Wade: Estimates of incidental kill of dolphins by the purse-seine tuna fishery 



351 



areas corresponding to the western/southern area, 

 MPS rates were as much as 100% greater than those 

 in the area corresponding to the northeastern area 

 during 1973-76 and 1977-78; significant areal dif- 

 ferences were found in 1977-78 (Wahlen, 1986). If 

 similar differences in MPS by area existed during 

 1959-72, my estimates of mortality for western/ 

 southern spotted dolphins are negatively biased. 



For the same reason, my mortality estimates for 

 whitebelly spinner dolphins may also have a nega- 

 tive bias. Observations of MPS of whitebelly spinner 

 dolphins came only from the region of overlap with 

 eastern spinner dolphins; no observations of MPS 

 were from the outside region west of 120°W where 

 whitebelly spinner dolphins are known to occur and 

 where significant fishing effort occurred in 1970, 

 1971, and 1972 (Punsley, 1983). 



The three trips observed prior to 1971 were not 

 part of an established data collection program and, 

 therefore, may have been biased observations. In 

 1965 and 1968, two trips with tuna boats were ob- 

 served by scientists, who collected dolphin specimens 

 and also recorded MPS data (Smith and Lo, 1983). 

 These data were not based on random samples, but 

 there is no obvious reason why the data from tuna 

 vessels on which scientists were allowed to collect 

 specimens would tend to have different mortality 

 rates. However, it is not certain a priori in which di- 

 rection bias would have occurred. Data from the third 



Table 3 



Average numbers of dolphins killed (M) in purse seine sets 

 in the eastern tropical Pacific by year for the 20 observed 

 trips between 1964 and 1972, for small (<600 tons carry- 

 ing capacity) and large (>600 tons carrying capacity) ves- 

 sels making successful (>l/4 ton yellowfin tuna) and un- 

 successful (<l/4 tonyellowfin tuna) sets on dolphin, pooled 

 over whether the backdown dolphin release procedure was 

 used or not. The number of observed sets (n) and the num- 

 ber of trips (n lr ) are given. Numbers of successful sets here 

 are greater than yearly totals calculated from Table 2 be- 

 cause that table excludes sets for which the use or not of 

 backdown was not recorded. 



Vessels and year n tl 



Successful 

 sets 



M n 



Unsuccessful 

 sets 



M n 



Small vessels 



1964 

 1965 

 1968 

 1971 

 1972 

 Total 



Large vessels 



1971 

 1972 

 Total 



1 

 1 

 1 

 3 

 6 

 12 



60.9 

 25.9 

 130.2 

 116.6 

 56.5 

 62.4 



41.1 

 36.9 



37.4 



20 

 35 

 13 

 19 

 103 

 190 



16 

 117 

 133 



60.0 

 2.6 

 4.0 



12.7 

 3.7 

 5.9 



0.0 

 0.4 

 0.4 



1 



11 



2 



3 



16 

 33 







12 

 12 



