Abstract. The spotted dol- 

 phin (Stenella attenuata) is found 

 throughout much of the eastern 

 tropical Pacific Ocean. A previous 

 study evaluated morphological 

 variation in skull morphology, but 

 now specimens are available for a 

 greater portion of the range. Also, 

 corrections have been made in 

 data and an assessment has been 

 made evaluating repeatability of 

 character measurements. We reas- 

 sessed geographic variation in 30 

 cranial features (26 morphometric 

 measures and 4 tooth counts) 

 based on 611 museum specimens. 

 All characters except two tooth 

 counts showed statistically signifi- 

 cant geographic variation, while 

 21 of the 30 characters exhibited 

 significant sexual dimorphism. 

 Males were larger in most charac- 

 ters; females were larger in some 

 length measurements involving 

 the rostrum and ramus. As in pre- 

 vious analyses, inshore S. attenu- 

 ata were found to be very distinc- 

 tive, so subsequent analyses fo- 

 cused on offshore spotted dolphins 

 from 29 5° latitude-longitude 

 blocks. Mantel tests and matrix 

 correlations for 19 of the 30 fea- 

 tures demonstrated significant "re- 

 gional patterning," whereas 22 of 

 the characters were shown to have 

 "local patterning." Principal-com- 

 ponents, canonical-variates, and 

 cluster (UPGMA and function- 

 point) analyses also were em- 

 ployed to assess geographic varia- 

 tion. In the eastern portion of the 

 range, the subdivision between 

 northern and southern offshore S. 

 attenuata found in the previous 

 investigation was confirmed. In 

 general, blocks to the west (includ- 

 ing one encompassing part of the 

 Hawaiian Islands) were more like 

 the southern blocks than those of 

 the northeast. Morphological pat- 

 terns were similar to those found 

 in a number of environmental vari- 

 ables, particularly water depth, so- 

 lar insolation (January), sea sur- 

 face temperature (January and 

 July), surface salinity, and thermo- 

 cline depth (winter and summer). 

 Present management units are 

 inconsistent with the pattern of 

 cranial variation; spotted dolphins 

 from west of lat. 120°W probably 

 should not be pooled with those to 

 the east, as they show closer affin- 

 ity with the Southern Offshore 

 unit. In addition, the boundary 

 between the Northern and South- 

 ern units should probably be moved 

 north to about lat. 5°N. 



Manuscript accepted 12 October 1993 

 Fishery Bulletin 92:324-346 (1994) 



Reexamination of geographic 

 variation in cranial morphology 

 of the pantropical spotted dolphin, 

 Stenella attenuata, in the 

 eastern Pacific 



William F. Perrin 



Southwest Fisheries Science Center. National Marine Fisheries Service, NOAA 

 PO Box 27 I . La Jolla, CA 92038 



Gary D. Schnell 

 Daniel J. Hough 



Oklahoma Biological Survey and Department of Zoology. University of Oklahoma 

 Norman, OK 73019 



James W. Gilpatrick Jr. 

 Jerry V Kashiwada 



Southwest Fisheries Science Center, National Marine Fisheries Service, NO/V\ 

 PO Box 27 I , La Jolla, CA 92038 



Spinner and spotted dolphins (Sten- 

 ella longirostris and S. attenuata) 

 have broadly overlapping ranges in 

 the eastern tropical Pacific Ocean 

 (Perrin et al., 1983). Information 

 concerning geographic variation of 

 these species is of both intrinsic 

 scientific and practical interest. 

 Dolphins in the region are killed as 

 a result of purse-seining for yellow- 

 fin tuna (Thunnus albacares; Allen, 

 1985). The tuna often are found in 

 association with these two dolphin 

 species (or with Delphinus delphis), 

 and fishermen set nets on the dol- 

 phin schools to capture tuna found 

 below the dolphins. Estimates indi- 

 cate that from 1985 to 1990 

 roughly 53,000 to 129,000 dolphins 

 were killed annually as a result of 

 fishing operations (Hall and Boyer 

 1987, 1988, 1989, 1990, 1991, 

 1992). Most recently the annual 

 kill has dropped to approximately 

 15,000 to 27,000 (Hall and Len- 

 nert 1 ' 2 ). Government regulations in 

 the United States set limitations 

 on U.S. vessels with respect to the 



extent of dolphin mortality that 

 will be permitted. Dolphins are 

 managed by defining a series of 

 management stocks. Data, such as 

 those on skull morphology, can pro- 

 vide insight into the underlying 

 population subdivision and may be 

 of considerable value in defining 

 geographic boundaries of biolog- 

 ically relevant management stocks 

 (Dizon et al., 1992). 



Douglas et al. (1992) have pro- 

 vided a detailed assessment of geo- 

 graphic variation in cranial mor- 

 phology of spinner dolphins. For 

 spotted dolphins, the most recent 

 geographic-variation analyses us- 

 ing skull characteristics were by 

 Douglas et al. (1984) and Schnell et 



Hall, M. A., and C. Lennert. 1992. Esti- 

 mates of incidental mortality of dolphins 

 in the purse-seine fishery for tunas in the 

 eastern Pacific Ocean in 1991. Int. What 

 Commn. meeting doc. SC/44/SM6, 5 p. 

 Hall, M. A., and C. Lennert. 1993 Inciden- 

 tal mortality of dolphins in the eastern 

 Pacific Ocean tuna fishery in 1992. Int. 

 What Commn. meeting doc. SC/45/SM1, 

 5 p. 



324 



