A TECHNIQUE FOR TAGGING 

 DEEPWATER FISH 



Mark-recapture data have been used extensively in 

 fishery science to estimate population size, survival/ 

 mortality rates, growth rates, and movement 

 parameters. Many devices and methods have been 

 used to tag fish (reviewed extensively by Laird and 

 Stott 1978). Virtually all tagging methods necessi- 

 tate bringing the fish to the surface for marking. For 

 fishes with physoclistic swim bladders inhabiting 

 deeper waters, raising them to the surface subjects 

 them to rapid changes in hydrostatic pressure and, 

 usually, temperature. Procedures used to obtain 

 healthy fish for marking include venting of excess 

 gases from the swim bladder and body cavity with a 

 hypodermic syringe (Gotshall 1964) and raising the 

 fish gradually to the surface to allow acclimatization 

 to changing pressure. Additionally, Phillips (1968) 

 attempted to mark California rockfish by using 

 detachable hooks with "Peterson type" plastic discs 

 fastened to the hooks with wire. However, these 

 methods are at best only moderately successful, as 

 well as time consuming, often expensive, and simply 

 impractical in some situations. 



In 1978 we began an investigation of the life history 

 and population dynamics of tilefish, Lopholatilus 

 chamaeleonticeps, in the Mid-Atlantic Bight. 

 Reasonable interpretation of these data requires 

 knowledge of tilefish movements. Because tilefish 

 are caught on longlines from depths of 73-254 m 

 along the outer continental shelf ( Grimes et al. 1 980) , 

 fishing operations usually kill or severely injure the 

 fish, thus making conventional marking at the surface 

 pointless. This note describes the design and evalua- 

 tion of a technique we developed for tagging tilefish, 

 and potentially other deepwater fishes, with tags 

 designed to detach from a bottom longline, thus elim- 

 inating the problems of pressure and temperature 

 changes caused by raising fish to the surface. 



Methods 



We intended to design a tag that could be lightly at- 

 tached to a longline, so that when a fish took a baited 

 tag the hook would become lodged in the jaw or lip, 

 detach, and thus mark the fish. We designed and con- 

 structed tags similar to the snoods or branch lines 

 used on commercial longline fishing gear (see Free- 

 man and Turner 1977 for a description of the gear). 

 These tags consisted of a 30 cm length of 23 kg test 

 monofilament line inserted through red vinyl tubing. 

 We crimped an 8/0 hook to one end of the tag, and the 

 other end was looped, crimped, and attached to the 



longline groundline at 4 m intervals. No addresses or 

 serial numbers were printed on the red vinyl tubing in 

 these preliminary experiments because our only pur- 

 pose was to determine if this tagging method was 

 functional. No reward was offered, but tags were 

 returned because we personally alerted most fisher- 

 men. This was possible because of the small size of 

 the fishery (i.e., about 25 vessels with most operators 

 already cooperating with our research program by 

 maintaining catch and effort logs) and also because of 

 the localized nature of the tilefish ports (i.e., only two 

 ports landed significant numbers offish). Because we 

 intended to evaluate only the tagging procedure, we 

 did not request biological data on tagged fish that 

 were caught. 



To determine the optimal tag design, we tested dif- 

 ferent hook types (straight and circle) and different 

 strengths of monofilament (0.9, 1.8, 2.7 kg test) for 

 attaching tags to the longline. The vinyl portion of 

 each tag was knotted to indicate the strength of mono- 

 filament (i.e., no knot for 0.9 kg, one knot for 1.8 kg, 

 and two knots for 2.7 kg test). 



We attached tags to longlines in two different se- 

 quences or "series" (one and ten) of attachment 

 strength and hook type. To prevent a patchy distribu- 

 tion of tilefish from biasing the frequency of removal 

 of tags of various hook types and attachment 

 strengths, the "one-series" tagging consisted of one 

 tag with a particular sequence of attachment strength 

 and hook type (e.g., one 0.9, one 1.8, and one 2.7 kg 

 monofilament with straight hooks; one 0.9, one 1.8, 

 and one 2.7 kg monofilament with circle hooks, etc.). 

 To make identification of hook type and attachment 

 strength easier when we observed tagging longlines 

 from a research submersible, the "ten-series" consist- 

 ed of 10 tags with a particular sequence of attach- 

 ment strength and hook type (e.g., ten 0.9, ten 1.8, 

 and ten 2.7 kg monofilaments with straight hooks; ten 

 0.9, ten 1.8, and ten 2.7 kg monofilaments with circle 

 hooks, etc.). 



Longlines fitted with detachable tags were coiled in 

 galvanized tubs, transported to the fishing grounds, 

 and set voluntarily on two occasions by cooperating 

 commercial fishermen. On one tagging operation (at 

 east Hudson Canyon, 23 August 1979, lat. 

 39°38'05"N, long. 72°16'35"W, 117 m) conducted 

 simultaneously with a gear evaluation study (Grimes 

 et al. 1982a), an onboard observer recorded the num- 

 bers of tags of various attachment strengths and hook 

 types remaining on the longline after retrieval. On the 

 other tagging operation (west Hudson Canyon, 17 

 September 1979, lat. 39°20'30"N, long. 72°26'30"W, 

 137 m), the longlines were set and retrieved by com- 

 mercial fishermen who returned the gear for us to 



FISHERY BULLETIN: VOL. 81, NO. 3, 1983. 



663 



