Prevalences of human taeniasis since 1915 in Korea are summarized in Table 1. There are other records not summarized in this table because of insufficient detail of description. In a report of Inoba (1924), prevalences of Taenia tapeworms among Koreans were quoted as 16% from author Handa, 7-8% from Mine, without description of locality or date. Probably they were small sized surveys done before 1924. Somewhat detailed prevalence data of Taenia tapeworms among Koreans were accumulated until 1929, so it was not difficult to understand about the tapeworm status of those days. In 1930s, research almost ceased, except focal surveys started again from 1945. It was, however, not until the 1960s that more substantial surveys for parasites began. Most of the data collected in the 1960s were from local surveys, although those by Soh et al. (1961) and Lee et al. (1966)' were relatively large scale. In 1969, Seo et al. conducted the first nationwide parasitological survey, including most of the provinces, and thus it could be regarded as one of the most important reports after the Second World War. It was revealed that 0.7% of Koreans were positive for taeniid eggs. Carefully prepared is this survey, dealing with total 40,581 individuals including adequate sample sizes for male/female comparisons, as well as with military soldiers. The only limitation of this survey is that there was a serious bias for the 5-29 year-old age group, comprising 96% of all subjects, making it difficult to call it a real nationwide survey representing all age groups of Koreans (Table 1). Thereafter nationwide surveys were conducted every five years, revealing taeniid egg positives of between 0.02 and 1.9%, with fluctuation varying greatly, causing difficulties in figuring out trends (Table 1).

Morphological similarity between taeniid eggs makes it difficult to determine the species ratio between T. solium and T. saginata. Gravid proglottids spread eggs irregularly through the proctostoma (Mazzotti, 1947; Rijpstra et al., 1961), and distribute them unevenly into stools, causing a lower detection rate than real cases (Lee, 1970; Rim, 1980). Possible ways to make differential diagnosis are observing the number of main uterine branches, and the motility of gravid proglottids, as well as observing the hooklets on the scolex. Kang et al. (1965a) reported that T. saginata is more common than T. solium, with as much as 86.9% out of 327 adult tapeworms collected from Cheju inhabitants after treatment with bithionol (Table 2). Lee et al. (1966) observed T. saginata comprised 73.1% of adults among 26 taeniasis patients in Chollabuk-do; Cho et al. (1967), 86.7% of 105; Rim et al. (1980), 55.8% of 199 in Seoul. Chungchongbuk-do, Kyongsangnam-do and Cheju-do; Joo et al. (1985), 73.8% of 84 in Seoul and Cheju-do; and Hong et al. (1983), 100% of 26 of Sorokdo (Island, a leprosarium), Chollanam-do. Summarizing the data during the past 25 years makes it clear that T. saginata is dominant by an average of 79.4% (53.3-100.0%) over T. solium (15.4%) among Korean people, which makes the approximate ratio of 1 : 5 for T. solium and T. saginata (Table 2).

Most of the reports on the metacestode surveys in intermediate hosts for Taenia tapeworms are on the Cysticercus cellulosae in pigs in Korea (Inoba, 1924; Kim, 1984a, 1984b; Kim, 1985; Jang and Kang, 1989). On the other hand, only limited reports are available for C. bovis, the larval tapeworm of T. saginata, from cattle. Inoba (1924) reported prevalences of 17.8-21.3% among 731 cattle in Seoul and Kongju. Nakanishi (1926) reported 37.6% of C. bovis positive cases in 408 Korean native calves from all over the country. Additionally, prevalences of 37.1% (56/151) were reported in Kyongju, 61.1% (22/36) in Masan, 14.6% (3/22) in Kimhae, 40.1% (8/20) in Ulsan, 40.0% (4/10) in Milyang, 0% (0/2) in Chongju, 42.9% (3/7) in Sariwon, 28.6% (2/7) in Chongdo, 100.00% (3/3) in Taegu, 40.0% (2/5) in Namhae 50.0% (2/4) in Tongnae, 0% (0/1) in Kojedo, and 34.8% (48/138) of C. bovis from areas of southern Korea. Most of the calves (62.1%), in the report, revealed light infections with under 10 metacestodes per calf; however, more than 100 were also found in some calves (2.6%), with a maximum infection of 10,462 in one calf. Infective organs were the heart, striated muscle, tongue, masseter muscle, diaphragm, lung, sub-cutaneous tissue, omentum, serosa, scrotum and stomach wall as well as a few cases from lymph nodes, esophagus, liver, kidney, pancreas, bladder, thymus, intestinal wall, testicles, epididymis, spleen, breast, uterus, ovary, spinal cord and eyes. Another paper by Isshiki (1960) reported retrospectively 30.9% of metacestodes from cattle back to 1936, and 5.1% back to 1943 in Cheju-do. Most recently, there was at least one local episode of measled cattle in Kongkun-myon, Hoengsong-Gun, Kangwon-do in November, 1975. The inhabitant who found the measled beef could not eat it, and should have discarded it (personal communication with Dr. Seung-Yull Cho in 1990).

All these data reveal that larval T. saginata tapeworms in cattle was common in Korea.

Since the 1960s, substantial epidemiological studies on Taenia tapeworm infections were performed in Cheju-do, Chollabuk-do or other provinces; and for more than 30 years thereafter, most of the researchers repeatedly felt that the textbooks do not coincide well with the tapeworm life cycles in Korea. Most of the epidemiological studies on consumption of foods showed continual inconsistency with the species ratio of Taenia tapeworms. Especially on the occasions of special ceremonies like weddings, funerals and meetings, the residents in rural villages used to slaughter the pigs by themselves; which is known as 'Churyum'. Eating the raw liver is the most favorite (96.7%) habit, and other organs, like the apex of the spleen, lungs, kidneys, pancreas and tongue are eaten preferably raw. In those villages, however, eating the raw beef is rare (Park and Chyu, 1963). Cho et al. (1967) suggested that T. saginata infected people more frequently even though they ate the raw pork (25.6%) or the raw liver of pigs (37.2%) more frequently than eating the raw beef (15.7%) or raw liver of cattle (11.4%) among 2,104 subjects in Cheju. Kim (1980) reported that most of the rural people with taeniasis in Cheju-do ate raw liver of pigs and cattle. Kim (1982), in a survey on the habit of eating raw foods and parasite infections, found the majority of people (32.6%) ate the raw liver of pigs.

Summing up the ratio of pork-eaters versus beef-eaters, it was approximately 1:1 (7.1% : 6.1%) out of 12,477 individuals subjected to questionnaire surveys. Additionally, those who consume pig viscera may reach as high as 25.0% of the population in Korea. Therefore the data can be analysed as pig-eaters versus cattle-eaters, in other words, pork plus viscera eaters of pigs versus beef-eaters, (7.1 + 25.0) : 6.1 = 32.1 : 6.1 (approximately 5:1), as seen in Table 3.

A more specific breakdown of Korean eating habits for the visceral organ of pigs, among 7,357 respondents, were: liver 24.0%, intestine 5.4%, lung 0.8% and spleen 0.7%. Of these, average 10.2% (10.1-11.4) were also liver-eaters of cattle (Kang et al., 1965a; Cho et al., 1967).

Between 1992 and 1995, a survey on the eating of the raw liver of pigs and cattle, covering eight provinces in South Korea, revealed that an average of 18.7% (1,349/7,211) of Koreans had experiences of eating raw liver of pigs; and 19.8% (1,425/7,211) for eating cattle liver. The report also determined that Cheju inhabitants (27.2%) eat more raw pig liver than those of any other province (Fig. 1). Some of the pig's livers were obtained from neighbors, pubs or meat shops, but most (80%) were obtained by slaughtering pigs on their own. Twenty-two percent of subjects had experienced eating raw pig liver more than 11 times. Most of the cattle's livers, in contrast, were purchased from meat shops (42%) as well as from pubs, neighbors or department stores. The percentage of people who had experienced eating raw liver of cattle more than 11 times was 25% (Eom, 1995).

These collective data give us a clear understanding on the inconsistency between food consumption patterns and the prevalence of T. saginata. The still unchanged habits for eating the raw liver of pigs and cattle in the 1990s Korea is noteworthy.

Obviously the most distinct puzzle, recognized by Hong et al, (1983), was in Sorokdo Island, at a leprosarium, where there were no cattle, although many T. saginata-like tapeworm occured among inhabitants; and only pigs were reared. Considering that the small island is isolated from the peninsula, it is difficult to understand what led the authors to speculate that the worms might be transfered from other provinces, or from land.

Asian Taenia was first identified in Taiwan aborigines, soon after it has been also found to be distributed in many Asian countries including Korea, Indonesia, the Philippines and Thailand (Fan et al., 1988, 1989a, 1989b; Zarlenga et al., 1991; Bowles and MacManus, 1994; Fan and Chung, 1998).

Historically, Huang et al. (1966) first refused to refer to the Taiwan Taenia as T. saginata because it didn't seem to come to human through beef, as most textbooks indicate. Consequently, Chao et al. (1979) observed that the eggs of Taiwan Taenia could infect the liver of pigs, but not the muscle of cattle. Consequently Chung et al. (1987) found that wild boars and Lanyu pigs were intermediate hosts in Taiwan. Eom and Rim (1992a, 1992b) in Korea also observed that this special tapeworm caused human taeniasis through the eating of the raw pig liver. The natural infection rate was revealed to be 1.01% in Korean domestic pigs (Eom and Rim, 1992a). Further more, Eom et al. (1992) observed that the Asian Taenia metacestodes caused viscero-tropic cysticercosis experimentally in new internal organs of pigs, such as the omentum, lung and serosa, as well as in an organ already known to be infected, the liver. Eom and Rim (1993) proposed a new species for this worm. Taenia asiatica, with the following taxonomic evidence: T. asiatica has a rostellum without hooklets, and this feature discriminates it from the classical T. saginata, along with some other morphological characteristics, such as the large number of uterine twigs, and the posterior protuberance of gravid proglottids. T. asiatica metacestodes (Cysticercus viscerotropica) also differ morphologically from T. saginata metacestodes (C. bovis) in having wart-like formations on the external surface of the bladder wall (Eom and Rim, 1993).

Traditionally the armed T. solium Linnaeus (1758) and the unarmed T. saginata Goeze (1782) were classified on the basis of the presence or absence of the hooklets (arm) and the rostellum on the scolex. The armed tapeworm, T. solium, has both of these structures. T. asiatica, on the other hand, looks phenotypically similar to classical T. saginata, which provoked debates on the taxonomic status at the level of species or subspecies.

Taxonomic disagreement on this tapeworm induced a redescription of Fan et al. (1995), where it was considered as a subspecies, T. saginata asiatica, on the basis of morphological similarities. Molecular methodologies were applied to this debate by comparing nucleotide sequences of the mitochondrial cytochrome C oxidase I (COI) and 5' 28S rDNA, and exhibited an intimate relationship of the Taenia species or subspecies, which supported the insistence of subspecies (Bowles and McManus, 1994; Fan and Chung, 1998).

Further studies on the morphology and genotype demonstrated that this parasite should be considered a distinct species, which is closely related to T. saginata and more distantly so to T. solium (Zarlenga and George, 1995; Galan-Puchades and Mas-Coma, 1996; Galan-Puchades and Fuentes, 2000; Hoberg et al 2000; Schantz, 2000).

Classical T. saginata infects humans through beef, and T. solium through pork, but T. asiatica through the viscera of pigs; in other words, the life cycles are different (Fig. 2).

The natural infection status of T. asiatica metacestodes (C. viscerotropica) was reported by Eom and Rim (1992a) from the liver of pigs inspected in Chongju, Chungchongbuk-do, Korea. All pigs were six months old cross-bred pigs of Yorkshire, Landrace, Duroc or Hampshire breeds. Live metacestodes were found in only 0.01% (3/25,358) of the examined livers, and were distributed randomly in lobes of livers (Fig. 3). The liver Livers were infected with one to ninety-six metacestodes. Most of the metacestodes, however, were either calcified (87.1%) or highly degenerated (12.9%). When metacestodes were alive, host tissue capsules were transparent. The capsule with degenerated metacestodes contained pus cells or mucopurulent fluid in the space between the thick yellowish host tissue capsule and the non-motile bladder worm.

If we postulate that Asian tapeworm metacestodes develop from eggs in 1 month, the metacestodes recognized in the liver had been infected for 1 to 5 months. In this short period of infection, all but a very few died. This is not a common feature in metacestodes of either T. solium or T. saginata, which live in the muscle of pigs or cattle for longer than a year. Even if very few T. asiatica metacestodes are alive in the liver, and the liver can still be a source of human infection. Chances of human infection should be very low, if any. This fact of rapid death of T. asiatica metacestodes in the livers of the pigs raises the question that there may be other tissues where T. asiatica metacestodes maintain patent infection for a longer period.

The infectivity of metacestodes of T. asiatica in human hosts was confirmed in a volunteer (Eom and Rim, 1992b). The metacestodes used in the experimental infection were collected from the livers of naturally infected domestic pigs at an abattoir in Chongju City (Eom and Rim, 1992a). The scolex, recovered after treatment, was found to be unarmed. The number of main uterine branches, varying from 16 to 21, was similar to that of T. saginata. With the eggs of T. asiatica from the experimentally infected Korean volunteer, the authors infected and observed the infection sites of the metacestodes in the visceral organs of pigs and cattle.

In the experimental study in pigs, T. asiatica eggs infected both of the hepatic and extrahepatic organs of pigs. Viscerotropic cysticercosis was observed in the pigs, with a majority of the non-calcified metacestodes in the livers (hepatic) and a minority in other organs (extrahepatic) such as the omentum, the lung and the serosa. Living metacestodes were observed both in hepatic and extrahepatic infection (Fig. 4).

In the experimental study using cattle, T. asiatica eggs infected calves in their livers but all were found to be calcified, measuring diameter less than 1.5 mm. T. saginata eggs infected calves in the muscle only (Fig. 4).

There is no doubt that Korea is one of the endemic areas for T. saginata, recalling the reports on C. bovis between the years 1924 and 1943 from calves in Cheju and many other provinces, with infection rates revealed to be 5.1-37.6% in cattle. The most recent observation appears to be in the measled cattle which was discarded by villagers in Kongkun-myon, Hoengsong-gun, Kangwon-do in 1975; this anecdote was reported to a Korean parasitologist. Considering all of the previous evidence, as well as the Korean habit of eating raw beef, and that no parasitologists living today had ever seen larval tapeworms in cattle by themselves, it is not difficult to speculate that Korea has had classical T. saginata in the past. But the recent researches by morphological or molecular biological methods after 1993 reveal that most of the tapeworms found are T. asiatica, and not T. saginata.

This situation leads to curiosity about the presence of T. saginata in recent times. Even if it is possible to find T. asiatica among the old slide specimens, labeled as T. saginata, preserved in many parasitology laboratories in universities and institutes scattered in Korea, the differential diagnosis is not easy, due to the morphological similarities between T. asiatica and T. saginata. An alternative method of identification, with the use of recently developed molecular biological methods, also has limited application due to DNA damage by long-term preservation of the worms in formalin.

More recently, Eom et al. (2000) reported a T. saginata case collected from a local hospital in Pusan. The Taenia worm from the case was transfered to and analysed in the Department of Parasitology, Chungbuk National University School of Medicine. The two terminal proglottids transfered did not show posterior protuberances, which are one of the differential diagnostic characters. One of them was 9 by 4 mm and the other was 11 by 4 mm, in measurements of terminal proglottids. The numbers of uterine main branches were not countable, because most of the eggs were discharged out of the uterus. Random amplified polymorphic DNA-polymerase chain reaction was done on isolated genomic DNA, which showed main bands in 1.95, 1.59, 1.00, 0.80 and 0.66 kb with a primer, oligonucleotides A-08 5'-GTG ACG TAG G-3'; and multiple bands of 2.07, 1.82, 1.59, 1.13 and 0.86 kb with another primer, A-20 5'-GTT GCG ATC C-3'. These bands were identical with T. saginata from Poland and Belgium but were different from T. solium originating from Honduras and China, as well as from T. asiatica from many countries.

This subject, a Korean woman aged 46-years, had no experience of eating raw liver of pigs, but prefered raw beef without any experience of traveling to other countries a few years before and after observing the expulsion of terminal proglottids by herself. With overall consideration of these data, T. saginata thus is distributed sympatrically with T. asiatica in Korea.

Natural infection of T. asiatica metacestodes in Korean domestic pigs was reported in 1992 by Eom and Rim. The infectivity of those metacestodes was also confirmed in a human volunteer (Eom and Rim, 1992). If we assume the visceral organ of pigs as a source of Taenia tapeworm infection, we can now explain the epidemiological inconsistency as follows. The species ratio of 'armed tapeworm versus unarmed tapeworm = 15.4 : 79.4 = 16 : 84' and fits well to 'the food-consuming ratio of pork versus pig's viscera plus beef = 7.1 : (25.0 + 6.1) = 7.1 : 31.1 = 19 : 81' (Table 2 & 3; Fig. 5). Or being more simple, armed tapeworm versus unarmed tapeworm is 16 : 84 = (approximately) 1:5. And pork versus pig's viscera versus beef is 7.1 : 25.0 : 6.1 = (approximately) 1 : 4 : 1. In other words, pork : (pig's viscera + beef) is 1 : (4 + 1) = 1 : 5. Conversely, the 5 : 1 ratio of pig-eaters versus cattle-eaters can be explained as follows. Pig-eaters consist of pork-eaters and viscera-eaters, and cattle-eaters consist only of beef-eaters, which makes (1 + 4) : 1 = 5 : 1. This figuring can be shown with simple numbers in brackets in Fig. 5. Again, the pig's viscera plays an important role in the transmission of T. asiatica, with a relative ratio value, 4, which unravels the inconsistency between 1 : 5 and 5 : 1. Here we can now understand that pigs play their role both as an infective source of T. solium with pork, and T. asiatica with visceral organs. This hypothesis gives a definite role of the pig's viscera in the tapeworm epidemiology in Korea (Table 3).

Still no natural evidence of live C. viscerotropica In the liver of cattle is available. It needs further evaluation yet, but 10.2% (10.1-11.4) of people eat raw liver of cattle (Kang et al., 1965a; Cho et al., 1967) and this may receive due consideration for its importance in T. asiatica epidemiology in future, because human infection is possible with live metacestodes recovered from the liver of experimental cattle in Taiwan (Chao et al., 1988). This hypothesis gives a good fit between the species ratio of armed tapeworm versus unarmed tapeworm, at 16 : 84, and the eating habits, i.e., pork : pig's viscera plus cattle's liver plus beef was 7.1 : (25.0 + 10.2 + 6.1) = 7.1 : 41.3 = 15 : 85, theoretically though, if we assume that transmission rates to humans are equal for the three different species.

Taenia tapeworm infections among humans correspond closely to the eating habits of intermediate hosts in each country. In Korea, we can count the eating habits of 'Churyum' and 'Semmai' those rationalize our hypothesis 1 and 2. 'Churyum' is one of Korean habits of eating widely spread among rural people in Cheju or other provinces where they eat raw liver of pigs while they are still warm right after slaughtering during village ceremonies. The other term, 'Semmai' is also called 'Senmai' which means [Senezi (n) n. Fat around the intestines: epiploon, fatty skin around the intestine of animals, fat of the epiploon: thin layer of fatty tissue on the cud pouch or rumen, semeze = semezi (provided by Prof. Won-Kll Park, Academic Society of Korea and Mongolia)]. The terms implicate the possible source of T. asiatica infection by eating 'Semmai', the fat, contaminated with extrahepatic metacestodes that could be introduced to human and cause tapeworm infections. In other words, this Mongolian term has been used in Korea, especially in Cheju-do, since the 13th century when Mongolian soldiers temporarily ruled. The habit of eating raw Semmai alone or attached to the stomach, spleen or any other visceral organs may cause the infection with T. asiatica. The omentum and fat tissue around the stomach, spleen and intestine harbor sometimes many hidden, hardly recognizable, transparent metacestodes. Eating the raw spleen itself may have no meaning but eating the omentum or fat tissues around the organs together with fat may induce the tapeworm infection. We can say the same is true for the other visceral organs such as the stomach, intestine, kidney and pancreas.

Of the three tapeworms in Korea, the distribution ratio is uncertain now, but it may be possible to speculate a ratio of T. solium : T. asiatica : T. saginata is about 7.1 : 25.0 : 6.1 = 19 : 65 : 16 (approximately 1 : 4 : 1) according to hypothesis-1; or 7.1 : (25.0 + 10.2) : 6.1 = 15 : 73 : 13 (approximately 1:5:1) according to hypothesis-2.

Evaluating the application of these hypotheses, we can get a clear understanding about the epidemiologicalal enigma, which was addressed continuously since the 1960s in Korea, by simply acknowledging the existence of T. asiatica.

The custom of eating raw viscera is not new in other Asian countries also, like Taiwan and Thailand where Asian tapeworm is endemic (Cross et al., 1991; Fan et al., 1989b). Usually the small intestine and the flesh of wild or domestic animals are eaten raw in these countries, and most Asian countries are believed to have a similar kind of Asian tapeworm phenomenon as in Korea.

An isolated island Sorokdo (a leprosarium) is a small place showing the most simple pattern of Taenia epidemiology in Korea, and possibly, in the world. This unique place of leprosy patients' residence has a single kind of Taenia tapeworm with a single kind of intermediate host, pig. So distinct was this epidemiological pattern with unarmed Taenia tapeworm, the pig was suspected to be an intermediate host of this tapeworm, and T. asiatica was identified later, without difficulty in retrospective, though; this place is considered to be the most simple place in this study (Fig. 6A).

The most similar Taiwanese place to Korean Sorokdo Island is Lanyu Island, of 30 square miles. As is known to western hemisphere with a familiar name, Orchid Island is a land of the Yami tribe, who moved to this small island two hundred years ago from South-East Asia, via the Philippines. Since 1960 when Hsieh first reported a beef tapeworm, it was known as one of the endemic areas of T. saginata. Quite recently, however, it was recognized as Asian Taenia, which was transfered to humans through Lanyu native pigs, a confirmed intermediate host (Chung et al., 1987). It is a simpler place, with wild animals and native goats, than mainland Taiwan; but it shows a rather more complicated epidemiological pattern than Korean Sorokdo Island, because it has cattle, the intermediate host of beef tapeworm (Fig. 6B).

In contrast, the mainland Taiwan is an island far bigger than either Sorokdo Island or Orchid Island, with many intermediate hosts as well as with complicated eating habits of raw foods. In Taiwan, Taenia tapeworms are prevalent as high as 17% and all of the tapeworms are unarmed, and thus T. solium is not available (Liu et al., 1981; Chan et al., 1987; Chung et al., 1987; Chung et al., 1988; Lin et al., 1988; Fan et al., 1989a, 1989b). It shows a more complicated epidemiological pattern than on Lanyu Island, in having many kinds of possible intermediate hosts as well as with complicated eating habits of raw foods; but there is no T. solium (Fig. 6C).

It is strongly considered that the pork tapeworms and the beef tapeworms occur together in Korea where the pigs and cattle are also distributed, and as well, both of them are eaten raw. Considering this epidemiologicalal complexity, it must have been difficult for parasitologists to determine what kinds of tapeworms are transfered through what kinds of intermediate hosts. As for the reason for more beef tapeworm infections, the suggestion that raw beef was consumed more frequently was not persuasive at all, because of the rarity of T. saginata metacestodes in intermediate hosts. Evidently, this erroneous speculation was closely related to the presence of T. asiatica that shares the pigs as intermediate hosts with T. solium; Cheju-do Island had been a very complex place to understand.

We can rearrange from the simplest to the hardest aspect of environmental complexity in understanding Taenia epidemiology as follows: Sorokdo Island < Lanyu Island < Mainland Taiwan < Cheju-do Island (Fig. 6A, B, C & D). It may sound paradoxical, but Cheju-do is one of the best places to study because it is separated from the land, has a small area, but has all complex factors on Taenia epidemiology just as in other Asian countries which have a similar infection pattern to Korea.

Thus the explanation for the epidemiological inconsistency, which may possibly occur in other Asian countries also, is likely to depend on recognition of the unarmed tapeworm, T. asiatica.