Bottlenose Dolphins

Species of the genus Tursiops inhabit temperate and tropical coastal and offshore waters of the world They are among the most popular species displayed in aquariums and often met in their natural environments in coastal waters Thus, they have offered us opportunities to study their morphology, behavior, and social structure However, the taxonomy of this genus has not yet been fully settled One of the reasons for this is the difficulty of obtaining specimen materials from their broadranging habitats, which is a common problem in cetacean studies Another factor making their taxonomic study difficult is the existence of numerous local populations with various degrees of differentiation, as is usually the case in coastal small cetaceans Under these circumstances, scientists have been able to identify differences between nearby populations but have had difficulty until recently in identifying and evaluating differences between geographically distant populations This problem is being resolved by the use of genetics, which allows objective evaluation of differences The genus contains two ecotypes-one inhabiting coastal waters and the other offshore waters-each of which contains numerous populations (Reeves et al. 2004), but the two ecotypes do not match up with the two species currently recognized in the genus

One of the two currently recognized species, T. aduncus (Ehrenberg 1828, 1833), described from a specimen from the Red Sea, is known to inhabit coastal waters from the Indian Ocean to central Japan, including the coasts of Australia and Southeast Asia It is smaller and has a relatively longer rostrum than the other species of the genus T. truncatus and the adults often have spots (Gao et al. 1995; Kasuya et al. 1997, in Japanese; Rice 1998; Wang et al 1999, 2000a,b; Moller and Beheregaray 2001; Best 2007)

T. truncatus (Montagu 1821) was described based on a specimen from Great Britain This species has greater body length, a relatively shorter and stouter rostrum than T. aduncus, and is not spotted It inhabits offshore waters of the tropical and temperate regions of the Indian Ocean and the South and North Pacific (including Japanese waters), the North and South Atlantic, the Mediterranean Sea, and the Black Sea It also inhabits coastal waters not inhabited by T. aduncus Both species are known from Japanese waters

The two species, T. aduncus and T. truncatus, interbreed easily in aquariums and produce fertile offspring (Hale et al. 2000) This suggests that they have not established a genetic mechanism to prevent interbreeding but that the isolation is achieved mainly by difference in habitat preference The means of inhibiting interbreeding between such species will work differently between the sexes Males are more likely to move outside of the ordinary range of a population and have a

greater opportunity to interbreed with other species or populations It will be valuable to further examine both mitochondrial and nuclear DNA for a better understanding of species and population structure within the genus Tursiops

It was difficult to distinguish between the two species of Tursiops based on single morphological characters or measurements, but multivariate statistics applied in South Africa, China, and Australia successfully delineated the two species in comparisons between populations in neighboring habitats but not always between distant populations (Ross 1977; Wang et al. 2000b; Kemper 2004) This has suggested the possibility of some still undetermined taxonomic variations within each of the two species Two forms, that is, coastal and offshore types, are known for T. truncatus in the western North Atlantic, where T. aduncus does not occur The coastal type in Florida is smaller than conspecifics in offshore waters and also differs in skull morphology and physiology The offshore type exhibits blood physiology better suited for deep diving (Hersh and Duffield 1990) Three populations, often called “communities,” are known for the coastal T. truncatus on the west coast of Florida, each of which occupies tens of kilometers of the coastline (Wells and Scott 1990a; Weigle 1990) These coastal communities of T. truncatus apparently occupy a similar ecological niche as T. aduncus in other oceans, that is, in the coastal waters of the Indian Ocean and western Pacific Future studies will reveal further details of geographical variation within the genus Tursiops

Wang et al (1999) analyzed mitochondrial DNA (mtDNA) to examine the geographical distribution of the two species of Tursiops in waters around eastern Asia and confirmed the presence of T. aduncus in the Taiwan Strait, Hainan Island, and Indonesia and T. truncatus in the Taiwan Strait, Hong Kong, the east coast of Taiwan, Mauritania, and Brazil Thus, waters from the Taiwan Strait to Hong Kong are inhabited by both species Before its prohibition in 1990, the dolphindrive fishery in the Pescadores Islands off the west coast of Taiwan took both species of Tursiops (Wang and Yang 2007) Only T. truncatus has been reported from the Yellow Sea and the Bohai Sea, which are located to the north of Taiwan, and only T. aduncus from the waters south of Hong Kong (Zhou and Quian 1985; Wang et al. 2000a) However, in view of the segregation of the two species by the 30 m isobath off eastern Australia, there still remains the possibility of finding T. truncatus in the offshore waters of the South China Sea

Perrin et  al. (2007) compared partial mtDNA controlregion sequences and morphology of the type specimen of T.  aduncus with those of Tursiops spp from other regions and found that mtDNA of the type specimen showed a perfect agreement with that of dolphins referred to the same species from the Atlantic coast of South Africa but that it did

type specimen of T. aduncus were also found to be closer to T. aduncus from South Africa (33 individuals) and Taiwan (19 individuals) than T. truncatus from South Africa (9 individuals) and Taiwan (50 individuals)

Before the study by Perrin et al. (2007), Natoli et al. (2004) analyzed mtDNA for a total of 11 samples of the two species of Tursiops from the Mediterranean, North Atlantic, Gulf of Mexico, Bahamas, West Africa, South Africa, and the Chinese coast It was unfortunate that they did not analyze Japanese samples Their conclusions were as follows: (1) T. truncatus and T. aduncus are distantly related, which was expected, (2)  T.  aduncus from South Africa and those from China represented two distantly related groups, and (3) T. truncatus in the Gulf of Mexico and US east coast formed one group and those in the Mediterranean and offshore waters of eastern and western North Atlantic formed another group Their results suggested that the genus Tursiops contains three groups and possibly more It is possible that future studies will identify two species or subspecies within the current T. aduncus and the same in the current T. truncatus

LeDuc et al (1999) proposed that T. aduncus was genetically closer to Stenella frontalis in the Atlantic than to T. truncatus The taxonomy of these genera has long relied upon easily identifiable external or skeletal morphology, which is likely subject to convergence, and restructuring of the classification using more objective genetic methodology is under way

The common names currently in use are “common bottlenose dolphin” for T. truncatus and “Indo-Pacific bottlenose dolphin” for T. aduncus The latter species was identified in Japan in 1965 by Nishiwaki (see Kasuya and Yamada 1995, in Japanese), although the presence of the species was fully confirmed later (Section 113) The Japanese common name hando-iruka or bando-iruka is used for the common bottlenose dolphin and minami-hando-iruka or minami-bandoiruka for the Indo-Pacific bottlenose dolphin

The Japanese common bottlenose dolphin, T. truncatus, can be distinguished from the Indo-Pacific bottlenose dolphin, T. aduncus, by greater body size, absence of spotting on adults, and shorter and more robust rostrum (Figure 111) Wang et al (2000a) compared the external morphology of 40 common bottlenose dolphins and 17 Indo-Pacific bottlenose dolphins Since his sample combined measurements of both sexes as well as juveniles, only maximum figures are cited in the following The maximum body size for the common bottlenose dolphin was 2955 cm, which exceeded by 28 cm the corresponding figure of 2680  cm for the Indo-Pacific bottlenose dolphin The maximum length of the rostrum of the common bottlenose dolphin was 130 cm, while that for the Indo-Pacific bottlenose dolphin was 137 cm Although the absolute difference was negligible, the former species has a more robust rostrum Tooth and vertebral counts also showed some differences, with range overlaps (Wang et  al. 2000b;

Kemper 2004; Table 111) The mean tooth diameter of the common bottlenose dolphin was 71  mm with a range of 60-84  mm (n = 12), while that of Indo-Pacific bottlenose dolphin was 62 mm with a range of 50-75 mm (n = 12), but the difference was not statistically significant (Kemper 2004) A small difference was probably masked by large individual variation Gao et al (1995) reported the maximum body size of the common bottlenose dolphin at 330 cm and that of the Indo-Pacific bottlenose dolphin at 254 cm The former figure was about 30 cm greater than the maximum size of the same species reported by Wang et al. (2000a) as cited earlier The difference is presumably due to inclusion of common bottlenose dolphins from the Bohai and Yellow Seas in Gao et al (1995) as reported by Zhou and Qian (1985) The common bottlenose dolphin has broad geographical variation in body size reflecting environmental temperature or latitude The largest individuals are found among populations in European waters, where males reach 380 cm and females 367 cm (Wells and Scott 1990a)

External morphology was observed and body lengths measured for bottlenose dolphins taken by the drive fisheries on the coast of the Izu Peninsula (34°36′N-35°05′N, 138°45′E-139°10′E) and at Taiji (33°36′N, 135°57′E) and culling operations at Katsumoto on Iki Island (33°45′N, 129°45′E), during the 11  years from 1973 to 1983, but none of them

appeared to represent the Indo-Pacific bottlenose dolphin (Table 112) Iki Island is located in Tsushima Strait between the East China Sea and the southern Sea of Japan Modal body lengths were 290-299 cm for both sexes, and the maximum body lengths were 336 cm (male) and 320 cm (female) These figures are close to those reported from the Yellow Sea and Bohai Sea but about 30  cm greater than the corresponding figures reported by Wang et al (2000a) from Taiwan and to the south Ogawa (1936, in Japanese) reported tooth counts of 21-24 in each jaw (total 88-96) with an average total count of 90 for four Tursiops sp killed off the Sanriku Region of the Pacific coast of Japan in latitudes 37°44′N-41°30′N, which

were closer to the range of the common bottlenose dolphin than that of the Indo-Pacific bottlenose dolphin shown in Table 111

While examining common bottlenose dolphins listed in Table 112, Kasuya et  al. (1997, in Japanese) collected several teeth at the center of the lower tooth row for age determination Some smaller teeth were collected near the end of the tooth row from animals that had lost most of the teeth, presumably due to old age The teeth were sectioned longitudinally and the diameter measured Using only measurements of teeth from 203 females (over 10 years of age) from 10 schools with acceptable sample size (8-30 individuals/school), they got average diameters in a range of 72-83 mm for each

TABLE 11.2 Materials for Japanese Common Bottlenose Dolphins Analyzed by Kasuya et al. (1997, in Japanese)

Comparison of Tooth and Vertebral Counts between Two Species of the Genus Tursiops

203 (range 53-100 mm) Comparison between the samples from the Pacific coast (Izu and Taiji) and Iki did not find a significant difference in mean diameter These values were about 1-2 mm greater than those reported by Kemper (2004) for both species (see the preceding text in this section), presumably reflecting different body size Although there were some statistically significant differences observed between some schools from the Pacific coast, Kasuya et al. (1997, in Japanese) did not consider them important

The common bottlenose dolphin inhabits Japanese waters south of 43°N, or approximately south of southern Hokkaido (Figures 112 and 113) The Indo-Pacific bottlenose dolphin is resident in coastal waters around Amami Island (28°15′N, 129°20′E) (Miyazaki and Nakayama 1989; Rice 1998), Mikura Island (33°53′N, 139°35′E) (Kogi et  al. 2004), the Bonin Islands (27°N, 142°E) (Ogasawara Whale Watching Association 2003, in Japanese), and Tsuji-shima Island (32°45′N, 130°20′E) in western Kyushu (Shirakihara et  al. 2002) (Figure 112) Further investigation may find it to occur in other locations

Shirakihara et  al. (2002) identified 178 individuals in a population of Indo-Pacific bottlenose dolphins off Tsuji-shima Island from 1994 to 1998 and estimated the total population size at 218 after correcting for unidentifiable individuals and those that died during the study period I had the impression during a shipboard observation off Tsuji-shima that bottlenose dolphins there are spotted to a lesser degree than those

the same species around Mikura Island has distinct spots

The oceanographic environment and recent occasional records of the species suggest that resident populations of the Indo-Pacific bottlenose dolphin may have occurred along the Pacific coast of the Japanese islands south of Tokyo (ca 35°50′N), but none have been recorded there It is a possibility that such resident populations could have been exterminated by the local fisheries before being identified by scientists In the 1960s, a small-type whaler and whale meat dealer in Taiji talked about a dolphin species called hau-kasu taken by the local fishery The name meant a “hybrid with the spotted one” This species was said to be similar to the common bottlenose dolphin, which was locally called kuro (black), but had spots on the body Thus, the fishermen thought them to be a hybrid between the common bottlenose dolphin and the kasuri-iruka (pantropical spotted dolphin) Kasuya and Yamada (1995, in Japanese) assumed considering additional information that the species had white lips and that it was probably the roughtoothed dolphin, Steno bredanensis However, the possibility remains that the hau-kasu was the Indo-Pacific bottlenose dolphin (Section 27)

There have been several cases where a small number of Indo-Pacific bottlenose dolphins visited a Japanese coastal area and resided there for a period Around 1976, one group lived for a time off Mi-irihama in Kamogawa City (35°07′N, 140°06′E) and then disappeared A group of six with juveniles has inhabited the waters of Sunosaki in Kamogawa City since 1998, and four of them were identified as previously observed at Mikura Island (Fujita 2003, in Japanese) In July 2007, a male was caught in a trap net in Kamogawa City and transported to Kamogawa Sea World This individual was reported to be from Mikura Island (Saeki 2007, in Japanese), but the relationship with those in Sunosaki was not reported

According to Miki Shirakihara (personal communication in April 2008), most of the T. aduncus off Tsuji-shima traveled in the spring of 2000 southward for a distance of about 50-60 km to live in waters off Naga-shima Island (29°10′N, 130°10′E) for about one year Then, leaving behind several tens of dolphins in this new habitat, most returned to their original habitat off Tsuji-shima to merge with old group members Some members of the Tsuji-shima group were found in a group of several Indo-Pacific bottlenose dolphins that recently settled off Notojima (37°07′N, 137°00′E) on the Sea of Japan coast, which required travel of about 1100 km northeast along the coast of the sea There must be more unreported similar cases

These local movements of Indo-Pacific bottlenose dolphins along the coasts of Japan suggest a mechanism of maintaining population size One of the direct results is to establish a new colony through emigration from a large, perhaps oversized, population Another result is to supply new members or genetic variability to a small local population The carrying capacity of a single coastal habitat may not be sufficient for a population of viable size, which is unknown but must certainly be over 200 Loss of a few individuals by emigration will not harm the survival of the local fully grown group,

members and new genetic variation to other smaller groups If adult males temporarily visit nearby colonies to mate, that will also function to increase the genetic variability of the recipient group

The Indo-Pacific bottlenose dolphin off Japan may have survived through repetition of extirpation and recolonization of sites and emigration, immigration, and genetic exchange between colonies This structure containing a network of such numerous smaller colonies is called a metapopulation, which results in increasing population stability as a whole Due to the limited carrying capacity of local habitats, many local colonies would have trouble maintaining population stability, but the physical and genetic interactions among the colonies enhance the stability of the whole complex of small populations Further efforts to monitor the process of establishing new colonies along the coasts of Japan and the process of exchanging members between them will improve our understanding of their population structure

Currently, we do not have records of resident populations of common bottlenose dolphins in Japanese coastal waters, although further effort will be required to confirm the absence There are cases of resident populations of the species inhabiting tens or hundreds of kilometers of the coast of California in the eastern North Pacific (Wells et al 1990) Similar cases are also known along the coast of Florida in the western North Atlantic, where each of the local population is called a “community” Some males of a community are known to temporarily visit other communities to mate, and some individuals of both sexes move permanently to nearby communities with an annual immigration rate of 2%–3% (Wells 1991; Connor et  al. 2000) These common bottlenose dolphins occupy a habitat that is used by the Indo-Pacific bottlenose dolphin in other oceans This is an indication of strong adaptability in the genus Tursiops

Bottlenose dolphins, Tursiops spp, have been sighted around Japan in waters above 11°C in surface temperature (Table 101) The northern limit of their distribution in winter was thought to be in the Iki Island area near Tsushima Strait, which connects the East China Sea and the Sea of Japan, and at 36°N off the Pacific coast of Japan, expanding in summer to the coast of southern Hokkaido (Kasuya 1980; Miyashita 1986a, both in Japanese, Figure 113) However, the recent identification of a small resident group of Indo-Pacific bottlenose dolphins off Notojima on the Sea of Japan coast expanded the known winter range to 37°N Knowing whether this small population at the northern limit of the range survives requires monitoring

Fishery statistics give some additional information on distribution Miyazaki (1980a) confirmed year-around landings of bottlenose dolphins at Taiji on the Pacific coast of central Japan, where all the Tursiops examined by scientists were common bottlenose dolphins Catch statistics of a drive fishery at Arari (34°50′N, 138°46′E) on the west coast of the Izu Peninsula covering January 1950 to April 1957 record drives of hasunaga, which is believed to be the common bottlenose dolphin (Section 382), in March, May,

July, September, and December These fishery data indicate year-around distribution of the common bottlenose dolphin at least in waters south of the Izu Peninsula

Miyashita (1993) reported the occurrence of bottlenose dolphins during summer (from June to September) in a large area extending from the Japanese coast to the central North Pacific at 175°E based on data obtained through systematic sighting cruises conducted by the Far Seas Fisheries Research Laboratory of the Fisheries Agency in 1983 through 1991 These sighting surveys did not distinguish between the two species of the genus, but since the survey placed weight on the offshore area and the survey track lines almost avoided waters inside of the 100 m isobath, the possibility of including Indo-Pacific bottlenose dolphins must be negligible These data, which are used in Figures 112 and 113, showed the distribution of the common bottlenose dolphin in June to extend down to latitudes 22°N-23°N off the east coast of Taiwan Sighting effort south of these latitudes did not find the species, suggesting an absence of the species in offshore Pacific waters south of Taiwan This apparent southern limit moved in July to latitudes of 25°N-26°N and remained there through August to September Because the survey was limited to the south of 35°N in June, Miyashita (1993) did not provide information on the northern limit of the species in that month In July, the survey covered latitudes from 12°N to 45°N and recorded common bottlenose dolphin from 25°N

northern limit remained the same through midsummer from August to September

Miyashita (1993) confirmed an almost continuous distribution of the common bottlenose dolphin from the Pacific coast of Japan to 175°E, but distribution of the species to the east of this longitude remains to be studied The density of the species was not uniform within this western North Pacific area The density tended to be lower in more eastern waters, and there was an apparent distribution gap around 30°N, 165°E (Figure 112) This needs to be confirmed in future surveys

The Fisheries Agency at one time collected sightings of cetaceans from fishery inspection vessels of Nagasaki Prefecture that operated in the East China Sea and the Iki Island area as part of research activities conducted in relation to the dolphin-fishery conflict in the Iki Island area It recorded bottlenose dolphins, presumed to be common bottlenose dolphins, in the East China Sea in all months except August, with a peak frequency in April and May (Tamura et al 1986, in Japanese) Similar records collected by Fisheries Agency vessels (Miyashita 1986a, in Japanese) confirmed the presence of the species in the East China Sea throughout the year, with the range extending from northern Kyushu to the Bohai Strait/western Yellow Sea in the northwest and to the southern Ryukyu Islands (ca 24°30′N, 124°00′E) in the south In the Sea of Japan, reported sighting of common bottlenose dolphins ranged from Tsushima Strait to off Aomori Prefecture (40°25′N-41°15′N), which did not disagree with the results of whale-sighting surveys presented in Figure 113 The northernmost record of the species in the Sea of Japan was off southern Hokkaido (42°N-43°N) The offshore distribution of the species in the Sea of Japan was limited to within about 120 nautical miles (about 220 km) off the coast Only Dall’s porpoises were sighted further offshore (Tamura et al 1986, in Japanese) Warmer-water species such as the bottlenose dolphins, Pacific white-sided dolphin, and false killer whale occurred in the Sea of Japan only in the coastal waters of southern Korea and Japan (Tamura et al. 1986, in Japanese) Miyashita (1993) sighted bottlenose dolphins in June on the Sea of Japan coast between Tsushima Strait and Ishikawa Prefecture (c. 37°30′N, 137°15′E), but they were not recorded from October to December to the northeast of Ishikawa Prefecture Seasonal north-south movement of the species is suggested between 37°N and 43°N The residence of a small group of Indo-Pacific bottlenose dolphin in Notojima is at the northern limit of the genus in the winter months

The age of bottlenose dolphins has been determined using the growth layers in dentine and cementum of the teeth The common bottlenose dolphin is one of the best-studied species on the formation of growth layers in the teeth, and the annual deposition of the growth layers has been confirmed using wild or aquarium-reared animals of known age as well as

and Cornel 1990) Kasuya et al. (1986), using common bottlenose dolphins born in aquariums and those taken by fisheries of less than 4 years of estimated age, compared growth layers in dentine and cementum and established a standard for reading the annual growth layers in the teeth Because deposition of dentinal growth layers ceases at a certain age, as in many other delphinids, the ages of old individuals have to be determined using cemental layers Kasuya et al. (1997, in Japanese), on which the following description is based, counted layers in each tissue of common bottlenose dolphins three times and accepted the middle figure as the true value If the cemental age exceeded the dentinal age, the former was used as the age of the individual All the ages between n and n + 1 were expressed as n + 05 years (n being an integer) This was for the convenience of mathematical analysis and did not signify the precision of the readings Because parturitions of Japanese common bottlenose dolphins occur from February to October with a peak in June (see Section 11423) and most of the specimens used by Kasuya et  al. (1997, in Japanese) were obtained during January through April, this grouping of ages would not cause a significant bias

11.4.2.1 Gestation Period Many common bottlenose dolphins have been successfully maintained in aquariums, and their reproductive activities have been monitored with the help of modern technologies A surge of estrous hormone or its metabolites in blood or urine indicates ovulation, and continuation of a high level of progesterone suggests pregnancy Ultrasonography enables monitoring the growth of Graafian follicles and fetuses Yoshioka et  al. (1986) using such methodology confirmed that female common bottlenose dolphins were receptive only during a one-or two-day period before and after an ovulation in an aquarium (total receptive period was two to four days) Monitoring of 77 pregnancies in US aquariums revealed that their gestation periods were within a range of seven days on both sides of a mean of 370 days (the 77 records were within a range of 14 days) (Asper et al. 1992) I expect a similar gestation period exists for common bottlenose dolphins around Japan

Body lengths of neonates born in Japanese aquariums and measured within 11 days from birth were in a range of 116-140  cm (mean 128  cm, n = 20) (Kasuya et  al 1986), which was about 18 cm greater than the corresponding figure of 100-120 cm (mean 1095 cm, n = 21) of neonates of presumed coastal forms stranded along the Texas coasts (Fernandez 1992, cited in Urian et al. 1996)

11.4.2.2 Fetal Growth The fetal growth of cetaceans consists of an early phase of slow curvilinear growth and a later phase of linear growth The period of the latter is a set proportion of the total

point when the extended linear growth line cuts the axis of time is estimated at about 50 days, or 135% of the total gestation period for species with a gestation period of about 1 year (Section 852) Based on this and neonatal length of 128 cm, the fetal growth rate in the linear phase is estimated at about 04  cm/day or 123 cm/month for Japanese common bottlenose dolphins Thus, the relationship between body length (BL, cm) and fetal age (t, days) is expressed by

BL = [(t − 0135G)X]/[G(1 − 0135)]

where G is an average gestation period of 370 days X is the mean neonatal length of 128 cm

This equation does not apply to the early phase of fetal growth, which lasts for about 2 months from conception

Kasuya et  al (1997) examined common bottlenose dolphins taken from January to April by drive fisheries at Taiji on the Pacific coast and at Katsumoto on Iki Island They found fetal body length ranging from 10 to 120 cm and estimated the dates when these fetuses could have reached the mean neonatal body length if they had not been killed by the fishermen (Figure 114) One school taken at Arari on the Izu Peninsula in August (Table 112) was omitted from this analysis because it contained only two fetuses at 13 cm and 40  cm, but inclusion of that school would not change the following conclusion The parturition dates estimated from fetuses over 30 cm in body length were similar between schools within the same geographical region, that is, the Pacific and Iki Island areas The dates ranged from March to October for the Pacific sample and from February to October for the Iki Island sample, with a peak in June for both areas The mean parturition date was July 6 (SD = 51 days, n = 46) for the Pacific sample and June 30 (SD = 65 days, n = 60) for the Iki sample (Kasuya et  al. 1997, in Japanese) Thus, the

Pacific and Iki areas Fitting fetuses smaller than 30  cm to the linear growth curve would estimate the parturition date as earlier than the true one, with greater bias being caused by the smaller fetuses

Steiner and Bossley (2008) recorded parturition dates for a resident population of Indo-Pacific bottlenose dolphins in an estuary area near Adelaide, Australia They recorded 45 births during the period of 1989-2005, which occurred in 6 months from December to May (no births occurred from June to November) There was a peak in February (11 births) followed by January (7 births) The latitude of the study site was about 35°S, which was nearly opposite to the latitude of Japanese specimens (c. 33°35′N-33°45′N), and the mating peaks were about 4 months apart

Urian et  al. (1996) compared the parturition seasons of common bottlenose dolphins between wild individuals in Florida and Texas and between wild and aquarium animals from each of the regions They found that the difference in parturition season was small between the east and west coasts of Florida Births occurred in any month of the year, with a main peak from March through September This pattern was very similar to that observed off Japan However, parturitions on the Texas coast were limited to a short period from February to April In order to investigate the regional difference in breeding seasonality, Urian et  al. (1996) monitored females that experienced over five parturitions in an aquarium environment for a possible change in breeding season in the artificial environment They obtained an unexpected result; the females retained the seasonality of the original habitat for at least up to the fifth parturition Although the mechanism controlling the seasonality of breeding remained unknown, they found a tendency of parturition occurring slightly in the later part of the season and greater standard deviations for females that lived in an aquarium These results indicate that the breeding seasonality of wild individuals not only is retained for a long period in an aquarium but also suggests a possibility that the seasonality of wild animals may be gradually lost during longer life in an artificial environment

Breeding seasonality of Japanese common bottlenose dolphins in an artificial environment has been recorded for three females in Kamogawa Sea World (Yoshioka et al. 1986; Yoshioka 1990, in Japanese) These females were introduced from the wild in 1971-1978, presumably selected from drive hunts on the Izu coast Their body lengths were 273-288 cm at the time of transfer, but their maturity was not determined from these measurements Hormonal levels were analyzed 3-12 years after the transfer into the aquarium, or in 19821984, when two measured 282 and 285 cm and had already experienced parturition and another measured at 294 cm was considered sexually mature One of the three females underwent several estrogen surges and subsequent rises of progesterone level in June-August of 1982 and 1983, but these did not result in pregnancies The authors concluded that these estrous periods were not followed by conception This female did not show such a cycle in 1984 The second female had estrous cycles in June-August of 1982 and 1984 The third

started estrogen and progesterone cycles in April 1984 and repeated the cycle until October The second female exhibited several months of pseudo-pregnancy after the estrous cycles, but the other two females did not show any features of pregnancy in spite of copulations during estrus These records indicate that the breeding seasonality of females in the wild is retained for over 10 years in an aquarium

Kamogawa Sea World also produced information on reproductive seasonality in a male common bottlenose dolphin (Katsumata et al. 1994, in Japanese) The specimen was caught at Futo (34°54′N, 139°06′E) on the west coast of the Izu Peninsula and brought to the aquarium in March 1985, when it was 292  cm in body length It was monitored for serum testosterone level during 6 subsequent years, ending in 1991, when it measured 306 cm The male could not be definitely determined to be mature based on body length at capture, 292 cm, but it was judged sexually mature in 1991, when it measured 306 cm All 24 measurements of testosterone level during the first year in the aquarium were below 1 ng/mL, which was interpreted by Katsumata et al. (1994, in Japanese) as an indication of physiological pathology after introduction to the aquarium, but the possibility of the animal being sexually immature cannot be fully excluded in my view In 1986, or in the second year at the aquarium, there were 13 measurements of testosterone level, with a peak of 10-20 ng/mL in July-August Only 3-8 measurements were available for the subsequent years, which were insufficient for identifying annual peaks Combining all the measurements of the male revealed low mean testosterone level of 5 ng/mL in January-February, a high value of over 15 ng/mL in April-August, and a gradual decline to November when it again reached a level of around 5 ng/mL The seasonality of testosterone level in the captive male agreed with the mating season of females in the wild

Mortality rate differs between the sexes throughout life; thus, the sex ratio can change during fetal and postnatal life The sex ratio in our samples may also be influenced by gear selectivity in fisheries or by behavioral difference between the sexes

The sex ratio of common bottlenose dolphins taken by drive fisheries is shown in Table 113 The fetal sex ratio of the Pacific sample is 22:22 and that of the Iki sample 20:37 Although the latter apparently indicates an excess of female fetuses, it is not statistically different from parity The data are insufficient to see change in sex ratio with fetal growth

Males slightly exceeded females in number among postnatal individuals below 10 years of age, but the difference is not statistically significant The proportion of females increased with postnatal age in both the Pacific and Iki samples, that is, 455% (<10 years), 620% (10-20 years), and 586% (>30 years) for the Pacific sample and 483% (<10  years), 658% (1020 years), and 703% (>30 years) for the Iki sample This is likely to be due to a higher mortality rate in males The age of the oldest female exceeded that of the oldest male in both

samples, 425 versus 395 years in the Pacific sample and 455 versus 435  years in the Iki sample, which also suggests a slightly higher survival rate in females

Kasuya et al. (1997, in Japanese) obtained the age composition of a total of 15 schools of common bottlenose dolphins taken by drive fisheries on the Pacific coast and by a culling operation in the Iki Island area (Kasuya 1985) These schools are listed in Table 112, and the age composition is shown in Figure 115 and Table 113

Age frequencies were compared between the two geographical samples by grouping the dolphins into three age groups of 0-10, 10-20, and >20 The frequencies in the Pacific sample were 99, 108, and 111, or ratios of 1: 109: 112, while those in the Iki sample were 236, 111, and 91, or 1: 047: 039 The Iki sample contained an extremely large number of juveniles below 10 years of age Such an age composition occurs if (1) the population has a high mortality rate, (2) the population is increasing with a high reproductive rate, or (3) a school of immature individuals has been taken by chance I  have the impression that the last is the most likely cause This problem will be further discussed in Section 1148

As the age frequency is shown in Figure 115 on a logarithmic scale, the slope constitutes an “apparent mortality rate,” which is a combination of mortality and increasing (or decreasing) growth rate of the population The apparent mortality rate is equal to the real “mortality rate” of the population only if the population size remains constant with balanced mortality and recruitment Because the common bottlenose dolphin may change in social behavior with age and the obtained age composition can be biased, caution is required in estimating the mortality rate from age composition If we tentatively assume that the species exhibits a uniform mortality rate through life, which is certainly incorrect, and that 1% survive to 45 years, the maximum age observed for the species, then 97% is obtained as the average annual mortality rate for ages 0-45 years An assumption of 5% surviving 45 years gives a mortality rate of 64%

The observation of a resident community of common bottlenose dolphins in the coastal waters of Sarasota on the west coast of Florida during 1970 to 1987 revealed that the population was stable at a size of about 100 individuals and resulted in estimation of several parameters of population dynamics by Wells and Scott (1990b) They estimated the survival rate of calves in their first year at 0803 (mortality rate of 0197) and the annual mortality rate of individuals over age 1 year at 0038 The average mortality rate for the whole age range must be between the two extremities, or between 0197 and 0038 They also calculated the annual recruitment rate to

Sex Ratio of Common Bottlenose Dolphins off Japan, Obtained from 7 Drives on the Pacific Coast and 4 Drives at Iki

tality rate of individuals over age 1 year given earlier, 0038, if the population remained absolutely stable The annual mortality rate estimated for individuals over 1 year of age included emigration to nearby communities This emigration rate was not separated from mortality, but it could have been of similar level of immigration from other communities, which is known to be about 002 They calculated annual birth rate at 0055, which was the number of newborns of both sexes produced by individuals of both sexes of all age classes If the average mortality rate for all age classes were equal to this birth rate, the population would be stable Dividing 0144, which is the recruitment rate to age 1 year relative to the population of adult females calculated by Wells and Scott (1990b), by 0803, a survival rate of newborns to age 1 year, gives 0179, which is an approximate annual parturition rate of adult females This annual parturition rate gives an average calving interval of 56 years This annual parturition rate is smaller than the annual pregnancy rates of 033-041 estimated for Japanese populations (see Section 11484) This suggests that the Sarasota population is almost stable with low rates of reproduction and mortality in a small habitat with limited predators Although the annual pregnancy rates estimated for Japanese populations do not, in a strict sense, distinguish between annual conception rate and annual parturition rate, the effect would be negligible in the comparison between the two oceans

The age composition of Japanese common bottlenose dolphins shows a deficiency of some age classes Both sexes aged 15-125 years, which could actually include ages from 1 to 13 years, are apparently underrepresented in the Pacific sample Since lactation in the population lasts for 17 years on average (Table 1111) and both sexes attain reproductive ability by the age of 13 years (Tables 116 and 119), the underrepresented age classes roughly correspond with those after weaning and before sexual maturity This can probably be interpreted as a result of growth-related behavioral change, where weaned immature individuals tend to live with individuals of similar age Thus, the age composition of the sample depends on the chance of fishermen finding and driving particular types of schools It is also a possibility that this may be due to geographical segregation or difference in school size, neither of which has been confirmed for this species

Although the Iki sample does not exhibit a clear deficiency of ages just after weaning, it still exhibits a deficiency of females aged 55-105 years and males aged 55-135 years, which corresponds to the range of age around the attainment of sexual maturity Thus, it is likely that dolphins of both sexes just before sexual maturity are underrepresented in the Iki sample

It has been reported that striped and pantropical spotted dolphins between weaning and sexual maturity tend to leave the mother’s school to live with individuals of the same growth stage (Kasuya et  al. 1974; Miyazaki and Nishiwaki 1978; Kasuya 1999; see also Section 10514) Common bottlenose dolphins in a resident population on the Florida coast are known to form groups of individuals of the same age, sex, and reproductive status (Shane et al. 1986; Wells 1991) In this

suckling calves, but more often they live with other mothercalf pairs Calves, particularly males, at age 5-6 years start to spend time apart from their mothers and finally leave their mothers to live with juveniles of the same sex This behavior of common bottlenose dolphins has some similarity to what has been inferred for Japanese striped dolphins

Female common bottlenose dolphins off Florida rejoin their mothers at sexual maturity, which suggests an evolutionary step toward the formation of a matrilineal social system It is easy for the young females of this community to locate and join their mother’s school at sexual maturity, because they live in a small community in a semienclosed habitat Such long-term observations have not been achieved for a population of the same species in offshore open waters, and it remains to be confirmed if females in such offshore populations have the opportunity to return to their mother’s group at sexual maturity

Males in the Florida community at sexual maturity behave quite differently from females They do not join their mothers at sexual maturity but may live alone or live in a coalition with several adult males and visit female schools Such male coalitions may be formed by joining with other adult males or by maintaining a coalition after the juvenile period Such male coalitions have also been observed in Indo-Pacific bottlenose dolphins in Shark Bay in Western Australia and some other waters and are attracting the attention of behavioral biologists attempting to determine the benefit of such a life (Connor et al 2000) Some hypotheses are that the coalitions (1) protect against predators, (2) allow inferior males in the group the opportunity to mate, and (3) allow males to corral a female for mating The answer is still unknown

Figure 116 shows the body-length frequencies of common bottlenose dolphins off Japan The Pacific sample is represented by 244 dolphins from 7 schools driven at Taiji during January through April and 37 from a group driven at Arari in August, and the Iki sample is represented by 446 dolphins in 7 schools driven from January through March at Katsumoto at Iki Island as a part of fishermen’s activities in the “dolphinfishery conflict” (Section 34) (Table 112) The main season of the sample, from January to April, is about 3-6 months after the end of the preceding parturition season, which extends from February to October with a peak in June (Figure 114)

The largest male was measured at 328 cm in the Pacific and 336 cm in the Iki Island area, and the largest female at 318 cm in the Pacific and 320 cm in the Iki Island Area Males grow about 10-15 cm greater than females Although the Iki sample contained a dolphin of larger size than the Pacific sample, the difference is statistically insignificant

The body-length frequencies reveal three modes, that is, individuals of 130-140  cm, 180-210  cm, and over 220  cm Since neonates of this species off Japan are born at 116-140 cm (Section 1142), the smallest peak contains neonates born at the beginning of the parturition season The second group,

180-210 cm, contains individuals born in the preceding parturition season that are 3-13 months of age The largest group contains individuals of 2 or more years of age It becomes difficult to separate age groups based on body length after the age of 2 years, because growth rate decreases and individual size variation increases with age The age composition of the Pacific sample suggests underrepresentation of juveniles, which is also apparent in the body-length frequencies, particularly among males (Figure 116)

The usual way to construct a mean growth curve is to base it on body lengths and estimated ages from catches or bycatches obtained during a certain length of time As stated elsewhere, the mean growth curve thus created reflects the real mean growth curve only when the growth pattern in the population has remained stable during the whole lifetimes of the animals in the sample, that is, from the date of birth of the oldest individual to the date of the last sample Information from aquarium-reared known-age individuals is free from this problem but is often accompanied by the problems of insufficient sample size and evaluating the effect of the artificial environment It should also be noted that a mean growth curve does not represent the growth of any single individual, that is, an individual growth curve and the mean growth curve are likely to be quite different

An attempt has been made to resolve the first problem by using repeated measurements of wild individuals in Sarasota, Florida (Read et  al 1993) Body lengths measured periodically were plotted on ages of animals known from birth or aged using teeth extracted from the individuals of unknown birth date These materials were obtained mostly from young individuals below 10 years of age, covered less than 7 years, and lacked data on animals less than 1-year old (mother-calf pairs were not examined to avoid disturbance of their bond) The growth curves of individual dolphins thus obtained were used to construct a mean growth curve (Figure 117)

It appears that fast growth was uniform for 2-5  years, although the absolute growth rates were variable Then, the growth rate declined at individually variable ages around 5  years in females and at a slightly higher age in males This age variation contributes to the formation of a smooth upwardly convex mean growth curve at around age 5 years Almost linear growth of a young individual was also reported for a common bottlenose dolphin of unknown age that

approached humans on the English coast; it grew almost linearly from 2286 to 2709  cm during 400  days, for a mean growth rate of 32 cm/month (Lockyer and Morris 1987)

Kasuya et  al. (1986) reported growth records of three aquarium-reared common bottlenose dolphins taken from Japanese waters (Table 114) By assuming a neonatal length of 128 cm, a body-length increment of 74-93 cm, or increase of 58-73%, in the first year after birth, is estimated

Perrin et  al (1976) obtained the following interspecies relationship between mean fetal growth rate (X, cm/month), mean growth rate in a postnatal period that is equal to the gestation time (Y, cm/month), and mean neonatal body length (Z, cm):

Log (X − Y) = −133 + 0997 log Z

The mean monthly growth rate, Y = 477, is obtained for Japanese common bottlenose dolphins, with values of Z = 128 and X = Z/12 = 1067 The value of Y suggests a body-length increment of 572 cm in the first year after birth, which is at the lower bound of the range calculated from aquarium specimens

One of the growth models fitted by Read et al (1993) to their mean age-length relationship is

Female: L = 2492 exp(−0423 exp(−0314t)

Male: L = 2664 exp(−0422 exp(−0164t)

where L denotes body length in cm t age in years

These equations suggest that female length will exceed that of males at ages 3-10 years, but that male asymptotic length is about 17 cm greater than that of females It should be noted that a mean growth curve can be skewed by biased age composition or limited age range of input data In the previous case, the input data are mostly from animals below age 15 years The mean growth equation fitted to such data may correctly represent the mean growth of the young ages but may not represent real mean growth of older individuals Extrapolation of

such an equation to ages outside the range of the input data, for example, below the age of 1 year, should be avoided Read et al. (1993) averaged body lengths of calves between ages 05 and 14 years and estimated the average length of calves at 1 year of age, which were 184 cm for females and 183 cm for males Assuming mean neonatal length at 1095 cm (Section 1142) gives an estimate of about 68% as average body-length gain during the first year after birth for the coastal population of common bottlenose dolphins off Florida, which is similar to the 58%–73% estimated for Japanese waters

The age-body length relationships for Japanese common bottlenose dolphins are shown in Figures 118 and 119, where ages have been estimated by counting growth layers

Monthly and Annual Growth of Aquarium-Reared Common Bottlenose Dolphins, Calculated Assuming Neonatal Length of 128 cm, a Month of 30.4 Days, and a Year of 365 days

tum The figures show only mean body length, sample size, and range of one standard deviation on each side of the mean Those who are interested in growth equations are advised to attempt the task using the data in Table 115 The rapid increase in mean body length of juveniles ceases at around 5 years, and slower growth continues to around 10 years Agerelated increase in body length appears to cease at around 15 years in females and 20 years in males, which is expected to agree with ages where all the individuals cease growing

The mean body length at cessation of growth, or asymptotic body length, was calculated as an average body length of individuals over 20 years of age, which gives the following figures (Kasuya et al. 1997, in Japanese):

Males, Pacific area: 3053 cm (n = 32, SE = 15) Males, Iki area: 3052 cm (n = 27, SE = 24) Females, Pacific area: 2880 cm (n = 48, SE = 17) Females, Iki area: 2937 cm (n = 63, SE = 09)

Males did not show geographical differences in asymptotic lengths, but females showed statistically significant difference between the Pacific and Iki samples (p < 001) Females in the

TABLE 11.5 Age-Length Key for Japanese Common Bottlenose Dolphins, with Mean Body Length, Sample Size (N), and Standard Deviation (SD)

This can be interpreted as evidence that common bottlenose dolphins in these two geographical regions belong to different populations, although it is unclear whether the size difference is genetically determined or a reflection of difference in nutritional conditions

Comparison between the Pacific and Iki samples showed that mean body lengths of Iki sample were always greater than those in the Pacific sample of the same age below 11 years (males) or 12 years (females) Except for 37 individuals obtained in August 1973 at Arari on the Pacific coast, materials in the comparison presented earlier were obtained in the same months and almost the same years, that is, January-March in 1979-1980 (Iki) and January-April in 1981-1983 (Pacific) Such a geographical difference in juvenile growth could be expected to occur if food availability in the Iki Island area were better than in the Pacific area in the 1970s and early 1980s

11.4.7.1 Identification of Maturity In determining the sexual maturity of male reproductive organs, it is necessary to take into account the effect of reproductive seasonality and distinguish between a seasonal cycle and unidirectional development toward maturation As Dall’s porpoises mate in a few months of the summer and adult males almost cease spermatogenesis in winter, their testicular weight and diameter of the seminiferous tubules almost double in the mating season (Section 948) Full-grown males of the short-finned pilot whale off Japan are spermatogenic throughout the year, while some males in early stages of maturation become spermatogenic only seasonally Kasuya et al (1997, in Japanese) could not identify seasonal change in the male reproductive tracts of common bottlenose dolphins due to the limitation of the samples to a four-month period from January to April, which corresponded to the time of the lowest mating activity (January and February) and the early mating season (March and April) (Section 1142) However, I expect that the seasonal change in male reproductive tracts of common bottlenose dolphins will not be as distinct as in the Dall’s porpoise, because they have a much broader mating season extending for 8-9 months of the year (Figure 114) and the period of spermatogenic activity will extend longer than the mating season

In classifying the maturity stages of testicular tissue of the common bottlenose dolphin, Kasuya et al. (1997, in Japanese) adopted the same categories used by Kasuya and Marsh (1984) for short-finned plot whales (Section 12433), where maturity was classified into four histological stages by the proportion of spermatogenic seminiferous tubules at the midcenter of a testis A seminiferous tubule was determined as spermatogenic if spermatocytes, spermatids, or spermatozoa were present, and a testis was determined as “immature” if no tubules were spermatogenic, as “early maturing” if more than 0% but less than 50% of the tubules were spermatogenic,

the tubules were spermatogenic, and “mature” if 100% of the tubules were spermatogenic The three types of cells occurred in all tubules identified as spermatogenic The absence of spermatozoa would not cause significant bias in determining age at sexual maturity because it takes only a few weeks for a spermatocyte to develop into a spermatozoon

11.4.7.2 Relationship between Age and Testis Weight Figures 1110 and 1111 show the relationship between the age and the weight of a single testis The testis weights are plotted on a logarithmic scale This method covers a broad weight range and illustrates early testicular growth in detail but has the drawback of expressing a positive linear relationship between the weight and age for a curve with the slope that decreases with increasing age

In the Iki sample (Figure 1111), testis growth was slow at ages between 05 and 75 years, when weight increased from around 6 g to around 45 g This corresponded to an increase of 5-6 g/year or about a seven fold increase in the 7 years Then at around age 8 years, the testis started a rapid growth to reach about 200 g at 12 years of age, which was equivalent to an average increase of 39 g/year and a four-to fivefold increase during the 4 years With exception of one male aged 205, all the testes weighing 150 g or more were classified as “mature” After the age of 12 years, the testis continued a similar annual weight increase until age 20, when it reached about 550 g, which was equivalent to an average increase of 44 g/year or a two-to threefold increase in the 8  years Thus, testicular weight gain was almost constant at around

39-44 g/year for ages 8-20  years Age-related testicular growth appeared to cease at the ages of 20-25 years

The general pattern of testicular growth of the Pacific sample (Figure 1110) appeared to be the same as that of the Iki sample, but the Iki individuals almost always had heavier testis compared to Pacific animals of the same age The average testis weight of full-grown males from the Iki Island area was about 200 g greater than in the Pacific sample This is further discussed in the following in relation to body length

11.4.7.3 Relationship between Body Length and Testis Weight

Figures 1112 and 1113 show the relationship between body length and weight of a single testis plotted on a logarithmic scale These figures appear slightly different from Figures 1110 and 1111, because increase in body length is small after the attainment of sexual maturity Comparison between the Iki and Pacific samples showed that (1) both started rapid testicular growth at a similar testicular weight (about 200 g) but at a smaller body length in the Iki sample (260 vs 285 cm) and (2) the Iki specimens in general had heavier testes than individuals of the same body length in the Pacific sample Noting that the two samples were collected in similar months of the year and that the breeding season was the same between the two ocean areas, this can be interpreted to mean that males of the species in the Iki Island area start puberty at a smaller body length and perhaps at a younger age There is as yet no explanation of why adult males in the Iki area have heavier testes, but it does indicate that the bottlenose dolphins in the two ocean areas belong to different populations

11.4.7.4 Age and Body Length at the Attainment of Sexual Maturity

Males start rapid testicular growth at around 8 years of age The Iki dolphins had testis weight of about 45 g at age 8 years and attained about 200 g at age 12  years This rapid testis growth is observed at body length of 275-310 cm This bodylength range corresponded also with the stage where the

maturing,” and “mature” coexisted and is interpreted as the period when males were rapidly growing toward sexual maturity There was an apparent weight increase after body length of 300 cm, which could be a reflection of a positive correlation between testicular weight and body size Such was also observed in short-finned pilot whales (Figure 1212)

The period before age 8 years corresponded to an immature stage, when growth in testis weight was slow and spermatogenesis did not usually occur However, some individuals underwent spermatogenesis in limited seminiferous tubules at the end of this stage (ie, “early maturing” individuals) This was followed by the period of rapid testicular growth where some individuals underwent spermatogenesis in over 50% of the tubules (ie, “late maturing” individual) By 11 years, almost all the males attained the “mature” stage where 100% of tubules were spermatogenic However, some males remained in the “late maturing” stage until age 20 years in both the Iki and Pacific samples Thus, the process of male maturation seemed to proceed slowly, lasting several years and with broad individual variation, and it was difficult to determine clear subdivisions within it, which is similar to the case in other delphinids

Table 116 presents the relationship between maturity stage and age The youngest males in “early maturing” stage was 55  years old in both the Iki and Pacific samples, those in “late maturing” stage 85 years old (Iki), and those in “mature stage” 95 years old (Iki) or 115 years old (Pacific) Thus, the minimum length of the period from the beginning of spermatogenesis to the attainment of the “mature” stage was estimated for the Iki sample at 4 years Such estimation was not possible for the Pacific sample due to insufficient sample size This was the case for precocious males Slower maturation

(both Iki and Pacific) These individuals could have reached the “early maturing” stage at age 115 Such individuals in the Pacific sample reached the “late maturing” stage at 125 years at the earliest or 175 years at the latest (Pacific) and would attain the “mature” stage at some age after 185 years (Pacific) or 205 years (Iki) These were cases, supported by age and maturity data, of males that entered puberty at a late age and proceeded to maturation with reasonable speed, suggesting that some males needed almost 10 years to achieve the maturation process Although it is not supported by data, if a male that started the puberty at the earliest age (55 years) and proceeded to maturation process at the slowest speed, it would take almost 15 years to reach the mature stage at age 205 Such an extreme case is probably rare; male common bottlenose dolphins off Japan usually spend 4-10 years from early puberty to full sexual maturity

The mean age at the attainment of sexual maturity is estimated with acceptable sample size only for the Iki sample One method often used for this purpose is to fit a model to the relationship between age and the proportion of males sexually mature and to calculate an age when 50% are mature Sigmoid or linear models have been used for this purpose This method is not preferred because the estimate is not free from model-specific bias (DeMaster 1984) Kasuya et  al (1997, in Japanese) used the simple method of Hohn (1989), which is believed to be free from the bias This method is to add to the age of the earliest “mature” individual the proportions of nonmature individuals in the older age classes In the case of the Iki sample, this method calculates the mean age at the attainment of sexual maturity as

90 + 075 + 05 = 1025 years (SE = 0382)

TABLE 11.6 Testicular Maturity and Age of Male Common Bottlenose Dolphins in Japan

In the same way, the age at the attainment of the “late maturing” stage is calculated as 950 years (SE = 0456), where the sum of the proportions of “late maturing” and “mature” stages is used in the calculation Takemura (1986a, in Japanese) obtained 115 years as a mean age at the attainment of sexual maturity of males in the Iki Island area, but the definition of “sexual maturity” was not clarified

Information was insufficient for the Pacific sample to analyze age-related change in male maturity (Table 116) The oldest “immature” male was aged 105 and the youngest “mature” male 115, but some males could have attained the “mature” stage at higher ages, as indicated by males of the “late maturing” stage at ages 125 years and 175 years Similar exceptional late maturing cases occurred in the Iki sample, as in short-finned pilot whales off Japan (Figure 1212) Kasuya et al. (1997, in Japanese) ignored these exceptional individuals and estimated that the Pacific individuals of the species usually attain the “mature” stage at ages 11-13 years This was based on the estimation of the mean age at the attainment of the “mature” stage, 1220 years (SE = 0543), and that at the attainment of the “late maturing” stage, 11  years (SE = 0), using the method of Hohn (1989) It is possible that males in the Iki Island area attain sexual maturity at a younger age than those in the Pacific

The mean body length at the attainment of sexual maturity was calculated as the point where the proportion of mature individuals reached 50% The estimate varies with the age range of the sample used in the calculation, for example, either use of the whole sample with age range from 0 to the oldest individual or use of only individuals between the youngest mature and the oldest immature The former method results in a smaller figure than the latter because (1) the dolphins exhibit broad individual variation in body length at sexual maturity, (2) increase in body length ceases soon after the attainment of sexual maturity, and (3) therefore numerous old and relatively small mature individuals will be included in the calculation The first method also suffers from any historical change in growth

Kasuya and Marsh (1984) and Kasuya et  al (1997, in Japanese) questioned the value of the first method as a way of obtaining a growth parameter for a particular dolphin population Kasuya et al. (1997, in Japanese) extracted subsamples of individuals aged from 45 to 115 years (Iki) and from 45 to 185 years (Pacific), which nearly correspond to the ages where the process of maturation proceeds (Table 117) In the case of the Iki sample, the “early maturing” stage occurred at body length 260-309 cm and the “late maturing” stage at lengths over 290 cm Half of the individuals are expected to be at “late maturing” or “mature” stage when they are between 290 and 309 cm; thus, the mean body length at the attainment of the “late maturing” stage is about 300 cm (for this task both the “late maturing” and “mature” stages should be combined) In the same way, the average body length where 50% of individuals are “mature” is expected to be between 309 and 320 cm Takemura (1986a) obtained a body length of 2991 cm as the body length where 50% of males were sexually mature in the Iki Island area This figure is close to the mean body length at the attainment of the “late maturing” stage of Kasuya et al. (1997, in Japanese), although the method of calculation and the definition of maturity may not be the same

Males of the Pacific area were estimated to attain the “mature” stage at body length 290-299 cm Sample size was insufficient for further analyses of their growth parameters

11.4.7.5 Male Reproductive Ability The histology of 1-2 cm2 tissue samples taken from midlength of testes of common bottlenose dolphins revealed various levels of spermatogenesis (Kasuya et al. 1997, in Japanese) Some precocious males had some seminiferous tubules that were spermatogenic (ie, mature tubules) at age 55 years and body length 260-269 cm, but other individuals still had some immature tubules at age 205 years and body length 300-309 cm The average age when all the seminiferous tubules in the examined tissue were spermatogenic (ie, “mature” stage in the definition given earlier) was 1025  years (Iki sample) or 1220 years (Pacific sample) This definition of maturity stage

Testicular Maturity and Body Length of Male Common Bottlenose Dolphins in Japan

these “mature” males participated in mating and produced offspring One can only expect correlation between the testicular histology and behavior of the male

At some stage of the development of sexual maturation, a male cetacean will start to approach females for mating However, before these efforts end with mating and production of offspring, some social or physiological conditions may need to be satisfied If there is competition between males, a male must either win the competition or sneak in to mate with receptive females Whether an estrous female accepts every approaching male or chooses a particular male for mating will depend on the social structure, but we do not know whether a female chooses a certain male when there are multiple courting males Copulation must be concluded with ejaculation sufficient for fertilization, which would have some relevance to testicular histology Such a reproductive environment for males will not be the same between wild and aquarium environments The status of males who actually participate in reproduction is often called “social maturity” Some attempt has been made, without a clear conclusion, to match the state of social maturity with histological growth stages (Miyazaki and Nishiwaki 1978; Miyazaki 1984; Section 1055)

Reproductive success and its individual variation among males remain as one of the important fields of cetacean biology to be investigated Matching anatomical information on reproductive tracts against either reproductive behavior or genetic paternity evidence will help us reach understanding, but such information is currently unavailable for the wild common bottlenose dolphins around Japan To substitute for such data, Kasuya et al. (1997, in Japanese) compared testicular histology against spermatozoan density in the complex lumen of the epididymis, which functions for storage of spermatozoa produced in the testis They made a smear from the epididymis, air-dried it, and stained it with a water solution of toluidine blue for microscopy The density of spermatozoa was classified into five grades of “absent” to “extremely dense,” which was of a density found only in some epididymides but not in any testicular smear This followed the classification used for short-finned pilot whales (Section 12433 and Table 128) A difference of one density grade may sometimes be insignificant A possible but unconfirmed effect of preceding ejaculations on the epididymal sperm density was ignored

The epididymal smear was compared against testicular anatomy and the age of male common bottlenose dolphins (Table 118) The testis and epididymis were taken from the same side of each individual About 23% of males classified as “immature” through testicular histology had a small amount of spermatozoa in the epididymal smear, which was an evidence of spermatogenesis in some portion of the testis not sampled for testicular histology Similar cases, low levels of sperm production unidentified by testicular histology, were also reported in short-finned pilot whales aged 5-6  years (Kasuya and Marsh 1984) and in pantropical spotted dolphins aged 2 years (Kasuya et al. 1974) Testes of higher maturity

stages in general had high epididymal sperm density, but the two testicular maturity stages of “late maturing” and “mature” showed no difference in epididymal sperm density This suggests that males of the two maturity stages have a similar physiological capacity of reproduction In other words, these males can participate in reproduction if the opportunity is available

It should be noted that about 23% of males at the “late maturing” or “mature” stages had no spermatozoa or an extremely low level of spermatozoa in the epididymis One of the possible explanations is that most of the sample was obtained during January to April, which covered the period of the lowest mating activity (December-January) to the beginning of the mating season that extended from February to October Considering that one aquarium-reared adult male showed a seasonal testosterone cycle with the lowest level in January and February (Section 1142), it remains possible that at least some males of the species cease or decrease spermatogenic activity in the nonmating season Yoshioka et  al. (1993) reported an interesting experiment in which a male common bottlenose dolphin was trained to ejaculate in response to a signal from the trainer At the beginning of a series of such guided ejaculations, the semen did not contain spermatozoa, but the sperm density increased after more ejaculations This  suggests that external stimuli can

Relative Spermatozoa Density in Epididymal Smear of Common Bottlenose Dolphins from Iki Island, Viewed against Histology of Testis, Weight of Single Testis Removed from Epididymis, and Age

the testis to the epididymis This may happen in wild males of the species in response to visual or acoustic stimuli from other individuals

The density of spermatozoa in the epididymis was compared with testicular weight (Table 118) The epididymal sperm density tended to be lower with the testes of less than 100 g compared to heavier testes, but there was no clear density difference, and the mean density indices remained the same among testes weighing over 100 g It is noted that with the exception of one testis, all the testes weighing over 150 g were at the histological “mature” stage and that all the testes weighing less than 150 g were at the histological stage of “late maturing” or a lower stage (with the exception of one individual) (Figures 1110 through 1113) The lower bound for “mature” testis was 150 g and that for “late maturing” was about 100 g From this, it can be concluded that common bottlenose dolphins in the Iki Island area have the physiological capacity of reproduction if a testis weighs over 100 g or if testicular histology identifies them as “late maturing” This is the minimum requirement for reproductive capacity; males not satisfying either of these criteria are unlikely to be reproductive

This conclusion does not mean that male reproductive capacity is the same for testes weighing from 100 to 800 g or more It remains to be investigated whether males with larger testis have higher levels of testosterone and are more active sexually It seems to be a possibility that testis size has a positive correlation with body size and that males with larger testes are more likely to acquire more mating partners or that males with larger testes produce a larger quantity of spermatozoa and overwhelm ejaculates of other males It is generally believed that larger testes will benefit reproductive success in random mating or in a polygamous community (Ralls and Mesnick 2002) The mating system of common bottlenose dolphins is close to random mating, as in the case of striped dolphins (Section 10553)

The density of spermatozoa in the epididymis increased up to the age of 15 Then the correlation was lost (Table 118), which suggested that all the males over 15 years of age had a physiologically equal capacity to reproduce Some males at ages 10-15 had a sperm density grade of 2 or more, suggesting that they had a similar reproductive capacity to that of individuals over 15 years of age

The observations on common bottlenose dolphins in the Iki Island area can be summarized: (1) males having a testis of over 100 g or a testis at the histological maturity stages of “late maturing” or “mature” are physiologically capable of reproduction, (2) males aged 15 or older have the same capacity, and (3) males below age 15 can be considered to have the reproductive capacity if they satisfy the first condition The same species in the Pacific area tended to have smaller testes and sexual maturity appeared to be reached about 3 years later than in the Iki individuals However, the maturity criterion of 100 g will not cause a significant bias for the Pacific individuals because testicular development around the attainment of sexual maturity appears rapid

11.4.8.1 Identification of Sexual Maturity Biologists have classified female cetaceans as sexually mature if they have experienced ovulation, which is believed to be identifiable by the presence of a corpus luteum or corpus albicans in the ovaries Most female cetaceans conceive at the first ovulation; a few become pregnant at the second or third ovulation Although this is easily applicable by fishery scientists working on whale carcasses killed by a fishery, it is difficult to apply by scientists working on live wild cetaceans It takes about one year or more for the first ovulation to be followed by the first parturition, which can be more significant for population dynamics Further information on cetacean ovaries and morphology of the corpus luteum and albicans are available in Perrin and Donovan (1984) Recently, a question has been raised about the persistence of the corpus albicans of ovulation in some small toothed whales (see Sections 10561 and 11485)

In principle, pregnancy should be confirmed by the presence of a fetus in the uterus, but it can be determined with fair accuracy by the examination of the histology of the endometrium (Section 12445) Stretch marks in nonpregnant uteri are possibly useful for the identification of past pregnancy but require further calibration Lactation is usually identifiable through visual examination, but histological examination of the mammary gland increases accuracy Light brown coloration of the mammary glands of nonlactating baleen and sperm whales allowed visual determination of past lactation, but the method is not applicable to dolphins due to the small thickness of their mammary glands

Kasuya et al. (1997, in Japanese) confirmed field records of pregnancy and lactation using histology of the endometrium and mammary gland There were some cases where a fetus was lost before examination by scientists and pregnancy was determined by histology of the endometrium In principle, they identified female maturity by detecting a corpus luteum or albicans, but it was also determined by the presence of lactation or pregnancy for a small number of females where ovaries were removed by the fishermen The following is based on the study by Kasuya et al. (1997, in Japanese) conducted using materials listed in Table 112

11.4.8.2 Age at the Attainment of Sexual Maturity Both immature and mature females occurred at ages 55-85 in the Iki sample and 65-125 in the Pacific sample (Table 119) The mean age at the attainment of sexual maturity was calculated in the same way as for males to obtain a value of 691 (SE = 053) for the Iki sample and 919 (SE = 0716) for the Pacific sample It seems probable that common bottlenose dolphins in the Iki Island area attained sexual maturity about 2 years earlier than those in the Pacific area around 1980

Takemura (1986a, in Japanese) analyzed the reproduction of female common bottlenose dolphins by combining data of the 1979-1981 seasons contributed by my group and his own data for 1982-1985 and estimated the mean age at the attainment of

sexual maturity at “slightly below 7 years,” which was in good agreement with the results of Kasuya et al (1997, in Japanese)

11.4.8.3 Body Length at the Attainment of Sexual Maturity

The proportion of females sexually mature increased with increasing body length However, the body length where the number of immature and mature individuals were equal in a sample, that is, the body length where 50% are mature, was not the same as the body length where half of the individuals of a cohort attained maturity, that is, the mean body length at the attainment of sexual maturity, because small cetaceans cease growing soon after attaining sexual maturity and the growth curve is upwardly convex The former estimate is influenced by the age composition of the sample, that is, a sample biased toward older individuals will have smaller body length at 50% maturity The latter method is a more suitable indicator of growth of a species

In order to estimate the mean body length at the attainment of sexual maturity in the same way as used for males, Kasuya et al. (1997, in Japanese) selected individuals ranging from the age of 1 year below the youngest mature individual to the age of 1 year above the oldest immature individual, that is, 45-95 years for the Iki sample and 55-135 years for the Pacific sample (Tables 119 and 1110) Such selected samples revealed that immature and mature individuals coexisted at body lengths 260-299 cm in the Iki sample and 270-289 cm in the Pacific sample Around 1980, female common bottlenose dolphins attained sexual maturity at these body-length ranges Females in the Iki Island area attained sexual maturity at a greater body length and with greater individual variation than the Pacific individuals

The mean body length at the attainment of sexual maturity was estimated as a length where 50% of the females were sexually mature in the selected subsample mentioned earlier, which

was about 290 cm for the Iki sample and about 280 cm for the Pacific sample (Table 1110) Females in the Iki island area matured at a younger age (see above) and at greater body length

Takemura et al. (1986a, in Japanese) obtained 2725 cm for females in the Iki Island area as a body length where 50% of females are sexually mature This calculation included all the age classes of the sample collected in 1979-1985, and so the estimate was understandably lower than the mean body length at the attainment of sexual maturity obtained by Kasuya et al. (1997, in Japanese), that is, ca 290 cm, for the reason mentioned earlier

11.4.8.4 Female Reproductive Cycle Females of the short-finned pilot whales are known to exhibit distinct age-related decline in reproductive capacity (Tables 1213 and 1214), but striped dolphins show only a slight decline in pregnancy rate and a slight increase in resting females with age (Section 10511) Kasuya et al. (1997, in Japanese) presented the oldest ages of the common bottlenose dolphins at each reproductive stage:

Females ceased reproduction at the age 335-385 years, or 7-9 years before the age of the oldest females in each sample, but it seems to be premature to draw a firm conclusion on this topic from the limited analysis of this small sample Table 1111 shows age-related change in pregnancy rate for the Pacific and Iki samples If high pregnancy rate for ages

Sexual Maturity and Age of Female Common Bottlenose Dolphins off Japan

Sexual Maturity and Body Length of Female Common Bottlenose Dolphins off Japana

below 15 years were omitted from the analysis because the age classes included females in their first pregnancy, then agerelated change in pregnancy rate was not evident in the limited sample of ages above 15 years Thus, age-related decline in reproductive capacity is not evident in this species

Annual pregnancy rate is the probability of a sexually mature female becoming pregnant in a year, and the apparent pregnancy rate is the proportion of pregnant females in the total of sexually mature females Therefore, if sample bias is ignored, the annual pregnancy rate is calculated by dividing the apparent pregnancy rate by the gestation period in years As the gestation period of the common bottlenose dolphin was estimated at 1 year, annual pregnancy rate was 408% for the Iki sample and 333% for the Pacific sample The mean calving interval is the reciprocal of the annual pregnancy rate, that is, 245 years for the Iki sample and 3 years for the Pacific sample However, the standard errors of the annual pregnancy rates were about 4%, and the difference in pregnancy parameters between the two samples was not statistically significant (01 < p < 02)

The calculations of annual pregnancy rate and mean calving interval are valid only when the sample represents the composition of the population It cannot be applied to samples from fisheries that may selectively take particular reproductive stages (modern whaling prohibited lactating females) or operate in particular seasons (many fisheries operate in a particular season, and cetaceans often breed seasonally) Although the drive fisheries for dolphins operated in particular seasons, the drive was almost nonselective except for a possible bias in the opportunity for encountering particular types of schools

Both mating and parturition take place in an 8-to 9-month period from February/March to October among common bottlenose dolphins off Japan Kasuya et al. (1997, in Japanese) used samples taken by the drive fisheries mostly from January to April (Table 112), which was about 6-10 months after the previous parturition/conception peak and agreed with the time from the lowest mating activity to the beginning of the parturition/conception season (Figure 114) Therefore, their sample could cover all females that conceived in one mating season and probably were not biased by a seasonal effect on pregnancy rate, although it is uncertain whether the samples correctly represented the proportion of lactating and resting females We do not have information on the weaning season of bottlenose dolphins off Japan, which is expected to occur in a season of high food availability

The mean lengths of the lactation and resting periods were calculated by ignoring the uncertainty about the weaning season (Table 1111) The mean lactation period was 131 years for the Iki sample and 173  years for the Pacific sample Kasuya and Miyazaki (1981, in Japanese) reported that a female calf of 182 cm killed at Katsumoto on Iki Island had only milk in the stomach and no trace of solid food This calf was one of the individuals analyzed by Kasuya et al. (1997, in Japanese) and estimated at age below 1 year from body length Cornell et  al (1987) stated that common bottlenose dolphins born in an aquarium started taking solid food at age 3-5 months and the solid food occupied a major portion of the diet at age 9-12 months They also reported breeding intervals of 21-31 months with an average of 23 years for a female kept together with a male and their offspring This probably

Reproductive Status of Female Common Bottlenose Dolphins off Japan and Mean Reproductive Cycle Calculated Assuming Gestation of 12 Months

mated calving interval of common bottlenose dolphins off Japan, 245-300  years, is slightly longer than this but will reflect a difference in the nutritional environment

Bottlenose dolphin calves do not necessarily separate from their mothers at completion of weaning, but we do not have good data to estimate the length of the mother-calf bond for the Japanese population in the wild The age composition of the species taken by Japanese drive fisheries (Figure 115) has a trough at age 15 years (both sexes in the Iki sample and females in the Pacific sample) or 25  years (Pacific males) This trough is most probably created by the move of weaned calves to schools of immature individuals and suggests that departure of calves from the mother’s school starts within less than 6  months from the end of suckling, which lasts 131-173 years on average, or before the next estrus of their mothers

One of the resident communities of about 100 common bottlenose dolphins along the Florida coast was monitored for over 25  years Females had calving intervals of 2-10  years with an average of 5 years (Scott et al. 1996) The mother-calf bond was gradually loosened after 4-5 years, but some bonds lasted for 7-8 years (Wells et al. 1987) Japanese populations of the same species seem to have a much shorter mother-calf association and a shorter female reproductive cycle The difference is a reflection of habitat difference, that is, an open offshore habitat versus a closed inshore one The life history of the Florida community has achieved a balance with a low reproductive rate and low mortality rate in inshore water with fewer predators, while the Japanese populations have taken a strategy of high reproductive rate to match high predation and probably hunting pressure

The Indo-Pacific bottlenose dolphin inhabits coastal waters from Africa to central Japan A population of the species in an estuarine area near Adelaide is one of the well-studied populations, containing 74 individuals (Steiner and Bossley 2008) Juvenile survival rate to weaning was only 54%, and 9 females were observed to have weaned their calves successfully Their calving interval was 3 years (5 cases), 4 years (2 cases), 5 years (1 case), and 6 years (1 case) The average of the 9 intervals was 38 years In addition to these 9 cases, one female underwent estrus while nursing her calf and gave birth to the next calf when the elder calf was 19  years old The younger calf always accompanied the mother, but the older calf accompanied its mother with lower frequency after the birth This was probably a case of exceptionally early weaning due to early estrus of the mother and subsequent conception Weaned calves of this community usually accompany their mothers until the birth of the next calf, that is, for 28 years on average (pregnancy is assumed at 1 year) There were five cases where calves died before weaning, which caused early estrus, and the subsequent calving interval was only 17 years on average, which was strongly influenced by the timing of calf death The total average of the 15 observed calving intervals was 29 years, which was close to the average calving interval in Japanese populations of common bottlenose dolphins This suggests that the calving interval of Japanese

although we do not have a way to confirm this

Another study of the Indo-Pacific bottlenose dolphin was reported by Kogi et  al. (2004), who worked with a group inhabiting waters near Mikura Island (33°5′N, 139°36′E), about 200 km south of Tokyo Some of the members of this group were known to have visited or emigrated to the coast of Kamogawa and Tateyama (Section 113) near Tokyo or Toshima Island (35°44′N, 139°47′E), about 130  km south of Tokyo There were 19 cases of successful weaning followed by conception, which resulted in calving intervals of 3-5 years with an average of 35 years The total 26 known calving intervals, including 7 cases of death of calves before weaning, resulted in calving intervals of 1-6 years with an average of 34 years Using the data in Kogi et  al. (2004), I calculated a mean calving interval of 31 years for the seven cases where calves died before weaning Thus, calving intervals were slightly shorter if the calves died before weaning These seven cases included two cases of very short calving intervals of 1 and 2 years, due to the death of neonates at an early stage followed by conception within the same season or that of the next year Kogi et al. (2004) recorded two adult females that did not breed in their study period of 8 years, inclusion of which could increase to some degree the mean calving interval calculated earlier Neonates of Mikura Island always accompanied their mother when they were young but started to be separated at age 3-6 years with an average of 35 years

The age when the mother-calf bond dissolved in the Mikura community was similar to the mean calving interval This suggests that the end of the mother-calf bond and the next parturition are related and that, as suggested by Connor et al. (2000), the separation can be initiated by the near-term mother Connor et al. (2000) stated that calves often tended to live apart from their pregnant mothers and that females in near-term pregnancy intentionally avoided their calves The time between the end of suckling and the end of the mothercalf bond is still unknown Weaning is a slow process of switching nutrition from milk to solid food, so it is difficult to confirm the end of suckling in the wild

Although the difference was statistically insignificant, the data of Kasuya et al. (1997, in Japanese) suggested that common bottlenose dolphins in the Iki Island area had a shorter calving interval than those in the Pacific area, which could have been due to an about a 5-month shorter lactation period and a 15-month shorter resting period (see Table  1111) Such differences as well as a growth difference in the juvenile stage between the two ocean areas (Section 1146) were likely sensitive to the nutritional environment; it was possible that common bottlenose dolphins in the Iki Island area lived in a better nutritional environment than those in the Pacific area (Section 1151)

11.4.8.5 Ovulation Rate Belugas develop Graafian follicles during pregnancy They do not disappear with ovulation but develop during pregnancy into numerous accessory corpora lutea, which are

nancy (Brodie 1972) Such a case is known among cetaceans only in the beluga, although it is not rare among mammals The corpus luteum of ovulation degenerates into a corpus albicans and is believed to persist for life in a form indistinguishable from the degenerated corpus luteum of pregnancy (Takahashi et al. 2006) Cetacean biologists formerly believed that the total number of corpora could be used as an indicator of past ovulations However, a recent study on an Indo-Pacific bottlenose dolphin that was monitored for the history of estrus and pregnancy in an aquarium suggested that only the corpus luteum of pregnancy would remain in the ovary as a corpus albicans (Brook et al. 2002) This important suggestion needs further confirmation (Section 1056)

Keeping in mind this question, I will introduce the analysis of Kasuya et al. (1997, in Japanese) Figure 1114 presents the size distribution of corpora lutea and albicantia with female age, where two or more age groups are combined into one For example, females aged 55 and 65 years are represented at age 5 years, which actually included animals at ages between 5 and 7 (Section 1141) The diameter was calculated as a geometric mean of three dimensions The geometric mean is better suited for analysis than the arithmetic mean as an indication of volume of the compressed flat corpus albicans Observing that the diameter of degenerated corpora albicantia peaked at around 5 mm with downward tailing to 2 mm and that the peak height increased with increasing age, Kasuya et al. (1997, in Japanese) concluded that the corpus albicans decreased in size with time to a diameter of around 5  mm on average; that is, the corpus albicans lasted in the ovaries for life However, for their conclusion to be more convincing, they should have confirmed that the rate of age-related increase of corpora was explained by the annual ovulation rate This may not be an easy task because the age composition of the female sample should also be taken into account

corpora albicantia were unidentifiable from surrounding tissues with the naked eye

The average annual rate of accumulation of corpora in the ovaries was calculated by regression of the total number (y) of corpora on age (x, years), at 0458 (SE = 0052) for the Iki sample and 0435/year (SE = 0033) for the Pacific sample (Figure 1115) The two figures were not significantly different statistically The same method applied to each school separately gave a range of 044-055 (mean 048) for the Iki sample and 022-059 (mean 046) for the Pacific sample, and again there were no significant differences among schools (Kasuya et  al. 1997, in Japanese) The regression equations were calculated using individuals aged over 10 years so that all were sexually mature Inclusion of younger individuals causes a bias; immature individuals must be excluded from the calculation or dealt with as having zero corpora The time difference from 1 corpus to 2 corpora is not the same with that from no corpus to 1 corpus

The average annual ovulation rates thus calculated represent the true figures only if both age at sexual maturity and annual ovulation rate remained stable during the period covered by the sample, or from around 1940 to 1983 If during the 40-year period there was a decline in age at the attainment of sexual maturity or an increase in annual ovulation rate, the calculation will bias the recent ovulation rate downward

Kasuya et  al. (1997, in Japanese) compared the average annual ovulation rates estimated for the past 40  years with annual pregnancy rates during the sampling period, that is, the late 1970s to the early 1980s The average number of ovulations per pregnancy is obtained by dividing the annual ovulation rate by the annual pregnancy rate:

Iki sample: 0458/0408 = 112 Pacific sample: 0435/0333 = 131

This means that the proportion of ovulations followed by pregnancy was 891% in the Iki Island area and 765% in the Pacific area These figures are apparently reasonable and do not suggest rejection of the possibility that both the corpus luteum of pregnancy and that of ovulation remain as corpora albicantia in ovaries of the common bottlenose dolphin This does not necessarily mean that Kasuya et al. (1997, in Japanese) identified all the ovulations as corpora

Table 1112 compares life history parameters of common bottlenose dolphins between the Iki and the Pacific samples, both collected from the catch of driving operations The Iki sample of 446 individuals was from 2634 in 7 schools driven into a harbor near Katsumoto, Iki Island, in January-March from the Shichiriga-sone Bank (33°56′N, 129°30′E) between Iki Island and the Tsushima Islands The sea surface temperature

lower bound of the temperature of waters inhabited by this species (Table 101); Dall’s porpoises occurred about 150 km northeast of this area (Section 931 as illustrated in Fig 1 of Miyashita and Kasuya 1988) Thus, the Iki sample was collected in waters close to the northern limit of the range of the species in the season in the East China Sea and the Sea of Japan (Figure 113)

The Pacific sample of 281 individuals was from 540 in 9 schools driven in Arari on the Izu coast and Taiji in Wakayama Prefecture Taiji is located about 300 km southwest of Arari The Arari sample (37) was taken in August, but the other 244 were obtained from 8 schools driven in Taiji in January-April The northern limit of the range of this species off the Pacific coast of Japan extends in summer to latitudes 42°N-43°N off southern Hokkaido and retreats in winter to around latitude 35°N, which is also close to the southern limit of Dall’s porpoises in winter It is possible but has not been confirmed that the Arari and Taiji samples were from a single population

The distribution pattern of the common bottlenose dolphin around Japan suggests that there is little opportunity for the schools inhabiting the Arari/Taiji area to meet with schools in the Iki Island area in winter, but it is possible that they meet in summer in the Tsugaru Strait area (41°30′N, 140°30′E), which is situated between Hokkaido and Honshu and connects the Sea of Japan and the Pacific Thus, sample locations alone cannot be evidence that the Iki and Pacific samples belong to separate populations However, some differences in life history parameters (Table 1112) suggest that these samples represent different populations

I classified life history parameters listed in Table 1112 into the following three categories based on expected response to environmental variation, such as in nutrition:

1 Parameters that are likely to show a quick response to environmental change

2 Parameters in which response will occur quickly at the individual level but will take time before being identified at the population level

3 Parameters that are less affected by environmental change or those for which we do not know the direction of likely change

The first group of parameters includes annual pregnancy rate and growth in the juvenile stage

The second group includes age at the attainment of sexual maturity, body length at the attainment of sexual maturity, and asymptotic body length For example, nutritional improvement will quickly result in the improvement of body size of juveniles, but it will take nearly 10 years before its effect is identified as a change in body length at the attainment of sexual maturity

The third group includes breeding season, age at physical maturity, age at cessation of testicular growth, testicular weight in adults, and annual ovulation rate Improvement of nutrition of adult females will increase the proportion of estrus followed by conception, will shorten the resting period,

and will reduce neonatal mortality, but it is unknown how the total of these responses would change annual ovulation rate

Among the seven characters with difference between the two samples, five characters belonged to the first and the second groups, and all the differences observed were in the direction expected to occur if the Iki population had a more favorable nutritional environment No geographical difference was identified in asymptotic body length This suggests that some environmental change toward the improvement of nutrition of the common bottlenose dolphin in the Iki Island area occurred recently or during the 1970s to 1980s

11.5.2.1 Food of Common Bottlenose Dolphins off Japan

First, the nutritional requirement of common bottlenose dolphins is reviewed here Tobayama and Shimizu (1973, in Japanese) kept 11 common bottlenose dolphins in an outdoor tank and examined per capita food consumption and change in body weight The daily food requirement to maintain body weight varied from 608% of body weight for a young animal measuring 236 cm (n = 1) to 354% for larger animals

measuring 290-310 cm (n = 5) For the latter five individuals, the daily food requirement decreased from about 45% to 3%, while the water temperature increased from 115°C to 243°C. The lower bound of the temperature range of the experiment was the lowest sea surface temperature recorded during sightings of this species in the wild in the western North Pacific (Table 101)

Takemura (1986b, in Japanese) reported the stomach contents of 56 common bottlenose dolphins, most of which were collected from those driven at Iki Island for culling purposes in February 1983-1985 or after the collecting activities of Kasuya et  al. (1997) during 1979-1980 (Section 34) He removed food remains in the first and the second stomachs using a 32-mesh sieve Many of the examined stomachs were almost empty, and most of the food remains were limited to fish otoliths and squid beaks, which will reflect time at the start of drive operations as well as the length of time between drive and slaughter Thus, species composition could be determined but not the weight of the prey ingested

Out of the 56 common bottlenose dolphins examined by Takemura (1986b, in Japanese), 15 had fish remains, 24 had squid remains, and 1 had shrimp remains (Table 1113) The major food items consumed by common bottlenose dolphins in the Iki Island area were fish and squid This was the same

Life History Parameters of Common Bottlenose Dolphins Compared between the Pacific and Iki Samples

among three species of dolphins taken in the area: common bottlenose dolphins, false killer whales, and Pacific whitesided dolphins The Risso’s dolphins taken in the area were found with only squid in the stomachs The cephalopod species listed in Table 1113 were squids except for one octopus found in the stomach of a false killer whale The common bottlenose dolphins had preyed on numerous species of fish and squid The importance of fish and squid in the diet of common bottlenose dolphins is also known for the same species in coastal waters from Florida to Texas, but there was some geographical variation in the proportions of the two taxa (Barros and Odell 1990)

The common bottlenose dolphins consumed the highest variety of food items among the four species of dolphins culled in the Iki Island area (Table 1113) This result may partly be exaggerated by the fact that materials examined by Takemura (1986a, in Japanese) included stomachs from a small number of specimens incidentally killed in a seine net fishery, but it is true that the species lives on a variety of food items The jaws and teeth of bottlenose dolphins are intermediate between the extremely large and strong teeth of killer whales and false killer whales, and the delicate beak of Stenella spp appears to be adapted for feeding on a broad spectrum of food species

Takemura (1986b, in Japanese) reported total lengths of fish consumed by dolphins in the Iki Island area: 10-100 cm for bottlenose dolphins and 10-40 cm for Pacific white-sided dolphins Although he did not find yellowtail in his sample, Kasuya (1985) reported that one Pacific white-sided dolphin and four false killer whales had remains of yellowtail in the stomach The estimated total length of the yellowtail was 37 cm for the Pacific white-sided dolphin and 60-87 cm for the false killer whale

Geographical variation in food habits of the common bottlenose dolphin was reported by Suisan-cho Chosa-kenkyu-bu (Investigation and Research Department of Fisheries Agency) (1969, in Japanese), based on the results of the studies by scientists of the Saikai Regional Fisheries Research Laboratory (SRFRL) conducted in response to the first incident in the dolphin-fishery conflict in the Iki Island area that started around 1966 The second phase of the conflict started in 1978; the catches were investigated by my team in 1979-1980 (Kasuya 1985; Kasuya et  al. 1997, in Japanese) and by a Fisheries Agency Team in 1981-1985 (Tamura et al 1986, in Japanese; Takemura 1986a,b in Japanese) (Section 34) Scientists of the SRFRL collected the stomach contents of common bottlenose dolphins in the East China Sea and northern Kyushu area; grouped the contents into three major groups of squid, pelagic fish, and bottom fish; and examined geographical and seasonal variation in the composition (Table 1114) Their findings were that the dolphins had consumed mainly bottom fish in coastal waters and pelagic and bottom fish in offshore waters Squid made up 16%–28% of the total food items consumed

All the available information on stomach contents of common bottlenose dolphins indicates a broad spectrum of prey and their great adaptability in feeding Thus, a list of food items obtained in a particular season or in a particular area may not apply to a different season or area It would also be difficult to evaluate the nutritional environment based on information on the availability of some limited food items in the ocean

11.5.2.2 Effect of Dolphin Fisheries If the size of a population of common bottlenose dolphins decreases due to hunting, per capita food availability will increase and the nutritional environment will be improved I have suggested the possibility that overhunting of striped dolphins off the Izu coast resulted in the improvement of their growth and reproduction (Sections 1057 and 10512) The broad prey spectrum of the common bottlenose dolphin and

Food Habits of Common Bottlenose Dolphins in the Iki Island Area Indicated by the Number of Individuals Found with Particular Food Items

Geographical Difference in Food Habits of Common Bottlenose Dolphins Indicated by Proportion of the Number of Food Items Identified in the Sampled Stomachs (Individual Variation within the Sample Was Ignored)

gest that depletion of dolphin species other than common bottlenose dolphins would also help improve their nutritional environment

The history of exploitation of bottlenose dolphins in Japan was reviewed by Kasuya (1996, in Japanese) based on then published statistics, including those in Kasuya (1985) and Kishiro and Kasuya (1993) This is revisited where using some additional statistical data Common bottlenose dolphins have been hunted at least since the 1960s at Nago (26°35′N, 127°59′E) in Okinawa, Goto Islands (33°N, 129°E) in the East China Sea, Tsushima Island (34°30′N, 129°20′E) and Iki Island in Tsushima Strait, and Taiji and Izu on the Pacific coast of central Japan

The community of Nago operated an opportunistic dolphin drive historically, but it was replaced in 1975 by a crossbow fishery (Section 28) The statistics for 20  years from 1960 to 1994 (except for 1971 and 1973) are available in Kasuya (1996, in Japanese) Additional statistics for 1960-1975 are available in Miyazaki (1980b, in Japanese) and those for 1960-1982 in Uchida (1985, in Japanese) Some disagreement exists among these sets of statistics, but it is negligible These statistics are compiled in Table 213 for 1993 and later seasons when the fishery was operated with a catch quota and in Table 322 for years before that date Before 1981, when the catch of bottlenose dolphins was sporadic, Nago recorded a total maximum estimated take of 209 common bottlenose dolphins in 21 years, or an annual mean of 10 This level of take was unlikely to have had a significant effect on the sample of Kasuya et al. (1997, in Japanese) collected in the far north, that is, the Pacific waters off Taiji and Izu or waters around Iki Island Since 1981, when take of the species became a regularly activity, Nago recorded the annual takes of 0-77 co mmon bottlenose dolphins with a total of 224 individuals in the 12 years 1981-1992 The annual average was only 19 The crossbow fishery was placed under a quota system in 1993: initially 10 bottlenose dolphins plus takes of several other delphinid species (Table 213)

Nagasaki Prefecture, including Goto, Tsushima, and Iki, had a long history of dolphin fisheries (Sections 33 through 36) Referring to the catch statistics of bottlenose dolphins in Nagasaki Prefecture since 1965, Kasuya et  al (1997, in Japanese) stated an impression that the catch could have been large enough to result in the local depletion of the species The records of dolphin drives for 1944-1966 in Nagasaki Prefecture are available in Table 37 The table lists a total take of 9977 dolphins in 33 drives of 4 species during the 22-year period, which excludes one drive in 1963 for which number and species of dolphins were not reported Twentyseven drives (82%) of the 33 drives were in the 9 years 19581966, suggesting the possibility that statistics for the earlier 14 years 1944-1957 were incomplete Out of the 33 drives, 10 were made on bottlenose dolphins, 2 on the mixtures of bottlenose dolphins and false killer whales (90 in total), and 1 on a mixture of bottlenose dolphins and Risso’s dolphins (380 in total) By assuming that half of the animals in the

that 3498 dolphins, or 562% of those of known species, were bottlenose dolphins (13 drives) By assuming the same proportion for 3738 small cetaceans of unknown species (20  drives), a total take of 5599 bottlenose dolphins, or an annual mean of 243, is estimated for the 23 years from 1944 to 1966 This is not corrected for the possible large catches during 1944-1957, which includes the peri-World War II period (Sections 22, 23 and 383)

The statistics for subsequent years and prior to 1979 when I started sampling (Kasuya 1985; Kasuya et  al. 1997, in Japanese) are available in Table 34, which included 1116 bottlenose dolphins as the total take of the species in Nagasaki Prefecture during the 12 years from 1967 to 1978 This figure plus an estimated additional 525 gives a figure of 1641 bottlenose dolphins, or an annual mean of 137 taken during the 12 years in Nagasaki Prefecture The additional 525 was estimated by applying the proportion 562% obtained earlier to 934 dolphins recorded in Table 34 as nezumi-iruka, which is currently unidentifiable to species

Thus, the mean annual catch of bottlenose dolphins in Nagasaki Prefecture in 1944-1966 could have been about 250, which is likely an underestimate due to incomplete statistics, and about 140 during 1967-1978 These figures are rough estimates of the fishing pressure on the species in the Tsushima Strait area before the drives for culling purposes that started in 1972 and expanded in 1977 with a large-scale drive operation under the leadership of the Iki fishermen and with financial support by the Fisheries Agency of Japan and the prefectural government (Section 34)

The abundance of bottlenose dolphins was estimated at 35,000 in winter for waters surrounded by 126°E-131°E and 25°N-35°N, which includes the eastern half of the East China Sea, the Tsushima Strait area, and the Pacific waters around the Okinawa Islands (Table 1115; Miyashita 1986b, in Japanese) This estimate needs to be treated with caution before use for management purpose for two reasons: (1) a large coefficient of variation and (2) inclusion of the Pacific waters around the Okinawa Islands It is questionable whether the dolphins that winter in the Okinawa Islands area belong to the same population that winters in the Iki Island area in Tsushima Strait further north They might instead belong to another more southern population Even if the possibility is accepted that the Okinawa dolphin might migrate in summer to the Tsushima Strait area or to the coast of Taiji or the Izu Peninsula, uncertainty remains whether they mix with dolphins wintering in these northern areas

If the abundance of bottlenose dolphins is assumed to be proportional to size of the area of distribution, which is certainly incorrect, and if those in the Okinawa Islands area are excluded, the abundance of the species in the eastern East China and the Tsushima Strait area is nearly two-thirds of the central estimate obtained earlier, that is, 23,000 This figure and the broad confidence interval suggest the possibility that the annual take has exceeded 1% of the population and suppressed the abundance to some degree Exploitation of other dolphin species could also have made some contribution to

improving the nutritional environment of bottlenose dolphins in the Iki Island area

Bottlenose dolphins along the Pacific coast of Japan were hunted by fisheries along the Izu coasts and off Taiji Statistics for Taiji in Wakayama Prefecture are given in Table 317 for 1963-1994 and those for the entire Wakayama Prefecture in Table 318 These two sets of statistics are from different sources and show some contradiction Table 317, which I believe to be the  more reliable, shows a total catch of 666 bottlenose dolphins (annual range of 3-103 and average 39) during the 17 years from 1963 to 1979, when the annual catch of short-finned pilot whales was 52-479 (average 166) and that of striped dolphins 331-2397 (average 916)

Because bottlenose dolphins were not preferred by the local people, the catch of the species was at a relatively low level in Taiji before 1980, which was the only location of significant dolphin hunting in Wakayama Prefecture The current drive fishery in Taiji started in 1969 for short-finned pilot whales and expanded to striped dolphins in 1973 (Section 35) It was in 1980 that the fishermen found a new market for bottlenose dolphins for use as animal food in zoos and increased the catch of the species With the increase of supply of bottlenose dolphins, sales for human consumption in the local market apparently increased (Kasuya et al. 1997, in Japanese) For these reasons, the catch of bottlenose dolphins increased beginning in 1980 and reached a peak of 1670 in 1987 Then the catch, reaching the quota that started in 1993 only in the year 2000, declined to less than 200 in recent

the fishery (Table 319 and 63)

The major target of the dolphin-drive fisheries on the Izu coast in Shizuoka Prefecture was striped dolphins, and the catch statistics were quite incomplete until 1971 Only Arari left catch statistics from January 1942 to April 1957, which were published in Kasuya (1976, in Japanese) and Kasuya (1996, in Japanese) and also cited in Table 314 During the period of about 14 years, catch figures are available by species for only about 8 years from 1950 to 1957, when a total of 44,672 dolphins of all species were caught, including 300 bottlenose dolphins (annual range 0-103, mean 43) (Table 314)

The statistics for the catch of dolphins in the entire Shizuoka Prefecture, which was almost the same as the catch on the Izu coast, are complete only for years since 1972, when the Fisheries Agency started collecting statistics (Table 316) During the 9 years from 1972 to 1980, Shizuoka Prefecture recorded a take of 274 bottlenose dolphins (annual range 0-120, mean 30)

Thus, the average annual catch of bottlenose dolphins along the Pacific coast was less than 50 in Shizuoka (Tables 314 and 316) or Wakayama Prefectures (Table 317) and the total average annual catch did not reach 100 during the years before 1981 when sampling by Kasuya et al. (1997, in Japanese) began The catch of the species along the Pacific coast underwent a sudden increase to an annual mean of 569 during 1980-1984 with the contribution of the Taiji operation This catch increase could have had some effect on pregnancy rate of the samples collected in 1981-1983, but the time could have been insufficient for other life history parameters estimated from the sample to be affected

The abundance of common bottlenose dolphins was estimated at 37,000 individuals in summer (Table 1115) for the Pacific area north of Yakushima Island (30°23′N, 130°35′E) and west of 145°E, which includes waters within 300400 km off the Pacific coast of Japan This range was much broader than the operation range of the drive fisheries, which extended at most to about 40  km from Taiji port (Kishiro and Kasuya 1993), and we do not know the population structure of the species within the broad geographical range used for the abundance estimation However, it is possible to say that the abundance of common bottlenose dolphins in the Pacific area was of similar magnitude as or slightly greater than that of the same species in the eastern East China Sea and Tsushima Island area This leaves the possibility that the common bottlenose dolphins in the Iki Island area received greater fishing pressure during 1960-1980 than the population of the same species off the Pacific coast of Japan The striped dolphin population off the Pacific coast of Japan declined during the 1960s to the 1970s and could have influenced the feeding environment of the common bottlenose dolphins, which remains to be investigated

In summary, it is fairly sure that some differences existed in the late 1970s and early 1980s in the life history parameters between populations of common bottlenose dolphin in the Iki Island area, the eastern East China Sea, and the Pacific area off Japan The differences identified were in annual

Abundance of Common Bottlenose Dolphins around Japan Estimated by Shipboard Sighting Surveys Assuming g(0) = 1

body length of immature individuals but not on other life history parameters such as breeding season and body length and age at the attainment of physical maturity The former group of parameters was more likely to respond to changes in the nutritional environment or in population density than the latter, and the difference was in the direction expected if nutritional environment was better in the Iki Island area Although this apparently agreed with greater absolute catches of the species by local fisheries in the Iki Island area than those in the Pacific area before the time the samples were collected, full understanding of the background of the life history differences must await further information on the abundance of the species, the geographical range of the exploited populations, and the structure of the ecosystem, including other dolphin species as well as major fishery resources