314 



Fishery Bulletin 98(2) 



production of empty and blastodisc egg cases preceded 

 that of nutritive egg cases in the sandtiger shark and 

 that each oviducal gland produced one egg case per 

 day. We observed that blastodisc egg cases preceded 

 the nutritive egg cases in the shortfin mako (no. 2 in 

 Table 2). Therefore, the 2.7-3.3 cm TL embryos would 

 be about 40-50 days old. They were at a very early 

 stage of development because the yolk sac was almost 

 intact, and oophagy may not begin until the embryos 

 are about 3^ months old and 5-6 cm long. 



The gestation period of the shortfin mako is about 

 twice as long as that of the sandtiger shark (9 

 months; Cliff, unpubl. data), which has a term litter 

 mass (13 kg), about half that of the shortfin mako 

 (25-40 kg). Scant ovary data for the shortfin mako 

 and extensive ovary data for the sandtiger shark 

 (Gilmore et al., 1983; Cliff, unpubl. data) suggested 

 that the masses of ovulating ovaries are similar in 

 the two species. The shortfin mako would therefore 

 require about twice as long to nourish a litter double 

 the weight of that of the sandtiger shark. 



Our interpretation of reported great white shark 

 data suggested that the gestation is longer than one 

 year and probably similar to that of the shortfin 

 mako. Parturition is in late spring (Klimley, 1985; 

 Francis, 1996; Uchida et al., 1996). Uchida et al. 

 (1987; 1996) suggested that a 5.55-m-TL, 1970-kg 

 female captured in mid-February had aborted an 

 entire litter of near-term embryos. However, the 

 presence of almost 200 egg cases in the left uterus 

 (weighing 9 kg) and the robust appearance of the 

 specimen, probably due to a large liver, suggest that 

 this great white shark was at an early stage of ges- 

 tation. Further support for our gestation estimate of 

 about 18 months was the capture of a great white 

 shark with a midterm litter of nine embryos in the 

 summer of 1934 (Norman and Fraser, 1938). Length 

 and mass of these embryos (61 cm TL, 5.4 kg as 

 suggested by Ellis and McCosker, 1991) were as 

 expected for midterm embryos with a large yolk- 

 filled stomach (M/TL3=23.8 kg/m^, Fig. 6) and birth 

 was expected 9 months later in spring. 



Gestation periods longer than 12 months were 

 reported for the whitetip reef shark, Triaenodon 

 obesus (13.5 mo.), the tiger shark, Galeocerdo cuvier 

 (13-16 and 15-16 mo.), the dusky shark, Carcha- 

 rhinus obscurus (22 mo.), and the spiny dogfish, 

 Squalus acanthias (18-24 mo.) (Compagno, 1984; 

 Uchida et al., 1990; Randall, 1992; Musick et al., 

 1993; Crow'^). The basking shark, Cetorhinus maxi- 



inus, probably has a gestation period longer than 12 

 months, but we agree with Pauly'^ that the reported 

 gestation period of 3.5 years by Parker and Stott 

 ( 1965) was based on fallacious reasoning. 



Parturition in Northern and Southern Hemispheres 



Based on a few full-term litters and a large number of 

 neonates, parturition occurred mainly in late winter 

 to midspring in both hemispheres (Fig. 3A). Pratt 

 and Casey (1983) reported "late spring parturition" 

 in the western North Atlantic, but April should be 

 called early spring; and the data in their Figure 3 

 suggest that parturition begins as early as late Feb- 

 ruary (late winter). Other large pelagic and near- 

 shore sharks have seasonal parturition in late spring 

 to early summer in both hemispheres (e.g. Bass 

 et al., 1973; Pratt, 1979; Klimley, 1985; Randall, 

 1992; Musick et al., 1993; Francis, 1996; Crow^^). 

 Parturition of the porbeagle peaks in June— July in 

 both hemispheres which Francis and Stevens (2000) 

 found puzzling. 



Reproductive cycle and resting period 



Given a gestation period of about 18 months and a 

 distinct seasonal spring parturition, the reproduc- 

 tive cycle for shortfin makos is either 2 or 3 years. 

 There were insufficient data to determine its dura- 

 tion conclusively, but a 2-year cycle would only allow 

 a recovery period of some 6 months, which would 

 appear to be insufficient. Our temporal UWI data 

 (Fig. 3B) support a 3-year reproductive cycle, which 

 would mean an 18-month resting period. 



There is conflicting evidence about the need for 

 a resting period after parturition in elasmobranchs. 

 For many medium to large sharks a resting period 

 has been documented. Peres and Vooren (1991) 

 reported a 3-year reproductive cycle for the soupfin 

 shark, Galeorhinus galeus, off southern Brazil with 

 a gestation of 12 months. The gestation for the 

 whitetip reef shark is 13.5 months and the reproduc- 

 tive cycle is 2 or 3 years (Uchida et al., 1990). Musick 

 et al. ( 1993) suggested that the sandbar shark, Car- 

 charhinus plumbeus (with one-year gestation), must 

 have a resting period of at least one year, requiring 

 a 2-year reproductive cycle. The 2-year reproduc- 

 tive cycle for the sandtiger shark with 8—9 mo. ges- 

 tation period is well documented (Branstetter and 

 Musick, 1994; Cliff, unpubl. data). Tagging and cap- 



i*Crow, G. L. 1995. The reproductive biology ofthe tiger shark. 

 Galeocerdo cuvier, in Hawaii: a compilation of historical and 

 contemporary data. Program and Abstract ASIH and AES 

 Annual Meeting, University of Alberta, Edmonton, Alberta, 

 Canada, June 15-19 1995, 229 p. 



'^ Pauly, D. 1978. A critique of some literature data on the 

 growth, reproduction and mortalities of the lamnid shark Ceto- 

 rhinus maximus (Gunnerusl. Report to the Pelagic Fisheries 

 Commission, International Council for the E.xploration of the 

 Seas, CM 1978/H17, Copenhagen. 



