MFR PAPER 983 stored in a raw box or are carried di­ rec tl y to a steam cooker. Cooked menhaden enter a screw press which produce a liquor and press cake. T he liquor i separated into o il and stickwa­ ter by centrifuging or ettling, and the Menhaden Tagging and Recovery: tickwater is evaporated to condensed soluble. The pre s cake is conveyed Part II-Recovery of Internal Ferromagnetic through a rotating dryer to a scrap shed Tags Used to Mark Menhaden, for storage. Later, scrap i either ship­ ped in bulk or ground into meal before Genus Brevoortia shipping.

MAGNET INSTALLATIONS All companies in tall magnets to re­ R.O. PARKER, JR. duce meta llic contaminant in the prod­ uct and prevent damage to equipment. Most companies have two-pole plate magnets in chutes conveying materia l to grinders, and some pl ants have drum ABSTRACT magnet turning the termina l end of Plate and rotating grate lIIagnets in stalled ill \'(/riolls locations in redllctioll conveyor belt. Electromagnet a re plants effectil'ely recOl'er illternaljerrolllagnetic tags IIsed to IIwd menhaden . Tag' al 0 used in some pl ants.

recOl'ery efficiency rates for large (a dlllt) tags range /rom 14 to 97 percellt lI'ith (t In addition to company-owned mag­ mean of64 percellt alldfor slIIall UIII'ell ile) tags range /rOIllS to 86 percellt lI 'ith a nets, we in sta lled a variety of magnets at mean of 39 percellt. Magnet illstallations do not intel/ere lI 'ith the redllctioll various locations in the plants to in­ operatio/1, bllt facilitate it by relllo l'ing scrap metal. crease tag recovery efficiency a nd minimize the delay between processing and recovery. In drag line and chutes we in stalled two-pole and four-pole INTRODUCTION the low efficie ncy and hi gh cost of de­ plate magnets (Figure I). The pulling tector syste ms limit their u e. powerofthese magnets (304.8 mm long) Since Rounsefell and Dahlgren (1933) This pa per de cribes the recovery of demonstrated the use of magnets for re­ at a di stance of 50.8 mm for a standard internal ferromagnetic tags from men­ test piece of76.2 x 25.4 x 3.2 mm cold­ covering metallic tags in herring. many haden pl ants by magnets. Two sizes of scientists have successfully used this rolled steel was 141.8 g and 106.3 g. tags are recovered: a large tag, 14.0 x More recent installations of plate mag­ method of recovering tags from indus­ 3.0 x 0.5 mm, u ed in ad ult menhaden tnal fish, i.e., species converted to nets have been of the two-pole type with and a small tag, 7.0 x 2.5 x 0.4 mm, meal, oil, and condensed solubles (Fry, a pulling power of 5 10.3 g. In many used in juvenile menhaden. 1937; Fridricksson and Aasen, 1950; plants where scrap drops from con­ veyors, we in ta il ed rotating grate Stevenson, 1955; Aasen, A ndersen, PROCESS ING PROCEDU RE Gulland, Madsen, and Sahrhage, 1961; magnets (Figure 2). These grates are Klima and Bayliff. 1961). In addition to At least 96 percent of the menhaden cylindrical with several magnetic bar, magnets, electronic detectors have landed annually are proces ed by reduc­ 25.4 mm in diameter, spaced 76.2 mm been used to detect tagged fi sh among tion pl ants (Reintjes, 1969). About two apart. The, cylinders are rotated with thousands of fish landed (Da hlgren, dozen plants are active each year a long electric motors at 12 rpm. We aban­ 1936; Stevenson, 1955; Parker, 1972). the Atlantic and Gulf coasts of the doned stationary grate magnets after in­ These recoveries provid e the same in­ United States and convert mi ll ions of itial tests because they often became formation as magnet recoveries, such as menhaden daily to meal, oil, and con­ clogged. Magnet install atio ns do not in- migration pattern and mortalit y rates, densed solubles. and in addition permit the determination As the catch is un loaded, it is weighed R .O . Pa rker , Jr. is a Fisher y of: growth rates, validity of aging tech­ automatica ll y before being transferred Biologist on the staff of the NMFS niques, best tag injection location, and to the plant by a drag line o r screw con­ Atla ntic Estua rine F is heries time for tag incisions to heal. However, veyor. In the plant, fish are temporaril y Center, Beaufort, NC 28516.

36 Figure 1.-Two·pole and four·pole plate mag· 3). Others are covered with a metal nets in a conveyor chute. plate quipped with a handle, permitting all material to be lifted at once (Figure 4). Metal and fish scrap are scraped into buckets and the magnet number (iden­ tifying the plant and magnet location) and date are recorded. The material i spread on the floor and a magnetic sweeper is passed over it several time to concentrate the metal (Figure 5). The metal is searched by hand for tags (Fig­ ure 6). If there is a large quantity of scrap metal, the bulk of it is separated from the tags with sieves (Figure 7). Recovered tags are sealed in small en­ velopes identified by magnet number and date. Later, the tags are taped to forms on which all recovery information is transferred. The forms are checked and sent to computer processing. Ten of thousands of tags are recovered an­ nually (Coston, 1971). EFFICIENCY TESTS Tests are conducted weekly through­ out the fishing season to monitor mag­ net and plant tag recovery efficiencies under various conditions. From the e tests we evaluate magnet installations and estimate the number of field tags entering the plants. Magnet installation efficiency is determined by scattering 100 tags in the scrapjust before it pa ses terfere with the reduction operation, but facilitate it by removing crap metal. All magnet installations have been Figure 2.-A rotating grate mag· net removes metal from scrap as ;=..-o-.====o:::;..:~,-",=~'="-.-,-""",; __-=",--_"""", de ignated as primary or secondary lo­ the scrap drops from a conveyor. cation ba ed on the length of delay be­ tween the time of landings and passage of fish scrap over the magnets. Primary installations, between the dryer and scrap shed , search scrap the day fish are landed, except during heavy landing when fish must be stored for a day or two in raw boxes. Secondary in stall a­ tions, usually in the scrap shed just ahead of the grinders, search scrap at irregular intervals, sometimes months after fis h are landed.

TAG RECOVERY Magnets are usually cleaned daily. To faci litate cleaning, some magnets are attached with hinges and a latch (Figure

37 Figure 3 (above).-Hlnged magnets facilitate cleaning.

Figure 4 (top rlght).-A metallic magnet cover with a handle permits all material to be lifted at once.

Figure 5 (below).-Metal and fish scrap are separated with a magnetic sweeper.

Figure 6 (center rlght).-Metalls searched by hand for tags.

Figure 7 (boftom rlght).-Sieves help separate scrap metal and tags.

38 over the magne ts. The pe rcent of te t covery efficienc ie s range fro m 14 to 97 S Dep. (ommer. "0 ...... "all \Iar f I h tags recovered o n the magne ts i a en . Dala Kep. M. II ~ p. pe rcent with a mean of 64 percent. 0 Dahlgren. E H 19311. runher Jelelopmcnh on the mea ure of the e ffi c ie ncy of the in sta ll a­ tests of magne ti c in sta ll ation e ffi c ie n­ taggong oflhe Pacific herring. Clul''''' 1',,11,/11/ 1 Con.,. 1 I ~~9-~47 ti o n. T ag recovery e ffi c ie ncy at a pl ant c ies were conducted wit h s ma ll tags Fridrii-."on. . anJ 0 \ .. 'en. 19~() I he i dete rmined by distributing 100 tagged since these tags we re not de ve loped un­ Nor\\eglan- lcelanJ,c herring lagging C\PCrI men I,>. Repon o. I Rep. om. h,h \I" fi sh in the ho ld of a fi shing vessel, in the til ollr third year of tagging. Seasona l I nve,>l. 9( II ):41 p. conve yor syste m be tween dock a nd recovery efficie nc ies for s ma ll tags Fry. D. H .. J r. 1937. \ Iagnetlc reco'er) of Ihh tag. CalI f. Fhh. Game ~3:119-1~4. pl ant, o r in raw boxes. Because of the ra nge from 5 to 86 pe rcent with a mean Klima. E.F .. and w H. Ba~lilT 1961 Taggong "I delay between prima ry a nd second ary a nchoveta" (Celell~rtlulil 1II1'.lIIn'IUI/ on Gulf or of 39 perce nt. In some pla nts la rge Panama. P roc Gulf Cirobb Fi,h In" in sta ll ati o n recoveries, weekl y pl ant tag numbers of tags a re lost in cooke rs a nd 13: 151-156. Parker. R.O .. J r. 19n An electronIc Jeteclor } ,. recovery e ffi c ie ncy is determined from dryers in additio n to losses thro ugh lem for recovering Intcrnall~ lagged menhaJen. test tags recovered on primary in ta ll a­ magnet ine ffi c ie ncy. genu,8rel'oorl;(/. .S. Dep. Commer .• ~() \ \ Tech. Rep. 1 F S RF -654. 7 p. tio ns o nl y. Seasona l tag recovery ef­ Reontjes. J .W . 1969 ynop'>I'> of biological Jata on fic ie nc ies a re based on both primary a nd LITERATURE CITED the Atlantic menhaden. Brt'\'oort;a /vrallllll\ U. Fish Wild!. erv.. ire 3~0. ,(j p. ( \1,,, econdary recoveries. T est tag a nd fie ld FAO Fish. Synop. 0.42). A asen, 0 .. K . P . A ndersen. J . G ulla nd. K .P. Rounsefe ll . G .A .. and E. H Dahlgren. 19,1. l ag­ recoveri es a re recorded a li ke. Madsen. and D . Sahrhage. 1961. ICES herring ging e\periment, on the PacIfic herring. CIUI'''(/ tagging expe riments in 1957 and 1958. Rapp. pallas;i. J . Con,. 8:371-384. Magne t in sta ll atio n effic ie nc ies fo r P.-V . Re un . Cons. Perm. I nl. Ex plo r. Mer. 152. Stevenson. J . . 1955 . T he movement of hernng in large tags range fro m 18 to 100 pe rcent 50 p. Brilish o lumbm waler, a, determIned b~ lag­ Costo n. L.C. 197 1. Summary of tag, released and ging. Rapp. P.-V. Reun . on, Perm Inl F- \­ with a mean of 7 1 percent; seaso na l re- recovered from Atla ntic menhaden. 1966-69. plor. Mer. 1 40(~) :33-3 4 .

MFR Paper 983. The paper above is from Marine Fisheries Re­ view, Vol. 35, Nos. 5-6. Copies of this reprint, in limited numbers, are available from 083, Te chnical Information Division, Environ­ mental Science Information Center, NOAA. Washington, DC 20235.

The National Ma rine F isheries Service (N~lFS) does not approve, rec­ ommend or endorse a ny proprietary product or proprietary material mentioned in this publication. No r eference shall be made to N:\-lFS, or to t his publication furnished by NMF S, in any advertising or sales pro­ motion which would indicate or imply that Nl\1FS approves, recommends or endor ses any proprieta r y pr oduct or proprietar y material mentioned herein, or which has as its pur pose a n intent to cause directly or indirectly the adver tised product to be used or purchased because of this N~lFS publication.

39