Acceleration rates, analyses of body movements, and pertinent anatomical features are given for various members of the Scombridae, Istiophoridae, and Xiphiidae and are discussed in light of the modern work on smaller freshwater teleosts and marine cetaceans.Acceleration values up to 43.1 m/sec^are given for Thunnus albacares and Acanthocybium solanderi. Nearly 100% of the propulsive thrust comes from the caudal fin. There are two main intervertebral joints which are concerned with locomotion, a pre-peduncular and a post-peduncular joint with the body and peduncle held almost stiff, a condition similar to that described for marine cetaceans. The caudal fin achieves its highest transverse speed near or past the axis of progression and the fin's maximum angle of attack occurs before the fin crosses the axis of progression. The caudal fin's angle of attack has an average value of 32.4°as the fin crosses the axis of progression and the average value for all observed fin positions is 29°. Values as high as 100°were attained.The aspect ratio (spanV surface area) of the caudal fin ranges from 10.26 (Istiophorus greyi) to 4.19 (Xiphias gladius). The higher values approximate those of efficient, high speed airfoils.The positions of the joints and the rigidity of the vertebral column are discussed for the Scombridae, Istiophoridae, Xiphiidae, and Lepidocybium fiavo-bruntieurn. It is suggested that the retention of autonomous epurals, hypurals, and uroneurals functions to allow camber in the otherwise fused caudal skeleton. The bony peduncular keel is shown to act as a pulley for the great lateral tendons as well as to somewhat stiffen the peduncle and to serve as a support for the fleshy peduncular keels. Data concerning the shape, length, origin, insertion, and action of the 2-shaped myomeres are given for the various fishes. Attention is focused on the relative role played by the red and white muscle, the tendons, and the intrinsic caudal musculature in locomotion.A subordinal classification is presented for the scombroid fishes based upon some of the data presented in this study. I MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 6

Seven species of the family Istiophoridae and Xiphias gladiu6 were identified using only features of their rostrum. In the Istiophoridae, two rostral regions were emphasized, one-fourth and one·half the distance between the distal tip and the orbital margin of the lateral ethmoid bone. Characters studied in each region were the depth and width of rostrum and height, width, and position of nutrient canals (as seen in cross-section). Characters studied without reference to region were the distribution of denticles on both dorsal and ventral surfaces of the bilI and position of the prenasal bone. In the Xiphiidae, the only characters studied were the depth and width of the rostrum at the level of the dermethmoid bone and the presence and placement of central chambers as seen in radiographs. A total of 32 characters were analyzed as ratios using both multivariate and uni variate statistics. The rostrum of X. gladiw was separated from the Istiophoridae by its flat shape, Tetrapturus angwtirostris from all other istiophorids by its widely separated nutrient canals, and the complex of T. audax/T. pfluegeri/ Makaira nigrican6jM. indica from the complex of lstiophorus platypteru6jT. al bidw by having a smaller area of denticles on the dorsal surface. Tetrapturu6 pfluegeri was separated from T. audux, M. nigricans, and M. indica by having a longer denticle-free midline on the ventral surface of the rostrum. Tetrapturus audux was separated from M. nigricans and M. indica by the location of its nutrient canals. The complexes of Makaira nigrican6/M. indica and 1. platypte. ru6/T. albidu6 were each separated using multivariate discriminant analysis. We show the study has application in identifying rostral fragments found as fossils and impaled in animate and inanimate objects such as marine turtles and wooden ships and should have application wherever rostral fragments are found.

' have questioned the validity of certain fish swimming speed estimates because these require that either completely la.mina.r dow or very high power factors must be postulated in order to explain the Anjmal s ' performance. Speeds in excess of 45 km./h are included. in this suspect category. Walters' showed. that certain morphological features indicate that soombroid fishes may attain velocities of 10 body-lengths/sec, and the large species should be able to attain maximum speeds conaiderably in excess, of 45 kmJh.~~ .An opportunity of measuring scombroid swimming speeds arose when the junior author was invited to partici pate in a tunabehaviour research cruise offthe Pacific coast of Costa Rica. The measuring equipment consisted of a spinning rod and a spinning reel containing fishing line marked 'with iron powder. A magnetic pick-up was mounted near the tip of the rod, 80 that when a hooked fish pulled line from the reel, the magnetio field was disturbed by the iron powder markings. The resulting electrical signal was fed into a tape·recorder. The magnetic pick·up consisted of a monaural variable reluotance phonograph cartridge (General Electric Model 4G-050) modified by removal of the stylus assembly and mounted in a 1·6 om x 1·6 em x' 1.6 em aluminium box (Fig. 1). The line passage was a 1·6 em length of 2 nun inside diam. plastic tube into which the magnetic poles projected. The entire box. except for the line p..

No abstract

SynopsisBillfishes have long been known to impale a great variety of objects, but there are only two brief, obscure records of marine turtles being speared. Details are presented on these two, as well as on two other confirmed records; data from two additional unconfirmed records are also presented. In total, three species of marine turtles are known to have been impaled by three species of billfishes; a fourth species of fish and a fourth species turtle are listed in an unconfirmed case. Records come from the eastern and western Pacific as well as the eastern Atlantic. Of the four confirmed cases, the turtles survived in two, and apparently died as an effect of the spearing in the other two. In three confirmed cases only the impaled rostrum was encountered, and in one confirmed case the entire fish was found, with its rostrum piercing the turtle. There is no obvious advantage -or clear disadvantage -involved in impaling turtles. It is argued that these attacks are accidental, and the result of attempts made by the billfish to capture prey that are near the turtle. These spearings indicate that the chelonians serve as shelters for prey animals on the high seas, and thus, are further evidence of the pelagic existence of marine turtles. The impalings are evidence of a singular ecological role of the turtles -as live fish aggregation devices.