944, fish collections were made at 19 field stations on the Mississippi River between Caruthersville, Missouri, and Warsaw, Illinois, and, in 1946, at 12 field stations on the river between Burlington, Iowa, and Dubuque, Iowa, fig. 1. Table 1 contains a list of the field stations, in- clusive dates of fishing operations at each [325] 326 li.i.iNois Natural History Survey Bulletin Vol. 26, Art. 4 station, pool numbers, and the distances of stations from Dubuque. I he sur\e> covered M)b miles of the Mississippi below the mouth of the Mis- souri River and 383 miles above this point. The lower section extended from the MISSISSIPPI RIVER FIELD STATIONS LEGEND • 1944 ILLINOIS - MISSOURI STATIONS O 1946 ILLINOIS- IOWA STATIONS • 1946-47 POISON CENSUS OF BACKWATErJ A LOCK AND OAM CARUTHERSVILLE STA.«S TENNESSEE Fi^' !•—The Mississippi River between Caruthersviile, Missouri, and Dubuque, Iowa; shown is the location of field sampling sta- tions used during the fisheries survey of 1944 and 1946. The river distance between Caruth- ersviile and Dubuque is 689 miles. mouth of the Missouri Ri\er to Caruth- ersviile and is referred to in this paper as the MR-C section. The upper section extended from Dubuque to the mouth of the Missouri River and is designated as the D-MR section. In both years of the survey, a key sta- tion was established for checking seasonal (iifterences in catches. In 1944, Grafton was the key station and, in 1946, Andalu- sia. Three sainpling periods were spent at each of these stations. In the fall of 1944, trammel net ex- periments were conducted at Quincy and at Grafton, Illinois. The data from these experiments are treated separately in this paper. When, in this paper, reference is made to mesh size of nets and seines, the size of the mesh in inches is indicated ; for ex- ample, a 1 -inch-mesh wing net refers to a wing net having mesh of 1-inch square measure. Square measure mesh size is used throughout this publication. The ap- proximate stretch measure of the webbing can be calculated by multiplying the square measure by two. The following fishing devices, some of which could not in 1944 or 1946 be used legally by commercial fishermen, were tested by the survey party: Trammel nets, length of each 80, 100, or 150 yards (mesh of inner net lYo, 1%, 2. or 3 inches), depth 5 or 6 feet Seines, length of each 100. 150, or 200 yards (mesh 1 inch), depth 10 feet Hoop nets (mesh 1 inch, hoop diam- eters 3V2 or 4 feet; mesh 21/4 inches, hoop diameters 4 or 41/2 feet ; mesh 3 inches, hoop diameters 4 or 41/2 feet) Wing nets with and without leads (mesh 1 inch, hoop diameters 2yo to 4^/2 feet ; mesh 1 V2 inches, hoop diameter 3 feet; mesh 2V^ inches, hoop diameters 31/^ and 4y2 feet) Trap nets (mesh lYi inches) Basket traps (opening IV2 inches) Trot lines The number of each of the various kinds of fish caught and the total lengths of the individuals and their weights, as well as the dates and hours of setting and raising gear, were recorded separately for each tj'pe of fishing device (except trot lines) and each mesh size of net (except trammel net) and seine used. The data July, 1955 Starrett & Barnickol: Commercial Fishing Devices 327 from trot line sets were too meager for valid analysis. The hoop diameters of wing nets and hoop nets were not con- sidered in this study, and catches made in these nets were not separated on the basis of hoop size. The common and scientific names of fish caught are listed in table 2. Characteristics of Area The Mississippi River in the MR-C section is more turbid and swifter than in the D-MR section. The MR-C sec- tion is free of locks and dams, whereas the D-MR section is canalized by a series of 14 locks and dams maintained for navi- gation. Detailed descriptions of the physical and fish-faunal characteristics of the Caruthersville-Dubuque section of the Mississippi River and the field stations established during the 2-year survey are contained in the earlier report prepared by Barnickol & Starrett (1951). In test-net collections, sport fishes were present in greater numbers in the D-MR section of the river than in the MR-C section. In the MR-C collections, small- mouth buffalos and black buffalos were more abundant than bigmouth buffalos. Bigmouth buffalos were more common in the D-MR collections than in the MR-C collections. Blue catfish were found up- stream as far as Lock and Dam No. 19 at Keokuk, Iowa. Fish of this species were taken in greater numbers in the Missis- sippi River below the mouth of the Mis- souri than above. Flathead catfish were common in both sections of the Missis- sippi. In 1944, more channel catfish were in test-net collections between Warsaw and the mouth of the Missouri River than in collections from the MR-C sec- tion. Commercial Fishing Activities At the time of the survey, commercial fishing was conducted along much of the Mississippi River between Caruthersville and Dubuque. According to Starrett & Parr (1951:6), in 1950 there were 248 residents of Illinois who worked either part or full time as professional commer- cial fishermen on the Mississippi. These Table 1.—Stations at which fish collections were made in 1944 and 1946 fisheries survey of the Mississippi River between Caruthers- ville, Missouri, and Dubuque, Iowa, inclusive dates for each collection, and location of each station. 328 Illinois Natural History Survey Bulletin Vol.26, Art. 4 Table 2.—Accepted common, scientific, and local names of fishes occurrin;* in Mississippi River test-net or other fisheries survey collections made between Caruthersville, Missouri, and Dubuque, Iowa, 1944 and 1946.* Accepted Common Name July, 1955 Starrett & Barnickol: Commercial Fishing Devices 329 Table 2.—Concluded Accepted Common Name 330 Illinois Natural History Survey Bulletin Vol.26, Art. 4 succeedinji sections. The terminology rec- ommended by the committee for commer- cial fishing devices is used in this paper. Angling Devices Angling refers to the capture of fish with either single or multiple hooks at- tached to a line. Jugs or Floats (Single Hook).— Some fishermen on the Mississippi River use jugs or floats with a hook and bait at- tached to each, fig. 2. The jugs or floats are floated downstream and are often very effective in taking large catfishes, blue and flathead. These fishing devices provide sport as well as meat, especialh' when a 25- or 30-pound catfish takes the bait. Trot Lines (xVIultiple Hooks).— Trot lines and throw lines are multiple- hook fishing devices common on the Mis- sissippi. Throw lines are favorites among sport fishermen, but they are seldom used by commercial fishermen. Trot lines are of some importance as commercial tackle, fig. 3. 'i'hey are fished mainly for cat- fishes in the Caruthersville-Dubuque sec- tion and to a lesser extent for carp, bull- heads, and freshwater drums. In 1947, 2.8 per cent of the reported Illinois commercial catch from the Missis- sippi was taken with trot lines, also known as set lines (Upper Mississippi River Conservation Committee 1948, third sec- tion :19). In this same year, 10.4 per cent of the reported Missouri commercial catch from the Mississippi was taken on trot lines (set lines). At the time of the survey, regulations relative to the number of hooks that a fisherman might use varied among the states. A trot line consists of a heavy cord to which are tied, at intervals of 2 to 3 feet, short drop lines, to each of which is tied a sing'e hook. The ends of the cord Fi^. 2.—Rigging up floats on the Mississippi River for blue catfish. Float fishing, or jug fishing, is more popular on the Mississippi below the mouth of the Missouri River than above. I July, 1955 Starrett & Barnickol: Commercial Fishing Devices 331 are secured to stumps or other supports. Each drop line may be a foot or more in length ; the length depends upon the depth at which fishing is to be done. Live min- nows or small fish (young carp are pre- ferred by many fishermen), cut fish of va- rious kinds, crayfish, spoiled clam meat, and grasshoppers are the common baits used on trot lines for catfish. Dough-balls or corn are the usual baits for carp. The length of trot lines and number of hooks used seem to vary with the accessi- bility of bait, number of men to run the lines, and fishing conditions. Evermann (1899:292) reported that he had learned of a trot line 12 miles long in Louisiana. In the Caruthersville-Dubuque section of the Mississippi, the lines seldom exceed 3.00Q yards. During the survey, only a few trot line sets were made. The search for live bait would have consumed more time than could be allotted to one phase of the test- fishing research program. An investiga- tion devoted solely to studying trot line catches in the Caruthersville-Dubuque area would be of value in managing the fishery. Enconipassment Devices Seines are the only type of encompass- ment tackle of importance to commercial fishermen in the Caruthersville-Dubuque section of the Mississippi River, fig. 4. Dip nets, another type, are used by hun- dreds of sport fishermen along this river, but are of little importance as commer- cial tackle. Most of the commercial seining in the Caruthersville-Dubuque section of the river is done with short seines. The seines commonly used there are 200 yards in length. In some parts of the river, longer seines are used. A short seine can be op- erated by two or three men, whereas a longer seine requires more men for effi- cient operation. The owner of a large seine should be in a position to handle large catches and sell them in distant mar- kets. The operator of a small seine usu- ally can sell his catch locally. Prior to making a seine haul, a com- mercial fisherman usually cruises near his selected "seine hauls" or water areas he knows that have bottoms clean enough for seining. If he believes fish are using the Fig. 3.—Trot lines used by commercial fishermen. On the Mississippi River, trot lines are used chiefly for catfishes. Usually the hooks and lines are secured in a homemade box as shown above. 332 Illinois Natural History Survey Bulletin Vol. 26, Art. 4 area, he makes preparations to lay out the seine. A seine haul of 200 or 300 yards can usually be completed within a couple of hours. Handling a big catch in a large seine may require several days. The survey crew made 25 seine hauls in the Mississippi River between Win- field, Missouri, and Cordova, Illinois. These hauls were made with 100-, 1 50-, and 200-yard seines of I -inch square mesh. Time and personnel limited the seining operations to one mesh size. A summary of the survey seine hauls is given in table 3. Commercial and preda- tory fishes represented 89.3 per cent of the total weight of all fishes taken with the seine. Sport fishes occurred regularly in the hauls; they represented 11.7 per cent of the total numbers and 33 per cent of the total weight of fish in the seine hauls. Crappies made up 82.1 per cent of the number of sport fishes. Only insignificant numbers of other sport species were taken in the seines. A seine is an efficient fishing device when operated by experienced commercial fishermen. Commercial seines are of larger mesh sizes than the seines employed in the survey. Fishermen generally try to use seines selectively by making hauls in waters they believe are being used by species they desire. The commercial hauls usually are made for carp, buffalofishes, freshwater drums, and paddlefish. How- ever, fishermen using commercial seines frequently catch only a few carp and buf- falofishes in a haul and many gizzard shads and gars. The seine is an important commercial fishing device on the Mississippi River Table 3.—Composition of catches made with 1-inch-mesh 100-yard, 150-yard, and 200-yard seines in fisheries survey of the Mississippi River between Winfield, Missouri, and Cordova, Illinois, 1944 and 1946. Kind of Fish Commercial Carp Bigmouth buffalo Smallmouth buffalo Black buffalo Paddlefish Channel catfish Blue catfish Flathead catfish Carpsuckers Suckers Freshwater drum Subtotal Sport White crappie Black crappie Bluegill Yellow walleye Sauger White bass Yellow bass Subtotal Predatory Shortnose gar Longnose gar Bowfin Subtotal Forage Gizzard shad Mooneye, goldeye Subtotal Total Number of Fish Taken 96 16 50 1 51 121 27 1 148 1 88 600 114 28 2 2 6 16 5 173 391 71 6 468 181 51 232 1,473 Per Cent OF Total Number Taken 0.1 3.4 8.2 1.8 0.1 10.0 0.1 6.0 40.7 7.8 1.9 0.1 0.1 0.4 1.1 0.3 11.7 26.6 4.8 0.4 31.8 12.3 3.5 15.8 100.0 Weight, Pounds 221.54 26.63 25.16 5.93 139.80 48.56 17.04. 0.19 96.05 1.68 29.78 612.36 24,96 5.16 0.24 1.99 5.16 5.02 0.83 43.36 416.06 134.73 17.93 568.72 85.06 12.58 97.64 1,322.08 Per Cent OF Total Weight 16.8 2.0 1.9 0.4 10.6 3.7 1.3 7.3 0.1 2.2 46.3 1.9 0.4 0.1 0.4 0.4 0.1 3.3 31.5 10.2 1.3 43.0 6.4 1.0 7.4 100.0 July, 1955 Starrett & Barxickoi. : Commercial Fishixg Devices 333 ^a Fig. 4.—Commercial fishermen shipping a seine after completing a haul. throughout the year. In the D-MR sec- tion, large hauls of carp and buffalofishes often are taken from under the ice during the winter months. The seine enables some fishermen in both the D-MR and the MR-C sections of the river to take paddlefish, carp, buffalofishes, and fresh- water drums during periods when few fish are being caught in other devices. Entrapment Devices Entrapment devices used in some or all parts of the Caruthersville-Dubuque area of the Mississippi River at the time of the sur\ey included basket traps, hoop nets, wing nets, and trap nets. Basket traps are constructed of wooden slats and have flexible, wooden throats, fig. 5. Fishermen usually bait these traps with old cheese scrapings and often in the spring they put a live gravid female cat- fish in each trap to serve as a decoy for unsuspecting males. Basket traps are set in current and are fished in water rang- ing from 4 to 15 feet in depth. A weight is secured to the tail line of each basket trap to anchor it in place. Some fisher- men fish two basket traps or a basket trap and a hoop net by joining the tail lines. The basket trap is an important fishing device for commercial fishermen of Illi- nois and Iowa who fish the Mississippi. The state of Missouri does not permit the use of this device. In 1950, Illinois com- mercial fishermen of the Mississippi River took 52.3 per cent of their reported cat- fish catch in basket traps (Starrett 5: Parr 1951:5). The local terminolog}- often applied to hoop nets, wing nets, and trap nets can be quite confusing. In one area, a hoop net, fig. 6. may be called a buffalo net, in an- other a fyke net, and in still another a fiddler net. Hoop nets are defined b^- the Upper ^Mississippi River Conservation Committee (1946:13) as a "group of de- vices" "constructed of vegetable fiber net- 334 Illinois Natural History Survky Bulletin- Vol.26, Art. 4 ting without fore ba\ . leads, or wings." ferenccs in mesh size and hoop diameter. The group "includes' hoop net, bait net, The fine-mesh nets with small hoops are buffalo net. fiddler net, and fyke net." used for catfish, and the nets with large The main differences in the local termi- hoops and large meshes are fished for buf- nology among these nets result from dif- falofishes and carp. Fig. 5.—A Mississippi River commercial fisherman removing a catfish from a basket trap. In the upper tub is cheese bait. Fig. 6.—A hoop net stretched to show the construction of the net. The hoop net does not have wings. July, 1955 Starrett & Barnickol: CoMMnRciAL Fishing Devices 335 Hoop nets are fished in the currents of rivers and in depths that cover the nets entirely. Often these nets are baited with cheese scrapings and other wastes pur- chased from cheese manufacturers, fig. 7. Frequently they are fished in 15 or more feet of water. Each of the.-e nets is an- chored by a wire cable or rope attached to a weight, a stake, or a basket trap. The fisherman setting the net takes a mental fix of his location by noting several land- marks. The following day he checks his bearings and locates the approximate place his net was set the previous day. He drags the bottom with a grappling hook until he strikes the net cable, fig. 8. He then re- Fig. 7.—A commercial fisherman baiting a hoop net with cheese. This fisherman demon- strates a common method of fishing for catfish on the Mississippi River. 336 Illinois Natural History Survey Bulletin Vol. 26, Art. 4 Fig 8—Searching the bottom of the Mississippi River with a grappling hook for a hoop net and basket trap set. Experienced fishermen quickly locate and retrieve their deep-water sets with a grappling hook. Fig. 9.—Commercial wing net sets. Wing nets are particularly effective for carp and buf- falofishes. trieves the net and removes the fish from the net. Hoop nets of 1-, 21/2-, and 3-inch mesh were used in the survey. The 3-inch- mesh nets were fished only at the Du- buque station. The wing net, fig. 9, is identical in con- struction to the hoop net insofar as the net proper is concerned. Attached to the first hoop of the wing net are two pieces of netting called wings, which give the net its name. The wings are set at about 45 degree angles to the main axis of the net and are secured in position with poles. The wing net is fished either with or with- out a lead, a piece of netting extending outward from the first hoop and continu- ing the main axis of the net. It is fished in sloughs, backwaters, and sluggish sec- tions of rivers; also in shallower water than the hoop net. It is held in position by poles. On the Illinois River, the wing net is the net most commonly used by fishermen, whereas on the Mississippi the hoop net is the one most generally fished. The wing net is not popular among com- July, 1955 Starrett & Barnickol: Commercial Fishing Devices 337 mercial fishermen who operate on the Mis- sissippi River. Wing nets with and without leads were used in the 1944 survey. In 1946, all wing nets were fished without leads. The meshes used in this study were 1 inch, iy2 inches, and 214 inches. The 11/9 -inch- mesh wing nets w^re used only in 1944. The catch data for the 2i/2-inch-mesh wing nets fished with leads have been combined with the catch data for these nets fished without leads. About 70 per cent of the commercial fish catch reported from the Illinois sec- tion and 59 per cent of that reported from the Missouri section of the Mississippi River for 1947 were taken in hoop nets (Upper Mississippi River Conservation Committee 1948, third section :19). The reports do not differentiate between hoop nets and wing nets ; however, the wing net catch comprised only a small fraction of the total annual yield. The trap net is quite similar to the wing net in construction and operation, fig. 10. It difiers from the wing net in that it has a double rectangular wooden frame in front of the first hoop. It is staked out in the same fashion as the wing net. It may be fished either with or with- out a lead. During the survey, trap nets were fished at only three stations. The mesh used was II4 inches. The nets were fished in localities identical to those fished with wing nets. Analyzing Entrapment Catch Data, —In the Mississippi River survey, entrap- ment devices were tested more than any of the other devices. The unit of measurement usually em- ployed in analyzing net catch data is the net-day, and the efficiency of a fishing de- vice is measured by the number of fish and/or the number of pounds of fish it catches per net-day. One net fished for 24 hours is termed one net-day. In this investigation, the number of net-days fished (or trap-days in the case of basket traps) was recorded for each entrapment device. In some instances, when the catch per net-day of a certain species of fish in one type of net is compared with the catch of this same species in another type of net, the difference is obviously significant. In other instances, it may not be clear as to whether the catch difference is significant, unless the data are tested statistically. In this study, tests for significance of differ- ences in the efficiency of various entrap- ment devices were made by using the chi- square method of analysis (Snedecor 1946:16). The chi-square was computed from the following formula: 2 _ (Xi- mi)2 , (X2 - m2)2 mi + X^ and X2 are the actual catches in numbers of commercial-sized fish and m, and m, are the expected catches. The ex- pected catch for each net of the two nets being compared is that part of the actual total catch of the two nets which is di- rectly proportional to the total number of net-days fished by the net. In table 4 are given catch data on commercial-sized carp, as presented in table 12, and below the table an equation that tests whether there is a significant difference in catch of commercial-sized carp (15 inches or more total length) between 1 -inch-mesh wing nets (without leads) and li/o-inch- mesh wing nets (most without leads*) at the 0.05 level with 1 degree of freedom. The value 21.09 indicates statistically that at the 0.05 level the catch of com- mercial-sized carp is significantly greater in the ll/^-inch-mesh wing net than in the 1 -inch-mesh wing net. A chi-square value of 3.841 or more is considered significant at the 0.05 level with 1 degree of free- dom. The fish samples were taken at the same stations and the nets were fished si- multaneously, thereby eliminating station and seasonal differences with respect to species. However, at each station various habitats were netted, and as a result prob- ably different segments of the population at each station were sampled. This pat- tern of netting in no way affects the anal- ysis, since the information sought was, in many cases, for the purpose of giving a comparison of efficiency and selectivity of two nets designed for different habitats. For example, the hoop net usually is fished in deeper water than is the wing net and is always fished in the current; the wing net usually is fished in quiet water. Hab- itat differences of flowing and quiet w^ater are discernible in the catch data, table 5. In the backwater areas and sloughs of the Of 156.10 net-days. 8.84 net-days were with leads. 338 Illinois Natural History Survey Bulletin Vol. 26, Art. 4 Table 4.—Data (from table 12) on catches of wing nets of two different mesh sizes used in fisheries survey of the Curuthersville-Dubuque section of the Mississippi River, 1944, the data selected and presented to allow comparison of the efficiency of the two mesh sizes in taking carp of commercial sizes. Net Number of Net-Days Per Cent of Total Number OF Net-Days Number of Commercial- Sized Carp Caught Number of Commercial- Sized Carp in Expected Catch 1 -inch-mesh. . l}^-inch-mesh Total. . . . 508.12 156.10 664.22 76 24 100 413 194 607 461.32 145.68 607.00 The following equation is based on the chi-square (x'') formula on page 337. , (413 - 461.32)2 (194 _ 145.68)2 461.32 + 145.68 = 5.06 + 16.03 = 21.09 Fig. 10.—Trap net. This net is used by some fishermen who operate on the Mississippi River. Photo courtesy of Dr. John Moyle, Division of Game and Fish, Minnesota Department of Conservation, St. Paul. July, 1955 Starrett & Barnickol: Commercial Fishing Devices 339 upper Mississippi, there is usually a more abundant population of sport fishes than in the river channel. In figs. 11 and 12, the total catches for all species of fish taken with entrapment devices are presented graphically. Spe- cies composition differences between the two sections of the river sampled in 1944 and 1946 account for most of the catch differences reflected in the graphs for the 2 years of test-netting. Differences be- tween 1944 and 1946 in the number of sport fish taken per net-day are apparent in the graphs. The data in table 5 are not suited for detailed analysis, since fishing with the different types of gear was not in all cases done simultaneously. In order to have some standard for comparing the efficiency and selectivity of the various entrapment devices, it was de- sirable to select a device that was fished at all the stations. The only entrapment device that met this requirement was the 1 -inch-mesh wing net without a lead. In the following catch analyses, the data for the 1 -inch-mesh wing net without a lead are used from only those stations at which one or more of the other devices were fished. The analyses concern largely the catches of the nine most important com- mercial and sport fishes occurring in the Caruthersville-Dubuque section of the river: carp, bigmouth buffalo, smallmouth buffalo, freshwater drum, channel catfish, flathead catfish, black crappie, white crappie, and bluegill. In tests for significant differences in numbers of a species caught by various en- trapment devices, only fish of commercial or desirable sizes were included. The fol- lowing minimum total lengths were used : 15 inches for carp, buffalofishes, and channel catfish; 10 inches for freshwater drum; 18 inches for flathead catfish; 8 inches for crappies ; 7 inches for bluegill. ^ 3 m 'y/y\ m WING NET 1-INCH MESH WITH LEAD [x] COMMERCIAL FISHES H SPORT FISHES ^PREDATORY FISHES [l] FORAGE FISHES WING NET 2 1/2-INCH MESH HOOP NET 2 1/2-INCH MESH _^L BASKET TRAP Fig. 11.—The number of fish, all sizes, taken per net-day in entrapment devices fished in the Mississippi River between Caruthersville and Warsaw in 1944. 340 Illinois Natural History Survey Bulletin Vol. 26, Art. 4 .2i Is is U ? u « a -I- 2 2-5 «^ .5 *^ c ca — js £ a 3 2 uS X '» w ^ a -O A' C S 'S .- E «> ^ k o ^ » « E 3 » O .2'^ " *C OS e a ^ >3 c f s.- c i«.!2 » S » JJ .2 E H « c QQ 5 csos) (S.VEQ-J3VJ OOSC) cf'-OCA) HS3N-H3"I-fi3 (SXBQ-J3fj i6'S0l) (sXeq-jsjvj 98-S58) (sXBQ-ja^ T8-981) (sXbq-J3^ c9-/.z.e) (sXEQ-jax Or9Sl) (sXbq-j3s^ Of'-SZ.l) ijsaj^-qDui-i I ISI \0 oo o O •^ ro oo I |0tN| I I |0->0 I i°:"i 1 1 rO •* PO r~ O fS PO — ^ I I lO — OPOO-* I I I I |0 — ©J ododdd o d ^ I l=! I I I Io *t I I *toooo oo *- I Tfc M I I I I oo oo ooo I I I I £ I I £ I I I I \0 '— oo 0\ I I I O — Ovo dodo I 19o I I o ooo oo- oo *^o *- ooooooooo M I I't I *^oooo- ^ o-o ofct u . 4^ 3 o wo c u 5 c . V > ^:: -71 >< y 2 £•&« ^^ C/5 2- < 2: V, — — X '^^ O .-3 t- lU re li vj - 2- O >^ W? c))^- O 30- 3 C O 3 E '-^ u c to }J July, 1955 Starrett & Barxickol: Commercial Fishing Devices 341 o ' ' o o o o • o o ' o ^ I I I I ° I I ^ o o I I o o do o 1d 342 Illinois Natural History Survey Bulletin Vol. 26, Art. 4 In comparing the efficiency and selectiv- ity of the various entrapment devices, it was necessary to consider the average size of the fish taken, as well as the num- ber of commercial-sized or desirable-sized fish caught. A small-mesh net might take as many fish of commercial species as a large-mesh net ; however, the catch of the small-mesh net might be largely of fish under commercial size. The large-mesh net taking only a few fish under commer- cial size would be a more efficient net to the fisherman since it would require less time for the handling and releasing of small fish. The mean of the total lengths of indi- JFig. 12.—The number of fish, all sizes, taken per net-day in entrapment devices fished in the Mississippi River between Burlington and Dubuque in 1946. Table 6.—Data (from table 13) on catches of wing nets of two different mesh sizes used in fisheries survey of the Caruthersville-Dubuque section of the Mississippi River, 1944, the data selected and presented to allow comparison of the sizes of carp taken by the two mesh sizes. Net July, 1955 Starrett & Barnickol: Commercial Fishing Devices 343 Table 7.---Composition of catches made in 74 basket trap sets in fisheries survey of the Mississippi River between Cape Girardeau and Louisiana, Missouri, 1944. Kind of Fish Commercial Carp Blue catfish Channel catfish Flathead catfish Freshwater drum Predatory Shortnose gar Total Number of Fish Taken 2 18 32 9 1 11 73 Per Cent of Total Number Taken 2.7 24.7 43.8 12.3 1.4 15.1 100.0 Weight IN Pounds 2.52 8.70 32.24 12.98 0.75 5.25 62.44 Per Cent of Total Weight 4.0 13.9 51.7 20.8 1.2 8.4 100.0 Table 8.—Composition of catches of commercial-sized or desirable-sized fish taken in basket traps and in 1-inch-mesh wing nets without leads in fisheries survey of the Mississippi River between Cape Girardeau and Louisiana, Missouri, 1944. Kind of Fish Carp Bigmouth buffalo. . Smallmouth buffalo Freshwater drum. . . Channel catfish. . . . Flathead catfish. . . . Bluegill White crappie Black crappie Basket Traps (74.0 Net-Days) Pounds per Net-Day 0.01 0.34 0.09 Number 1 16 2 Number per Net-Day 0.01 0.22 0.03 1-Inch-Mesh Wing Nets Without Leads (236.5 Net-Days) Pounds per Net-Day 0.53 0.26 1.93 Number 229 4 12 171 31 99 21 187 249 Number per Net-Day 0.97 0.02 05 0.72 0.13 0.42 0.09 0.79 1.05 Chi- Square Value 51.71* 2.60 26.85* * Denotes a significant difference, in numbers of commercial-sized or desirable-sized fish taken, between basket traps and 1-inch-mesh wing nets at 0.05 level with 1 degree of freedom. viduals of each species taken in each type of net was computed from measurements of all the fish of that species taken in that t3pe of net. The actual total length of each individual was converted to the near- est one-half inch. The mean of the total lengths and the standard deviation of length of the individuals of a species taken in a particular type of net indicates ap- proximately the size of fish that can be expected in future catches of that net. The presence or absence of dominant year- classes of the various species would be ex- pected to change the species composition and the mean of the total lengths of indi- viduals in the catches from year to year. However, such changes would probably not materially affect the basic differences in efficiency and selectivity of the devices discussed here. Frequently the differences in the means of the total lengths of fish taken by two types of nets were not apparent. To de- termine whether the sizes of the indi- viduals of a species taken in one type of net were significantly dififerent from the sizes of the individuals of that species taken in another type of net, a statistical test was desirable. The t-test, derived from a Snedecor (1946:81) formula adapted to groups with dilierent numbers of individuals, appeared to be a satisfac- tory test for significance of size differ- ences of fish taken in two types of nets. The formula used in computing the t-test for significance is as follows: 344 Illinois Natural History Survey Bulletin Vol. 26, Art. 4 t •= Xi - Xj (si)' + (St)' X, and "Xo are the mean total lenj^ths of the Hsh ; n, and n. represent the num- bers of individuals, all sizes, of the species taken in the gear ; and s, and So represent the standard deviation. The degrees of freedom were determined by using the geometric mean : d.f. = V ni • n; - 1 From actual net data, table 13, an ex- ample is given below table 6 to test if a significant difference exists between the size of carp taken in 1 -inch-mesh wing nets and the size of those taken in the 11/4;- inch-mesh wing nets. The t-test below table 6 shows that, at the 0.05 level and with 561 degrees of freedom, there is a significant difference in size between the carp taken in the ly^- inch-mesh wing nets and those taken in the 1 -inch-mesh wing nets. The 11/2" inch-mesh wing nets, on the basis of this test, catch carp which are significantly larger than those taken in the 1 -inch-mesh wing nets. Catches With Basket Traps.—Not a great deal of fishing was done with bas- ket traps during the survey; however, 74 sets were made in the spring and early summer months of 1944. In these sets, the catch was dominated by catfish, table 7. Of a total of 73 fish taken with basket traps, 80.8 per cent were catfish. Chan- nel catfish made up 54.2 per cent of the number and 59.8 per cent of the weight of the catfish catch. The shortnose gar, the only other fish taken in numbers with basket traps, amounted to 15.1 per cent of the number of fish taken with these traps. No sport fish was taken in the basket traps set by the survey party. However, the writers have observed a few crappies taken in baited basket traps set in one of the bot- tomland lakes of the Illinois River. A comparison of the catches of 1 -inch- mesh wing nets with catches of basket traps may be made by referring to tables 8 and 9. Fishing was done at the same stations. The channel catfish was the only species of fish taken in greater numbers per net-day in basket traps than in 1 -inch- mesh wing nets ; the difference was not significant. The catch of flathead catfish was much greater in 1 -inch-mesh wing nets than in basket traps. The difference Table 9.—Size ranges and means of the total lengths of fish taken in basket traps and in 1 -inch-mesh wing nets without leads in fisheries survey of the Mississippi River between Cape Girardeau and Louisiana, Missouri, 1944. July, 1955 Starrett & Barnickol: Commercial Fishing Devices 345 in size between channel catfish and flat- head catfish taken in 1 -inch-mesh wing nets and these same kinds of fish taken in basket traps w'as not significant. The basket trap is considered by com- mercial fishermen of the upper Missis- sippi as an efficient device for catching catfish. It provides the fisherman with a device that usually does not take sport fish and that can catch channel catfish as effi- ciently as does the 1 -inch-mesh wing net, which is illegal for use in the upper Mis- sissippi. The basket trap was found to be one of the most selective fishing devices tested on the river. Catches With Wing Nets.—The catches made with wing nets during the survey tend to indicate that commercial fishermen of the upper Mississippi might well increase their catches by using these Table 10.—Composition of catches of commercial-sized or desirable-sized fish taken in 1-inch-mesh wing nets with leads and in 1-inch-mesh wing nets without leads in fisheries survey of the Mississippi River between Caruthersville and Hannibal, Missouri, 1944. Kind of Fish 346 Illinois Natural History Survey Bulletin Vol. 26, Art. 4 nets more. Wing nets are particularly adapted to fishing backwater and slough areas. In the canalized section of the up- per Mississippi, there are now more slough and backwater areas than in years previous to the construction of dams. In periods of low water, when the current is too sluggish for hoop nets, wing nets can be used to advantage. In that part of the survey made in 1944, several 1 -inch-mesh wing nets were fished with leads and others were fished without leads. In table 10 are listed, for each of nine species, the num- bers of commercial fish of commercial sizes and sport fish of desirable or usable sizes caught in these nets. The catch per net-day of buffalofishes of commercial sizes was somewhat greater and the catch of crappies and bluegills of desirable sizes was much greater in the nets wath leads than in the nets without leads. With re- spect to each of the other species of fish listed, the difiference between the catch Table 12.—Composition of catches of commercial-sized or desirable-sized fish taken in I'/j-inch-mesh wing nets (most without leads) and in 1-inch-mesh wing nets (without leads) in fisheries survey of the Mississippi River between Grand Tower and Warsaw, Illinois, 1944. Kind of Fish 1^-Inch-Mesh Wing Nets (156.10 Net-Days) Pounds per Net-Day Number Number per Net-Day 1-Inch-Mesh Wing Nets (508.12 Net-Days) Pounds per Net-Day Number Number per Net-Day Chi- Square Value Carp Bigmouth buflFalo . . Smallmouth buffalo Freshwater drum. . . Channel catfish. . . . Flathead catfish. . . . Bluegill.... White crappie Black crappie 3.09 0.09 0.38 0.08 1.05 194 3 72 6 40 12 51 66 1.24 0.02 0.46 0.04 0.26 0.08 0.32 0.42 2.18 0.04 0.08 0.39 0.20 0.78 413 6 13 255 59 86 49 319 496 0.81 6.01 0.03 0.50 0.12 0.17 0.10 0.63 0.98 21.09* 0.43 0.71 7.78* 4.14* 0.63 21.17* 46.29* • Denotes a significant difference, in numbers of commercial-sized or desirable-sized fish taken, between the two types of nets at 0.05 level with I degree of freedom. Table 13.—Size ranges and means of the total lengths of fish taken in l'/2-'nch-mesh wing nets (most without leads) and in 1-inch-mesh wing nets (without leads) in fisheries survey of the Mississipp' River between Grand Tower and Warsaw, Illinois, 1944. Kind of Fish SOS a, £ rt 1J^-1nch-Mesh Wing Nets o ^ ^:< M S O OJ ^ U 1> a- fc rt 1-Inch-Mesh Wing Nets H^i; 1^ u -O rtl Bj Carp Bigmouth buffalo Smallmouth buf- falo Freshwater drum Channel catfish. . Flathead catfish. Bluegill ... White crappie. . . Black crappie. . . 54.5 75.0 0.0 80.0 60.0 50.6 40.0 61.4 52.4 356 4 29 90 10 79 30 83 126 14.96 16.88 10.47 11.23 16.20 18.38 6.57 8.39 8.08 7.0-22.7 12.6-22.3 7.7-14.0 6.6-16.8 11.1-21.1 12.3-31.0 5.6- 8.0 5.7-12.4 6.1-10.9 2.37 4.45 46. 46, 1.78 1 93 3.20 3.71 0.63 1.50 1.07 7.8 43.4 24.0 29.5 33.1 54.9 70.9 13 166 588 246 292 148 581 700 14.47 14.92 11.36 9.45 12.85 16.28 6.43 7.94 8.43 5.2-30.7 10.0-21.6 5.6-28.0 3.8-16.4 2.8-24.5 7.0-41.2 4.2- 8.2 4.6-15.2 4.5-12.0 3.34 2.88' 561 6 2.83 68 2.28 229 3.33 49 4.38 0.79 151 66^ 1.46 219 1.25 296 2.908* 0.829 2.242* 7.954* 3.240* 4,287* 1.058 2.564* 3.293* * Denotes a significant difference, in sizes of fish taken, between the two types of nets at 0.05 level for degrees of freedom indicated. July, 1955 Starrett & Barxickol: Commercial Fishing Devices 347 rate of the nets with leads and that of the nets without leads was not great enough to be statistically significant. Data from the sur\'ey indicate that 1 -inch-mesh wing nets with leads were the most efficient entrapment gear used for crappies and bluegills. That small-mesh wing nets were effective during the sur- vey in taking sport fishes is shown in table 5. With or without leads, 1 -inch-mesh wing nets were effective in catching pred- The species composition of the 1944 catches in the 1 -inch-mesh wing nets (without leads) and that of the 1 i/o-inch- mesh wing nets were quite similar, fig. 1 1 and table 5. The catch per net-day of the 114-inch-mesh nets, however, was less than that of the 1 -inch-mesh nets. The in- creased mesh size of one-half inch tended to reduce the take of small fish. The wing nets of 1 i/o-inch-mesh took, per net-day, fewer sport fishes of all sizes. Table 14.—Catches of entrapment devices used in fisheries survey of the Mississippi River between Caruthersville, Missouri, and Dubuque, Iowa, 1944 and 1946. For each device is given the percentage of the catch represented by each of the various classes of fish: commercial, sport, predatory, and forage. 348 Illinois Natural History Survey Bulletin Vol.26, Art. 4 of 1-inch mesh than in the nets of larger mesh. The writers are unable to explain why the mean length for one species of crappie was greater in nets of 1-inch mesh and the mean length for the other species was greater in the nets of larger mesh. In the nets of 1-inch mesh, 24.0 per cent of the channel catfish and 29.5 per cent of the flathead catfish were of com- mercial sizes; in the nets of U/L'-inch mesh, bO.O per cent of the channel catfish and 50.6 per cent of the flathead catfish were of commercial sizes, table 13. The number of commercial-sized fish per net- day was greater for the channel catfish in the nets of 1-inch mesh and for the flat- head catfish in the nets of li/>-inch mesh, table 12. The 21/4-inch mesh was the largest mesh tested in wing nets. The 2 years of test fishing indicate that very few sport fishes are taken in nets of this mesh size, Table 15.—Composition of catches of commercial-sized or desirable-sized fish taken in 2'/2 -inch-mesh wing nets and in 1 -inch-mesh wing nets, both types without leads, in fisheries sur\ey of the .Mississippi River between Burlington and Dubuque, Iowa, 1946. July, 1955 Starrett & Barxickol: Commercial Fishing Devices 349 Table 17.—Composition of catches of commercial-sized or desirable-sized fish taken in 1-inch-mesh hoop nets and 1-inch-mesh wing nets, both types without leads, in fisheries survey of the Mississippi River between Burlington and Dubuque, Iowa, 1946. Kind of Fish 1-Inch-Mesh Hoop Nets (105.97 Net-Days) Carp Bigmouth buffalo . . Smallmouth buffalo Freshwater drum. . . Channel catfish. . . . Flathead catfish. . . . Bluegill White crappie Black crappie Pounds per Net-Dav Number 0.48 0.34 0.10 0.16 9 32 8 5 1 9 6 Number per Net-Dav 0.08 0.30 0.08 05 0.009 0.13 0.06 1-Inch-Mesh Wing Nets (855.86 Net-Days) Pounds per Net-Dav 1.10 0.04 0.04 0.23 0.10 0.27 Number 238 13 16 210 53 42 97 1,228 560 Number per Net-Day 0.28 0.02 0.02 0.25 0.06 0.05 0.11 1.43 0.65 Chi- Square Value 13.65* 1.22 0.28 0.06 9.97* 133.34* 57.12* * Denotes a significant difference, in numbers of commercial-sized or desirable-sized fish taken, between the two types of nets at 0.05 level with 1 degree of freedom. Table 18.—Size ranges and means of the total lengths of fish taken in 1-inch-mesh hoop nets and in 1-inch-mesh wing nets, both types without leads, in fisheries survey of the Missis- sippi River between Burlington and Dubuque, Iowa, 1946. 1-Inch-Mesh Hoop Nets 1-Inch-Mesh Wing Nets I' c- U ^ K.IXD OF hlSH I, c ° «J CJ.5c/D ;a-. c rt Carp 100.0 Bigmouth buf- falo Smallmouth buffalo 0.0 Freshwater drum 49 . 2 Channel catfish. . 12.9 Flathead catfish. 11.4 Bluegill 50.0 White crappie. . . 64.3 Black crappie. . . i 66.7 „ 350 Illinois Natural History Survey Bulletin Vol.26, Art. 4 carp avoid small-mesh nets and traps, but that they are taken readi'y in larjje-mesh nets. He writes. "Channel cat and bull- heads seem to go most readil\ into small- mesh nets and basket traps, probably be- cause, being largely nocturnal and hiding during the day, they go into the nets for concealment." In 1046 test fishing, all of the 172 carp taken in the 2V-j-'"^'h"'"P^l^ wing nets were of commercial sizes, table 16. At the stations at which the 2Vi,»-'nch-mesh wing nets were Hshed, only 78.5 per cent of the 303 carp taken in 1 -inch-mesh wing nets were of commercial sizes. The catch of commercial-sized carp in the 2V^- inch-mesh wing nets amounted to 1.76 pounds per net-day and in the 1 -inch-mesh nets to 1.10 pounds per net-day, table 15. The data indicate that the 1 -inch-mesh Table 19.—Composition of catches of commercial-sized or desirable-sized fish taken in 2'/2-inch-mesh hoop nets and in 1-inch-mesh wing nets, both types without leads, in fisheries survey of the Mississippi River between Burlington and Dubuque, Iowa, 1946. Kind of Kish 2^-Inch-Mesh Hoop Nets (730.42 Net-Davs) Pounds per Net-Day Number Number per Net-Day 1-Inch-Mesh Wing Nets (855.86 Net-Days) Pounds per Net-Day Number Number per Net-Day Chi- Square Value Carp Bigmouth buffalo . . Smallmouth buffalo Freshwater drum. . . Channel catfish. . . . Flathe.-id catfish. . . . BluegiU..... White crappie Black crappie 1,10 0.19 0.19 0.34 0.01 0.84 150 45 59 152 2 58 3 1 0.21 0.06 0.08 0.21 0.003 0.08 0.004 0.001 1.10 0.04 0.04 0.23 0.10 0.27 238 13 16 210 53 42 97 1,228 560 0.28 0,02 0.02 0.25 0.06 0.05 0.11 1.43 0.65 8.42* 23.30* 32.22* 2.35 39.74* 5.80* 1,037.54* 474.19* • Denotes a significant difference, in numbers of commercial-sized or desirable-sized fish taken, between tlie two types of nets at 0.05 level with 1 degree of freedom. Table 20.—Size ranges and means of the total lengths of fish taken in 2'/2 -inch-mesh hoop nets and in 1-inch-mesh wing nets, both types without leads, in fisheries survey of the Missis- sippi River between Burlington and Dubuque, Iowa, 1946. July, 1955 Starrett Si Barnickol: Commercial Fishing Devices 35] wing nets are much more efficient in tak- ing catfishes than are the 2i/4-mch-mesh wing nets, table 15. In 1946, only 4 flat- head catfish were taken in the 2l/^-inch- mesh wing nets and 179 (42 of them of commercial sizes, table 15) in the 1-inch- mesh wing nets, table 16. The fish of this species in the 2y2-inch-mesh wing net col- lections ranged in total length from 18.8 to 37.0 inches, and in the 1 -inch-mesh wing net collections from 5.6 to 37.6 inches. More commercial-sized buffalofishes per net-day were caught in the 2l/2-inch- mesh wing nets than in the 1 -inch-mesh wing nets, table 15. In the 2y2-inch-mesh wing net catches, 97.8 per cent of the big- mouth buffalos and 68.6 per cent of the smallmouth buifalos were of commer- cial sizes, table 16. In the 1 -inch-mesh wing net sets, only 39.4 per cent of the bigmouth buffalos and 5.8 per cent of the smallmouth buffalos were of commercial sizes. More freshwater drums of commer- cial sizes were caught per net-day in the 1 -inch-mesh wing nets than in the 21/2- inch-mesh wing nets, table 15. On a pounds-per-net-day basis of commercial- sized drums, the catches of the two types of net were almost identical. All of the freshwater drums taken in the Zy^-inch- mesh wing nets were of commercial sizes, whereas only 33.2 per cent of these fish taken in the 1 -inch-mesh wing nets were of these sizes, table 16. The test-net fig- ures indicate that a fisherman using 2^/2" inch-mesh wing nets will handle fewer undersized freshwater drums than one using nets of a smaller mesh size. As indicated in preceding paragraphs, the efficiency and selectivity of wing nets of the mesh sizes used in the survey varied with species and sizes of fish. The small- mesh (1-inch) wing nets were more se- lective for crappies, bluegills, and cat- fishes than the 214-inch-mesh nets, table 15. The 21/2-inch-mesh nets were more efficient than the small-mesh (1-inch) nets in taking commercial-sized carp and buffalofishes, table 15. Catches With Hoop Nets.—In the 1946 test fishing, sport fishes comprised 53.4 per cent of the entire catch with 1 -inch-mesh wing nets (without leads), whereas they comprised only 10.4 per cent of the catch with 1 -inch-mesh hoop nets, table 14. No attempt was made to fish hoop nets and wing nets of the same mesh in the same habitat to determine if the absence or presence of the wings influ- enced the catch. In tables 17 and 18 are listed compara- tive data on nine species of fish caught in the 1946 test fishing in 1 -inch-mesh hoop nets and 1 -inch-mesh wing nets (without leads) used at the same stations, al- though not necessarily in similar habitats. The numbers, per net-day, of commer- cial-sized carp and usable-sized bluegills and crappies caught in the wnng nets were significantly greater than the numbers, per net-day, of these fish caught in hoop nets, table 17. The numbers, per net-day, of commercial-sized freshwater drums and catfishes did not differ greatly between the two nets. The weights, per net-day, of commercial-sized carp and flathead cat- fish were greater in the wing net catch, and the weight, per net-day, of commer- cial-sized freshwater drums was greater in the hoop net catch. No buffalofish of commercial size was taken in the 1 -inch-mesh hoop net sets. The catch of these fishes in the 1 -inch- mesh wing nets was small. The mean of the total lengths for each of three species, carp, smallmouth buffalo, and freshwater drum, was significantly greater for individuals taken in the 1 -inch- mesh hoop nets than for those taken in the wing nets of the same mesh, table 18. The mean of the total lengths for flathead cat- fish was greater for individuals taken in the wing nets. The small number of carp (nine) taken in the hoop nets casts doubt on the value of the test for fish of this spe- cies. The mean of the total lengths for bluegills and for crappies was approxi- mately the same for individuals taken in hoop nets as for those taken in wing nets. In tables 19 and 20 are listed compara- tive data on nine species of fish taken in the 1946 test fishing in 2V2-inch-mesh hoop nets and 1 -inch-mesh wing nets (without leads) fished at the same sta- tions, but in most cases in different habi- tats. The number of commercial-sized buf- falofishes taken, per net-day, was much greater in the 2V2-inch-mesh hoop nets than in the small-mesh wing nets, table 352 Illinois Natural History Survey Bulletin Vol.26, Art. 4 19. The 2' 2*'nt:h-mesh hoop nets were less efficient in taking commercial-sized bigmouth buffalos than were the lYAnch- mesh wing nets, table 21. The catch per net-day of smallniouth buffalos in the 2yo-inch-mesh hoop nets was identical with that in the 2i/2"i"^'li""i^sh wing nets. Freshwater drums often frequent wa- ters too deep and swift for wing nets. However, such waters are usually suitable for hoop net fishing. Perhaps that is why in 1946 the weights and numbers, per net- day, of commercial-sized drums were higher in the 2i4-inch-mesh hoop net catch than in the 2V2-ir'ch-mesh wing net catch, table 21. The mean of the total lengths for drums caught in 1946 in 21/2- inch-mesh wing nets was greater than for drums caught in the 2i/2-inch-mesh hoop nets, table 22. The test-net survey data indicate that the 2V2-inch-mesh hoop nets are not effi- Table 21.—Composition of catches of commercial-sized or desirable-sized fish taken in 21/2 -inch-mesh wing nets and in 21/2 -inch-mesh hoop nets, both types without leads, in fisheries survey of the Mississippi River between Burlington and Dubuque, Iowa, 1946. July, 1955 Starrett iS: Barxickol: Commercial Fishing Devices 353 cient devices for catching channel cat- nets did not take as many sport fishes per fish, table 5. However, these large-mesh net-day as the 1 -inch-mesh wing nets, ta- nets were found to be the most efficient hies 5 and 23 ; however, they took more devices tested for taking commercial-sized than the nets of 2l/'2-inch-mesh. The trap Table 23.—Composition of catches of commercial-sized or desirable-sized fish taken in IV4 -inch-mesh trap nets (with leads) and in 1-inch-mesh wing nets (without leads) in fisheries survey of the Mississippi River between Andalusia, Illinois, and Dubuque, Iowa, 1946. Kind of Fish Carp Bigmouth buffalo . . Smallmouth buffalo Freshwater drum. . . Channel catfish. . . . Flathead catfish. . . . Bluegill White crappie Black crappie IJ^-Inch-Mesh Trap Net (50.53 Net-Days) Pounds per Net-Day 2.78 0.29 0.05 Number 37 14 2 2 20 6 Number per Net-Day 0.73 0.28 0.04 0.04 0.40 0.12 1-Inch-Mesh Wing Net (200.50 Net-Days) Pounds per Net-Day 0.96 0.26 0.04 0.15 Number 43 2 56 5 9 21 131 137 Number per Net-Day 0.21 0.01 0.28 0.02 0.05 0.10 0.65 0,68 Chi- Square Value 34.45* 0.00 0.32 1.84 4.31* 22.32* * Denotes a significant difference, in numbers of commercial-sized or desirable-sized fish taken, between the two types of nets at 0.0? level with 1 degree of freedom. flathead catfish. Of the flatheads taken with these nets in 1946, 95.1 per cent were 18 inches or more in total length, table 20. In the 1-inch-mesh wing net collections of 1946, only 23.5 per cent were 18 inches or more in total length. The mean of the total lengths of flatheads taken in 1946 in the 2V2-inch-mesh hoop nets was 26.19 inches, table 20, as com- pared with 15.39 inches in the 1-inch mesh wing nets and 13.20 inches in the 1-inch-mesh hoop nets, table 18. In the 1946 test-netting, the mean of the total lengths for the 58 flatheads taken in the 21/2-inch-mesh hoop nets was about the same as the mean for the 4 flatheads in the 2l/:>-inch-mesh wing net collections, table 22. The catch of sport fishes in the 21/2' inch-mesh hoop nets amounted to less than 0.01 fish per net-day in 1946, table 5. In this same year, the catch of sport fishes was 5.43 fish per net-day in the 1-inch-mesh wing net collections. Catches With Trap Nets.—Trap nets were fished at only three stations for a total of 50.53 net-days. The nets were all of 11/4-inch mesh. As indicated by fig. 12, the catch with these nets included both commercial and sport fishes. These net sets made during this investigation were too few in number to give any con- clusive evidence regarding usage of these nets on the Mississippi. Entanglement Device The trammel net is the only entangle- ment device that was used in the survey. This device is composed of three separate nets secured to a single top and a single bottom line. The two exterior nets (for purposes of explanation called here left- exterior and right-exterior) are made of heavy twine and have a mesh size of 8 to 10 inches. The inner net is a small-mesh gill net loosely sandwiched between the two large-mesh outer nets, fig. 13. A fish swimming into a trammel net from the left side passes freely through the left-exterior net and strikes the loosely hung inner net with enough force to carry a portion of the inner net with it through a mesh opening of the right-exterior net. The inner net, passing through a mesh of the right-exterior net, forms a pocket in which the fish is enclosed. Similarly, a fish approaching the trammel net from the right side passes through the right-exte- rior net and becomes trapped in a pocket 354 Illinois Natural History Survey Bulletin Vol.26, Art. 4 Fig. 13.—View of a trammel net, showing the two exterior large-mesh nets and the inner small-mesh net. formed when the inner net is carried through a mesh opening of the left-exte- rior net. The fish is held in the pocket until released by a fisherman, fig. 14. Trammel nets can be fished in all types of habitats found on the Mississippi. The method by which the trammel net is fished differs with habitat. If a backwater or quiet stretch of the river is to be fished, the net is set, fig. 15. If the river channel is to be fished, the net is floated or drifted downstream. Trammel net sets and floats were made during the survey at some of the field sta- tions in the D-MR section of the river. A summary of the trammel net fishing data is presented in table 24. Except for data collected from experimental tram- mel netting at Grafton in October, 1944, and at Quincy in November of the same year, the data relative to catches made in trammel nets of various lengths and mesh sizes were combined. Data from the ex- perimental trammel netting at Grafton and at Quincy were not included in the previous report on the survey (Barnickol & Starrett 1951). In 1947, trammel nets accounted for 25.1 per cent of the commercial catch from the Missouri section of the river and 14.5 per cent from the Illinois section (Upper Mississippi River Conservation Committee 1948, third section :19). Catches With Set Trammel Nets. —The set method of trammel netting is the one most generally used by commer- cial fishermen of the upper Mississippi River. When this method is used, the net is set around a school of fish. Some- times it is laid out in a zigzag or spiral fashion. Frequently the shore line is used as a backstop for the net. After the net has been set around the fish, a disturb- ance in the water is created with plungers, fig. 16. Usually the trapped fish become excited and, when they attempt to escape July, 1955 Starrett Si. Barnickol: Commercial Fishing Devices 355 by swimming toward the deep w^ater, they hit the net. The trammel net can be fished select- ively for commercial species by a fisher- man familiar with the habits of these fishes. Carp and buffalofishes tend to school in large numbers when spawning or feeding. These fishes ripple the surface of the water or stir up the bottom, revealing their presence to the commercial fisher- man searching for signs of fish activity. If the fisherman believes there are quite a few fish of a desired species present, he proceeds to set his trammel net around them. On three occasions, selective net- ting for one or two commercial species was done by the survey crew. The catches made in these sets are listed in table 25. Carp and buffalofishes dominated these catches. No sport fish was taken in these sets. Trammel net catches taken by com- mercial fishermen and examined by the writers have been found to contain largely carp and buffalofishes. Commercial fishes, most of them carp and buffalofishes, dominated the 116 test sets made in 1944 and 1946 with trammel nets, table 26. The trammel net seems to be a very inefficient method for taking catfish. Only three channel catfish and two flathead catfish were netted in the 116 sets. Bowfins and gars were caught regularly in the sets. These fishes are sold com- mercially in some areas along the river. More than 90 per cent of the carp, big- mouth buffalos, and freshwater drums taken in 73 sets of trammel nets in 1946 were of commercial sizes, table 27. Sport fishermen have long believed and argued that the trammel net is a destroyer of sport fishes. Their belief probably has been influential in outlawing the use of the trammel net in most midwestern wa- Fig. 14.—A pocketed i\sh in a trammel net. The inner net forming the pocket around the fish hangs over the cord of one of the exterior nets. 356 Illinois N.atur.al History Survey Bulletin Vol. 26, Art. 4 L^B Fig. 15.—Fishermen setting a trammel net around a school of carp on the Mississippi. ters. Sport fishes amounted to 17.3 per cent of the catches in trammel nets set at the various stations tested along the river in 1944 and 1946, including the late fall, 1944, experimental catches at Grafton, table 26. The majority of the river sets were made with a net having a li/^-inch- or 2-inch-mesh inner net and usually were not made in sites indicating the pres- ence of carp or buffalofishes. Many of the sets were made in areas that appeared especially favorable for sport fishes. In the 116 sets made with trammel nets of various mesh sizes, only 7 bass (Microp- Table 24.—Summary of trammel net sets and floats made in the Mississippi River by the fisheries survey party between Grafton, Illinois, and Dubuque, Iowa, 1944 and 1946.* July, 1955 Starrett & Barxickoi. : Commercial Fishing Devices 357 Table 25.—Examples of trammel net sets in which catches of the fisheries survey party were dominated by a single species of fish, Mississippi River, 1946. Kind of Fish Burlington, Iowa April 17, 1946 (One Set) Number Pounds Muscatine, Iowa May 28, 1946 (Two Sets) Number Pounds Carp Bigmouth buffalo. . . Smallmouth buffalo. Carpsuckers Freshwater drum . . . Total 31 1 32 0.00 88.21 2.10 0.00 0.00 90.31 149 1 7 2 7 166 ST1.19 3.50 10.39 2.60 9.62 553.90 Fig. 16.—A commercial fisherman using a plunger to create a disturbance in the water in an effort to drive fish into his trammel net. 358 Illinois Natural History Survey Bulletix Vol. 26, Art. 4 Table 26.—Composition of catches of trammel net sets and trammel net floats made in the Mississippi River by the fisheries survey party between Grafton, Illinois, and Dubuque, Iowa, 1944 and 1946. Included are catches of the experimental sets made at Grafton and of the experi- mental floats made at Quincy in the fall of 1944 and all other sets and floats except a few that were omitted because of discrepancies in methods of fishing. Kind of Fish Trammel Net Sets (116) Number of Fish Per Cent of Total Number Weight, Pounds Per Cent of Total Weight Trammel Net Floats (108) Number of Fish Per Cent ofTotal Number Weight, Pounds Per Cent ofTotal Weight Commercial Shovelnose sturgeon. . Carp Bigmouth buffalo. . . . Black buffalo Smallmouth buffalo. Bullheads. Channel catfish Flathead cattish Suckers and redhorses Carpsuckers Freshwater drum .... Subtotal Sport Largemouth bass. . . . Spotted bass Black crappie White crappie Bluegill Warmouth Sauger Northern pike White bass Yellow bass Subtotal Predatory Longnose gar Shortnose gar Bowfin Subtotal Forage Gizzard shad Goldeye Subtotal Total 584 135 10 32 14 3 2 4 67 50 901 1 6 117 76 20 2 1 1 2 2 5 94 51 150 35 3 38 1,317 0.0 44.3 10.3 0.8 2.4 1.1 0.2 0.1 0.3 5.1 3.8 68.4 0.1 0.5 8.9 5.8 1.5 0.1 0.1 0.1 0.1 0.1 17.3 0.4 7.1 3.9 11.4 2.7 0.2 2.9 100.0 0.00 1,986.54 376.63 23.76 45.42 11.68 8.37 4.45 7.56 84.89 63.74 2,613.04 0.87 6.50 61.35 43.04 8.35 0.98 1.52 3.49 3.45 0.87 130.42 12.85 140.98 171.41 325.24 30.33 2.00 32.33 3,101.03 0.0 64.1 12.1 0.8 1.5 0.4 0.3 0.1 0.2 2.7 2.1 84.3 tr.* 0.2 2.0 1.4 0.3 tr.* 0.1 0.1 0.1 tr.* 4.2 0.4 4.6 5.5 10.5 0.9 0.1 1.0 100.0 383 28 2 2 7 7 14 96 539 5 5 14 5 19 1 1 564 67.9 4.9 0.0 0.0 0.4 0.0 0.4 1.2 1.2 2.5 17.0 95.5 0.0 0.0 0.0 0.0 0.0 0.0 0.9 0.0 0.0 0.0 0.9 2.5 0.9 0.0 3.4 0.0 0.2 0.2 100.0 473.38 135.21 0.00 0.00 3.94 0.00 10.42 11.32 15.55 16.80 109.83 776.45 0.00 0.00 0.00 0.00 0.00 0.00 8.54 0.00 0.00 0.00 8.54 57.49 8.66 0.00 66.15 0.00 0.83 0.83 851.97 55.5 15.9 0.0 0.5 0.0 1.2 1.3 1.8 2.0 12.9 91.1 0.0 0.0 0.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 1.0 6.8 1.0 0.0 7.8 0.0 0.1 0.1 100.0 * ir. (trace) indicates that the species was taken but that the take averaged less than 0.1 per cent of total weight. Table 27.—Summary of catches of three important commercial fishes in 73 trammel net sets made in the fisheries survey of the upper Mississippi River, 1946. Kind of Fish July, 1955 Starrett (St Barxickol: Commercial Fishing Devices 359 terus spp.) and 20 bluegills were caught. Crappies amounted to 84.6 per cent of the 228 sport fishes taken in the 116 sets. Experimental Sets at Grafton.— During the regular test-netting program, no specific test was made of the efficiency rious mesh sizes of the inner net in the taking of sport fishes, as well as commer- cial and other species. The experimental trammel net was composed of three 50-yard lengths sewed together to form one net 150 yards in Table 28.—Summary of the catches of fish taken in 27 sets made with a 150-yard experi- mental trammel net in three bottomland lakes (Flat, Silver, and Royal) near Grafton, Illinois, October, 1944. Figures in parentheses indicate the numbers of fish gilled. Kind of Fish 13^-Inch-Mesh Net Number Caught Per Cent Gilled 2-Inch-Mesh Net Number Caught Per Cent Gilled 3-Inch-Mesh Net Number Caught Per Cent Gilled Commercial Channel catfish . Black bullhead. . Brown bullhead. Carp Smallmouth buffalo Bigmouth buffalo Black buffalo . . . Carpsuckers Subtotal Average Sport Largemouth bass Spotted ba.ss. . . . White crappie. . . Black crappie. . . Bluegill Warmouth Yellow bass Subtotal Average Predatory Shortnose gar. . . Longnose gar . . . Bowfin Subtotal Average Forage Gizzard shad. . . . Goldeye Subtotal Average Total Average 0(0) 12(1) 0(0) 1(0) 5(1) 0(0) 2(0) 3(0) 23 {2) 1 (0) 5(2) 8 (0) 53 (8) 4(1) 1 (1) 2(0) 74 {12) 46 (7) 2(1) 11 (0) 59 {8) 8(3) 1(0) 9 {3) 165 {25) 0.0 8.3 0.0 0.0 20.0 0.0 0.0 0.0 8.7 0.0 40.0 0.0 15.1 25.0 100.0 0.0 16 2 15.2 50.0 0.0 13 6 37.5 0.0 33 3 15 2 2(0) 0(0) 1 (0) 6(2) 3(0) 13(5) 3(1) 4(0) 32 {8) 0(0) 1 (1) 6(0) 13(4) 0(0) (0) 0(0) 20 (5) 2(0) 0(0) 6(0) 8{0) 11 (1) 0(0) 11 {1) 71 {14) 0.0 0.0 0,0 33.3 0.0 38.5 33.3 0.0 25 0.0 100.0 0.0 30.8 0.0 0.0 0.0 25.0 0.0 0.0 0.0 9.1 0.0 9 1 19 7 0(0) 0(0) 0(0) 1 (0) 1 (0) 5(1) 1 (0) (0) 8 (1) 0(0) 0(0) 1 (0) 1 (0) 0(0) 0(0) 0(0) 2{0) 0(0) 0(0) 0(0) 0{0) 0(0) 1 (0) 1 {0) 11 {1) 0.0 0.0 0.0 0.0 0.0 20.0 0.0 0.0 12 5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 9.1 and selectivity of various mesh sizes of the inner net. However, such a test was con- ducted in late October, 1944, in a series of 27 sets made with a 150-yard experi- mental trammel net in several bottomland lakes near Grafton, Illinois. As these lakes were known to have high popula- tions of sport fishes, particularly crappies, they appeared to be especially suited for testing the efficiency and selectivity of va- length. The mesh size of the inner net of the first section, or length, was 11/2 inches, of the second 2 inches, and of the third 3 inches. This experimental net per- mitted the sampling of a site with three mesh sizes in one set. Each 50-yard sec- tion of netting is assumed to have had an equal chance to catch fish and, on the basis of this assumption, the following analysis has been made. 360 Illinois Natural History Survey Bulletin Vol. 26, Art. 4 The catcli of commercial species in the 27 experimental sets was low, since no effort was made to search out such spe- cies. In the 21 sets, 247 fish were caught, tahle 28. The section with the 1 i/o-inch- mesh inner net caught 165 fish, or 66.8 per cent of the number of fish taken ; the section with the 2-inch-iTiesh inner net took 71 fish, or 28.7 per cent of the total; and the section with the 3-inch-mesh in- ner net took only 11 fish, or 4.5 per cent of the total. Of the 165 fish taken in the section with the 1 i/i>-'nch-mesh inner net, 44.8 per cent were sport fishes, 35.8 per cent were garfishes and bowflns, and 13.9 per cent were commercial species. Crappies comprised 82.4 per cent of the number of sport fishes taken in the section with the 1 V2"'n'^h-mesh inner net. Of the remain- ing 17.6 per cent of the sport fishes, six were bass {Micropteriis spp.), four were bluegills, two were yellow bass, and one was a warmouth. Bullheads were the principal commercial fishes taken in the section with the 1 V^-inch-mesh inner net. Seven buffalofishes, one carp, and three carpsuckers formed the remainder of the commercial catch. Commercial species, principally buffalo- fishes and carp, made up 45.1 per cent of the number of fish caught in the section of the experimental trammel net with 2- inch-mesh inner net. In this section of the net, sport fishes comprised 28.2 per cent of the catch ; crappies comprised 95.0 per cent of the number of sport fishes. The section with the 1 V2"'rich-mesh inner net caught 3.7 times as many sport fishes as did the section with the 2-inch-mesh in- ner net. Only 2 garfishes were taken in the section with the 2-inch-mesh inner net and 48 in the section with the IV2" inch-mesh inner net. Of the 1 1 fish caught in the section with the 3-inch-mesh inner net, 72.7 per cent were commercial species. This sec- tion of the net caught but 2 sport fishes. Of the three mesh sizes used in the ex- perimental trammel net, the 2-inch size appeared to be by far the best for com- mercial fishing. The section of the net having an inner net of this mesh size had the highest catch of commercial species, it had few garfishes, and it had a catch of sport fishes that was low when it is con- sidered that the net was fished in waters known to contain a high population of these fishes, table 28. If sport fishes of larger sizes had been abundant in the waters fished, perhaps the catch of these fishes would have been much higher in this section of the net. The section with the 3-inch-mesh inner net would probably have taken a high catch of large commer- cial species if fished in a habitat being used by such fishes. This section of the net took very few sport fishes of the sizes then present in the Grafton waters. In conjunction with the trammel net- ting experiment at Grafton, 12 net-days of fishing were done with 1 -inch-mesh wing nets. Ninety-one bluegills were taken in these wing nets; they ranged in size from 3.8 to 7.6 inches total length. The mean of the total lengths of these bluegills was 5.7 inches. The four blue- gills taken in that section of the experi- mental trammel net having a lV2"inch- mesh inner net ranged from 7.0 to 7.2 inches in length. Evidently most of the bluegills present in the Grafton waters in the fall of 1944 were too small to be taken in the 1 l/2"inch-mesh net. If blue- gills of 7 inches and longer had been ex- tremely abundant, the catch in the ex- perimental trammel net probably would have been higher. The mean of the total lengths of bluegills taken in the Missis- sippi River during the 2-year survey with all types of fishing devices was between 6.1 and 6.5 inches. In the 116 trammel net sets made during the survey, only 20 bluegills were caught, table 26. On the basis of the above discussion, it appears that trammel nets having inner nets with meshes of II/2 inches or larger are quite inefficient in taking bluegills in the Ca- ruthersville-Dubuque section of the Mis- sissippi. In the 12 net-days of fishing with wing nets at Grafton, 285 crappies were taken. These fish ranged in total length from 4.4 to 11.4 inches; the mean of their total lengths was 7.6 inches. Crappies caught in the section of the experimental net hav- ing an inner net of fi/o-inch mesh ranged in total length from 7.4 to 12.2 inches; the mean of their total lengths was 9.3 inches. Crappies taken in the section hav- ing an inner net of 2-inch mesh ranged July, 1955 Starrett & Barnickol: Commercial Fishing Devices 361 ".^ n 11 " inclr • r to al 'engths was 10 8 .iiciies. 1 he two crappies taken in the sec- tion having an inner net of 3-inch mesh were each more than 1 1 inches in total length. The mean of the total lengths of white crappies taken at the various sta- tions in the 2 years of test-netting with various commercial devices was between 7.5 and 8.0 inches, and of black crappies between 7.1 and 8.4 inches. As was the case with the bluegills, most of the crap- pies were too small to be taken in the trammel nets having meshes of 1 1/4 inches or larger. How much the crappie popu- lation of the Caruthersville-Dubuque sec- tion of the Mississippi fluctuates from year to year in abundance of individuals of different size or age groups is not known ; however, the differences in indi- vidual lengths between the 1944 and the 1946 samples from the river were not great (Barnickol & Starrett 1951:317). It was shown by test-netting that, al- though trammel nets will catch sport fishes, if a mesh of 2 inches or larger size is used, the catch of these fishes is nom- inal, even when the nets are fished in waters having a high population of crap- pies and such other common sport fishes as are found in the Caruthersville-Du- buque section of the Mississippi River. Grilling Experiment With Set Trammel Net.—Some critics of the trammel net have claimed that the net not only captures large numbers of sport fishes, but that sport fishes are killed in the net by gilling. Gilling occurs when a fish strikes the inner fine-mesh net and forces its head through the netting far enough to allow the net twine to become lodged beneath one or both gill covers. The gills may be injured while the fish is trying to escape or while a fisherman is dislodging the fish from the netting. Re- lease of a gilled fish requires of the fish- erman more time and patience than does release of a pocketed fish. Pocketed fish are seldom injured in the net. In the trammel net experiment at Graf- ton in October, 1944, a record was kept of the section of the net in which each fish was caught, as well as whether the fish was pocketed or gilled in the net. The number and per cent of fishes that were gilled are listed by species and size •"" '> ..-, .,-h''r^ »h" fish were caught, a..ie 2 5. Gilling was re'atively greater among the buffalofishes than among any of the other commercial species. The high percentage of gilled buffalofishes might be due to the terrific force with which these fishes hit a trammel net. A fisher- man trying to hold buffalofishes alive for a period of time in a crib or holding pond might experience a higher mortality among the gilled fish than among the pocketed fish. Persons who have fished gill nets know that the sunfishes, including the crappies, the bluegill, and the basses {Micropterus spp.), are ordinarily a difficult group to take in large numbers in these nets, whereas the perches and similarly shaped fish are taken readily in them. A low per- centage of gilling was recorded for the crappies in the section of the trammel net having a 1 Vi^-inch-mesh inner net, table 28. Of 61 crappies taken in this net, only 8 were gilled. These gilled fish ranged in total length from 8.2 to 10.2 inches, and the mean of their total lengths was 8.8 inches. The crappies that were pocketed ranged in length from 7.4 to 12.2 inches, and the mean of their total lengths was 9.4 inches. Although the mean of the total lengths for the pocketed crappies was higher than that for gilled fish, some of the pocketed fish were smaller than any of the gilled fish. The percentage of gilled crappies in the section of the experimental net with the 2-inch mesh was higher than in the section with the 1 V2'inch mesh ; however, there were twice as many gilled crappies in the section with the IV^'inch mesh as in the section with the 2-inch mesh. No spoit fish was gilled in that section of the net with the 3-inch mesh. Catches With Floated Trammel Net.—Current and a clean river bottom are requirements for fishing with a tram- mel net that is to be floated or drifted. The floats on the net must be of such buoyancy and the weights of such weight as to hold the narrow axis of the net in a nearly vertical position. The fisherman lays his net in an area of the channel he desires to fish and lets the current carry the net downstream over a clean sand or gravel bottom. In the D-MR section of the Mississippi 362 Illinois Natlrai. History Survhv Bulletin Vol. 26, Art. 4 River. Hshinp; with the trammel net floated is limited mainl\ to the upper ends of the pools where there is enough current to drift the net. The float method of fishing with the trammel net is limited not only in use but in the kinds of fishes it takes. The shovel- nose sturgeon, or hackleback, is the prin- cipal species taken by the float method. It may be noted in table 26 and fig. 17 that this species of sturgeon was not taken in the survey collections made with set tram- mel nets. The trammel net float was the only method by which the survey crew was able to take shovelnose sturgeons in numbers. The only shovelnose sturgeon taken by the survey party with other gear or methods of fishing was a single speci- men in a hoop net at the New Boston sta- tion. However, in the MR-C section of the river, sturgeons are taken on trot lines by commercial fishermen. Trammel net floating is an important method available to commercial fishermen interested in fishing for shovelnose stur- geons. This method of fishing makes it possible to harvest a fish crop that might otherwise be lost to man. The freshwater drum was the only other species taken by the float method in large enough numbers to be of importance to commercial fisher- men. Sport fishes are seldom taken in the Mississippi River by the float method. In 108 floats, only 5 sport fishes were taken, of which all were saugers, table 26. 80 70 60 50 40 2; 30 20 10 H TRAMMEL NET SETS TRAMMEL NET FLOATS oz -lUJ iLlOXE03 Fig. 17.—The relative numbers of commercial, sport, predator, and forage species of fish taken in trammel net sets and in trammel net floats on the Mississippi River between Grafton, Illinois, and Dubuque, Iowa, in 1944 and 1946. The graph is based on the total number of fish taken by each of the two methods of fishing, table 26. July, 1955 Starrett & Barnickoi, : Commercial Fishing Devices 363 Experimental Floats at Quincy.— An experiment with trammel net floats was conducted in the Mississippi River near Quincy, Illinois, in early November of 1944. The experimental net used was 100 yards in length and consisted of two sections, one 50-yard section with a 1V2" inch-mesh inner net and the other 50-yard section with 2-inch-mesh inner net. The two sections were sewed together to form one continuous net. In 33 floats, 197 fish were taken, table 29. The catches from these floats are in- cluded also in table 25. Shovelnose stur- geons comprised 51.3 per cent and fresh- water drums 30.5 per cent of the number of fish in these catches. Other commer- cial fishes amounted to only 6.6 per cent of the number of individuals in these catches. Garfishes amounted to 8.6 per cent. Five saugers were the only sport fishes taken in the 33 experimental floats. These fish were taken in the section of the net with the 1 l/^-inch-mesh inner net. The float method of trammel netting, as dem- onstrated by the 33 experimental floats and the other 75 floats made during the 2-year survey, presents no problem rela- tive to sport fishes. The section of the experimental net with the IV2 -inch-mesh inner net took 3.8 times as many shovelnose sturgeons as the section with the 2-inch-mesh inner net. The sturgeons taken in the lV2"ir'ch mesh were smaller than those taken in the 2- inch mesh. The mean of the fork lengths* of shovelnose sturgeons was 25.5 inches for those taken in the fiA-inch mesh, and 26.8 inches for those taken in the 2-inch mesh. One 8.4-inch shovelnose sturgeon was taken in the 1 i/o-inch mesh ; the re- mainder taken in this mesh ranged from 22.5 to 31.2 inches fork length. In the section of the net with 2-inch mesh, the sturgeons caught ranged in fork length * Fork length is the measurement from the tip of the snout to the base of the caudal filament. Table 29.—Summary of the catches of fish taken in 33 floats with a 100-yard experimental trammel net (50 yards with I'/j -inch-mesh inner net and 50 yards with 2-inch-mesh inner net) in the Mississippi River near Quincy, Illinois, November, 1944. Figures in parentheses indicate the numbers of fish filled. Kind of Fish IJ^-Inch-Mesh Net Number Caught Per Cent Gilled 2-Inch-Mesh Net Number Caught Per Cent Gilled Commercial Flathead catfish Channel catfish Carp Freshwater drum Smallmouth buffalo Shovelnose sturgeon Carpsuckers Blue sucker Subtotal Average Sport Sauger Subtotal Average Predatory Shortnose gar Longnose gar Subtotal Average Forage Goldeye Subtotal Average Total Average 4(0) 1 (0) 0(0) 37(1) 0(0) 80 (3) 2(0) 2(0) 126 {4) 5(0) 5{0) 4(1) 7(1) 11 {2) 1 (0) 1{0) 143 (6) 0.0 0.0 0.0 2.7 0.0 3.8 0.0 0.0 3.2 0.0 25.0 14.3 18.2 0.0 4.2 1 (0) 0(0) 1 (0) 23(1) 1 (0) 21 (1) 0(0) 1 (0) 48 m 0(0) 0{0) 1 (1) 5(0) 6{1) 0(0) 0{0) 54 (3) 0.0 0.0 0.0 4.3 0.0 4.8 0.0 0.0 4 2 0.0 100.0 16 7 0.0 5 6 364 Illinois Natural History Survey Bulletin Vol. 26, Art. 4 from 23.0 to 30.6 inches. Of the stur- geons taken in this section of the net. 76.2 per cent were of commercial sizes (25 inches or more fork length) ; of those sturgeons taken in the section with 1 Vi>- inch mesh, 61.2 per cent were of commer- cial sizes. The numher and percentages of the fishes that were gilled in the experimental floats are listed in table 29. None of the five saugers taken during the experiment was gilled. In the section with the 11/2- inch-mcsh inner net, 4.2 per cent of the fish were gilled ; in the section with the 2-inch-mesh inner net, 5.6 per cent were gilled. Discussion During the past 50 years, there have been increasing numbers of sport fisher- men in Illinois and other states. Many of these fishermen view with suspicion the operations of commercial fishermen. Some even believe that commercial fishermen have been responsible for the decline of sport fish populations in certain localities. Their beliefs probably have developed as a result of hearsay, casual contact with the commercial fish industry, and lack of realization that many environmental changes have taken place in the past cen- tury and that some of these have had an eflFect on the fishery of the Mississippi River and other waters. The effect of commercial fishing devices of illegal mesh size on certain sport fishes was demonstrated by Bennett (1948:411) in his study at Fork Lake, a small artifi- cial lake of 1.38 acres stocked with blue- gills and largemouth bass. After nearly 314 years of being cropped heavily with small-mesh wing nets, Fork Lake still contained a large population of fish. If it was not possible in a 1.38-acre lake to re- duce a sport fish population to a low level by concentrated effort with commercial gear of a mesh smaller than legal size, then it certainly does not appear logical that commercial fishermen, using nets of legal-size mesh, could remove enough sport fishes from the Mississippi River to affect the sport fishery, even if it were lawful to keep the sport fish taken in com- mercial devices. It was demonstrated in the survey of 1944 and 1946 reported here that com- mercial fishing devices can be fished on a selective basis. Of the fishing devices tested in the survey, basket traps and floated trammel nets were found to be the ones most selective for commercial species. Other devices were found to be quite se- lective when fished for a particular species of fish. The type of fishing device, the mesh size, and the fishing site influenced the species composition of the catch. Missouri statutes no longer specify minimum size limits for any commercial fish, except catfish. Illinois and Iowa still have minimum size limits on certain com- mercial species. Analysis of data from the Mississippi River survey of 1944 and 1946 indicates that, where minimum size limits of commercial fish are necessary, sizes of fish caught can be controlled by using nets of certain mesh sizes. The use of mesh size as a means of controlling the size of fish in the commercial catches re- duces the labor of complying w'ith the law by eliminating the necessity for measur- ing the fish in the catches. Practices and policies of fish manage- ment for the Caruthersville-Dubuque sec- tion of the Mississippi River should not eliminate either sport fishing or commer- cial fishing. Prohibitive measures aimed at either kind of fishing affect the best use of the fishery. Too frequently, laws gov- erning this dynamic river fishery tend to be static and thereby defeat their original purpose to benefit the fishery. Laws es- tablished to aid in the management of the river fishery should not be adopted with an attitude of permanency. These laws should be changed whenever scientific findings indicate they are no longer useful. Summary 1. Data on fishing with various types of commercial fishing devices were col- lected during a fish survey of the Missis- sippi River between Caruthersville, Mis- souri, and Dubuque, low^a, in 1944 and 1946. Particular emphasis was placed on determining the selectivity and effective- ness of the commercial fishing devices of various mesh sizes used on the river. 2. During the survey, the following types of commercial fishing devices were used : seines, trammel nets, basket traps, Tulv, 1955 Starrett & Barnickol: Commercial Fishing Devices 365 wing nets, hoop nets, trap nets, and trot lines. 3. Twenty-five seine hauls were made by the survey crew with 1 -inch-mesh seines of 100-, 150-, and 200-yard lengths. Carp, buftalofishes, channel catfish, pad- dlefish, carpsuclcers, and freshwater drum were the principal commercial species taken with seines. Shortnose gar was the principal predatory fish, gizzard shad the principal forage fish, and white crappie the principal sport fish. Sport fishes amounted to 11.7 per cent by number and 3.3 per cent by weight of the total catch with seines. Crappies comprised 82.1 per cent of the number of sport fishes taken with seines. 4. Of the fish taken in the survey with basket traps, 80.8 per cent were catfishes, 15.1 per cent were shortnose gars. No sport fish was taken in the basket traps. 5. One-inch-mesh wing nets with leads were found to be more efficient in taking crappies, bluegills, and buffalofishes than were nets of the same mesh fished without leads. 6. The number of commercial-sized channel catfish caught per net-day was larger in the 1 -inch-mesh wing nets than in wing nets of larger mesh sizes. 7. The number of commercial-sized carp and buffalofishes caught per net-day was larger in the nets of large mesh sizes than in the nets of 1-inch mesh. 8. The efficiency of wing nets in tak- ing sport fishes decreased as mesh size in- creased. 9. Wing nets took more sport fishes per net-day than did hoop nets. 10. The catch of freshwater drums of commercial sizes was larger per net-day in the 2Vi2-inch-mesh hoop nets than in the 214-inch-mesh wing nets. The catch of buffalofishes of commercial sizes was slightly larger in the 2Vo-inch-mesh wing nets than in the 2i/4-inch-mesh hoop nets. The 214-inch-mesh hoop nets were the most effective entrapment devices used for flathead catfish of commercial sizes. 11. Three trammel net sets were made on a selective basis for carp and buffalo- fishes. Trammel nets having an inner net of 11/4-inch mesh took a larger number of sport fishes than did trammel nets hav- ing inner nets with mesh sizes of 2 or 3 inches. 12. Trammel net float fishing was found to be selective for shovelnose stur- geons and freshwater drums. Only five sport fishes were taken by this method of trammel netting. 13. In 27 experimental trammel net sets at Grafton, 15.2 per cent of the fish taken in the section with an inner net of li^-inch mesh were gilled ; 19.7 per cent taken in the section with an inner net of 2-inch mesh, and 9.1 per cent taken in the section with an inner net of 3-inch mesh were gilled. No white crappie was gilled. In the section with the 1 i/o-inch-mesh in- ner net, 15.1 per cent of the black crap- pies were gilled ; in the section with the 2-inch-mesh net, 30.8 per cent of the black crappies were gilled. Of commercial spe- cies taken, buffalofishes had the greatest tendency to become gilled. 14. In 33 experimental trammel net floats made at Quincy, 4.2 per cent of the fish taken in the section of the net with a 1 i/)-inch-mesh inner net were gilled; 5.6 per cent in the section with a 2-inch- raesh inner net were gil'ed. LITERATURE CITED American Fisheries Society 1948. A list of common and scientific names of the better known fishes of the United States and Canada. Am. Fish. Soc. Spec. Pub. 1. 45 pp. Bailey, Reeve M. 1952. [Report of] Committee on Names of Fishes. Am. Fish. Soc. Trans. 81 ( 1951) : 324-6. 1953. [Report of] Committee on Names of Fishes. Am. Fish. Soc. Trans. 82(1952) :326-8. Barnickol, Paul G., and William C. Starrett 1951. Commercial and sport fishes of the Mississippi River between Caruthersville, Missouri, and Dubuque, Iowa. 111. Nat. Hist. Surv. Bui. 25(5) :267-350. 366 Illinois Natural History Survey Bulletin Vol.26, Art. 4 Bennett, George W. 1948. The bass-bluegill combination in a small artificial lake. 111. Nat. Hist. Surv. Bui. 24(3):377-412. Evermann, Barton Warren 1899. Report on investigations by the United States Fish Commission in Mississippi, Lou- isiana, and Texas, in 1897. U. S. Comm. Fish and Fisheries Commr. Rep. for 1898 :2G5- 310. Hubbs. Carl L., and Karl F. Lagler 1947. Fishes of the Great Lakes region. Cranbrook Inst. Sci. Bui. 26. 186 pp. Snedecor, George W. 1946. Statistical methods. (Fourth Ed.) Iowa State College Press, Ames, Iowa. 485 pp. Starrett, William C., and Sam A. Parr 1951. Commercial fisheries of Illinois rivers: A statistical report for 1950. 111. Nat. Hist. Surv. Biol. Notes 25. 35 pp. Thompson, David H. 1925. Some observations on the oxygen requirements of fishes in the Illinois River. 111. Nat. Hist. Surv. Bui. 15(7) :423-37. Upper Mississippi River Conservation Committee 1946. Second progress report of the technical committee for fisheries. 26 pp. Mimeo. 1948. Fifth progress report of the technical committee for fisheries. 23 pp. (Third section of Proceedings of the Fifth Annual Meeting, Upper Mississippi River Conservation Committee. 122 pp. Mimeo.) 1952. Eighth progress report of the technical committee for fisheries. (Pages 10—84 of Pro- ceedings of the Eighth Annual Meeting, Upper Mississippi River Conservation Com- mittee. 143 pp. Mimeo.) ^janss^tmKSJis^sfi':, E RECENT PUBLICATIONS A.—ILLINOIS NATURAL HISTORY SURVEY BULLETIN. Volume 2S, Article 1.—Characteristics of Residual insecticides Toxic to the House Fly. Willis N. Bruce. July, 1949. 32 pp., frontis. + 14 figs., bibliog. Volume 25, Article 2.—Eticct of Permanent Flooding in a River-Bottom Timber Area. By E. Yeager. August, 1949. 34 pp., frontis. -f 21 hgs., bibliog. Volume 25, Article 3.—Canada Geese of the Mississippi Flyway, with ipecial reference to,1| lUinoia flock. By Harold C. Hanson and Robert H. Smith. March, 1950. 144 pp., froi + 82 figs., bibliog. Volume 25, Article 4.—Biology of the White Crappie in Illinois. By Donald F. Hansen. | August, 1951. 56 pp., frontis. -f 13 figs., bibliog. Volume 25, Article 5.—Commercial and Sport Fishes of the Mississippi River Between Ca- ruthersville, Missouri, and Dubuque, Iowa. By Paul G. Barnickol and William C. Starrettj September, 1951. 84 pp., frontis. + 10 figs., bibliog. Volume 25, Article 6.—Tularemia, Weather, and Rabbit Populations. By Ralph E. Yeatter and David H. Thompson. June, 1952. 32 pp., frontis. -f 29 figs., bibliog. Volume 26, Article 1.—The Mayflies, or Ephemeroptera, of Illinois. By B. D. Burks. 1953. 216 pp., frontis. + 395 figs., bibliog. $1.25. Volume 26, Article 2.—Largemouth Bass in Ridge Lake, Coles County, Illinois. By Georgl^ W. Bennett. November, 1954. 60 pp., frontis. -f 15 figs,, bibliog. Volume 26, Article 3.—Natural Availability of Oak Wilt Inocula. By E. A. Curl. June, 195$.j 48 pp., frontis. + 22 figs., bibliog. B.—ILLINOIS NATURAL HISTORY SURVEY CIRCULAR. 32.—Pleasure With Plants. By L. R. Tebon. February, 1952. (Fourth printing, with ttrl visions.) 32 pp., frontis. -f- 9 figs. 38.—Windbreaks for Illinois Farmsteads. By J. E. Davis. February, 1954. (Fifth printiag^l with revisions by L. B. Culver.) 34 pp., frontis. + 27 figs. 39.—How to Collect and Preserve Insects. By H. H. Ross. June, 1953. (Fourth printing, witf^ alterations.) 59 pp., frontis. + 65 figs. 41.—How to Recognize and Control Termites in Illinois. By B. G. Berger. February, 11 (Reprinted without text revision, April, 1950.) 44 pp., frontis. 4* 32 figs. 42.—Bird Dogs in Sport and Conservation. By Ralph E. Yeatter. December, 1948. 64 pp., iProntis. + 40 figs. 43.—Peach Insects of Illinois and Their Control. By Stewart C. Chandler. December, 11 63 pp., frontis. + 39 figs. 44.—The Drug Plants of Illinois. By Leo R- Tehon. July, 1951. 135 pp., frontis. -f 262 figs. 45.—Housing for Wood Ducks. By Frank C. Bellrose. February, 1955. (Second printing, with revisions.) 47 pp., illus., bibliog. C—ILLINOIS NATURAL HISTORY SURVEY MANUAL. 2.—Fieldbook of Illinois Land Snails. By Frank Collins Baker. August, 1939. 166 pp., color frontis. + 170 figs., 8 pis. $1.00. 3.—Fieldbook of Native Illinois Shrubs. By Leo R. Tehon. December, 1942. 307 pp., 4 color pis. + 72 figs., glossary, index. $1.25., Lift of available publications, about 400 titles, mailed on request. Single copies of Illinois Natural History Survey publications for which no price is listed will be furnished free of charge to individuals until the supply becomes low, after which a nominal charge may be made. More than one copy of any free publication may be obtained without cost by educational institutions and official organizations within the State of Illinois; prices to others on quantity orders of these publications will be quoted upon request. Address orders and correspondence to the Chief Illinois Natural History Survey Natural Resources Building, Urbana, Illinois Payment in the form of money order or check made out to State Treasurer of Illinois, Springfield, Illinois, must accompany requests for those publications on which a price is set.