326 



Fishery Bulletin 89(2). 1991 



was given a gentle tug. If the tag ap- 

 peared loose, it was removed and the 

 animal was discarded. 



Recaptured animals were obtained from 

 commercial fishermen and stored in a 

 freezer. Animals were then thawed and 

 dissected in order to determine the loca- 

 tion within the animals of the tag anchor. 

 The carapace was cut diagonally on either 

 side of the protruding tag by inserting the 

 lower blade of a pair of scissors into the 

 epimeral line. The forward portion of the 

 triangular piece of cut carapace was then 

 lifted to uncover the end of the tag. Tags 

 were classified as being in one of four 

 positions (Table 1): (1) anchor lodged in 

 the dorsal musculature; (2) anchor loose 

 in the body cavity, e.g, nestled against 

 the hepatopancreas; (3) anchor lodged in 

 the soft, newly forming carapace under 

 the hard carapace, or attached to the 

 underside of the hard carapace; and (4) 

 anchor lodged in the basal leg muscula- 

 ture. The extent of any necrotic tissue 

 around the anchor was also noted (Table 

 2). 



Analysis of recapture data 



The logic of our analysis is as follows. We 

 don't know the proportion of animals 

 with the tag anchor placed in each of the 

 four locations, and we don't know the 

 magnitude of any initial tag loss or mor- 

 tality immediately (within one month) 

 following release of the tagged animals. 

 However, those recaptured animals at 

 liberty for more than one month which 

 did not molt provide an estimate of the 

 proportions of animals with tag anchors 

 in each location prior to molting. Similar- 

 ly, the recaptured animals which did molt 

 provide an estimate of the proportions 

 after molting. Consequently, differences 

 in the proportions reflect different tag 

 retention and/or survival. 



This analysis requires two assumptions: (1) tags in 

 all of the tagging locations affect the timing of molting 

 in the same way (if at all); (2) differential mortality and 

 tag loss among tagging locations is zero both prior to 

 and subsequent to molting (with the possible exception 

 of the period immediately after tagging and release of 

 animals). A number of studies have shown that molt- 

 ing can be a critical period for tag-induced mortality 

 and tag shedding (Restrepo and Hoenig 1988, Hurley 



et al. 1990). Hence, it is reasonable to presume that 

 if animals survive the first month at liberty with their 

 tag intact, then they are likely to remain viable until 

 the time of molting. 



Consider the recapture data for 1984 in Table 1. The 

 relative abundance of animals with tag anchors loose 

 in the body cavity and in the dorsal musculature, among 

 animals which did not molt, is 14:29 or 0.48:1. Among 

 animals which molted, the relative abundance is 1 : 16 



