772 



Fishery Bulletin 100(4) 



carrying these tags were recaptured at the same location 

 in the same purse-seine set. The respective differences 

 between estimates and actual values were 0.20 and 0.30 

 for latitude and 0.23 and 0.06 for longitude. We also noted 

 that the differences between estimated and actual lati- 

 tudes and longitudes were 1.95 and 0.29, respectively, for 

 the fish with tag 99-812, and 2.67 and 0.28, respectively 

 for the fish with tag 99-865 (Table 3). These two fish car- 

 rying these tags were recaptured by longline vessels, and 

 scientific observers were not aboard to verify recapture 

 positions. 



Aside from the large differences between the estimates 

 and actual recapture latitudes in Table 3 for tag numbers 

 99-801, 99-847, and 99-891, the majority of the estimates 

 were within 2 degrees of the actual latitudes. Tags 99-792, 

 99-835, 99-861, and 99-877 were the only tags that provid- 

 ed differences more than 1 degree between the estimates 

 and actual recapture longitudes. 



Fish with tags 99-792, 99-793, 99-804, and 99-826 were 

 at liberty during the autumn equinox. The geolocation 

 estimates for longitude were unaffected by this event, but 

 the estimates for latitude were unreliable for a few weeks 



