Finless Porpoise

Some languages distinguish between dolphins and porpoises, but the Japanese language does not; the word iruka applies to both the Delphinidae and Phocoenidae as well as some species of other families, such as the beluga and the river dolphins However, in this book, I use iruka only for the three families of the Delphinoidea (Delphinidae, Phocoenidae, and Monodontidae) and use kawa-iruka (river dolphins) for the taxonomically distinct groups of river dolphins (Chapter I)

The Taxonomy Committee of the Society for Marine Mammalogy proposed recognition of two species in the genus Neophocaena, that is, N. phocoenoides inhabiting waters from the South China Sea to the Indian Ocean and Neophocaena asiaeorientalis in waters of the Yangtze River, Yellow Sea, Korean coast, and Japan The former has a broad tubercular area on the back and the latter a narrower one Although I have no intention of rejecting the proposal, this chapter is based on the earlier usage of only one species in the genus The main reasons for this are that this chapter was drafted before the new taxonomic opinion become available; that classification is less important, in my opinion, for conservation biology than correct identification of units to conserve; and that we will be better able to revisit the systematics of the genus Neophocaena when we have accumulated more information on geographical variation within it (also see the “Note” at the end of this chapter)

The finless porpoise, Neophocaena phocaenoides (G Cuvier, 1829), is one of the three species of the family Phocoenidae inhabiting Japanese waters and is the only warmwater species It has been recorded in coastal and inshore waters, south of Hime near Mawaki (37°18′N, 137°12′E), on the west coast of Toyama Bay (37°N, 137°15′E) on the Sea of Japan coast, and south of Sendai Bay (38°20′N, 141°15′E) on the Pacific coast It usually inhabits waters shallower than 50 m The harbor porpoise also inhabits the continental shelf area within the 200 m isobath from northeastern Japan to the Okhotsk and Bering Seas (Taguchi et al 2010), which partially overlaps with the range of the finless porpoise (Kasuya 1994, in Japanese) Dall’s porpoise, another member of the family, also inhabits cold waters like the harbor porpoise, but it occurs in offshore waters, usually outside the continental shelf area

Among the three species of Phocoenidae off Japan, the  finless porpoise is identified by the smallest body size, the absence of a dorsal fin, and a broad rostrum on the skull The other two species have a more pointed rostrum when the skull is seen in dorsal view (Table 81) The relationship between rostrum width and skull length of the finless porpoises is shown in Figure 83

The finless porpoise is one of the smallest species of Cetacea together with the tucuxi, Sotalia fluviatilis, in the Amazon River (Caballero et al. 2007) and a subspecies of the spinner dolphin, Stenella longirostris roseiventris, in the Gulf of Siam (Perrin et al. 1989, 1999) Although some individuals of the species in Japanese waters exceed 190 cm in body length, those in the Indian Ocean grow to only 150-160 cm The finless porpoise has a round head and lacks the slender rostrum often seen among dolphins and river dolphins; with its blunt head it resembles the much larger short-finned pilot whale There are 16-20 teeth in each jaw, equipped with a crown of a flat spade shape, with exception of the anteriormost 1-2 pairs, which are pointed (Kasuya 1999) The tooth crown is often worn off, and the typical morphology is not seen in old individuals The height and shape of the dorsal ridge are geographically variable The entire body of the adult Japanese finless porpoise is grayish white with a paler ventral side

In Japan, the finless porpoise has been known by numerous names, presumably reflecting frequent contact with humans in the coastal habitat Yamashita (1991, in Japanese) collected over 20 local names of the species that appeared in the literature, which could be classified into five groups (Table 82) He attributed one of them, bozu (shaven head, or Buddhist monk) group, to the shape of the head, name or nami (smooth, or wave) group to their swimming style, suna (sand) group to the tubercles on the dorsal ridge, and same (shark) group to sharks, but he reserved conclusion on the origin of the gondo group I would attribute the origin of the gondo group to the shape of the head, which is proportionally large and blunt as also seen in the short-finned pilot whale (ma-gondo), false killer whale (oki-gondo), and Risso’s dolphin (hana-gondo)

Yamashita (1991, in Japanese) stated that names of the suna group are still in use in Tokyo and the nearby area, same group in Ise Bay (34°45′N, 136°45′E), bozu group in Okayama Prefecture in an eastern part of the Inland Sea, gondo group in Hiroshima Prefecture in a western part of the Inland Sea, and name group in the broad Inland Sea area I had the impression in the late 1970s, when I studied finless porpoises in the Inland Sea, that name or name-no-uo (name fish) were used in the eastern Inland Sea and de-gon or ze-gondo in the western Inland Sea (Kasuya and Yamada 1995, in Japanese)

The finless porpoise was first described, as Delphinus phocaenoides, by G Cuvier in 1829 (Le Règne Animal 1: 291, not seen) This was followed by several alternations of the scientific name Gray (1846) moved this species to

a newly erected genus Neomeris However, Palmer (1899) found that Neomeris had been preoccupied by a species that was believed to belong to the Coelenterata and created another new genus Neophocaena for the finless porpoise Thus, the scientific name of the species is expressed as N. phocaenoides (G Cuvier, 1892), with the author’s name in parenthesis

However, the history of the scientific name of this species was more complicated, with incidents of misprinted use of Meomeris (Gray 1847) and Neomeris (Coues 1890) or temporary resurrection of Neomeris Thomas (1922) and Allen (1923) agreed with the earlier taxonomists in the view that Neomeris is invalid, but believed that Meomeris created by

of using that genus name Soon after this, Thomas (1925) found that the Neomeris, which was once thought to belong to the Coelenterata, was in reality a calcareous alga and considered that Neomeris was valid for the finless porpoise This was accepted for almost 36  years, until Hershkovitz (1961) found that Neomeris was preoccupied by a polychaete in 1844 and once again advanced the once abandoned name Meomeris However, in the same year, due to modification of the International Code of Zoological Nomenclature to reject misspelled names, the current genus name Neophocaena came into use Further details are in Hershkovitz (1966) and Rice (1998)

The type specimen of the finless porpoise that was the basis of the taxonomic arguments was stated to have been from the Cape of Good Hope, but this was questioned and the specimen was thought to have been presumably brought from India (Allen 1923) No evidence exists on the presence of this species along the coast of South Africa (Rice 1977)

The Japanese finless porpoise was first described as Delphinus melas by Temminck and Schlegel (1844) on pages 14-16 of Fauna Japonica Les Mammifères Marins based on a specimen presumably from Kyushu on the East China Sea The authors and date of publication of this reference are from Kuroda (1934, in Japanese), but Hershkovitz (1966) considered the author and the date to be Temminck (1841) based on the same publication I am unable to determine the source of the difference

The distribution of the finless porpoise has been summarized by Kasuya (1999) and Amano (2003a) Reeves et al. (1997) recorded the occurrence of the species by range country, and Parsons and Wang (1998) offered information on the species in Southeast Asia The following description of distribution of the species is based on these works Finless porpoises are marine cetaceans inhabiting the coastal waters of tropical Asia They enter into mangrove swamps and may enter into some riverine habitats, for example, 60 km above the mouth of the Indus River, 40 km into the Brahmaputra River, and 20  km into the Yalu River An endemic population of the species occurs in a 1670 km stretch of the middle and lower reaches of the Yangtze River

Finless porpoises have often been stated to inhabit coastal waters up to a depth of 200 m or up to the outer edge of the continental shelf This does not correctly describe their habitat in Japanese waters Shirakihara et al. (1994) recorded the relationship between density of finless porpoises and water depth in Tachibana Bay (32°30′N-32°45′N, 130°45′E-130°10′E) and the connected Ariake Sound (32°30′N-33°20′N, 130°10′E-130°35′E) in western Kyushu The Ariake Sound is a shallow inland body of water approximately 90 × 20 km (maximum depth of about 50 m, but mostly less than 20 m)

Distinctive Skull Characters of Three Phocoenidae Species around Japan

TABLE 8.2 Vernacular Names of Finless Porpoise in Japan

openings is through Tachibana Bay, which is connected to the East China Sea through a broader opening Another narrow passage connects the Ariake Sound to the Yatsushiro Sea, which opens to the East China Sea through several narrow passes and is not inhabited by finless porpoises In Tachibana Bay, which has an area of about 20 × 30 km, the survey covered waters up to the depth of 70 m and found that finless porpoise density declined at depths greater than 50 m, but some animals occurred at even greater depth Shirakihara et  al. (1994) surveyed the entire Ariake Sound and recorded finless porpoises in the whole depth range but with higher density in shallower waters This observation suggests that the finless porpoise prefers waters shallower than 50 m

The Inland Sea, which is the largest habitat of finless porpoises in Japan, has an average water depth of 31 m and a maximum depth of about 98 m near Hayasui Pass (33°20′N, 132°00′E) at the southwestern entrance of the sea Most of the Inland Sea is shallower than 40 m From surveys conducted in the Inland Sea in 1976-1978, Kasuya and Kureha (1979) noted that sighting of finless porpoises was limited approximately within a depth of 40 m and that even within this depth range density rapidly declined with an increasing distance from the shore The sighting rate within 1 nautical mile (1851 km) from the shore was 257/survey of 10 nautical miles The rate declined to 119 in the 1-3 nautical mile range and to 024 in waters beyond 3 nautical miles from the nearest shore The 10-fold density decline within 3 nautical miles suggests that density is not uniform within 1 nautical mile of shore (Kasuya and Kureha 1979)

Based on sightings of finless porpoises during the cruise of Miyashita et al (1995) in the Yellow Sea in 1994, Reeves et al (1997) correctly stated that some finless porpoises occurred in waters 240 km from the shore However, it should be noted that there is shallow water extending from the Chinese coast eastward and the maximum depth in the Yellow Sea is located near the Korean coast or to the east of the middle point of the sea The midpoint between the two shores is over 200 km from the coasts and is only about 50 m deep The presence of finless porpoises in the offshore Yellow Sea should be considered to be due to the shallow water depth

Korean scientists repeated cetacean sighting cruises off the west coast of the Korean Peninsula (eastern part of the Yellow Sea) and reported high density of finless porpoises in the archipelago (eg, An et al. 2009) While the sighting track lines were evenly distributed from the Korean coast to the longitude of 123°25′E, finless porpoises occurred only to the east of c. 125°30′E or about 90 km from the west coast of the Korean Peninsula and within the coastal archipelago These porpoises were sighted at depths less than 50 m (personal communication from Y An) This indicates that the habitats of finless porpoises in China and Korea are separated by deeper water in the eastern Yellow Sea However, this does not preclude a possibility that the habitats meet on the northern Bohai coast

The distribution of finless porpoise is apparently affected by both distance offshore and water depth; thus, if water is

distant from the shore We do not know what kind of oceanographic factor might be functioning in the apparent correlation among finless porpoise, water depth, and distance offshore There have been statements that finless porpoises aggregate in a water channel between islands or near the tip of a cape (Kasuya and Kureha 1979) or that they prefer sand bottom or soft muddy bottom (Jefferson and Hung 2004), which most likely reflects distribution of their prey

The finless porpoise is known from the Persian Gulf to the coast of northern Japan Between these two limits, it is known from Pakistan, India, Sri Lanka, Bangladesh, the Malay Peninsula, Indochina, China, and the western and southern coasts of the Korean Peninsula It is also known from the northern coast of Sumatra, the islands of Bangka and Belitung, the northern coasts of Java, Kalimantan (except for the east coast), and the Tambelan Islands A record of the species from Palawan Island is now known to not be valid More survey effort will likely find the species on the Myanmar coast The incidental mortality of finless porpoises has been reported from fisheries along the west coast of Taiwan and in the Penghu Islands Although some of the carcasses stranded on the west coast of Taiwan might have represented bycatch by Taiwanese fishermen taken along the continental coast and subsequently discarded, it is also true that finless porpoises inhabit the Taiwanese coast (Wang and Yang 2007, in Chinese and English; Wang et al. 2008) Finless porpoises are known from the western and southern coasts of the Korean Peninsula, but they are absent from the east coast The eastern boundary of the finless porpoise in Korea seems to be located around Ulsan, which is seen in cruise reports by Korean scientists (eg, An et al 2006; An and Kim 2007)

In Japan, finless porpoises are known from western Kyushu to northeastern Japan Shirakihara et  al. (1992a) collected published records of finless porpoises in Japan and at the same time made a questionnaire survey of fishery cooperative unions (FCUs) The survey reported that on the Pacific coast of Japan, the northern limit of records of the species is on the northern coast (38°24′N) of Sendai Bay and along the coast of the Oshika Peninsula (38°16′–38°24′N, 141°25′E-141°33′E), which separates the bay from the Pacific Ocean On the Sea of Japan coast, there are records from the coast of northern Kyushu to Hime near Mawaki on the west coast of Toyama Bay The questionnaire survey could obtain information from a broad area in a short period and help create a basis for future surveys, but caution is needed because of possible inaccuracy in species identification, date, and location (location of the FCU was used as location of the finless porpoise) The survey suggested that the northern limit of the species is at Onagawa (38°26′N, 141°27′E) on the Pacific coast and at the northern coast of Sado Island (37°50′N-38°20′N, 138°12′E-138°30′E) off the Sea of Japan coast These limits show good agreement with the known range, although they slightly expand the range north of the confirmed records

The following three points should be noted about the results of the questionnaire survey: (1) except for the area around Kanmon Pass (33°56′N, 130°56′E), there were only a

(2) positive replies were also scarce for the Pacific coast facing the outer seas (ie, Kyushu, Shikoku, Kii Peninsula, and Shizuoka Prefecture), and (3) there were no positive replies for the islands south of Kyushu and along the islands on Tsushima Strait (ie, Iki Island and Tsushima Island) These suggest discontinuity in the habitats of Japanese and Korean finless porpoises and discontinuity of distribution even within the Japanese coastal habitat

Finless porpoises are classified into three subspecies (Rice 1998; also see the “Note” at the end of this chapter): N. phocaenoides phocaenoides inhabiting waters from the Indian Ocean to the South China Sea, N. phocaenoides asiaeorientalis (Pilleri and Gihr 1972) in the Yangtze River, and N. p. sunameri (Pilleri and Gihr 1975) in waters of the East China Sea, Yellow Sea, Bohai Sea, and Japan

Finless porpoises inhabit shallow coastal waters, and the distribution pattern is almost linear Additionally, their habitats are often separated by steep coastal topography, which is likely to limit genetic exchange between the habitats Marine species having such a distribution pattern would be less likely to have genetic exchange between remote habitats compared with other species that have 2D distributions The finless porpoise, which apparently has its origin in tropical waters, now has expanded its range to cold temperate waters such as the Bohai Sea, the Inland Sea, and Sendai Bay, where winter temperature can be around 5°C or less, and adapted to broadly variable environments Genetic isolation and selection pressure under different environments could have resulted in distinct local populations

The known geographical variation in the external morphology of the finless porpoise includes variation in body size, pigmentation, and shape of the dorsal ridge Individuals in the Indian Ocean have a pale area extending from the chin to the belly (Pilleri and Chen 1980), juveniles in the Yangtze River have a pale lip at the angle of the gape (Reeves et al 1997), and young individuals off Hong Kong have paler lips and throat region with gray in the remaining area, which darkens with growth (Parsons and Wang 1998) Contrary to this, newborns in Japanese water have an almost black pigmentation that becomes paler at the age of 4-6 months and changes to light gray in adults (Kasuya and Kureha 1979) Some of the other toothed cetaceans, for example, Risso’s dolphin, beluga, narwhal, and sperm whale, become paler with age, and the Amazon River dolphin is known to become darker if placed in less turbid water Before using body pigmentation for taxonomy or identification of local populations of finless porpoises, we need more information on the range of individual variation and on ontogenetic change

Finless porpoises tend to be smaller in tropical waters than in boreal waters The largest males known from the Indian Ocean were 150  cm long and the largest female 155  cm They were 201 and 200 cm long, respectively, in the Yellow Sea and Bohai regions where the sea surface may freeze in

168 cm; female, 151 cm), Inland Sea of Japan (male, 192 cm; female, 180 cm), and western Kyushu of Japan (male, 175 cm; female, 165  cm) are slightly smaller This suggests that Bergman’s rule might apply to the finless porpoise Although in some of the examples given earlier, females are larger than males, this is due to small sample size On average, males of this species grow larger than females (Kasuya 1999)

Instead of a dorsal fin, the finless porpoise has a dorsal ridge extending from the level of the shoulder blades to the midlength of the tail peduncle There is an area of numerous tubercles on the dorsal ridge Various hypotheses have been proposed for functions of the tubercle area, including “protection of the skin,” “for a mother to carry her calf,” “sensory organ,” or “for contact between individuals” (Kasuya 1999) Geographical variation of the shape of the ridge has attracted the attention of taxonomists (Figure 81) Finless porpoises in waters from Japan to the west coast of the Taiwan Strait (continental shore) throughout the Yellow Sea and East China Sea have a narrow ridge with a maximum width of 1-2 cm, height of 2-3 cm, and tubercles arranged in 4-10 rows The corresponding structure on freshwater individuals in the Yangtze River is much narrower, the 2-3 mm wide ridge has only 2-3 tubercle rows (Howell 1927; Wang 1999, in Chinese), but these porpoises may be grouped with the northern oceanic individuals in a narrow-ridged type See Section 834 for additional observations on the morphology of the dorsal ridge area of the narrow-ridged types

Ocean along the coasts of the South China Sea and South East Asia, finless porpoises have a broader tubercle area on the dorsal ridge, with a maximum width of 12  cm and greater number of tubercle rows, often exceeding 20 (Pilleri and Chen 1980; Wang 1999, in Chinese) Both the narrow-ridged and the broad-ridged types occur in waters around the Jinmen and Matsu Islands on the west cost of the Taiwan Strait (continental coast) (Parsons and Wang 1998; Jefferson 2002; Wang and Yang 2007, in Chinese and English) Jefferson (2002) has reported 4 individuals of the narrow-ridged type (04-06 cm in width) and 17 of the broad-ridged type (ridge 30-95 cm in width) collected in a fishing port on the Fujian coast, continental China, but it is unclear whether they were obtained in the same season or they might have been seasonally separated On the east coast of the Taiwan Strait (west coast of Taiwan), most sightings and strandings have been of the narrow-ridged type, with only one stranding of the broad-ridged type; the suspicion is that it may have drifted from the continental side of the strait or have been discarded by fishermen who had been fishing on the continental side (J Wang personal communication on January 18, 2008) A detailed description of the dorsal ridge and its geographical variation is available in Jefferson (2002)

Jefferson (2002) compared skull morphology between the two ridge types using specimens covering a broad geographical range from the Persian Gulf to Japan He did not identify a clear cline in skull length across the latitudinal range However, an analysis using the data in Jefferson (2002) and separating the two ridge types reveals a latitudinal cline in skull size The narrow-ridged-type porpoises from the waters of Japan and the Yellow Sea and Bohai Sea have greater skull length (about 215-25 cm) than the Yangtze individuals (21 cm) Among the broad-ridged types, those from the South China Sea (Vietnam, Hong Kong, Taiwan, Fujian Province) tend to have larger skulls (approximately 21-245  cm) than those in the southern region of Malaysia, Thailand, Mekong, India, and Pakistan (19-215  cm) Even within the broadridged types to the north of the Malay Peninsula (ie, excluding the Indian Ocean individuals), there seems to be a latitudinal cline from the smallest skulls (18-21 cm) on the Malay coasts to the largest ones (215-25 cm) on the coast of Fujian Province in China Analyses combining specimens of broader geographical ranges or combining the two ridge types could have masked the presence of the clines

Pilleri and Gihr (1972) were the first to recognize multiple species in this genus They found statistically significant difference in some of the mean skull measurements between specimens from the Indian Ocean and those from the Yellow Sea and the Yangtze River and proposed to deal with them as two separate species, that is, Neomeris phocaenoides and N.  asiaeorientalis, respectively The type locality of the latter was the Yangtze River, but their analysis included the Yellow Sea specimens Subsequently, the same authors studied Japanese specimens and named them N. sunameri (Pilleri and Gihr 1975) This taxonomy depended on arbitrarily selected morphological measurements and on statistical difference in

difference between two mean values Some minor morphological difference is likely to exist between geographically isolated populations, and they can be identified as different on average if sample sizes are sufficient, but taxonomic evaluation is another problem; mean differences do not necessarily indicate separate species Pilleri’s group also reported the earlier mentioned difference in the dorsal ridge between N.  phocaenoides and N. asiaeorientalis (Plleri and Chen 1980)

Amano et  al (1992) carried out multivariate analysis of skull measurements among specimens from three geographical areas, namely, the Indian Ocean plus the South China Sea, Yangtze River, and Japan, and found that individuals of the first group were separated from members of the latter two groups and that separation of individuals between the latter two groups was incomplete They concluded that the three groups are separate subspecies Although their analysis revealed greater differentiation of the Indian Ocean-South China Sea series (N. p. phocaenoides) than that between those from the Yangtze River (N. p. asiaeorientalis) and Japanese waters (N. p. sunameri), this was not reflected in their taxonomic conclusions Amano later (2003) reviewed the distribution and taxonomy of the species

Jefferson (2002) attempted to improve the taxonomy by combining osteology and external morphology He divided specimens that covered the whole distribution range into two groups by morphology of the dorsal ridge: broad-ridged type (N. p. phocaenoides) and narrow-ridged type (N. p. asiaeorientalis and N. p. sunameri) and carried out multivariate analysis of the skull morphology His result with the broad-ridged types showed that the Indian Ocean individuals and southern South China Sea individuals were almost inseparable and that individuals in the southern South China Sea and the northern South China Sea were nearly completely separated (with partial overlap) For the narrow-ridged type, he found that the series were only partially separated among three geographical areas (Japan, the Yangtze River, and the Yellow Sea plus the Bohai Sea)

The scientists who place weight on the morphology of the dorsal ridge and on geographical overlap of the range in the boundary area are of the opinion that these two types should be handled as separate species This idea comes from interpretation of the sympatric distribution as an indication of the presence of some mechanism of reproductive isolation Such a view was expressed by Wang et  al. (2008), who analyzed width of the dorsal ridge, microsatellite DNA, and mitochondrial DNA (mtDNA), based on 18 specimens from Hong Kong waters where only the broad-ridged type was known and 38 from area where both broad-ridged and narrow-ridged types were known Among the latter were 15 specimens from Fujian Province, 17 from the Matsu Island area, and 6 from the west coast of Taiwan They found that all the specimens were classified into one of the two ridge-width types with no intermediates The microsatellite analysis was able to classify all the specimens into the correct ridge types Out of the 69 alleles, 40 (58%) occurred only in one of the ridge types, which they interpreted as an indication of low genetic interchange

rosatellite analysis However, it is also true that half of the alleles were common between the two ridge types Wang et al. (2008) interpreted their results to indicate that separation of the two lines of finless porpoises occurred within less than 18,000 years ago or after the last glacial period, but apparently reserved judgment on the taxonomic position of the two lines (broad-ridged and narrow-ridged types) (see the “Note” at the end of this chapter)

In my view, determination of the mechanism of reproductive isolation between partially sympatric populations of cetaceans is an important research objective, and interpretation requires careful consideration The causes of reproductive isolation between such populations are not limited to physiological mechanisms but can be obtained if the breeding members of two populations are segregated, for example, by two populations overlapping only in the nonbreeding season or by segregation of breeding components (range overlap limited to nonbreeding components) Behavioral or cultural differences in some highly social species can also limit genetic exchange; this is thought to happen in killer and sperm whales

It appears that finless porpoises in the Yangtze River constitute a single population, but currently there is almost no reliable information on the boundary between the riverine and marine populations, whether their range overlaps, or how the boundary moves seasonally reflecting change in riverine discharge Many of the past studies classified their specimens simply into river and ocean series without giving geographical location or time of sampling (eg, Gao and Zhou 1995a,b) Such basic information is important for the improvement of our knowledge on local populations and better understanding of differences between the populations

Our knowledge on the general distribution of Japanese finless porpoises depends mostly on Shirakihara et al. (1992a), which is based on published records of the species and questionnaires sent to 2053 FCUs covering the entirety of Japan except for Hokkaido, where finless porpoises were not expected The questionnaire asked if the member fishermen had ever seen the finless porpoise (1382 FCUs replied) The northernmost positive responses were obtained from an FCU on the northern coast of Sado Island in the Sea of Japan and an FCU near Onagawa on the Pacific coast (Section 831) These records almost agreed with published records from Hime on the west coast of Toyama Bay in the Sea of Japan and the northern shore of Sendai Bay on the Pacific coast but slightly expanded the range to the north Although such a distribution is highly probable, it seems to be premature to reach any firm conclusions before scientists confirm the distribution

Shirakihara et al. (1992a) noted that FCUs with positive replies tended to be located on a shallow seashore without rocky bottom and listed five such sea areas: (1) Omura Bay; (2) Ariake Sound-Tachibana Bay; (3) the Inland Sea, Osaka Bay, and the Kii Channel; (4) Ise Bay-Mikawa Bay, and (5) the area extending from Tokyo Bay to Sendai Bay

(see Figure 82 for the locations) The species does not seem to inhabit the coasts of the Japanese Southwestern Islands (24°N-31°N, 123°E-131°E), which separate the western North Pacific and the East China Sea Although a white shark landed at Okinawa Island was found with two finless porpoises in different states of digestion in the stomach (Kasuya 1999) and one finless porpoise was stranded at the same island (see Section 834), these incidents should not be taken as evidence of occurrence of the species in the Southwestern Islands

Now, we will consider further details of the distribution of finless porpoises in Japan The Shimabara Peninsula is located between the earlier mentioned two habitats of the finless porpoise, Omura Bay and Ariake Sound-Tachibana Bay, and records of the species are scarce along the coast of the peninsula, which suggests that the two habitats on the west coast of Kyushu are discontinuous Distribution of finless porpoises in the western Inland Sea is continuous through Kanmon Pass (which connects the northwestern Inland Sea and the southern Sea of Japan) to the coasts of northern Kyushu and western Yamaguchi Prefecture in longitudes from 130°15′E to 130°55′E, where year-round occurrence of the species has been confirmed Therefore, the habitat of the species in the Inland Sea-Osaka Bay-Kii Channel extends to this area in the southern Sea of Japan (Shirakihara et al. 1992b; Nakamura et al 2003; Nakamura and Hiruda 2003, last two in Japanese) Akamatsu et  al (2008) placed underwater microphones in Kanmon Pass to detect the movement of finless porpoises for

from 2005 to 2006 Kanmon Pass has a water depth of about 10-30 m and a minimum width of 670 m They recorded sounds produced by 37 finless porpoises, including 33 recorded during the nighttime hours of 2000 to 0400 This time coincided with minimum vessel traffic Finless porpoises seemed to travel through the pass during the night, but it was not certain if this was related with traffic density or if the negative correlation was only an apparent one They did not find correlation between the current direction and finless porpoise travel, but they noted that out of 16 individuals, 14 swam with the current (the tidal current reverses periodically) This observation agreed with that of Kasuya and Kureha (1979) for Naruto Pass (34°15′N, 134°39′E), which connects the southeastern Inland Sea and Kii Channel and is known to have rapid tidal currents (see Section 841)

Finless porpoises in Omura Bay are separated by an almost empty area along the coast of northwestern Kyushu from individuals in the Kanmon area-Inland Sea-Osaka Bay-Kii Channel Thus, the porpoises in Omura Bay and Ariake Sound-Tachibana Bay may constitute two distinct populations Regarding distribution of finless porpoise along the Sea of Japan coasts, Shirakihara et  al. (1992a) recorded in their questionnaire survey only a small number of positive replies and identified only a few confirmed records of the species along the Sea of Japan coast east of the Kanmon area This means that finless porpoises are uncommon along the coast east of Yamaguchi Prefecture The limited number of records of the species from the Sea of Japan coast between Yamaguchi Prefecture and Hime in Toyama Bay area is likely to represent stragglers from the Kanmon area, that is, from those inhabiting the Inland Sea-Osaka Bay-Kii Channel area

Finless porpoises are continuously distributed from the Kanmon area to the Kii Channel via the Inland Sea and Osaka Bay (Figure 82), but they are rare on the Pacific coast of southern Kyushu, Shikoku Island (32°45′N-33°30′N, 132°45′E-134°30′E), and the Kii Peninsula (33°26′N-34°10′N, 135°05′E-136°20′E) The only exceptions were one record from Cape Shionomisaki (33°26′N, 135°45′E) and another from Taiji (33°36′N, 135°57′E) on the Pacific coast of the Kii Peninsula (Shirakihara et al. 1992a; see Section 834) Thus, individuals in the Kanmon area-Inland Sea-Osaka Bay-Kii Channel and those in Ise Bay-Mikawa Bay are discontinuous in distribution The latter two bays are connected to each other and then open to the Pacific through a common opening, and sighting of finless porpoises are continuous between the two bays Numerous stranding records and some sightings along the Pacific coast near the exit to the Pacific suggest that individuals in Ise Bay-Mikawa Bay move outside of the bays at least seasonally (Figure 82)

A question remains on the identity of finless porpoises in Suruga Bay (35°N, 138°40′E) about 170 km to the east of the Ise Bay-Mikawa Bay opening The presence of the species there has been known from early times (Kuroda 1940, in Japanese) Although this habitat appears to be separated by an empty area from Ise Bay-Mikawa Bay, there is insufficient information to confirm that

finless porpoises relates to the porpoises in the Tokyo Bay to Sendai Bay area We know that some finless porpoises occur year-round in Tokyo Bay north of Futtsu (35°19′N, 139°47′E) near the entrance, but there are no confirmed records of the species outside (or south) of Tokyo Bay: west coast of the Awa area (34°54′N-35°09′N, 139°45′E-140°13′E) or southern part of the Boso Peninsula (34°54′N-35°30′N, 139°45′E-140°25′E) Aerial sighting surveys by Amano et  al. (2003) did not find finless porpoises in this area, and the results of the interview survey by Suzuno et al. (2010, in Japanese) suggest a possibility of past decline of the species in the area as well as its absence along the coast in recent time This information suggests a hypothesis that finless porpoises in Tokyo Bay are separated from those found along the east coast of the Boso Peninsula and further north to Sendai Bay

The population structure of finless porpoises inhabiting the broad area extending from the southern tip (34°55′N, 139°53′E) of the Boso Peninsula to the bottom (38°24′N) of Sendai Bay also needs further consideration Amano et  al. (2003) suggested a possible discontinuity of distribution around Cape Shioya (37°00′N, 141°59′E) If this is accepted, there is a possibility of three local populations in the area from Tokyo Bay to Sendai Bay along the Pacific coast Yoshida et  al. (2001) suggested from the analysis of mtDNA a possibility of different populations in Sendai Bay and along the Pacific coast (35°15′N-37°N) south of Sendai Bay The population structure of finless porpoises in Tokyo Bay to Sendai Bay area remains to be investigated

The preceding section discussed possible local populations of Japanese finless porpoises based on the information on their distribution For the hypotheses thus created to be confirmed, limitation in interchange of individuals between the habitats or the presence of only limited genetic exchange between them must be confirmed If genetic exchange or mixing of individuals is sufficiently small to allow independence in population dynamics, individuals in each habitat should be dealt with as different populations for the purpose of conservation in management

Both genetic and osteological studies have been used for the analysis of population structure of Japanese finless porpoises As morphology can be influenced by both environment and the genome, two sets of samples from a single population in different historical periods might show significant differences if there were some large environmental change in the intervening period, such as in forage So, caution is needed in analyzing specimens collected over a broad historical time period

In order to identify local populations of Japanese finless porpoises, Yoshida et al (1995) analyzed skull morphology and Yoshida et  al. (2001) analyzed haplotype frequency of mtDNA; both were reviewed by Yoshida (2002) Yoshida et al. (1995) used a total of 146 skulls from five geographical regions: 8  skulls from Omura Bay, 74 from Ariake

Inland Sea, 11 from Ise Bay-Mikawa Bay, and 5 from Tokyo Bay to Sendai Bay area Samples of both sexes were combined after removing characters found with sexual dimorphism Season and year of collection of the specimens were not given in their study They first analyzed relative growth, using all the samples, of 15 measurements against the condylobasal length of the skull (analysis of covariance) and found 6 measurements with geographical difference in pair-wise comparison of the five sample groups One measurement compared between Omura Bay and Kanmon area-Inland Sea was an exception The most distinct difference was found in two measurements relating to the width of the skull: width of the rostrum at midlength and width of the skull across the zygomatic processes of the squamosal Specimens from Ise Bay-Mikawa Bay had these values significantly smaller than those for specimens from the other four geographical regions (Figure 83) Canonical discriminant analysis using skulls aged 4 years or older and using the earlier mentioned six characters (Figure 83) yielded the following results The Ise Bay-Mikawa Bay group was separated from all other groups The Ariake Sound-Tachibana Bay group was separated from the geographically close Omura Bay group and the Inland Sea-Kanmon group of considerable geographical distance away but partially overlapped with the geographically distant Tokyo Bay-Sendai Bay group The Omura Bay group and Inland Sea-Kanmon group were only partially separated from each other in the discriminant analysis; they are located near each other geographically The Tokyo Bay-Sendai Bay group, which covers an extended northern habitat of this species in Japan, overlapped partially with the Omura Bay

Kanmon group but separated completely from the Ise BayMikawa Bay group, which was the closest geographically

The earlier discussion can be summarized thusly: if we exclude animals from Omura Bay in western Kyushu and the Pacific coast north of Tokyo Bay, finless porpoises in each of the three areas of Ariake Sound-Tachibana Bay, Inland Sea-Kanmon, and Ise Bay-Mikawa Bay are individually identifiable to the location of origin based on their skull morphology Porpoises in Omura Bay show some morphological differences from those in the geographically close Inland Sea-Kanmon area and those in the more distant Tokyo Bay and northern area This result suggests the presence of a minimum of five local populations of finless porpoises in Japan Unfortunately, only five specimens from the broad Tokyo Bay-Sendai Bay area were available for this study; they had to be combined into a single group for the analysis

Yoshida et al. (2001) analyzed 345 base pairs of the control region of mtDNA of 174 finless porpoises from Japan, identified 10 haplotypes, and analyzed haplotype frequency in the earlier mentioned five geographical areas (Table 83) As mtDNA is maternally inherited, the analysis can detect males from another habitat visiting temporarily to breed only when they happen to be sampled directly; their offspring will bear the mtDNA of the resident mothers Detection of sex-biased genetic dispersal must also involve the analysis of nuclear DNA However, Yoshida et  al. (2001) concluded that sexbiased movement did not exist because no between-sex differences in haplotype frequency were detected and therefore combined the sexes for the analysis of haplotype frequency Difference in haplotype frequency was statistically significant

in all the pair-wise combinations of samples except for one case, the comparison between 8 samples from Omura Bay and 30 from the Inland Sea-Kanmon area, which could have been due to a small sample size for the Omura Bay population

Yoshida et  al. (2001) noted the presence of genetic isolation between Omura Bay and Ariake Sound-Tachibana Bay, which are only 60  km apart, as an indication of limited intermingling between the habitats They also stated that a female finless porpoise obtained at Taiji had a haplotype known from the Ise Bay-Mikawa Bay area, which was about 30-40 km west of the known range of the population The authors identified only two haplotypes for the 14 porpoises from the Tokyo Bay-Sendai Bay area, noted that all the 7  individuals from Tokyo Bay had a single haplotype (type  a) and that all the specimens from Sendai Bay (presumably 6 individuals) had another haplotype (type b), and suggested that there was a possibility of more than one local population in this extended area Their 14 samples from the Tokyo Bay-Sendai Bay area contained 1 sample from the Ibaraki Museum in Ibaraki Prefecture (35°45′N-36°50′N on the Pacific coast) This museum specimen was likely to have been from the nearby coast, so the total Sendai Bay sample could be six animals

In 2004, an emaciated finless porpoise (1296  cm and 192 kg) was stranded on the coast of the Motobu Peninsula (26°40′N, 127°50′E) on the East China Sea coast of Okinawa Island Yoshida et al. (2010) analyzed mtDNA of this animal and found that its haplotype did not match any of the 10 haplotypes found among Japanese specimens but matched that of one of the 17 haplotypes reported from finless porpoises inhabiting the Chinese coast from the Taiwan Strait to the Yellow Sea They concluded that the porpoise must have been a stray from the Chinese coast They mentioned only that the stranded individual had a narrow dorsal ridge, but it was clear in an accompanying photograph that there was a longitudinal groove on each side of the dorsal ridge Thus, the cross section of the ridge area shows a W shape, which is different from the feature in Japanese individuals but known from finless porpoises along the coasts of Korea and the Yellow Sea

The analyses given earlier lead to the conclusion that there are several local populations of finless porpoises in Japan that are currently almost isolated from each other

However, this does not necessarily mean that the isolation dates from long ago geologically The planet has had a series of climatic cycles The last glacial period that peaked in around 18,000-20,000 years before the present was followed by gradual warming and recorded peak warmth 5,000-6,000 years before the present This is called in Japan the period of jomon marine transgression because it occurred in the prehistoric Jomon cultural era known from cord-patterned pottery Then the climate became cooler with retreat of the ocean to the current status in marine geography around Japan Current global warming may reverse the climate again with a speed much faster than the natural cyclic changes of the past

While the climate was becoming warmer in the recent geographic past, some finless porpoises in the southern habitat could have immigrated to the northern habitat and established a new local population Some in the new population could have survived to the present through the subsequent cooling period There presumably was a founder effect, and there could have been new haplotypes arising only through rare mutations The currently surviving parent population would be expected to have greater genetic variation than the daughter population

The data presented by Yoshida et  al. (2001) are particularly interesting in this regard The number of haplotypes tended to decline moving north If we exclude the Omura Bay sample of small size and also for the time being the Tokyo Bay-Sendai Bay sample, which is likely to be from multiple populations, the Ariake Sound-Tachibana Bay sample from western Kyushu showed the highest variation with six haplotypes (with a ratio of 46:9:6:2:1:1), followed by a decreasing number of haplotypes in the Inland Sea-Kanmon sample (two haplotypes with a ratio of 27:3) and the Ise Bay-Mikawa Bay sample (two haplotypes with a ratio of 52:4) (Figure 82 and Table 83) This decreasing cline is followed by a single haplotype in the Tokyo Bay sample (n = 7) and another single haplotype in the Sendai Bay sample (presumed n = 6, see the preceding text in this section) This cline in haplotype frequency from southwest to northeast may reflect the history of expansion of habitat from south to north during the period of warming Another interesting point in the results presented by Yoshida et al. (2001) is the absence of common haplotypes among the three southwestern samples (three populations) and the two northeastern samples (presumed to be from more

Geographical Differences in Mitochondrial Haplotype Frequencies of Japanese Finless Porpoises

tion of local populations and history of low genetic exchange between nearby populations

In Japan, the term migration has been often used without a firm definition and even confounded with “dispersal” or “shift” In a strict sense, migration represents an activity where animals move toward a destination with other activities such as feeding or mating repressed during the travel Shift means seasonal expansion or retraction of the geographical range to meet seasonal change in the suitable environment Among cetaceans, gray whales and humpback whales are species known to migrate in the strict sense However, some gray whales are known to feed on the way to their destination in the Bering Sea Strict definitions are likely to have exceptions Seasonal movement of striped dolphins along the Pacific coast of Japan is apparently different from migration They inhabit the Kuroshio front area and areas to the south of it In early summer as sea temperature rises, their range moves north to reach latitude of 40°N-41°N in the summer, and in the autumn, it retreats to a southern boundary at 33°30′N-34°30′N Their reproductive and feeding activities continue throughout the movement This can be called a seasonal movement or shift rather than migration

Pilleri and Gihr (1972) were probably the first to mention seasonal movements of finless porpoises They stated that the density of the species in the Indus delta was highest in the winter season of October-April and then declined; they concluded that the porpoises moved offshore in the spring to feed on shrimps Wang (1984, in Chinese) reported an opposite case where finless porpoises in the Bohai Sea, Yellow Sea, and East China Sea approached the coast in March-April or May, resulting in density increase in coastal waters The freshwater population in the Yangtze River is thought to make seasonal shifts as water level changes

Shirakihara et al. (1994), based on 25 years of riding ferries, observed seasonal change in finless porpoise density in the Ariake Sound-Tachibana Bay area of western Kyushu The geography of this area is described in brief in Section 831 They recorded a clear seasonal change in sighting density inside the sound from a high of over 4 animals/100  km of searching from January through June to a low of less than 2 from July through November This seasonal fluctuation has some similarity with that observed in the Inland Sea by Kasuya and Kureha (1979) Shirakihara et al. (1994), however, observed an opposite density trend in Tachibana Bay and in the area connecting to the Ariake Sound, where density was lower from January through May and higher from July through November The sighting density in the peak month was only 1 individual/100  km of effort, which was much lower than the figure for inside the sound If we accept this observation, then we can infer that finless porpoises tend to spend winter in the Ariake Sound and move to Tachibana Bay in summer

by one or more of the following factors: (1) most of the population remained in the sound in summer, (2) the porpoises were dispersed in Tachibana Bay area in summer, or (3) sea conditions in Tachibana Bay were unfavorable for sighting

The results of aerial surveys by Yoshida et al. (1997) did not apparently agree with the interpretation mentioned earlier (Table 84) They repeated the surveys in the Ariake Sound, Tachibana Bay, and the surrounding area for finless porpoises during 5 months from May 1993 to May 1994, and they were able to confirm an offshore limit of distribution of the species at the 50 m isobath and its absence in the coastal waters between Omura Bay and Tachibana Bay and also in the Yatsushiro Sea, which is another enclosed bay connected to the Ariake Sound Their results confirmed that distribution of finless porpoises was continuous from the Ariake Sound to Tachibana Bay but that this Ariake Sound-Tachibana Bay habitat was unconnected with other habitats of finless porpoises in Japan Their abundance estimates for the Ariake Sound recorded a peak in May and a low in June, while those for Tachibana Bay reached a peak in February and a low in November Broad confidence intervals accompanying the estimates preclude firm conclusions on seasonal abundance change in these waters Thus, Yoshida et al. (1997) could not offer positive support for the inference of Shirakihara et al. (1994) on seasonal movement in the species Although support for seasonal change in the abundance of finless porpoises is inconclusive for the Ariake Sound or Tachibana Bay, there remains the possibility that winter abundance in Tachibana Bay (2416 individuals in February) is greater than the abundances in the rest of the seasons (458 in August and 398 in November; Table 84), which is the reverse of the trend suggested by Shirakihara et al. (1994)

TABLE 8.4 Abundance of the Finless Porpoise in the Ariake Sound-Tachibana Bay Populations and Their Seasonal Fluctuations, Estimated by Sighting Surveys and Assuming g(0) = 1

west of Ariake Sound-Tachibana Bay, has an area of about 35 × 15 km, and is connected to the East China Sea by two northern passages with widths of 200 and 10 m This bay was surveyed by Shirakihara et  al. (1994) using vessels and by Yoshida et al. (1998) from the air Shirakihara et al. (1994) apparently analyzed sightings recorded by the crews of highspeed ferryboats operating in Omura Bay; they concluded that finless porpoises were dispersed throughout the bay in spring but aggregated in coastal parts of the bay in other seasons They also interviewed fishermen along the coast of the East China Sea near the entrance to Omura Bay and found records of rare incidental mortality in fisheries but were unable to find anyone who had seen the species in the wild Yoshida et al. (1998) flew aerial surveys in the same bay in February, May, August, and November; they found that finless porpoises were dispersed widely in the bay in May but that the density was lower in the central part of the bay in other seasons, which apparently agreed with the results of Shirakihara et al. (1994) However, the lack of a definition of “spring” in Shirakihara et  al. (1994) caused me some difficulty in precise comparisons Abundance estimated by Yoshida et al. (1998) showed an apparent seasonal trend from 343 individuals in May to 203 in August, 104 in November, and 92 in February, but these estimates were not statistically different Using this result and the earlier mentioned interviews made by Shirakihara et al. (1994), Yoshida et al. (1998) concluded that finless porpoises in Omura Bay usually do not leave the bay They most likely have small-scale seasonal movements within the bay following prey species and remain within the bay year-round

The Inland Sea is the largest inland water of Japan, with a size of about 370 × 50 km or surface area of 14,300 km2, and contains about 3,000 islands The sea is connected to outer seas by four passes: Kanmon Pass (600 m wide and 47 m deep) at the northwest corner opens to the southwestern Sea of Japan; Hayasui Pass (14 km wide and 195 m deep) at the southwest corner opens to the Bungo Channel, which opens to the Pacific Ocean; Akashi Pass (36 km wide and 100 m deep) at the northeast corner opens to Osaka Bay, which then opens to the Kii Channel; and Naruto Pass (14 km wide and 90 m deep) at the southeast corner opens to the Kii Channel, which opens to the Pacific Ocean (Figure 82)

Over a period of about 25  years from April 1976 to October 1978, Kasuya and Kureha (1979) surveyed the distribution of finless porpoises in the Inland Sea They used low-speed passenger ferryboats or ferries for both passengers and cars but avoided high-speed passenger ferries The total number of ferry tracks used was 31 including some in Osaka Bay, but this figure depends partly on the method of counting tracks If a track is defined as from embarkation to disembarkation, the number is 34 tracks served by 28 different vessels One survey took about 2 weeks, and every track was surveyed as many times as possible within the survey Thus, the total surveyed distance was several times greater than the sum of the surveyed tracks

The density of finless porpoises in the Inland Sea varied depending on distance from the shore; the factor behind this

is unknown Combining all the survey data gave an average sighting rate of about 25 individuals per 100 nautical miles for waters within 1 nautical mile from the shore, which decreased to slightly over 10 in the 1-3 nautical mile range and to less than 5 in waters over 3 nautical miles from the shore (Figure 84) The sighting density also appeared to vary with distance from the eastern and western openings, with lower density in waters near the two extremities of the sea than in central waters This was constant among the three strata of distance from the shore and among seasons (Figure 85) in the late 1970s (Kasuya and Kureha 1979), but the distribution pattern changed in the late 1990s (Kasuya et al. 2002)

Kasuya and Kureha (1979) observed a seasonal density decline in the nearshore stratum (<1 nautical mile) of the middle part of the sea (areas 2, 3, and 4 in Figure 85) from about 10 in November-February or 12 in March-April to about 05 in May-September and felt that the density was likely to be high in the spring They also observed the estimated abundance of the species in the Inland Sea to be highest in April and lowest in October (Figure 86) This seasonal change is mostly a reflection of change in the nearshore stratum (<1 nautical mile) The abundance estimated in this study is now believed to be unreliable due to technical problems (see Section 842), but the estimated seasonal and geographical trends still reflect reality

The results of Kasuya and Kureha (1979) indicated that most of the finless porpoise population in the Inland Sea inhabits the nearshore area and the possibility that some porpoises may disperse either to the offshore area or to waters near the outer seas The latter was similar to the case suggested by Shirakihara et al. (1994), which was not confirmable in data from a subsequent study by Yoshida et al. (2001)

Kasuya and Kureha (1979) unsuccessfully attempted to confirm whether the finless porpoises in the Inland Sea move out in the early summer through any of the four passes Land-based observation of 18 hours extending over a 2-day period at Kanmon Pass (northwestern exit) in May and 7 ferry cruises across Hayasui Pass (southwestern exit) in March and April ended with sightings of no finless porpoises Fortyeight ferry cruises across Akashi Pass (northeastern exit) in March-April and 14 cruises in October across the same pass ended with sightings of only 6 finless porpoises in two groups in March Naruto Pass, the southeast exit, was surveyed in March and April-May, using ferryboats across the western side (Inland Sea side) of the pass and the eastern side (Kii Channel side) A total of 1392 nautical miles of survey on both sides of Naruto Pass recorded 233 finless porpoises Their density was higher in downstream flow in the pass, so the location of high density switched according to the direction of the tidal current, perhaps in relation to the distribution of prey organisms Our attempt failed to relate swimming

direction to seasonal movement of finless porpoises between the Inland Sea and nearby waters In the late 1970s, I had the impression that the species was extremely abundant in the region of Naruto Pass at least in the March-May season, which was in contrast with near absence observed by Kasuya et al. (2002) in the late 1990s (see Section 8441)

Various methods have been used to estimate the abundance of cetaceans, but they all have some biases due to the technique used or behavior of the animals A census counts all the members of a population and might be considered the most reliable, but the applicable situation is limited to the cases where the range of the population is small or all the members aggregate in a particular area at a time If all the members of a population cannot be encountered at a particular time, it becomes necessary to distinguish already identified individuals from

cable to particular species such as humpback or gray whales Although the eastern stock of gray whales might be thought as easy to census because almost all the members pass by a particular coastal location, the method also requires estimation of the number of individuals that travel during the night or on days of poor visibility and those that pass outside of the observers’ visual range Counting error by observers is also unavoidable

The mark-recapture method consists of marking or individually identifying animals in the population and then, after a period for mixing, determining the ratio of marked animals to unmarked animals in the population and thereby estimating the population size Individual identification can be based on external pigmentation, natural scars, artificial marks, or DNA fingerprinting Full mixing of marked and unmarked individuals is assumed in this method

Distance-sampling methods are another way to estimate whale abundance The method first determines the study area, such as the Inland Sea, samples a portion of area within the study area, estimates the density of whales in the sampled area, and extrapolates the density into the total study area The sampled area is usually a strip along the cruise track of a vessel or an airplane This method is widely used in cetacean management, and elaborated versions and theories based on them have been developed The study by Miyashita (2002, in Japanese) is useful for beginners to understand the principle of the method and that by Garner et al. (2002, in Japanese) is useful for more advanced users

When designing track lines for a sighting survey, seasonal movement of the targeted whales must be considered For example, if a vessel survey proceeds from south to north within the targeted area during a season when whales are expected to shift their range northward, then there is a risk of the observer traveling with northbound whales, and the survey will result in an overestimate If the survey proceeds in the opposite direction, the bias will be reversed Such risks are particularly great for shipboard surveys because they often last for several weeks, and correction of the bias is difficult It is a good idea to plan a survey in a season when the targeted population reaches the climax of seasonal movement

Another important point to be considered in designing sighting surveys is any geographical density gradient of the targeted species The area must be sampled randomly or systematically to avoid sample bias However, random sampling is usually impractical in track-line design, and unbiased systematic sampling has been applied by arranging the track lines evenly and parallel to the expected density gradient of the whales The study area can be divided into several subareas to allocate greater effort in a subarea with the expected high whale density Then the abundance is estimated for each subarea before generating an abundance estimate for the entire study area This helps increase the precision of the total abundance estimate This method was followed by aerial surveys for Japanese finless porpoises conducted in 2000 using funds from the Environment Agency (Seibutsu Tayosei Senta [Biodiversity Center] 2002, in Japanese) Track

of coastlines or distribution of islands The survey design of Kasuya and Kureha (1979), which used ferryboats connecting the islands in the Inland Sea, did not do this The track lines tended to be parallel with the coastline, that is, perpendicular to an observed density gradient of the finless porpoise, and were likely to cause bias in the abundance estimation Their attempt to divide the survey data into strata by offshore distance is believed to be insufficient to remove the bias

The third point to be remembered for cetacean sighting surveys is the fact that only a certain proportion of individuals present in the sampled strip are identified and recorded The proportion of undetected whales increases with increasing perpendicular distance from the track line and will be almost zero at a distance of 500 m or more in the case of the finless porpoise This decreasing trend is used to estimate sighting rate as a proportion of sighting rate at zero distance, which is expressed as g(0) and often assumed as 1 However, in reality g(0) can never be 1, and such an assumption biases the abundance estimate downward A value of g(0) = 03 was estimated for a shipboard survey of harbor porpoises, which are similar to finless porpoises in behavior and in body size aside from the presence of a dorsal fin (Hammond et al. 1995) This g(0) estimate was for a survey employing three observers; the value must be smaller for surveys with only one observer, such as that by Kasuya and Kureha (1979) The value of g(0) was estimated from data for duplicate sightings collected by placing two independent observer teams on a vessel

The estimate of g(0) is also affected by additional factors such as vessel speed, eye height of observers, number of observers and their sighting ability, sea condition, diving profile of cetaceans, and density and group size The values for aerial survey and shipboard survey will be totally different Kasuya and Kureha (1979) surveyed finless porpoises at Beaufort sea states below 3, but they still found that the decline of sighting rate with increasing perpendicular distance was more rapid in November-February (average Beaufort state of 12-17) than in March-June (average Beaufort state of 06-12) This was interpreted as an effect of sea-state difference

The response of cetaceans to survey vessels causes bias to the sighting rate and subsequently the abundance estimate Some dolphins and porpoises (eg, weaned young Dall’s porpoises) tend to approach sighting vessels and are likely to cause underestimation of perpendicular distance, compared with the case where the porpoises are indifferent to vessels, and to overestimate abundance On the other hand, a tendency to move away from the vessel (eg, by adult Dall’s porpoises) or to dive underwater before being identified by observers will bias the abundance estimate downward Usually, finless porpoises are not attracted to vessels but rather avoid them (Pilleri and Gihr 1973-1974) or dive beneath the bow (Zhou et  al. 1979) They have not been observed riding the bow, but they have been observed to (1) dive suddenly in front of the bow when approached by a vessel to a distance of 10-20 m or (2) ride the stern wave for a few seconds when the waves pass over the animal (Kasuya and Kureha 1979) Kureha (1976, in Japanese) noted that

some juveniles may occasionally ride the stern wave

The abundance of Japanese finless porpoises has been estimated by aerial or shipboard surveys, which were not necessarily free from the various types of bias mentioned earlier In particular, the results of Kasuya and Kureha (1979) using ferryboats in the Inland Sea now seem to have problems The comparison of two sighting densities obtained by similar methods, one by Kasuya and Kureha (1979) and the other by Kasuya et al. (2002), reveals a drastic decline in density during the 22-year period, which can be interpreted as an evidence of decline in abundance However, abundance estimated in recent years using aerial surveys (Shirakihara et al. 2007) is greater than that estimated by the early study of Kasuya and Kureha (1979) Accepting the decline in sighting density is reasonable because it derived from surveys using similar and simple methods less affected by methodology than the abundance estimates This led us to reject one of the two abundance estimates, that is, the smaller figure from the shipboard survey in the late 1970s (Kasuya and Kureha 1979) or the larger figure from the aerial survey in 2000 (Shirakihara et  al. 2007) I consider the earlier abundance estimate unreliable, because it was based on undeveloped sighting theory and survey methodology In particular, the g(0) value assumed at 05 by us for the shipboard survey could have been too great, causing downward bias in the estimate During this survey, I was alone on board almost all the time and collected all the data, including sightings of finless porpoises, vessel position, weather, and sea surface temperature measured using a bucket water sampler The more recent aerial survey used two observers and one data recorder and probably had greater sighting ability than my shipboard survey The greater eye height in the aerial survey probably compensated for the bias caused of the greater speed Additional technical deficits in the abundance estimation of Kasuya and Kureha (1979) are described in the following Caution is necessary in comparing abundance estimates from aerial and shipboard surveys

All the recent aerial surveys for Japanese finless porpoises described here have used the same methods and aircraft: a high-wing Cessna with four seats, a pilot in the right front seat, a recorder in the left front, and a pair of observers in the rear seats The sighting altitude was 500 ft (150 m), and the speed was 80-90 knots (148-167  km/h or 2470-2780 m/min) The most recent aerial survey was carried out by K Shirakihara’s group with the support by the Ministry of Environment and covered all the major habitats of the finless porpoise in Japan The Seibutsu Tayosei Senta (Biodiversity Center 2002, in Japanese) published the entire results of the project, which was followed by the publication by particular sections in scientific journals by individual participants The results of this survey are useful for comparing relative values of several finless porpoise habitats in Japan, but any problem in the methodology must affect all of them

are summarized in the following Unless otherwise stated, g(0) is assumed to be 1

8.4.3.1 Omura Bay A geographical description of Omura Bay is given in Section 841 This bay was surveyed twice, the first time in 1993-1994 by Yoshida et al (1998) and the second in 2000 by Shirakihara and Shirakihara (2002, in Japanese) as described earlier

Yoshida et  al. (1998) carried out four surveys when the Beaufort sea state was 2 or below They obtained the largest estimate of 343 in May and the smallest of 92 in February, but the four estimates were not statistically different from each other Therefore, they made a pooled estimate of 187 individuals using the 33 sightings of 53 porpoises in the four surveys Only sightings by the north-side observer were used, because the south-side observer was subject to glare from the sun, on either the left or the right side of the plane on its east/west tracks at an interval of 1 nautical mile (1852 m)

Shirakihara and Shirakihara (2002) created two sets of track lines across Omura Bay; each set had 13 parallel track lines with an interval of 1 nautical mile, and the track lines of the two sets were 05 nautical miles apart Their plan was to survey Omura Bay twice by randomly choosing one of the two sets for each survey and ended with choosing each of them They made the two surveys in April, which was close in time to May when the maximum figure was obtained by Yoshida et al. (1998), and recorded a total of 38 sightings of 54 individuals and obtained an abundance estimate of 298 finless porpoises in Omura Bay

The difference between the two estimates, 187 by Yoshida et al. (1998) and 298 by Shirakihara and Shirakihara (2002), was not statistically significant, and no change in abundance was estimated for the 7-year period However, such small populations must be managed with great caution for conservation purposes The surface area of Omura Bay of 320 km2 gives an average density of 058-090 porpoises/km2

8.4.3.2 Ariake Sound-Tachibana Bay A geographical description of the Ariake Sound-Tachibana Bay area is available in Section 831 The two connected sea areas are believed to be inhabited by a single population of finless porpoises, and the proportions of animals in the two areas may vary seasonally, although the details are still unknown Shirakihara et al. (1994) were the first to estimate the abundance of finless porpoises in the Ariake Sound portion They collected sighting data from three ferry tracks and from platform-of-opportunity fishing vessels and a university training ship The number of observers, 1 or 2, varied among surveys and vessels By combining these data that covered a whole year, they estimated an abundance of 2700 finless porpoises in the Ariake Sound This figure does not include Tachibana Bay

Yoshida et al. (1997) made repeated aerial surveys in the Ariake Sound-Tachibana Bay area covering 8 months of the year The parallel track lines were arranged west/east with an interval of 2 nautical miles As no seasonal difference in density

sighting data to obtain a total abundance of 3093 (Tables 84 and 85) With a total surface area of 2465 km2 (Shirakihara and Shirakihara 2002, in Japanese), the density of finless porpoises in the Ariake Sound-Tachibana Bay area is estimated at 125/km2 We should note that the total survey effort and the total surveyed area can be used in estimating abundance; however, for calculating density it is more appropriate to use the area finless porpoises are confirmed to inhabit Inclusion of offshore water not inhabited by the species biases downward the true density of the species in its habitat

Using the same method as Yoshida et  al. (1997), (Shirakihara and Shirakihara 2002, in Japanese) surveyed this area in 2000, sighted 157 finless porpoises, and estimated the abundance at 3807 (Table 85) The density was 154/km2

The three abundance estimates extending over the past 12  years were not statistically different from each other, but that does not necessarily mean that the population size remained stable during the period Rather, the interpretation should be that the statistical power was insufficient to detect change in the abundance if any did occur

A geographical description of the Inland Sea is available in Section 841 Kasuya and Kureha (1979) made the first attempt to study the distribution and abundance of Japanese finless porpoises there They made nine series of surveys from ferryboats during the 2 years and 7 months from April 1976 to October 1978 There was a single observer for most of the cruises The vessel speed of the 28 ferryboats that covered 34 segments (from embarkation to disembarkation) ranged from 85 to 185 knots with an average of 113 knots (simple average) and 123 knots weighted by cruised distance The abundance estimates from the nine surveys were not statistically different from each other (Figure 86) Two independent surveys in April resulted in the highest figures of about 3000 porpoises (with approximately 90% confidence interval of 2800-4400) and about 2450 (1700-2800) The estimate of g(0) was 05, yielding an estimate of abundance of 4900 finless porpoises in the Inland Sea It is unclear why they used the smaller figure of the two However, if the larger figure (3000) was used, the estimated abundance would be about 6000 individuals (Table 85)

TABLE 8.5 Abundance of Japanese Finless Porpoises Estimated by Sighting Surveys

inferior to recent methods of survey and analysis, and the resultant abundance estimates will have little value except for historical importance as a record of sighting density and distribution pattern of the species in the Inland Sea The problems in Kasuya and Kureha (1979) in addition to those stated earlier are as follows: (1) Many of the survey tracks were in parallel with the coastline or perpendicular to the expected density gradient, and more sighting effort was in high-density areas (although the data were stratified by offshore distance to decrease the bias) (2) A view ahead was not available on most of the vessels (left and right sides were observed), and the sighting angle was assumed to be 90° and the radial distance was used as perpendicular distance, which could have caused an underestimate (the true perpendicular distance has to be calculated by r sinα, where r is radial distance and α sighting angle) (3) The problematic estimation of g(0) at 05 and surveys by single observers could have overestimated the true g(0) value (underestimating the abundance)

The surveys by Shirakihara et  al. (2007) were part of a series of systematic surveys made in 2000 to estimate the abundance of finless porpoises in Japanese waters (Seibutsu Tayosei Senta [Biodiversity Center] 2002, in Japanese) They surveyed the Inland Sea and part of the Kii Channel near Naruto Pass at the southeastern corner of the sea but excluded its southwestern corner, which was deeper than 50 m (finless porpoises were not expected to occur there) The flight track lines were arranged in a north/south direction with an interval of 4′ (about 33 nautical miles or 61 km) The eastern part of Naruto Pass was added because it was a high-density area recorded by Kasuya and Kureha (1979), but the survey ended with sighting of only one group in that area They excluded Osaka Bay, most of the Kii Channel, and a small area north of Kanmon Pass (the southern Sea of Japan), where some finless porpoises of the Inland Sea population were known to occur

It is worthwhile noting that Shirakihara et  al (2007) also found a higher density of finless porpoises in the Suwo Nada area (approximately west of 132°10′E and north of 33°35′N) in the western Inland Sea than in the middle and eastern Inland Sea This was quite different from the results reported by Kasuya and Kureha (1979) but was the same as reported by Kasuya et  al (2002), which was clear evidence of historical change in the distribution pattern or local depletion of the species between late 1970s and late 1990s in the Inland Sea In observing a total of 22181 km in 60 track lines, Shirakihara et al. (2007) sighted 148 schools of finless porpoises (no other cetacean species were present) The mean school size was 156, which allows calculation of the total number of porpoises in the 148 schools at 231 The estimated abundance was 7,572 individuals for the surveyed area of 13,949 km2 (Table 85) The abundance in the western Inland Sea west of 132°10′E, with inclusion of Suwo Nada and Beppu Bay (33°15′N-33°30′N, 131°30′E-131°45′E) southwest of Suwo Nada, was 5677 or 75% of the total abundance in the Inland Sea

The density of the finless porpoises was 131/km2 in the Suwo Nada area in the western Inland Sea, 0506 in Beppu

132°10′E A relatively high density in the Suwo Nada area seems to reflect a less damaged environment with broad shallow seas It should be noted that the density was much lower in the islands area further east The distribution pattern was quite different from that recorded by Kasuya and Kuraha (1979) in the late 1970s

8.4.3.4 Ise Bay and Mikawa Bay Ise Bay is about 70 km long and 30 km wide, has a maximum depth of 38 m, and is connected to the Pacific with a 12-km wide opening at the southern end The shallower Mikawa Bay is 40 km long and 12 km wide, located to the east of Ise Bay and connected to Ise Bay with a 11-km wide entrance near the opening to the Pacific This Ise Bay-Mikawa Bay area was surveyed twice in the past The first was a collaborative study between the Toba Aquarium and the whale scientists of the Far Seas Fisheries Research Laboratory Twelve cruises on small vessels were conducted in a 35-year period from 1991 to 1995 (Miyashita et al. 2003, in Japanese) The track lines were arranged in a sawtooth pattern along the coast Two observers were placed on a sighting platform with eye height of 35-6 m, depending on the vessel The survey speed was 10-13 knots (185-241 km/h) Sighting was conducted in Beaufort sea state 4 or less However, later analysis revealed that the sighting rate of 80 animals/100 km at sea state 0-2 (252 porpoises) declined to 23 at sea states 3 and 4 (23 porpoises), and the authors excluded the data obtained at sea states 3 and 4 (23 porpoises and 1003 km of survey effort) from the subsequent abundance estimation

Miyashita et  al. (2003, in Japanese), using g(0) = 0899, estimated the finless porpoise abundance in Ise Bay-Mikawa Bay at 1046 (CV = 028) in April-June when the highest density was observed (Table 85) The value of g(0) was calculated as the probability of a finless porpoise surfacing on a track line within visual distance using surfacing intervals of a porpoise kept in an aquarium tank of 45 m depth It seems to me that the g(0) value was an overestimate (underestimating abundance) due to the following reasons: (1) the diving time of aquarium animals that live in a shallow tank and are fed from the surface must be shorter than that of wild animals that live in deeper water and feed in the water column or on the bottom and (2) it is impossible for observers to identify all the individuals that surface on the track line within their visual distance (some must be overlooked)

Miyashita et al (2003, in Japanese) found that the densities of finless porpoises in Mikawa Bay were more than twice higher than those in Ise Bay in any season and that the densities tended to be lower from October to March and higher from April to June They postulated that finless porpoises move out of the bays in winter (Table 86) However, the large coefficients of variation that accompany the density estimates throw some doubt on the conclusions The density differences between the two connected bays disagree with the results of a later study by Yoshioka (2002, in Japanese) Some anthropogenic factors are suspected to be behind the disagreement (see the following text in this section) However, it is premature to exclude the possibility of seasonal density changes in Ise and Mikawa Bays I observed finless porpoises on the Pacific

coast of the Shima Peninsula just west of the entrance of Ise Bay in April-June 2000 (Kasuya unpublished), and there are numerous records of stranding of the species along the coast (Figure 82)

After the shipboard survey in 1991-1995 (Miyashita et al. 2003, in Japanese), an aerial survey was conducted in May 2000 by Yoshioka (2002, in Japanese) as a part of a project of the Environmental Agency (Seibutsu Tayosei Senta [Biodiversity Center] 2002, in Japanese) This survey had east/ west flight tracks in Ise Bay and north/south tracks in Mikawa Bay The interval of the track lines was 2 or 3 nautical miles (37 or 56 km) The survey covered the Pacific coast near the entrance of Ise Bay within the 100 m isobath because earlier information indicated the presence of the species there, but the survey ended with no sighting of finless porpoises there and this area was excluded from subsequent analyses From the survey, Yoshioka (2002, in Japanese), assuming g(0) = 1, estimated the abundance of finless porpoise at 3038 in Ise Bay and 705 in Mikawa Bay (Table 85) and the density at 195/km2 in Ise Bay and 138 in Mikawa Bay

The aerial survey resulted in much higher abundance estimates than those estimated by Miyashita et al. (2003, in Japanese) and the disagreement was greater for Ise Bay than for Mikawa Bay As stated by Yoshioka (2002, in Japanese), this could be due to differences in methodology, that is, aerial versus vessel surveys I once participated in a survey of Ise Bay with Miyashita in the winter, when the survey vessel was unable to approach close to the coast because of (1) shallow water depth and (2) the presence of nets for alga culture in the winter season These factors could have caused downward bias in the abundance estimate, but there is some uncertainty about how the obstacles differed between the two bays

8.4.3.5 From Tokyo Bay to Sendai Bay The current Tokyo Bay (35°30′N, 139°50′E) is about 30 km wide and 54 km long from the northern end to the narrowest southern entrance, which is about 9 km wide (at 35°15′N) Tokyo Bay opens through this entrance to eastern Sagami Bay that opens to the Pacific It is full of vessel traffic, and most of the natural shorelines have disappeared through reclamation for human use Finless porpoises were known in Tokyo Bay since early times (Otsuki 1808, in Japanese) Strandings and sightings were reported from Yokosuka (35°17′N, 139°41′E)

and Ishikawa (1994, in Japanese) More recent sightings were made in 2000 (Amano et  al. 2003) and in 2003 (Kasuya unpublished) at the northern end of the bay (off Kasai), near the central bank (Nakanose), and eastern shores of the bay (off Kisarazu and Futtsu) Amano et al. (2003) made secondary sightings of five porpoises in two groups when returning from their aerial sighting survey transit (Figure 87); these

Estimated Densityof Finless Porpoises in Ise and Mikawa Bays Shown by Population Size/km2 and Accompanying CV in %

no estimate of abundance of finless porpoises in Tokyo Bay

A survey reported by Amano et al (2003) was also a part of the project of the Ministry of Environment in 2000 (Seibutsu Tayosei Senta [Biodiversity Center] 2002, in Japanese) Their survey covered an area from western Sagami Bay, or the west coast of the southern part of the Boso Peninsula south of Minato (35°15′N, 139°52′E), and the Pacific coast from the tip (34°54′N, 139°53′E) of the Boso Peninsula to northern Sendai Bay (38°20′N), which is a semicircular bay with north/south length of 18  km and a 40  km wide opening to the Pacific The total 34 survey tracks, arranged at an interval of 111 km, covered waters within the 60 m isobath (Figure 87), where finless porpoises were expected to occur from records within the 50 m isobath in other habitats

This was the first complete survey of finless porpoise habitat north of Tokyo Most of the sightings of the species occurred in two clusters on the Pacific coast, one from 35°18′N to 36°30′N and the other north of 37°20′N There were no sightings in eastern Sagami Bay (west coast of the southern Boso Peninsula), and there was only one sighting of two porpoises along the 90 km shoreline between the two clusters Amano et al. (2003) thought that the distribution gap between latitudes 36°30′N and 37°20′N likely represented a boundary between still unconfirmed local populations of the species While Amano et al. (2003) did not find the species on the east coast of the southern Boso Peninsula south of 35°15′N, the presence of the species there is known by frequent opportunistic sightings off Ohara (35°15′N, 140°24′E) and a stranding at Kominato (35°07′N, 140°12E′) reported by Ishikawa (1994, in Japanese) Thus, a firm conclusion cannot be reached from a single aerial survey, but the noted distribution gaps are worthy of further examination for the possibility of local populations in the areas, that is, local populations in Tokyo Bay, along the Pacific coast of Chiba-Kashima Nada (35°N-36°30′N) and along the coast of Fukushima-Sendai Bay (37°20′N-38°24′N)

Amano et  al. (2003) found that finless porpoises were present only inside the 40 m isobath and within 20  km from the shore The water depth at the sighting of the species increased with increasing distance from the shore up to 5  km offshore; then, the correlation was lost (Figure 88) With the exception of one sighting at a depth of 35-40 m, all the sightings occurred inside the 35 m isobath I interpret this to mean that finless porpoises were in waters over 5 km from the shore because the depth there was shallow The range of finless porpoises extends offshore if the depth is shallow (see Section 831)

Amano et al. (2003), using sighting effort within the 50 m isobath and 51 primary sightings, estimated the abundance of fi nless porpoises at 3387 (Table 85) and the density at 0502/km2 The exclusion of sighting effort between the 40 and 50 m isobaths, where no sightings occurred, could have narrowed the confidence interval of the estimate; however, average abundance would not change, because both effort and size of the study area decreased However, the use of the area of 6750 km2 that includes area between 40 and 50 m isobaths

could have biased downward the density of finless porpoises in their habitat

8.4.4.1 Background It was on September 25, 1997, that I noticed that something strange could have happened to finless porpoises in the Inland Sea As an adviser to the Ajia Kosoku Co Ltd (Asian Aero Survey Co Ltd), which wanted to investigate the feasibility of using an airplane for finless porpoise studies, I joined a Cessna flight I suspected that the company intended to prepare for a possible future project of the government We planned to carry out a basic technological test for the aerial survey of finless porpoises Our plane departed from a small airport near Osaka, crossed the Kii Channel, and flew in the direction of the eastern Inland Sea area that includes Naruto Pass, Shodo-shima Island (34°30′N, 134°15′E), and the Ieshima Islands (34°40′N, 134°30′E), where the species was abundant in the late 1970s (Kasuya and Kureha 1979) The flight of 3 hours and 50 minutes ended with unexpectedly low sightings of only three porpoises in a group near Shodo-shima I found it hard to believe that I failed to see even one porpoise in the Naruto Pass area, where in the late 1970s I had experienced such a high density that it was difficult to record I was also astonished by the destruction on the Ieshima Islands, where all the hills and trees behind the villages had been scraped off to sea level, which was a result of mining to produce reclamation materials for the construction of the new Kansai Airport The flight company made an additional independent flight the next October to Kawanoe (34°01′N, 133°34′E), further west of Shodo-shima Island, but it also ended with no sighting of finless porpoises

the entire Inland Sea

The Inland Sea finless porpoises were surveyed twice over an interval of about 22  years using ferryboats and similar sighting methods Although these surveys were not suitable for abundance estimation, they were able to indicate historical changes in the density and distribution pattern of the species The first survey covered about 25 years from April 1976 to October 1979 (Kasuya and Kureha 1979) and the second about 1 year from March 1999 to April 2000 Sighting data from the second survey were analyzed in comparison with the data from the first survey for changes in density and distribution pattern between the two surveys, and the results were published in Kasuya et al. (2002), with some additional analyses in Kasuya (2008, 2010, both in Japanese)

The Inland Sea was faced with numerous environmental problems in the 1970s, for example, frequent oil spills, increased red tides due to eutrophication by city sewage, damage by the red tides of cultured yellowtail, and progress of chemical pollution by discharges from industry and agriculture Newspapers concluded that chemical pollution was responsible for reported malformed fishes To respond to this situation, the government promulgated Provisional Measures for Protection of the Environment of the Inland Sea in October 1973 (came into effect in November of the same year) and then modified it in June 1978 to Special Measures for Protection of the Environment of the Inland Sea (came into effect in June 1979) The major difference between the two regulations was a change from controlling the concentration of pollutants in discharges to controlling total pollutant amount allowed to be discharged Efforts toward the objective included decreasing phosphorus in sewage by improving detergents for family use and constructing sewage processing facilities to decrease the amount of nitrogen discharged According to the Setonaikai Kankyo Hozen Kyokai (Association for Conservation of Environment of Inland Sea, 1996, in Japanese), the total number of oil spills reported for the Inland Sea, Osaka Bay, and the Kii Channel increased from 155 in 1970 to a peak of 874 in 1973 and declined to around 100 in 1994 The incidence of red tides increased from 79 in 1970 to a peak of 326 in 1976 and then decreased to 87 or fewer in 1995 and thereafter Although these improvements reflected effort toward the conservation of the Inland Sea, further improvement remains dubious

Under such circumstances, it was feared that many of the chemical pollutants would be transferred to the finless porpoises through the food web and be accumulated in the body Even if individual porpoises did not apparently suffer from acute poisoning by red tides or artificial chemical pollutants, chronic effects might affect the population trend through changes in mortality or reproductive rates We thought it necessary to record the current status of the finless porpoise population in the Inland Sea in order to identify possible future changes, and myself and K Kureha, both at the Ocean Research Institute of the University of Tokyo, started the first stage of a survey with the financial support of the World Wildlife Fund, Japan

The first survey by Kasuya and Kureha (1979) placed high priority on determining the seasonal change in the density and geographical density pattern in relation to the environment, and the plan was to repeat numerous ferry tracks within the limits of time and budget The recorded information included ferry track, sightings of finless porpoises (position, school size, growth stage, substructure within the school), and sea surface temperature The ferryboats used were mainly for both passengers and cars, but those for only passengers were also used provided that they were not high-speed ferries Sighting effort was carried out only when Beaufort sea state was below 3, which is when white caps start to be seen

Kasuya et al (2002) chose April-June for the second set of surveys in 1999 and 2000, because more finless porpoises were sighted during those months in the first surveys However, due to survey scheduling, there was a small amount of effort in late March and early July included They extracted data from the first set of surveys in the late 1970s only for those months and same ferry tracks for comparison between the first and second sets of surveys (Figure 89) During the approximately 22 years between the two surveys, bridges were constructed to connect some islands and some ferry operations were discontinued, but we placed the highest priority on selecting the same tracks and making multiple trips on them In addition to the second ferry-based survey, we used an additional vessel Seisui-maru, a training vessel from Mie University, for a total of 9 days on three cruises

The Seisui-maru started at Naruto Pass, the southeastern entrance, cruised through the Inland Sea toward the west, and completed the survey in Hayasi Pass, the southwestern exit In the Inland Sea, the track lines of this vessel included both duplication of ferry tracks and entirely new tracks to obtain information in areas of no ferry operation Sighting effort on the Seisi-maru was carried out with the participation of A Kawamura and K Shirakihara of Mie University, M Furuta and staff of the Toba Aquarium, and M Shirakihara and crew of the vessel; therefore, the sighting efficiency should have been much superior to that in earlier surveys, which were staffed by a single observer These data were combined for analysis

If we had found an apparent increase in the density of finless porpoises through comparison of the two sets of data, we could not have reached any conclusion, because the second survey contained cruises with higher sighting efficiency However, we obtained a lower density in the more recent survey, so a declining trend was judged valid It was mentioned earlier how the ferry tracks were selected However, one weak point of the study was the absence of exact data to confirm the assumption of Kasuya et  al. (2002) that vessel speeds were similar between the two surveys

The original sighting data from the first and second surveys are deposited at the National Science Museum in Tokyo for open use by any scientists who may wish to confirm the analysis or carry out new analyses

8.4.4.3 Diminished High-Density Area Now, let us compare the positions of sighting of finless porpoises between the first and second surveys, which were

about 22 years apart (Figure 89) Most of the track lines are expressed by two numbers connected by a hyphen, which indicate ports of embarkation and disembarkation, but short track lines such as that in Naruto Pass are expressed by a single number The solid lines in Figure 89 indicate track lines surveyed in both surveys, black circles indicate sighting positions in the first survey, and the white circles indicate those of the second survey However, in high-density areas they represent only the geographical range of sightings rather than number of sightings Some of the track lines are accompanied by almost continuous black circles indicating extended area of high sighting density in the first survey, but white circles (second-survey sightings) are much sparser in those areas This visualizes high-density areas that disappeared during the 22 years

Among the cruise tracks in the central and eastern Inland Sea (upper panel of Figure 89), Naruto Pass (1), around Shodo-shima Island (tracks 2-3, 2-4, 2-5, and 2-6), from

Omi-shima to Kure via Osaki and Kamagari (track 11-17), and Mitsu-hama to Yanai via Nakajima and Hashira-jima (track 18-22) are such areas of drastic decline in density of sightings The second cruise of the Seisui-maru made along the track from Kure (17), to Nasake-jima (20, 21), to Yanai (22) was of particular interest, because numerous spotters including ship’s crews searched for finless porpoises under ideal weather condition but ended with so few sightings that I could not believe that we were cruising the same sea that I surveyed 22 years before (bottom frame of Figure 89)

Sighting density in the Suwo Nada area was apparently higher than in the central and eastern Inland Sea, and the density decline was less pronounced This observation agrees with the recent distribution pattern of the species observed through aerial survey by Shirakihara et al. (2007) In addition to the finless porpoise, I had the impression that seabirds had also declined considerably from the late 1970s to 1990-2000, but I did not collect seabird data for analysis

historical change in the finless porpoise density in the Inland Sea Now, I will confirm it using mean number of finless porpoises sighted on each track line (Table 87) The survey track from Shodo-shima to Uno (track 2-6), for example, was traveled five times in the survey in the late 1970s with an average of 820 finless porpoises sighted (range 0-27), but eight trips on the same track in 1990-2000 resulted in an average sighting of 013 (range 0-1) The probability of such a difference arising by chance is 12%; thus, it could be interpreted to be due to a real change in the density of the species If the probability is less than 5%, the difference is accepted as real, stated as significant in Table 87 All of the 18 tracks available for this analysis (Table 87) showed an apparent density decline, and 11 of them showed a statistically significant difference The decline was not significant for seven tracks, but this result could be due to small sample size

Two tracks in the Suwo Nada area showed a relatively minor density decline The track between Yanai and Iwaishima (track 22-23) showed decline to 885% of the density in the late 1970s, and the difference was statistically significant Another track between Tokuyama and Takedazu (track 24-25) showed decline to 50% from the earlier survey, but the difference was not significant

8.4.4.4 Density Decline and Distance from the Shore Finless porpoises usually inhabit waters within the 50 m isobath, so they can be found in such offshore waters as the middle of the East China Sea, where depth is only about 50 m (Section 831) I have shown that in the Inland Sea of Japan their distribution is apparently influenced also by distance from the shore, although the degrees of contribution of water depth and offshore distance are unknown Table 88 shows the sighting density of finless porpoises in relation to offshore distance for both the first and the second surveys Data used here are from the same months and include those in Table 87 plus those from other cruise tracks available in Kasuya and Kureha (1979) and Kasuya et al. (2002)

In the case of the first survey in the late 1970s, the density in the nearshore stratum (<1 nautical mile) was 182/100 km in the central and eastern Inland Sea, 91 in the middle stratum (1-3 nautical miles), and 2-3/100  km in the offshore stratum (>3 nautical miles) with limited data It is apparent from these data that the density of the finless porpoise rapidly declined with increasing distance from the shore In the western Inland Sea (west of 132°10′E), the density in the nearshore stratum was 196, which was close to the corresponding figure from the central and eastern parts of the sea However, the middle stratum of the same area had a density of 165, which was considerably higher than the corresponding figure for the central and eastern Inland Sea This was likely to be a reflection of a shallow area extending far from the shore Extremely low density in the offshore stratum was the same in both regions From the earlier discussion, it is inferred that the density pattern of the finless porpoise in the late 1970s was in principle similar between regions of the Inland Sea

TABLE 8.7 Density Decline of Finless Porpoises in the Inland Sea Detected through Repetition of Similar Surveys about 22 Years Apart, the First of Which Was Conducted March 2-June 25 in 1976-1978 and the Second March 30-July 1 in 1999-2000

Now, we compare these results with those of the second survey in the central and eastern Inland Sea The densities from the second survey were only 3% of those in the first survey in the nearshore stratum, 20% in the middle stratum, and apparently zero in the offshore stratum, which means that the density declined in all the three strata during the 22-year period The density from the second survey was 06/100 km in the nearshore stratum and 18 in the middle stratum, and the difference was statistically significant (Kai square test, p <1%), which indicates that the density decline was drastic in the nearshore stratum and that the density gradient had reversed in recent years This probably reflects severe destruction of the coastal habitat

In the western part of the Inland Sea, the density in the nearshore stratum declined to 57% of that in the first survey and in the middle stratum to 49% These declines of moderate degree were also statistically significant (0 < 1%) For the offshore stratum, we do not have sufficient data to see changes in the 22 years

8.4.4.5 Decline in Population Size It is evident that the density of finless porpoises declined in the whole Inland Sea over 22 years Because the sea constitutes a major habitat of the population, the density decline must have reflected a decline in population size Kasuya (2008, in Japanese) attempted to estimate the degree of abundance decline using data in Kasuya et al. (2002) The method was to multiply sighting density by the area size to get indices of abundance for each stratum, that is, (no of individuals sighted)/(distance surveyed, 100 km) × (area size, 1000 km2), and then the indices of abundance for each stratum were

added to obtain a total abundance index Table 88 shows the process of the calculation, and Table 89 compares the results

The results suggest that the abundance of finless porpoises in the central and eastern Inland Sea in 1990-2000 was about 10% of the level of the late 1970s, while the corresponding level was 62% in the western Inland Sea The total abundance index in the late 1970s was 1512, while that in 1990-2000 was 457, which suggested that the Inland Sea population of finless porpoises declined to a level of about 30% of original size in the 22-year period

The proportion of finless porpoises in the western Inland Sea was 443% at the time of the first survey in the late 1970s, but it increased to 813% in the second survey in 1990-2000 This latter figure is in reasonable agreement with 75% obtained from the recent aerial survey of Shirakihara et  al. (2007) (Section 8433) I conclude that the abundance decline of finless porpoises during the 22-year period from the late 1970s to 1990-2000 was greater in the central and eastern Inland Sea and that the major habitat has shrunk to the western part of the Inland Sea and perhaps to the Kanmon area

The population of Inland Sea finless porpoises in 1990-2000 appears to have been only about 30% of that in the late 1970s, although evaluation of this figure requires some caution because (1) it is not accompanied by a confidence interval, (2)  it ignores the possibility of still containing upward bias due to greater number of observers in some recent cruises, and (3) it ignores unknown effects of vessel speed, which might have changed during the period This change has occurred

Regional and Topographic Effects on the Density Decline of Finless Porpoises in the Inland Sea

with a particularly intense density decline in the central and eastern part of the Inland Sea, where the decline was greater in the nearshore stratum than outside and the inshore/offshore density gradient reversed during the period

Finless porpoises live to the age of 20-25 years in both sexes (see Section 852) Although we do not know their natural mortality rate, it is probably higher than expected for species of longer lifespan such as short-finned pilot whales (females live to 62  years) and sperm whales (to around 70  years) If longevity is defined as the age when a cohort decreases to 1% of the 0-year level and assuming a constant mortality rate for finless porpoises by ignoring the common concept that natural mortality rates of mammals are age dependent with higher values in earlier and later stages of life, then the maximum longevity of 25 years gives an average annual natural mortality rate of 168% for the finless porpoise So, it is probable that the average natural mortality rate in the finless porpoise population is somewhere between 10% and 20% If a gross annual reproductive rate of the population (no of births in a year/population size) is equal to the mortality rate, the population will remain stable However, if the mortality rate increases by 25% points and the gross reproductive rate decreases by about 25 points, then the population would decline at an annual rate of 5%, which extrapolates to abundance reduction to about 30% of original in 23  years I suspect that such a situation could have taken place for finless porpoises in the Inland Sea

Kasuya (1997, 2008, 2010, all in Japanese) and Kasuya et  al. (2002) listed several plausible causes of such demographic changes in the finless porpoise population: (1) incidental mortality in fisheries, (2) physiological problems due to chemical pollutants, (3) habitat loss due to reclamation and sand extraction, and (4)  ship strikes They reserved conclusions on the possible effects of (5) interaction with fisheries through food resources, (6) red tides, (7) sound pollution, and (8) epidemics Shirakihara et al. (2007) noted habitat destruction due to sand extraction However, it seems to be more reasonable to consider that several factors jointly worked to reduce survival and reproductive rates of the finless porpoise population, as further discussed in the following

8.4.5.1 Incidental Mortality in Fisheries Shirakihara et  al. (1993) collected finless porpoises killed incidental to fishery operations through contact with FCUs and used them for the study of life history Among the 114 specimens they collected from the west coast of Kyushu and the western Inland Sea (off the east coast of Kyushu), 84 were known to have been killed in particular fishery operations (58 in bottom-set gillnets, 17 in surface drift gill nets, 7 in trap nets, 1 in a trawl net, and 1 in a drifting discarded net) Because these specimens were obtained through requests to the FCUs in advance of such incidents, there could have been some bias toward fisheries known to have such takes

In Ise Bay, the Toba Aquarium used drift gill nets for live capture of finless porpoises in the 1970s and 1980s but learned the method was likely to result in high mortality, switched to the use of a sardine seine net, and caught nine porpoises in November 2004 The seine net was 500 m long and operated by two vessels (Furuta et al. 2007, in Japanese) Kasuya et al. (1984) reported the incidental mortality of finless porpoises in seine nets for Spanish mackerel in Ise Bay and reported that the Toba Aquarium hired mackerel fishermen to catch live finless porpoises

The operation of gill nets, seine nets, and trawl nets is common in the habitats of Japanese finless porpoises Taking of a large number of finless porpoises, about 300 per year, was reported in fishing operations off the southern coast of the Korean Peninsula, most believed to have been killed in trawl fisheries (An et  al. 2010; IWC 2010) Although Shirakihara et al. (1993) listed limited incidents of finless porpoises taken in trawl fisheries, more investigation of this is needed

We do not know how finless porpoises move and mix within the Inland Sea However, if finless porpoises are killed by local fisheries, the local density change will be propagated in time to the entire Inland Sea population through the movement of individuals and distribution shift to lower density areas Thus, local damage to the species can be masked and its identification difficult

Table 810 lists the records of incidental mortality and stranding of finless porpoises in the last 5 years as collected by the Fisheries Agency Extremely high mortality has been

Regional Difference in the Abundance Decline of the Inland Sea Finless Porpoises during an Approximate 22-Year Period from 1976-1978 to 1999-2000

reported from Mie and Aichi Prefectures on the coast of Ise Bay and Mikawa Bay and from Yamaguchi Prefecture in the eastern Inland Sea and Kanmon area This reflects a high reporting rate due to conservation groups working on those coasts in association with local aquariums and universities Apparently low incidence of such reports in other regions should be considered to be due to a limited level of observation effort Yoshida (1994, in Japanese, cited in Yoshida et  al. 1997) estimated in his doctoral thesis that 37 finless porpoises were killed annually in fishery operations in the Ariake Sound-Tachibana Bay area in western Kyushu, while Fishery Agency statistics reported the incidental mortality of only around 5 per year in this area plus Omura Bay I once attempted in the early 2000s to collect the records of incidental mortality from bottom-set gill-net fishermen on Iwaishima Island (33°46′N, 131°58′E), a small island in Suwo Nada in the western Inland Sea, and had the impression that the incidental mortality of the finless porpoise caused by them alone could exceed the annual figure for the prefecture at the time reported by the Fisheries Agency

Reliable information on the mortality of finless porpoises in various types of fisheries is critically important for establishing conservation plans for the species It should be considered as an obligation of users of marine resources to collect the mortality data and take necessary mitigation actions However, it is hard to expect that all the fishermen will respect the Fishery Resources Protection Act (see Section 591) and report finless porpoises caught in their fishing gear to the government Rather, most fishermen will avoid the trouble of formalities and discard the carcasses overboard Some of those discarded carcasses will be stranded and so reported Government officials working in fishery management are not ignorant of this

fact, but it seems to me that they pretend otherwise to escape the burden of mitigation One example of mitigation is use of sound emitters, called pingers, which have been successfully attached to fishing nets to decrease the incidental mortality of harbor porpoises (Northridge and Hofman 1999)

Caution is required in evaluating figures on incidental mortality or stranding of porpoises because they can be influenced by factors in addition to population size and fishing activities, for example, by the concern of citizens, which is affected by the social atmosphere on conservation Kasuya et al. (2002) analyzed the trend of incidental mortality in fisheries and stranding of finless porpoises in the Inland Sea in the 29 years from 1970 to 1999, mostly based on data collected by the Institute of Cetacean Research They observed a low frequency of such incidents during the recent 14 years from 1986 to 1999; 2 cases of incidental mortality and 12 stranded carcasses were recorded with an annual fluctuation of 0-6 and an annual average of only one (6 records in 1998 and 0 in 1999 included) However, they indicated higher figures for the preceding 16-year period from 1970 to 1985 with a total of 62 (37 incidental kill, 16 stranded carcasses, and 9 deaths of unknown cause) with an annual mean of 39 and an annual fluctuation of 1-6 Kasuya et al. (2002) felt that the recent decline in such records was apparently inconsistent with recently increasing concern of local communities about cetaceans and suggested the possibility that it could reflect decreased abundance of the species in the Inland Sea that occurred in the mid-1980s

8.4.5.2 Ecological Interaction with Fisheries Only limited information is available on the prey items of finless porpoises in Japan Mizue et  al. (1965, in Japanese) reported horse mackerel, sardines, and squid in the stomachs

Recent Estimates of the Abundance of Finless Porpoise in Japan (from Table 8.5) and the Total Reported Figures of Strandings and Incidental Mortality of the Species during a 5-Year Period in 2000-2004

(1976, in Japanese) reported sand lance, squid, and crustaceans in the stomachs of finless porpoises in Ise Bay Shirakihara et al. (2008) analyzed the stomach contents of finless porpoises incidentally killed in fisheries in two populations off western Kyushu: (1) Omura Bay and (2) Ariake Sound-Tachibana Bay Their sample contained 9 individuals including 1 juvenile from Omura Bay and 78 individuals including 20 juveniles from Ariake Sound-Tachibana Bay The prey in the Ariake SoundTachibana Bay sample was more diverse than in the Omura Bay sample Food items in the former sample identified to species included three species of Cephalopoda (mainly Octopodidae, Sepiidae, and Loliginidae) and 26 species of Teleostei (mainly Clupeidae, Engraulidae, and Sciaenidae), in addition to a small number of crustaceans Prey in the Omura Bay sample contained six species of Teleostei, two species of Cephalopoda, and some crustaceans, with Gobiidae and Atherinidae in most stomachs The authors thought that the difference in the diversity of prey reflected the marine fauna of the two areas rather than the difference in their sample sizes They noted for the Omura Bay samples that the food of juvenile finless porpoises below 1 year of age was dominated by small bottom fishes of the families Gobiidae and Apogonidae and Cephalopoda, while adults did not consume Gobiidae, and they concluded that the juveniles of weaning age had food preferences different from those of adults Although this conclusion could be correct, it could also be true that an interaction between seasonal change of prey fauna and a seasonally limited weaning period could have affected their analysis Therefore, it is necessary to compare stomach contents between different growth stages using the samples obtained in the same season before firm conclusions can be reached

Barros et  al. (2002) reported the stomach contents of 31 finless porpoises stranded in the Hong Kong area to be composed of fish, squid, shrimp, and octopus (in a decreasing order) Among the fish families, Apogonidae was the most numerous, followed by Sciaenidae, Engraulidae, and Leiognathidae, and Loliginidae was the most common among the cephalopods These comprised 77% of the stomach contents examined From these observations, Barros et  al. (2002) concluded that finless porpoises in the Hong Kong area (1) feed in coastal waters (food items were all coastal species), (2) feed also on benthic organisms (Sepioidea, Octopoda, and Sciaenidae), and (3) also feed in the water column (Apogonidae, Engraulidae, and Trichiuridae) They also indicated that some of the prey items are species targeted by trawl fisheries in the region

The coastal marine environment in temperate latitudes inhabited by finless porpoise populations fluctuates greatly with season both in water temperature and prey fauna available to the porpoises Kasuya and Kureha (1979) recorded the range of surface temperature in the Inland Sea from 6°C to close to 29°C, and a slightly higher temperature was also recorded (Table 101) Sendai Bay, the northern limit of the species in Japan, experiences lower winter temperatures than the Inland Sea Japanese finless porpoises are probably enjoying the high productivity of shallow inland waters through

and ability to utilize a broad spectrum of seasonally variable food resources

Marine animal products from the Inland Sea, excluding products of aquaculture, recorded both a rise and then a decline in the 1965-1994 period (Setonaikai Kankyo Hozen Kyokai [Association for Conservation of Environment of Inland Sea] (1996, in Japanese)) The total increased from slightly under 200,000 tons in 1965 to 350,000 tons in 1982 and then declined to the previous level of 200,000 tons in 1994 Change in the harvested fish fauna accompanied the change in the total amount of fishery production; during the high-production phase, Japanese anchovy Engraulis japonicus and sand lance Ammodytes personatus increased and jack mackerel Trachurus spp and an octopus Octopus vulgaris decreased, but the latter two increased during the later low-production phase This probably reflected the abundance of phytoplankton feeders such as anchovy that increased responding to eutrophication of the Inland Sea in the 1970s and 1980s and decreased subsequently following decline in phytoplankton in the 1990s due to the improvement of the water quality

In view of the flexibility of prey preference of finless porpoises, it is unlikely that there was competition at the level of the entire Inland Sea between finless porpoises and fisheries for food resources and that the competition caused the observed decline in the abundance of the porpoises in the 1980s and 1990s However, this does not preclude the possibility of competition in a particular local habitat

8.4.5.3 Red Tides Red tide is so named because the seawater changes to red or brown due to an outbreak of unicellular organisms such as dinoflagellates and diatoms Fish may suffer mechanical damage to the gills or die from suffocation due to consumption of dissolved oxygen by carcasses of the red tide organisms However, marine mammals appear to undergo more damages from poisonous chemicals produced by some red tide organisms There are examples of mortality of humans from consuming mussels that accumulated saxitoxins from plankton and of humpback whales that consumed similarly poisonous mackerel Gray whales, bottlenose dolphins, manatees, sea lions, and sea otters have been reported to have died from ingesting brevitoxin and domoic acid from red tide Damage by ciguatoxin to Hawaiian monk seals is suspected (Van Dolah et al 2003) These acute cases are more easily identified than chronic effects

Four red tide incidents were recorded in the Inland Sea in 1950, followed by a slow increase to 18 in 1960 and then a linear rapid increase to a maximum of 299 in 1976 After the maximum in 1976, the incidence of red tides in the Inland Sea slowly declined to 188 in 1980, 170 in 1985, and 108 in 1990 The number of red tides subsequently remained almost stable at around 90-100 until 1994 (Setonaikai Kankyo Hozen Kyokai [Association for Conservation of Environment of Inland Sea] (1996, in Japanese) Although I am uncertain how accurate the counting of red tides is, the numbers mentioned earlier give some idea of the annual trend The red tides were reported from almost the entire range of the Inland Sea during

to shrink, and as of 1995, they were almost limited to the eastern half of the Inland Sea ranging from Kagawa and Okayama Prefectures to Harima Nada and to Osaka Bay in the east This improvement could have been a result of regulation of discharge of nutrients into the sea

Setonaikai Kankyo Hozen Kyokai (Association for Conservation of the Environment of the Inland Sea; 1996, in Japanese) also reported damage by red tide to aquaculture such as that of yellowtail and red sea bream The amount of the damage was 199 billion yen (33 billion yellowtail plus 5600 tons of bream) during the 24 years from 1972 to 1995 The red tide organisms that caused the damage were reported to have been Chattonella (8 incidents) in 1972-1988 and Gymnodinium (15), Heterosigma (8), Noctiluca (3), and Gonyaulax (2) in 19911995 Several species of the genus Chattonella are known to produce brevitoxin and those of Gymnodinium to produce saxitoxins (Van Dolah et al. 2003) However, no incidents have been reported of mortality of marine mammals or seabirds due to ingestion of marine algal toxins in the Inland Sea It will be valuable to investigate the cause of mortality of aquaculture fishes

I could not find any cases where marine mammals were impacted by algal toxins in Japan, although there were reported cases where scallops cultured off the Pacific coast of northern Japan became unsuitable for human consumption due to the accumulation of algal toxins It needs to be confirmed whether Japanese finless porpoises have or have not received acute or chronic damage from algal toxins If acute poisoning really happened to finless porpoises in the Inland Sea, it is likely to have been noticed The effect of chronic poisoning, if it has happened, could still be affecting the population dynamics of the current population Red tides that occurred everywhere in the Inland Sea during the 1970s and 1980s are currently limited to smaller areas, but if they are producing algal toxins, it could affect the density of finless porpoises in the broader range through the movement of prey animals as well as finless porpoises

8.4.5.4 Chemical Pollution The Inland Sea of Japan has a geographical structure likely to accumulate pollutants, because it opens to the outer sea through only four small openings and is surrounded by industrial, agricultural, and urbanized areas Nutrients and

of the pollutants discharged through these human activities, such as chlorinated organic compounds (eg, PCB, DDT, and their derivatives) are poisonous, persist in the sea for long period, and are taken into phytoplankton to be transmitted to zooplankton, fish, and marine mammals through the food chain The levels of concentration of such molecules increase with a trophic step and become particularly high in cetaceans because they are top predators, live a long life, and have limited physical ability to break down the chemicals

These pollutants in female cetaceans are transferred to the young through the placenta and milk, which means that females can lower their burden of the pollutants through reproduction and that pollutant level of newborns depends on the reproductive history of their mother On the other hand, the accumulation process of such pollutants in males tends to be simpler, with levels usually increasing with age, because they do not have such opportunities to shed the pollutants through reproduction This is the reason why males are often selected as an indicator of pollutant level in a population and why maximum figures are compared between populations

Kasuya et  al. (2002) identified high concentrations of PCB, DDT, and organic tins in finless porpoises in the Inland Sea (Table 811) and suggested the possibility to reduce the survival rate and reproductive rate of the population The pollutant levels exceed levels in toothed whales in the outer seas off Japan or those of belugas in the St  Lawrence estuary, which recorded the level of PCBs in blubber at 100 ppm in females and 250 ppm in males This beluga population was once hunted down to 600-700 (12%–14% of the initial level) by 1979 It did not show signs of recovery under protection that started in 1980, and the pollutants are suspected to be causing a high incidence of cancer identified (Martineau et al. 1994)

DDT is an insecticide that has been used worldwide and is still in use in some limited countries The density in marine waters is believed to be declining However, different from the situation for DDT, the level of PCBs in seawater does not show a declining trend (Reijinders 1996) One to 2 million tons of PCBs have been produced in the world since 1929, and 31% has already been discharged into the environment; 65% is believed to be controlled in some way, but the risk of

TABLE 8.11 Highest Recorded Levels of Persistent Pollutants in the Finless Porpoise in the Inland Sea

stopped in 1972, and their discharge into the environment is prohibited Dioxin is a by-product of chemical industries and incineration of city garbage

These chlorinated organic compounds are known to disrupt the normal function of hormones and the immune system in laboratory animals and to prevent fetuses and suckling calves from the normal development of reproductive functions Dioxins are the most powerful artificially produced chemicals, and their daily allowance level is 200 pg (10-12 g) for a person of 50 kg in Japan The pollutant levels of some Inland Sea fishes are so high that the consumption of one piece of sushi with 30 g of spotted shad is sufficient to exceed the daily limit, because shad in the Inland Sea contains 8-9 pg/g of dioxins Organic tin compounds have been discharged into the Inland Sea since the 1960s through maritime use for preventing fouling of vessels and nets of aquaculture by organisms and from terrestrial use for plastic products, floor wax, and cleaning industries One of the well-known environmental effects of organic tin compounds is the formation of a penis in a female snail It is also known to disrupt the functions of hormones and immune systems of mammals, with strong effects on animals in the early stages of development, which is similar to the effect of chlorinated organic compounds (Colborn and Smolen 2003)

We humans have discharged numerous chemicals not listed earlier into the environment, and some of them are suspected to have adverse effect on marine organisms Polycyclic aromatic hydrocarbons, one such example, are suspected to be another carcinogen of belugas in the St Lawrence estuary (Martineau et  al. 2003) The effects of persistent marine pollutants last many years after release and cover a greater area of the Inland Sea than red tides However, it is almost impossible to have a full understanding of the chronic effects of such chemicals on the health of finless porpoises or other cetacean species If a chemical is identified as harmful to laboratory animals, it is safe and scientific to assume the same effects on cetaceans and to take precautionary actions Conservation scientists are likely to be requested to provide firm evidence of malignant effects on wildlife, but such an attitude should be strictly rejected because it ends with the sole result of delaying effective action

8.4.5.5 Reclamation and Sand Extraction Finless porpoises usually inhabit waters inside the 50 m isobath and are rare in deeper waters, which are limited in the Inland Sea to a small area in the southwestern part of the sea An analysis of finless porpoise density in relation to offshore distance based on surveys in the late 1970s, when the population size was greater, revealed a sharp density decline from 26 individuals/cruise of 10 nautical miles for waters <1 nautical mile from the shore to 03 in waters >3 nautical miles from the shore (see Section 831) This indicates that finless porpoises prefer shallower or nearshore waters even within the 50 m isobath Such shallow seafloor is likely to have colonies of sea grass or algae and offer habitats for various marine animals consumed by the porpoises Subsequent surveys in 1999-2000, made about 22 years after the first one, revealed

the Inland Sea, (2) the decline was greater in the central and eastern portion of the sea and in the nearshore strata (<1 nautical mile from the shore), (3) as a result, the density gradient in the central and eastern region of the sea reversed between nearshore and intermediate (1-3 nautical miles offshore) strata, but (4) the degree of the decline in abundance was smaller in the western Inland Sea, and the earlier gradient pattern was still retained there in the 1999-2000 surveys (see Section 844)

One would expect the events of environmental destruction that occurred between the late 1970s and the late 1990s in nearshore waters of the Inland Sea and in greater magnitude in the central and eastern part of the sea than in the western part to be behind the finless porpoise decline Reclamation and sand extraction are such candidates It is efficient to take sand for construction from shallow coastal waters because of the higher quality and lower cost of the sand The activity destroys the tidal flat and coastal marine community preferred by finless porpoises, and recovery of the coastal marine community cannot be expected because of the increased water depth Such destruction further expands with time to nearby areas through sand migration to the deep hollows thus created As reclamation usually occurred in shallow waters near the coast, its effect was similar to or worse than that of sand extraction, because in some cases fill materials was supplied by sand extracted from nearby areas I suspect that such activities functioned to decrease the carrying capacity of the central and eastern Inland Sea for finless porpoises

Arita (1999, in Japanese) stated that sand from weathered granite was preferred as construction material available in the Inland Sea and that it was found in large quantities in narrow channels The top four prefectures that produced marine sand in the past 30 years were Kagawa, Hiroshima, Okayama, and Ehime in decreasing order, each of which produced 100130 million m3 of sand Yamaguchi Prefecture followed with 25 million m3 Matsuda (1999, in Japanese) stated that the extraction of marine sand started around 1960 and continued until 1999, when it was completely prohibited This sand extraction of nearly 40 years resulted in water depth increases ranging from 10-20 to 30-40 m and with destruction of habitat of the sand lance Fisheries in these prefectures recorded declines in landings of fish species living near the shore or on sandy bottom, such as sand lance, as well as landings of fishes that prey on sand lance (Matsuda 1999, in Japanese)

Shirakihara et al. (2007) identified two areas as the places where most of the sand extraction occurred These two areas agree with the top four prefectures listed by Arita (1999, in Japanese) One of them (A in Figure 810) is an island area between Hiroshima and Ehime Prefectures, and B is another island area between Okayama and Kagawa Prefectures Shirakihara et al. (2007) believed that the absence of sightings of finless porpoises in these two areas during their aerial surveys in 2000 was due to the destruction of their habitat through sand extraction When I surveyed from ferryboats connecting these islands in the late 1970s, I was impressed by a high density of finless porpoises between the islands (Figure 89; Kasuya and Kureha 1979), and the absence of the

species in the outside open waters was interpreted as a reflection of their preference for shallow habitat

During my survey activity in the late 1970s, I was able to identify sand extraction activities in the northern part of area A in Figure 810 (south of Takehara in Hiroshima Prefecture), but there were still numerous finless porpoises present After about 22 years, when I revisited the area in 1999, finless porpoises had totally disappeared (see track lines connecting points 11, 12, 13, 14, 15, 16, and 17 in Figure 89) The water depth south of Takehara was 5-18 m in 1961, but it increased to 35-48 m in 1995 (Shirakihara et al. 2007) Sand extraction for construction materials has increased the water depth of the channels between the islands and has created easy passages for vessels but seems to have destroyed habitats of the finless porpoise in the Inland Sea

Reclamation started in the Inland Sea in 1898 and created a total of 264 km2 of land by 1969 The reclamation was particularly intense after 1955 with demand for land for industrial use A conservation law in 1973 (see Section 8441) intensified restrictions, but there was additional reclamation of 974 km2 by 1995 (Setonaikai Kankyo Hozen Kyokai [Association for Conservation of Environment of Inland Sea] 1996, in Japanese) Over 350 km2 of shallow coastal habitat of the finless porpoise was lost since the late nineteenth century in the Inland Sea Natural seashores (sandy beaches and rocky shores) remaining in the Inland Sea make up only about 38% of the total coastal length, which is lower than the national average of 57%

I could not find any difference between the western Inland Sea and the rest of the sea in the proportion of natural seashore or the amount of reclamation A simple comparison of total amount of destruction would probably not have much meaning, but more important would be the proportion still remaining as a suitable habitat for finless porpoises of the habitat that originally existed in the Inland Sea The quality of the currently remaining habitat will differ between that of a group of numerous fragmented habitats and that of smaller number of larger habitat even if the total areas are the same A seaweed and sea grass bed in the Suwo Nada area (40 km2)

is the largest single remaining habitat patch in the Inland Sea This area is known for a relatively high density of finless porpoises The next largest habitat areas are found in Hiroshima Bay (23  km2) and in Iyo Nada (20  km2) where finless porpoises are rather scarce

Finless porpoises that have lived in habitat destroyed by human activities will not instantly die but will move to nearby habitats to create an apparent density increase in the new habitat Because this does not accompany an increase in carrying capacity, the increased density must return to the initial level Through this process, in the long term, the abundance of the Inland Sea population will decrease to match the new total carrying capacity

The direct influence of habitat destruction as described earlier might be limited to a particular geographical area However, we saw a drastic density decline of finless porpoise in the Naruto Pass area of the Inland Sea The only large-scale construction known in the area during the past 20-30 years was that of a large suspension bridge completed in 1985 It is 41 m above the sea surface and 876 m between piers and is unlikely to have had a significant effect on the distribution of finless porpoises (other than effects of activities during construction) The density decline in the Naruto Pass area must, at least in part, be attributed to physiological disruption due to chemical pollution and mortality in fishing gear, both of which have a final effect on the density in a broader range of the population An additional cause of the decline will be a still unconfirmed negative effect of local habitat destruction on the carrying capacity of finless porpoises in the broader area For understanding of these elements, we need better understanding of geographical movement of individual finless porpoises by season and by life cycle within the Inland Sea Currently, we do not have such information

8.4.5.6 Other Causes of the Decline Finless porpoises rarely move away from an approaching vessel but tend to dive in front of it (Kasuya and Kureha 1979)

1970s, some fishermen in Te-shima and Manabe-shima in the Shiaku Islands group (34°22′N, 133°40′E) in the Inland Sea at my request, retrieved carcasses of finless porpoises found during their fishing activities and buried them on the beach for my later retrieval during sighting survey trips Most of these carcasses had apparent wounds from ship’s screws, but it was not possible to determine if the injuries were caused before or after death

Installation of hot-bulb engines (also called “semidiesel engine”) on small Japanese fishing vessels started around 1920, and they were replaced by diesel engines in the 1960s This could have been the start of ship strikes on finless porpoises, but the early history is ignored here to concentrate on vessel traffic after the 1970s There was an increase of 26% during the period from 1972 to 1994 in the total tonnage of cargo vessels that entered ports in the Inland Sea In addition to this, there were increased operations of high-speed ferries including jetfoils and hydrofoils Thus, ship strikes of finless porpoises could have been increasing with time (Kasuya et al 2002) Underwater noise emitted by vessels, which depends on vessel size and propulsion type, also degraded the underwater acoustic environment of finless porpoises, although the effect cannot be evaluated at present

Many of the new environmental elements, such as monofilament gill nets, fast-moving ships, and vessel noise, were introduced into marine mammal habitat during the recent two or three generations of the species, and it seems premature to judge how the affected species respond and adjust their life to them (Tyack 2008)

Since the late twentieth century, there have been reports of mass mortalities of cetaceans around the world The causes include poisoning from red tide (Section 8453) and infection by influenza virus and morbillivirus (Geraci 1999) Hundreds of bottlenose dolphins were found dead on the east coast of North America and in the central North Atlantic in 1987-1988, and several thousand striped dolphins that died in the Mediterranean in 1990-1992 were believed to be victims of morbillivirus infection, but disruption of the immune system by accumulated chemical pollutants is suspected to have been a contributing cause

If mass mortalities of finless porpoises had occurred in the Inland Sea, they would have been noticed because the sea is almost a closed water system, has high fishing activity, and is surrounded by populated coasts We have no answer at present on why there have been no reports of mass mortalities of any cetaceans along the coast of Japan including the Inland Sea It is important to investigate the background of such situations and to be prepared for any future mass mortality

8.4.6.1 Industrial Use of Finless Porpoises The opinion on the suitability of finless porpoises for human consumption varies geographically Finless porpoises killed incidental to fishing activities in Korea have been sent to

Bay in western Kyushu discarded most of the finless porpoises found in their small fixed trap nets, but they also sent some of the carcasses to the fish market Mizue et al. (1965, in Japanese) purchased such carcasses for their osteological study This confirms that the species has been consumed on the west coast of Kyushu Later, in the 1970s, Prof Mizue of Nagasaki University told me that the trap-net fishery was discontinued due to change in fishery regulation and that there was no more incidental mortality of porpoises (Kasuya and Kureha 1979) This explains why Shirakihara et al. (1993) did not include samples from this source

The people of Ayukawa on the Oshika Peninsula, or on the east coast of Sendai Bay (the northern limit of this species in Japan), hold the species in quite a different light Nobody there eats finless porpoises because of a belief that it has a strong purgative effect N Kimura, the late director of the Ayukawa Whale Museum, once tried meat and blubber cooked together and experienced violent diarrhea within 10 minutes Then he repeated twice boiling the porpoise meat and poured off the oil before eating; this eliminated the purgative effect An early nineteenth century scholar in Sendai (west coast of Sendai Bay) wrote in his book Geishi-ko (On Natural History of Whales) that finless porpoises are unsuitable for human consumption because of high oil content (Otsuki 1808) Such a diarrheic effect of finless porpoise is also known from the Chinese coast (JY Yang 2007, personal information) Kasuya et al. (1985, in Japanese) noted that only the Satohama-Miyashita shell mound (see Section 21) produced remains of finless porpoises out of the 22 archaeological sites known with cetacean remains, including sites along the coasts of Tokyo Bay, Mikawa Bay, and the Inland Sea that were inhabited by finless porpoises, and hypothesized that it could be due to the unsuitable nature of the species for human consumption Out of five volunteer students who ate meat and blubber of finless porpoise off western Kyushu, only one experienced mild diarrhea (M Shirakihara personal communication, Kasuya 1999) Thus, the possible variation in effect needs to be investigated

Kaburagi (1932, in Japanese) stated that finless porpoises were hunted for oil in the Inland Sea and that the fishermen of Moriguchi (current Kamiura in 34°16′N, 133°03′E) made expeditions to Aba-shima Island (34°19′N, 132°57′E) area for finless porpoises and had several conflicts with fishermen of Tadanoumi north of Aba-shima The purpose of the hunt was to obtain oil for use as an insecticide in rice fields, but the Tadanoumi fishermen who fished in the Aba-shima area wanted to protect the porpoises Otsuki (1808, in Japanese) recorded finless porpoises as a source of whale oil, and Okura (1826, in Japanese) stated that oil from the finless porpoise, as well as other whale oils, can be used to kill locusts (most likely leafhoppers) in rice paddies It seems likely that finless porpoises were once hunted locally for oil

After World War II, there was a plan for small-type whaling for finless porpoises, which were considered by the people of the time as a nuisance in the Inland Sea, to make fish paste from the meat (The Asahi Shinbun of January 20, 1947) A  person at Tonosho in Shodo-shima ordered a

Seiki Seisakusho in Kochi, Shikoku In those days, this type of whaling was not regulated (Sections 41, 54 and 55) I do not know the results of the project, but it would seem to make more sense to operate small-type whaling for pilot whales and other dolphins in the Pacific rather than hunting such a small unprofitable species as the finless porpoise in the Inland Sea

While I was traveling the Inland Sea in the late 1970s, I  often heard from local people that shortly after World War II, they used to extract oil from finless porpoises found in their fishing nets The oil was used for light during electricity failures In the early 2000s, I was shown a small glass jar containing finless porpoise oil by a friend at Iwai-shima Island (33°47′N, 131°59′E) in the Suwo Nada region in the western Inland Sea The foul-smelling oil was extracted from a finless porpoise in early times to be used for burns, scalds, and cuts

The custom of using finless porpoises in the Inland Sea seems to have ceased in the 1960s or earlier, which is much earlier than the earlier described decline of the finless porpoise population in the sea

8.4.6.2 Listing of a Natural Monument and Fishery Regulations

Aba-shima Island in Hiroshima Prefecture is a small island of about 25 × 05 km situated off the northern coast of the central Inland Sea A sea area of 15 km radius centered on the southern tip of the island was listed in November 1930 among natural monument as a “sea where finless porpoises aggregated” (Section 593) Kaburagi (1932, in Japanese) explained the reason for the listing Tens or over a hundred finless porpoises aggregated in February to May in a small area surrounded by the islands of Kokuno-shima, Matsushima, and Kara-shima and the southern tip of Aba-shima The total size of this area is about 2 × 4 km The scenery was particularly splendid between the tips of Aba-shima and Kara-shima, a rock about 15 km to the south of Abashima About 20 fishing vessels from Tadano-umi Town used to operate a hook-and-line fishery for sea bream and sea bass in this area using finless porpoises as indicators of fish schools This kind of fishery was said to have lasted for about 200 years The role of the finless porpoises in the fishery was interpreted as follows: (1) finless porpoises feed on small fish such as sand lance that gather near the sea surface, (2) schools of the small fish descend to escape from predators (ie, finless porpoises), and (3) sea bream and sea bass attack the small fish from below, giving fishermen the opportunity to place their fishlines The fishermen erected a small shrine at the southern tip of Aba-shima and had a ritual ceremony for the finless porpoises yearly on January 26 (Kaburagi 1932, in Japanese) Kasuya and Kureha (1979) recorded later information from Tadano-umi FCU that this particular fishery was discontinued in the late 1960s due to disappearance of sand lance in the region This was shortly after the start of sand extraction (see Section 8455)

The Coastal Division of the Fisheries Agency stated that in 1989 it made a request to hand-harpoon and drive fishermen for dolphins and porpoises to refrain from hunting finless

Division, dated March 1, 1991) No cetacean fisheries of the time were located near a place that would allow hunting of finless porpoises, and no new fishing projects were planned for the species, so the real reason for the Fisheries Agency request is unclear to me It could have functioned to prevent local aquariums from hunting the species for display Then, in September 1990, the Director of the Coastal Division sent out notification No 2-1050 requesting that small cetaceans found in fishing gears such as fixed trap nets be (1) released if found alive or (2) buried if found dead and that (3) the incident be reported to the Coastal Division (Section 63)

The Act for Conservation of Endangered Species of Wild Animals and Plants was enacted in 1992, and its application rule of March 29, 1993, regulated the capture and international and domestic transportation of the finless porpoise (Section  592) Ministerial Order No 15 of April 1, 1993, added the finless porpoise, together with bowhead whale, blue whale, dugong, Pacific Ridley turtle, and leatherback turtle, to a list of species protected by the Fishery Resources Protection Act (Section 591) This prohibited hunting of the species and made it an obligation to report incidental mortality, which did not differ practically from the earlier regulations

All the Japanese protection measures for the finless porpoise prohibited only intentional capture of the species but did not have any provisions to decrease mortality incidental to fishing operations Such measures were preferred by the government because they required doing nothing They amounted to only a piece of paper, with no budget or personnel to implement them It is my view that such an action is the most passive conservation measure and does not deserve to be called protection The most urgent actions needed for finless porpoises in the Inland Sea, and perhaps for other populations in Japan, are to conserve their habitats and decrease mortality in fisheries

8.4.6.3 The Problem Is Not Limited to the Inland Sea It is believed that there are a minimum of five, and probably more, local populations of finless porpoises in Japanese waters The areas identified with the populations include (1)  Omura Bay, (2) Ariake Sound-Tachibana Bay, (3) the Inland Sea and nearby waters connected to it, and (4) Ise BayMikawa Bay and waters near the entrance of Ise Bay In addition to these, Tokyo Bay, the Pacific coast of Chiba Prefecture to Kashima Nada (34°54′N to 36°40′N), and the Pacific coast of Fukushima Prefecture to Sendai Bay (36°50′N to 38°24′N) are likely to support populations The Tokyo Bay-Sendai Bay area has not been well studied because of a paucity of available materials for DNA analysis The possibility of a small population in Suruga Bay still remains to be confirmed When there is the possibility of a local population, it is safer to establish conservation actions for it

Some of the local populations identified or suspected inhabit waters supporting high human activities, for example, Omura Bay, the Ariake Sound, Ise Bay, Mikawa Bay, and Tokyo Bay At least some of the hazardous human activities discussed earlier take place in all of these waters, including

traffic It is reasonable to assume that some of the remaining local populations are also experiencing decline in abundance so far unnoticed by scientists There are still some sporadic sightings of finless porpoises in Tokyo Bay, but it is unreasonable to consider that the abundance was at such a low level in the nineteenth century, when the coast was close to pristine, the waters were less polluted, and the vessel traffic was less Rather, it should be concluded that the population is one of the most endangered populations of finless porpoises in Japan The Inland Sea population of finless porpoise just happened to be surveyed by scientists and a decline confirmed there

I believe it is urgent to improve our understanding of the current status of Japanese finless porpoises and to establish a conservation plan for them Conserving the coastal marine environment suitable for finless porpoises will also contribute to the health of fisheries as well as the safety of fishery products for humans

Kasuya and Kureha (1979) concluded that parturitions occur in the Inland Sea from April to August with a peak in AprilMay This was based in part on 13 records of juveniles of 100 cm or less during late April to late June in a broad Pacific area from the Inland Sea in the southwest to Sendai Bay in the northeast, including Ise Bay, Mikawa Bay, Shimizu in Suruga Bay, and Tokyo Bay The additional evidence was

August/September during sighting surveys in the Inland Sea They felt it would also be of some significance that the proportion of single adults declined during the period; however, this was not independent from the increase in cow-calf pairs

Iwatsuki (2000, in Japanese) supplemented Kasuya and Kureha (1979) with additional samples and analyzed the seasonality of newborn finless porpoises for a broader geographical area She confirmed that there was no evidence to suggest different parturition seasons among finless porpoise populations inhabiting the Pacific coast of Japan and the Inland Sea She defined a newborn as a porpoise larger than the minimum known length of a calf (62 cm) and below the maximum known length of a fetus (913  cm) This seems to be a little strict and might have functioned to reduce the estimated range of the parturition season According to her analysis, all the newborns (11 individuals) occurred only in 3 months from April to July (a peak was identified in May and June) in the Inland Sea, and the 11 newborns were 37% of the total of 27 strandings recorded for the 3 months In Ise Bay-Mikawa Bay and the nearby Pacific area, all the 56 newborns were recorded in 5 months from April to August (a peak in May); they were 38% of the total of 145 stranded porpoises during the period I have tabulated in Table 812 all the records of strandings and incidental deaths accompanied by date and body length, where newborns were defined as 60-95 cm in body length (see Section 852)

The estimated parturition peak in April-June for the Inland Sea population shows reasonable agreement with a peak in copulations (April-May) observed in the Toba Aquarium in

TABLE 8.12 Stranding and Incidental Mortality of Japanese Finless Porpoises together with Records of Porpoises of Neonate Size (60-95 cm in Body Length) as an Indication of Parturition Season, Based on Records from November 16, 1929, to May 31, 1999

11 months is assumed) However, this parturition peak was inconsistent with a trend in cow-calf pairs, which increased in frequency until August/September in the Inland Sea In order to explain this apparent disagreement, Kasuya and Kureha (1979) assumed that finless porpoises not accompanied by calves tend to move outside the Inland Sea However, in view of the data collected by Iwatsuki (2000, in Japanese), tabulated in Table 812, such an assumption seems unnecessary We have a total of 68 records of newborns from the Inland Sea plus the Kanmon area connected to the sea and from the Ise Bay-Mikawa Bay area The monthly distribution of the newborns was 10 individuals (147%) in April, 31 (456%) in May, 21 (309%) in June, 2 (29%) in July, and 4 (59%) in August This indicates that the parturition season starts in April, peaks in May, and lasts until August Sixty percent of total births in a year occur in April and May and the remaining 40% in the 3 months from June to August, which explains the seasonal increase of cow/calf pairs toward August/September observed by Kasuya and Kureha (1979) in the Inland Sea Now, it is unnecessary to assume a different pattern of seasonal movement between growth and reproductive phases

Furuta et  al. (1989) reported 21 records of large fetuses and neonates of 515-90 cm from Ise Bay collected during 19661983 that were also included in Iwatsuki (2000, in Japanese) Seventeen of them were over 75 cm long (including two fetuses at 750 and 790 cm); 2 occurred in March, 11 in April, 3 in May, and 1 in June This was similar to the seasonality pattern in the Inland Sea A geographical difference of parturition

data set in Table 812 All of the neonates reported in a broad area ranging from the Inland Sea area to Sendai Bay along the Pacific coast of Japan occurred during March to August, with a peak in May in two populations with sufficient sample size, a population in the Inland Sea and its adjacent area and another in Ise Bay-Mikawa Bay and the adjacent area Thus, with an exception of one population off western Kyushu, there is currently no evidence to indicate geographical variation in the parturition season of finless porpoises in Japan

Mizue et al. (1965) thought that finless porpoises off western Kyushu had a parturition season in late August to early September, which was based on their own observation of one lactating female in late October and the information from fishermen that “newborn caves with umbilical fragments occurred in early September and later, which was preceded by occurrence of large fetuses” The specimen was presumably killed in the Ariake Sound-Tachibana Bay area, but the presence of two populations, one in Omura Bay and another in Ariake Sound-Tachibana Bay, was not known in those days I once erroneously rejected their conclusion in the analysis of breeding season of the species in the Inland Sea (Kasuya and Kureha 1979), which was due to my careless and inadequate reasoning that all the Japanese finless porpoise populations should have a similar breeding season I wish to apologize for my impudence and mistake

Shirakihara et  al. (1993) analyzed parturition season by two methods (Figure 811), which were (1) to assume neonates 84 cm or shorter as age zero and (2) to estimate parturition

growth rate of 0277 cm/day and average neonatal length of 786  cm (see Section 852) They applied these methods to two sample groups: western Kyushu (samples from two currently identified populations were combined) and the Inland Sea area plus other Pacific coasts Results of the two estimation methods agreed reasonably for the two sample groups, and the results for the Inland Sea plus Pacific coasts samples agreed with the conclusion given earlier However, the parturition season obtained from the western Kyushu sample apparently differed; parturition there extended from August to April with a main peak in November and December, or about 6 months apart from the rest of the populations, and agreed with the conclusions of Mizue et al (1965, in Japanese) The analysis of Shirakihara et al. (1993) showed a minor peak in March for the western Kyushu sample (Figure 811)

The analyses of Shirakihara et  al. (1993) suffer by having combined two populations later identified by Yoshida et al. (1995, 2001), those in Omura Bay and those in Ariake Sound-Tachibana Bay Their sample from the Ariake SoundTachibana Bay population was later reanalyzed by the same authors, and they concluded that parturition lasted from August to March with a major peak in November-December and an additional minor peak in March (Shirakihara et  al. 2008) Amano (2010, in Japanese) analyzed stranding records in 1901-2008 from the Omura Bay population and found that neonates below 90 cm occurred only in April and May; he concluded that the Omura Bay population had a parturition season that was similar to those in the Inland Sea and to the east Now, it is known that only one of the two finless porpoise populations in western Kyushu, the Ariake Sound-Tachibana Bay population, has a main parturition peak in autumn/winter and a minor peak in March, only the latter of which matches with the single peak identified for rest of the finless populations in Japan

Shirakihara et  al. (1993) reported a brief observation on seasonal difference in male reproductive system: all 11 adult males sampled in September-January had copious sperm in the epididymides, but 3 adults in June-July had less All the 14 adults had spermatozoa in the testes (no mention was made of abundance of spermatozoa in the seminiferous tubules) It is a reasonable strategy for males to start production of spermatozoa before the mating season and store a sufficient amount in the epididymides to allow the chance of finding estrous females during any part of the mating season The observation of Shirakihara et al. (1993) agrees with a mating peak in Autumn known for finless porpoises in the Ariake Sound-Tachibana Bay

Selection is believed to have worked to place parturitions in a season that would maximize the survival of offspring (Kiltie 1984) Survival of offspring will be affected by various factors, including climate, predators, nutrition of nursing females, and availability of suitable food for the calf during the period of switching from milk to solid food The two populations off the west coast of Kyushu described earlier inhabit almost similar latitudes, and the climate and oceanography do not differ much The only difference would be a smaller

rition season with the majority of Japanese finless porpoises It is also strange that individuals in Sendai Bay, the northernmost habitat of the species in Japan, breed in the same season as most of the Japanese populations further south

8.5.2.1 Body Length at Birth The mean body length of neonates was reported as between 786 and 816 cm (Table 813) The difference between these figures, in addition to being due to the limited sample size, is due at least partly to different definitions of neonate Geographical variation in neonatal length has not been identified in Japan

The estimation of mean neonatal length by Furuta et  al. (1989) was important because it used data from a single population (Ise Bay-Mikawa Bay population) and was based on neonates clearly identified by the presence of a remaining umbilical fragment that was usually 1-15 cm long The body length of neonates thus defined ranged from 765 to 850 cm (Table 813), and their 6 fetal samples ranged from 515 to 790 cm, which led Furuta et al. (1989) to conclude that finless porpoises in the Ise Bay-Mikawa Bay region were born at 75-80  cm Shirakihata et  al. (1993) combined data from two populations in western Kyushu (Omura Bay and Ariake Sound-Tachibana Bay populations) (Table 813) They defined neonates by the presence of at least one of the following characters: (1) vestigial sensory hair on upper jaw, (2) fetal folds, and (3) umbilical fragment Duration of these characters has not been investigated, but Furuta et  al. (1977, in Japanese) stated that the fetal folds last a minimum of 17 days after birth

The ratio of neonates to fetuses by body-length group increases with body length, and a length where they became equal in proportion has often been taken to be the average body length at birth (eg, Sections 943 and 1052)

TABLE 8.13 Records of Neonates of Japanese Finless Porpoises and Mean Neonatal Lengths Calculated from Them

of newborns, it is not applicable to the finless porpoise due to limitations of sample size Data used in the previous analyses were obtained mostly from stranded carcasses The incidence of perinatal mortality is not independent of birth size; for example, smaller newborns might be more liable to die Therefore, the mean body length at birth as estimated earlier can be biased, but we have no data to confirm this

8.5.2.2 Fetal Growth and Gestation Time If body length or the cube root of body weight is plotted on an axis of time measured after conception, mammalian fetal growth can be divided into two phases of a slower curvilinear growth phase in early development and a subsequent linear growth phase The date t0, which is the date when the back-extended linear growth cuts the axis of time, is a function of total gestation time (X), that is, t0 = 03X for species with X = 50-100 days, t0 = 02X for species with X = 100-400 days, and t0 = 01X for species with X > 400  days (Hugget and Widdas 1951) This general relationship is applicable to cetaceans (Lockyer 1984; Perrin and Reilly 1984), with the minor modifications of Laws (1959), who proposed 90% of t0 estimated using the relationship given earlier for cetaceans and using body length instead of the cube root of fetal weight As  the gestation period of finless porpoises is likely to be slightly shorter than 1 year, t0 is estimated by t = 09 × 015X Seasonal fetal size distribution gives an estimate of growth rate during the linear phase, which together with average body length at birth enables the calculation of the value of (X − t0); then the total gestation time (X) is calculated by dividing the value of (X − t0) by (1 − t0) However, we do not have data to use for direct estimation of fetal growth rate and a value of (X − t0) for the finless porpoise

An alternative way is to estimate gestation time and fetal growth rate using interspecies relationships of growth parameters The following relationship between gestation period (Y, month) and average body length at birth (X, cm) is known for the Delphinidae (Perrin et al. 1977)

log Y = 04586 log X + 01659

Using this equation and assuming average body length at birth of 80 ± 5 cm, Kasuya et al. (1986) estimated the gestation period of the Japanese finless porpoise at 109 months, with a range of 106-112 months

Another interspecies relationship that has been used for this purpose is that between fetal growth rate during the linear fetal growth phase (Y, cm/day) and average body length at birth (X, cm) for Delphinidae (Kasuya 1977)

Y = 0001462X + 01622

Assuming 80 cm as the average body length at birth for this equation, Kasuya et  al. (1986) obtained 0279  cm/day or 849 cm/month as an average growth rate of fetuses in the linear phase of growth This is used to calculate mean gestation time:

(80/849)/(1 – 015 × 09) = 109 (month)

very reliable because the average body length at birth is only an estimate for the finless porpoise, but fair agreement of the results by both methods, about 11 months, suggests that they are reasonable

Izawa and Kataoka (1965, in Japanese) reported the observations of a pair of Ise Bay finless porpoises kept in the Toba Aquarium during September 1963 to March 1965; annual copulation activity started in February and lasted to June Other observations of 13 individuals of both sexes in a tank at the same aquarium recorded a copulation peak in April and May, and one of the females that had been kept there since 1973 delivered a calf on April 17, 1976 (Furuta et al. 1977, in Japanese) These observations support the analysis presented earlier that Japanese finless porpoises (with exception of the Ariake Sound-Tachibana Bay population in western Kyushu) mate in spring and early summer, with gestation of about 1 year

Small-cetacean species exhibit common features of gestation time close to one year and neonatal size of slightly less than 1 m The neonatal length is probably close to a minimum body size required for a homoeothermic species to live safely in the aquatic environment, and the gestation length is an adaptation to the annual cycle of the marine environment Some cetacean species have achieved greater body size through either of two different ways Baleen whales have achieved it by increasing the fetal growth rate while retaining gestation time of about one year, which has been required to meet an annual cycle of feeding Some of the toothed whales such as the sperm whale and killer whale have increased size while retaining a relatively slower fetal growth rate and accepting a gestation period of over 1 year, with a possible cost of a reproductive cycle that does not synchronize with an annual cycle of the marine environment (Kasuya 1995)

8.5.2.3 Nursing Period and Weaning Season Females of some cetacean species such as the Dall’s porpoise (Section 946) and the harbor porpoise usually enter into estrus and conceive while nursing a calf, but the reproductive cycle of many other cetaceans usually follows the order of separate pregnancy, lactation, and resting The last two stages have greater variation in duration If the suckling calf dies, the female usually ceases lactation (adoption of another calf is known to occur in some cases) and enters into estrus in the next mating season for subsequent gestation that may starts earlier than in the normal case Some healthy well-fed females may conceive during lactation (thus the resting phase is skipped), but such lactation probably ceases before the next parturition, as in short-finned pilot whales (Section 1244) If a calf is allowed to continue suckling for a longer period, the next conception by the mother may be delayed This can happen to females of great age or to females kept together with their calves in an aquarium

Furuta et  al. (2007, in Japanese) reported the process of weaning for four calves born in an aquarium that lived to weaning They started taking solid food at the ages of 120, 132, 166, and 223 days, with an average of 160 days (53 months) Their daily food consumption increased for 5-6  months

started solid food on the 120th day (39 months), was observed suckling frequency during sampled hours The number of suckling bouts per hour of observation was 25-30 just after birth, decreased to 10-15 at the age of one month, and further continued to decrease until the 252th day (83 months) when observation was discontinued After an interruption of 104 days, or at the age of 357 days (117 months), this individual was observed for 8 hours but showed no sign of suckling This observation leads to a conclusion that finless porpoises from Ise Bay start taking solid food at the age of 4-7 months, followed by decreasing proportion of milk in the total nutrition with time, and end suckling, in an aquarium condition, at an age between 83 and 117 months The aquarium environment would not influence the start of weaning (53 months on average) However, the time at completion of weaning will require information on individuals in the wild

Shirakihara et al. (2008) examined the stomach contents of 17 juveniles from the Ariake Sound-Tachibana Bay population The body length of animals with only milk was below 995  cm, that of those with both solid food and milk was 995-107 cm (three individuals), and that of those with only solid food was 935-104 cm These data indicate that the porpoises start ingestion of solid food in the wild at body length 935-995 cm and that some individuals continue suckling until a body length of 107 cm is reached, although individuals may end suckling earlier The judgment of the absence of milk in the stomach requires caution because identification of a small amount of milk mixed with solid food is often difficult Shirakihara et al. (2008) estimated ages at the stages mentioned earlier, using the body length and growth equations of Shirakihara et al. (1993, see Section 8524), which suggested that the porpoises start taking solid food at ages between 3-4 and 5-6 months and that some individuals continue suckling up to an age of 9-10 months (from the oldest individual with a trace of milk) The case of extended suckling is close to the observations of Furuta et al. (2007) of an animal in captivity

Kasuya and Kureha (1979) reported seasonal change in the proportion of cow-calf pairs in the Inland Sea The proportion of such groups started to increase in spring and reached at a peak of about 30% in late summer and then suddenly decreased to about 10% accompanied by an increase in frequency of single individuals of calf size (Figure 812) Although the accuracy of the size estimation was not verified, they also recorded an increase in frequency of juveniles larger than the calves in winter They inferred from this observation that calves are born in spring and early summer and most of them become independent around October-November, but some accompany their mothers until the next spring or summer

These three observations on early growth of Japanese finless porpoises do not contradict each other, and it seems to be reasonable to conclude that most females breed synchronously and wean their calves before the mating season of the next spring This suggests, although inconclusively, that their breeding cycle is likely to be 1 or 2 years long

8.5.2.4 Growth Curve Furuta et al. (2007) assumed linear growth and got the following regression of body length (Y, cm) on age (X, days) for juvenile finless porpoises born in an aquarium:

Y 98 2 6 976X X 6 r 9 26= + £ £ =. . , , .0 0 0 0 00 0 0

This regression was based on 19 points representing an unstated number of calves If we accept this equation, it suggests a daily increment of 00976 cm and an average neonatal length of 98206 cm, which is unreasonably greater than the 80 cm estimated earlier The average body length at the age of 1 year of 1338 cm calculated from the equation presented earlier is only 36% greater than the neonatal length derived from the equation, which is unreasonably small Thus, it is necessary to investigate whether linear growth can be reasonably fitted to the growth during the period Six of the 11 points at the age of 0-150 days are below the line and 3 are above, all the 6 points at the age of 150-300 days are above the line, and the 2 points over 450 days are below the line This is evidence that linear growth is inappropriate for expressing the early growth of finless porpoises; a fit to an upwardly convex growth curve is more likely (Figure 813)

Kasuya and Matsui (1984) noted for five species of Delphinidae that body length increments in the first year

after birth were in the range of 55%–64% of neonatal length: 55% for the bottlenose dolphin, 60% for the spotted dolphin, 612% for the short-finned pilot whale, 613% for the common dolphin, and 64% for the striped dolphin Assuming neonatal length of 80 cm and applying this rule to the finless porpoise gives 124-131 cm as the body length at the age of 1 year

Perrin et al. (1976) obtained the following relationship for five species of toothed whales among average fetal growth rate (X, cm/month), average growth rate of neonate during the period equal to gestation time (Y, cm/month), and average neonatal body length (Z, cm)

Log(X − Y) = −133 + 0997 log Z

for finless porpoise, or 119 cm as the body length at the age of 11 months Extrapolation of this rate to 12 months gives 123 cm, which is logically expected to be an overestimate

Kasuya et al. (1986) made a rudimentary attempt to calibrate the accumulation rate of growth layers in the teeth of finless porpoises using six animals of known age: four neonates (age 0-28 days), a male of 1535 cm that died at the age of 864 days (about 2 years 4 months), and another male of 1595 cm that died at the age of 1719  days (about 4  years 8  months) The last two individuals were not accompanied by tooth samples for aging Of the four neonates of known age, the longer-lived individuals always had greater body lengths at death One possible explanation for this would be growth during their short life Another was that smaller neonates died earlier A daily growth rate of 02 cm/day was calculated by ignoring the latter possibility They aged another 8 individuals caught in fishing nets They were 100-132 cm in body length and aged at 0515 years from dentinal growth layers, which almost agreed with the body length at the age of 1 year calculated from the interspecies relationship given earlier This scanty information is the currently available basis of the estimated annual deposition of dental growth layers for the species

Later growth of the species was reported by Shirakihara et  al. (1993) and Kasuya (1999) using the ages determined by counting growth layers in the teeth and assuming annual deposition of the layers Growth layers are deposited in both dentine and cementum of young individuals, but deposition of dentine ceases after the age of about 10 layers in this species and aging must rely on reading of cementum layers, which is harder than reading the dentinal layers Shirakihara et al. (1993) combined specimens mostly from two populations off western Kyushu and a few specimens from the Inland Sea population They found no growth difference between the combined western Kyushu populations and the Inland Sea population, but they did not test for growth difference between the two populations in western Kyushu (those in Omura Bay and Ariake Sound-Tachibana Bay) The oldest individual in their series was aged 23 years for both sexes Body lengths of sexually mature male were in the 127-1745 cm range and those of females in the 135-161 cm range; the modal length was 150-159  cm for both sexes Male body lengths were slightly greater than those of females The early fast growth phase seemed to end at around 4-5 years, followed by slower growth that ended at around 10 years in females and at age slightly over that in males

Shirakihara et al. (1993) tried several growth equations for the finless porpoise and concluded that the following equations fit best to their data on body length (Y, cm) and age (X, years):

Male: Y = 1655 – 1572/(X + 1863)

Female: Y = 1577 – 1181/(X + 1624)

Kasuya (1999) listed maximum body size in Inland Sea/Pacific coast populations as 192 cm (males) and 180 cm  (females)

(males) and 158  cm (females) known from western Kyushu (two populations combined), but more data are needed before reaching a firm conclusion

The mean growth curve obtained earlier should be considered as an “apparent” mean growth curve Imagine a particular case where all human babies born in a town in a particular year are selected as a sample and their height is measured annually till the age of about 30 years when all the individuals of the sample are sure to have ceased growth If annual means of height were plotted on age, that would yield a “real” mean growth curve for the sampled population The same procedure can be also done using body weight as an indicator of body size, but the shape of the curve as well as age at cessation of growth will be different from that of body height

Due to technical reasons, such a real mean growth curve is not available for many cetacean species, including the finless porpoise, and instead we plot body length on age using data obtained in a limited length of time to construct a mean growth curve, which should be called as an “apparent” growth curve and is identical with the “real” mean growth curve only when the growth pattern of the particular population remains the same for a period of about 30 years or more, depending on the species If such an assumption is violated, the apparent mean growth curve is different from the real one For example, if sons and daughters grow larger than their parents, perhaps through the improvement of nutrition, the apparent mean growth curve will yield the erroneous impression that the mean body length diminishes at higher age, as observed for North Pacific sperm whales (Kasuya 1991) The opposite will occur if body size in a more recent generation diminishes

It should also be noted that the real mean growth curve may not be the same as the growth curve of any single individual In the human example presented earlier, some individuals will have an adolescent growth spurt at the age of 12-13 years and exceed the average size in the sample at that age, which is followed by a stage of slower growth Other individuals may have the growth spurt at a later age The presence of an adolescent growth spurt will be obscured on the mean growth curve by broad individual variation in its occurrence at age The ages at which an individual ceases growth is also variable among individuals, but the mean body length continues to increase until all individuals cease growing Thus, unless we have supplemental anatomical information such as state of fusion of the vertebral epiphyses (see Section 1345), analysis of a mean growth curve alone is likely to overestimate average age at cessation of growth Thus, interpretation of the “apparent” mean growth curve requires caution

I agree with the value of a growth equation as a way of expressing the mean growth curve, the biological value of which was evaluated earlier But overapplication of the equation is risky Each growth equation has its own peculiarities Even when it is judged by a scientist as acceptable to express the mean growth pattern of the whole age range of a cetacean species, the real fit may not be the same for a particular age segment of the sample Fitted growth equations also depend

to samples biased to younger individuals may not adequately show the growth of older individuals, and an equation fitted to samples of higher age may not properly represent the mean growth of younger individuals Thus, extrapolation of a growth curve to higher ages, where sample size is likely to be limited, for the purpose of estimating asymptotic body length should be done with caution

8.5.2.5 Body Weight Several equations have been proposed for the relationship between body weight (W, kg) and body length (L, cm) of finless porpoises Kasuya (1999) found no significant difference in the relationship among 6 individuals from Pakistan, 15 from China, and 21 from Japan and obtained for the total of 42 specimens a single equation:

W 1 816 1 L 65 3 L 187 4 2 477= ´ < <. ..0-

The equation can be expressed as log W = 2477 log L − 37409 This equation predicts an average body weight of 376 kg for animals of 140  cm and 523  kg for animals at 160  cm, or 141% difference, but Kasuya (1999) noted that 10 individuals at 140-160  cm had body weight variation of 263-482  kg, or 183% difference This large individual variation in body weight is possibly due to seasonal fluctuation in physiology

Kataoka et al. (1976, in Japanese) recorded large seasonal variation in the food consumption of finless porpoises (of both sexes) in their aquarium, from a minimum level of 2-3 kg/ day/individual in May-September to a high level of 5 kg in December-March Food consumption was negatively correlated with environmental water temperature These individuals had been in the aquarium for 3 years and shown active mating behavior (Izawa and Kataoka 1965, in Japanese), which suggests that they were sexually mature Kataoka et al. (1976, in Japanese) did not give precise biological data for these porpoises but estimated that they were around 160 cm in body length and about 56 kg in body weight

Furuta et al. (1989) analyzed the body weight of 29 finless porpoises from Ise Bay and nearby waters and found all to lie on a single regression line aside from one fetus of 515 cm They obtained the following equation for the 28 porpoises of both sexes:

log . log . .W L L= < <2 8405 4 5439 54 5 192-

This equation can also be expressed as W = 2858 × 10-5L28405 and gives a body weight of 521 kg at body length 160 cm, which is almost identical with the figure calculated from the equation of Kasuya (1999)

Shirakihara et  al. (1993) calculated the following two equations for finless porpoise samples from two populations in western Kyushu, excluding pregnant females:

Male: W 2 2 L L 174.5

Female: W 2

log . log . .