Archaeobotany of barnyard millet (Echinochloa) in the ...
Meiji University Premodern Japan Research Exchange, 0708 DEC 2009, USC
Prehistoric and Protohistoric Plant Use in the Japanese Archipelago Katsunori Takase （Faculty of Letters, Meiji University）
Introduction
This lecture outlines prehistoric and protohistoric plant use in the Japanese archipelago based on the latest information of plant archaeology. Slide 1 pictures useful plants which are examined today. I am focusing on crops of grass plants and beans including rice, broomcorn millet, foxtail millet, barnyard
millet,
barley,
wheat,
adzuki
and
soybeans. Slide 2 summarizes chronological and regional settings. I cannot talk about regional differences in detail due to time constraints. However, I would like to focus on diachronic change and periods of occurrence of useful plants. Please refer to the chronological table (Slide 2) as needed. Calendar years on this table are based on results of recent AMS radiocarbon dating and dendrochronology. In a framework of Japanese archaeology, prehistory means before the Jomon period, and protohistory means the Yayoi and the first half of the Kofun periods. Agriculture was the core economy of ancient Japan and paddy rice was the most important grain. Rice cultivation was the basis of the taxation system and rice has been continuously a major crop down to modern times in Japan. At the same time, however, it is notable that the ancient polity tried to increase 1
productivity
of
other
grains.
For
example, Prof. Shigemitsu Kimura has pointed out that there were three phases of policy concerning millets in classical Japan (Slide 3). In the first phase, until 820 C.E., foxtail millet was recommended. And in the second phase, barley
and
wheat
were
strongly
recommended. Then, after 920 C.E., various millets were recommended for cultivation. As is well known, various goods were paid as tribute to the ancient court. Millets and other useful plants were included. This table (Slide 4) shows various millets and useful plants that were cultivated in classical Japan. Cultivation not only of rice but also of some kinds of millet is evidenced by literary texts. According to Prof. Osamu Matsuda, about 160 species of plants are referred to in the Manyoshu, Kojiki,
Nihonshoki,
Fudoki,
Kokinwakashu,
Makuranososhi, and the Tale of Genji. Matsuda has noted that in total there are 38 useful plants mentioned in these texts,
with
representative
grains
including rice, barley, wheat, broomcorn millet, foxtail millet, and barnyard millet. Archaeological
records
further
confirm his argument. Many charred seeds of domesticated plants have been excavated
from
sites.
These
are
2
examples
of
carbonized
seeds
from
settlements dating from the 7th to the 9th centuries in eastern and northern Japan. Rice is usually dominant in a complex of domesticated plant seeds, but in the case of sites located in inland and hilly areas, seeds of millets and beans occasionally exceed rice seeds, as shown in Slide 5. There the picture on the left shows charred seeds from a settlement of the 7th century and located in the mountainous locale of western Tokyo. Foxtail millet is dominant at this site. The picture on the right shows seeds from Suginodo site in northern Japan. This site is a large settlement of the 8th to the 9th centuries. In this site, rice was the dominant crop; however, many kinds of millets have also been found there. Slide 6 shows additional seeds from the
Suginodo
site.
Charred
acorns
indicate that wild tree nuts were also used with domesticated plants. And I could also identify adzuki beans by observing an organ known as the primary leaf, a point on which I will elaborate later. Slide
7
also
shows
seeds
from
settlements of the 9th to the 10th centuries, and therein barley, beans, and acorns could be identified along with rice. To sum up, not only rice but also various millets and beans have been uncovered at ancient sites, even in 3
northern and eastern Japan.
Current
Status
of
Study
on
Prehistoric Plant Use in the Japanese archipelago
It is still difficult to know when and how these grains came to be used in the Japanese archipelago. There are two current hypotheses as to when domesticated plants such as rice and major millets came into use. The first hypothesis is that they were introduced before the Late Jomon. For example, authors of these highquality overviews (Slide 8) often argue that rice and millets were utilized before the Late Jomon, although farming at that time may not have played an important role in the economy. These books are very influential because they are useful to readers desirous of catching up on the latest knowledge of archaeology or paleoethnobotany. Therefore, this hypothesis has rapidly taken root in Japan during the last fifteen years. A second hypothesis is that rice and major millets were utilized after the second half of the Final Jomon or by the beginning of the Yayoi period. Its advocates take a cautious stance on the previous hypothesis. For example, they use AMS radiocarbon dating of charred seeds proactively to evaluate the reliability of plant remains. In addition, a new methodology known as the Replica SEM Method, which permits observation of seed impressions on potsherds, has afforded important results (Slide 9). In this method, replica of impressions on fragments of pottery are made using silicone rubber. Then observation of the replica with a scanning electron microscope enables much more accurate recognition of the character of the impression than does general microscopic observation. I will elaborate further on these achievements later. But first I would like to present an abstract of the diffusion process of farming in East Asia. Recent archaeological investigations in China have demonstrated that wild rice was first utilized in the Yangtze River basin prior to 9,000 or 10,000 years ago. After 7,000 years ago, domesticated rice became dominant, as shown by charred rice caryopses from archaeological sites in the Yangtze River basin. Slide 10 indicates the distribution of rice cultivation in each period. In contrast, millet cultivation also occurred before 8,000 or 7,000 years ago in the Yellow River basin. Much later, rice cultivation had diffused to the Shandon Peninsula by 2,500 B.C.E. from the south. In this region, wet rice cultivation was combined with millet farming. And still later the 4
technology of wet rice cultivation on the Korean
Peninsula
was
strongly
influenced by that on the Shandong Peninsula,
with
the
archipelago
also
being
Japanese affected
indirectly. Specifically, on the Korean Peninsula
millet
farming
and
rice
cultivation had already taken set form by 1,500 B.C.E. That makes it probable, as some archaeologists have expected, that rice and millet cultivation was introduced to the Japanese archipelago before the Late Jomon, dated to before 1,200 B.C.E. However to gain a more specific sense of when various grains were introduced to the Japanese archipelago, it is important to examine the archaeological data exactly. I turn to that task now.
Rice
In
last
decade
or
two
some
archaeologists have argued that rice cultivation goes back before the Middle Jomon. For example, Prof. Kazuo Hirose, Prof.
Kazuo
Miyamoto,
and
Prof.
Takehiko Matsuki have all stated in their books
that habitants
of the
Japanese archipelago started to use rice before the Middle or the Late Jomon. They
emphasize
the
existence
of
phytolith data and impressions of rice on pottery from before the Late Jomon. And Prof. Kazuo Miyamoto has opined 5
as well that charred seeds of rice indicate cultivation
before
the
Late
Jomon,
although the number of rice seeds found remains very small. Nevertheless there is always danger of contamination by tiny phytolith like plant opal in sediments of archaeological sites (Slide 11). To avoid this problem, fragments of pottery have been used as samples for plant opal analysis. If plant opal can be found in the pottery, its date is thought attributable to the date of the pottery. And Prof. Mamoru Takahashi has in fact reported that the plant opal of rice has been found in pottery dating from before the Middle Jomon. However, it is also important to recognize that phytolith is enough small to fly from the Chinese continent to the Japanese archipelago with yellow dust. As shown in Slide 11, a wide area of East Asia
is
covered
with
yellow
dust,
especially in spring. Yellow dust reaches as far as the Hawaiian Islands and sometimes can fly as far as Alaska. So it is likely that a small amount of rice opal is
found
in
Jomonperiod
sediment
simply because wild and domesticated rice existed on the Chinese continent at that time. Phytolith, of course, provides efficient data
when
they
are
quantitativelydetected. If there are more than five thousand plant opals of rice in 1g sediment from a site, it is possible to 6
estimate that there was a paddy rice field there.
However,
small
numbers
of
phytolith cannot be reliable evidence for proving the cultivation of rice in the Japanese archipelago. Then, what about charred seeds from archaeological sites? This table (Slide 12) shows the number of carbonized grass plant seeds found in sites from the Initial Jomon to the Middle Jomon. In this phase, only the genus Echinochloa—kinds of barnyard millet—was found in northern Japan. Data for barley is not reliable because the number of specimens is very small. Slide 13 shows a table summarizing the number of seeds from the Late Jomon to the Final Jomon. In this phase, the number of rice seeds increases. However, these
cases
still
seem
accidental,
considering the number of sites excavated in a year in Japan. In addition, it is notable that there is no example definitely dated to this period by physicochemical dating. Finally, Slide 14 includes a table showing the number of charred seeds from the first half of the Yayoi period. From this phase seeds of rice and millets can be found repeatedly in various sites in all areas except Hokkaido and Ryukyu. As these tables show, the number of rice seeds increase rapidly after the Yayoi period begins. Although more than two hundred charred seeds of rice have been 7
reported from the Late and Final Jomon, it is still hard to affirm the existence of rice cultivation on the basis of seeds. There is also danger of contamination by carbonized seeds because they move quite easily and naturally between sedimentary levels. And paddy rice fields have not yet been found before the Final Jomon. At present, carbonized seeds indicate that
fullscale
rice
cultivation
was
introduced after the end of the Final Jomon or during the Early Yayoi. This perspective is supported by paddy rice fields and implements of farming. Kazahari site in northern Japan is the only site where the radiocarbon date of carbonized seeds of rice was measured. The results were 2540±240BP and 2810± 270BP (Slide 15). These specimens were measured
in
the
early
1990’s
in
a
laboratory at the University of Toronto, and they represent one of the earliest examples of AMS radiocarbon dating of charred
seed
in
Japan.
Therefore,
calibrated ages have a wide range— from the 17th century B.C.E. to the 1st century B.C.E. These results indicate that charred seeds of rice from Kazahari cannot be indisputable evidence showing the existence of rice cultivation before the Late Jomon, because they might be assigned to the Yayoi period although more accurate periodization is not possible now. Present technology of AMS radiocarbon dating can provide results with a higher degree of accuracy. In recent years, the National Museum of Japanese History tried to determine the date of one specimen of charred rice from the same Kazahari site. Unfortunately, the result was 173±35BP, indicating a distinct case of contamination by modern rice. At present, there are no charred seeds of rice dated with certainty before the Late Jomon by physicochemical analysis. Researchers must also be quite careful in dating seed impressions on pottery because it is not easy to determine the chronology of the pottery. Slide 16 shows a fragment of a vessel with rice 8
impressions from Minamimizote site, in the ChugokuShikoku region. In popular books of Japanese archaeology, many authors have stated that this find constitutes reliable evidence of rice cultivation in the Late Jomon. But lately the fragment has been joined with others and restored as a large pot of the second half of the Final Jomon phase. It is not therefore evidence demonstrating rice cultivation in the Late Jomon. Next, let me introduce recent achievements of ReplicaSEM Method. Mr. Michihiko Nakazawa and Prof. Tsuyoshi Ushino have examined riceshaped impressions on pottery intensively using this method since 1998 (Slides 17, 18). As a result, it was revealed that ten of the eleven impressions were not made by rice, although formerly all impressions had been thought to be those of rice. Almost all of the impressions were made by small fragments of wood or unknown seeds of weed. It was an amazing result. In fact there is only a single reliable impression made by rice: that on a fragment of pottery from Ishigo site in the KantoKoshin’etsu region, which has been assigned to the second half of the Final Jomon (Slide 19). Using this method, researchers have continued to detect reliable data of the oldest rice in each region. This table (Slide 20) shows a chronological sequence of pottery; a black circular mark means the occurrence of reliable rice impressions. At present, such impressions never predate the first half of the Final Jomon in the western part of the Japanese archipelago. And in the Kanto and Tohoku regions, the number of impressions of rice suddenly increases after the Early Yayoi. And while I have searched intensively for impressions of seeds in the Kanto and Tohoku regions, I have found no reliable rice impressions before the Final Jomon. On the other hand I have already found more than 60 impressions of rice on pottery vessels dating from the Early and Middle Yayoi (Slide 21). As expected, the Yayoi period has proven to be an important epoch for rice cultivation in the eastern part of the archipelago. To sum up, there is no archaeological record showing the existence of rice before the Late Jomon. And according to seed impressions on pottery, the oldest rice seeds date from after the second half of the Final Jomon, or during the Early Yayoi.
An Insect Pest, the Maize Weevil
Another important problem highlighted by ReplicaSEM Method is that of an insect pest. Mr. Sumio Yamazaki has reported that 9
impressions of the maize weevil go back to at least the Late Jomon (Slide 22). He has also opined that cultivation of rice must also date back to the Late Jomon, since the maize weevil has a close relationship with rice. On the other hand, Prof. Hiromichi Ando asserts that this theory needs cautious examination because no one knows how the maize weevil lived before the introduction of domesticated crops in Japan, even if it is now seen as a rice
pest.
Meanwhile
pestology
has
studied the insect to learn how to prevent its destructive activities, but we do not yet know its longterm life history. Pestological study (Slide 23) has already clarified that the food of the maize weevil is restricted to rice，barley，wheat, and maize. They grow from egg to pula in one grain and therefore they need grains longer than their body length to become adult. Their reproduction also requires threshing because the tube for bearing eggs cannot penetrate the husks of the grain. Pestology shows that the maize weevil might be related to rice. Therefore, the presence of this insect may suggest utilization of rice before the Late Jomon. However, there is no entomological and pestological data as to whether they eat other kinds of nuts or grains, or as to conditions of reproduction. Archaeologists have to investigate this problem. Slide 24 shows results of a breeding experiment for maize weevils conducted by Prof. Hiroki Obata and Ms. Yasuko Senba. It looks like a crazy experiment, but I think it is important. According to this study, the maize weevil eats threshed rice, barley, wheat, broomcorn millet and acorn. But survival into a second generation occurred only in the case of threshed rice, barley, and wheat. If so, the existence of the maize weevil may well suggest the use of rice. Nevertheless, one more update needs noting: Prof. Obata recently stated that maize weevils fed with acorn were still alive although he once thought that they had all died. Our studies of the maize weevil and its significance for rice are 10
right back where they started.
Foxtail Millet
Until the 1970’s, some archaeologists believed that foxtail millet was cultivated during
the
Middle
KantoKoshin’etsu
Jomon region.
in
the Tiny
roundshaped charred seeds from sites of this period had been regarded as foxtail millet for a long time. However, Prof. Akiko Matsutani revealed that they were seeds of the beefsteak plant, also known as perilla. At present, the reliable data for foxtail millet is still very small. However, positive
evidence
has
been
reported
recently. This table (Slide 25) shows the number of seed from the Nakayashiki site of Kanto region, which is dated as the Early Yayoi. At this site some pit features including a lot of charred seeds of rice, millets, and horse chestnut were excavated. AMS radiocarbon dates of rice are consistent with remarks of investigation (Slide 26). Foxtail millet should therefore be assigned to the Early Yayoi. Furthermore last month Ms. Yuka Sasaki and others reported that plural impressions of foxtail millet could be seen on fragments of pottery from this site. In addition, many charred seeds of foxtail millet were found in a field cultivated during the Early Yayoi at ShoKuramoto site located on a campus of Tokushima 11
University,
ChugokuShikoku
region.
Current data seems to suggest that the Early Yayoi was the time when foxtail millet came to be utilized. Another reliable example of foxtail millet was excavated at Tagoyama site, Kanto region (Slide 27). This is a dwelling pit of the Late Yayoi. Carbonized wood and heataltered soil suggest that this house was burned down. In areas A and B, many carbonized seeds of rice, millet, and beans were found around the floor of the dwelling pit. The picture on the right strongly suggests that ears of rice were charred for some reason because rice husks face in the same direction in area A. In area B, on the other hand, many seeds of foxtail millet and beans were recovered by the waterflotation technique as shown in this table (Slide 28). In total, about two hundred thousand seeds of foxtail millet were found in this dwelling pit (Slide 28). This is the largest number of foxtail millet seeds from any single pre or protohistoric site in Japan, and one of the most reliable examples of foxtail millet use in the Yayoi period. Recently,
we
also
examined
AMS
radiocarbon dating of rice and foxtail millet from this site. Specimens are one caryopsis
rice
for
AMS
dating,
five
caryopses of rice for stable isotope analysis of
carbon
and
nitrogen,
and
sixty
caryopses of foxtail millet for AMS dating and stable isotope analysis. The results of 12
radiocarbon dating, seen in Slide 29, demonstrate that the rice and foxtail millet date from the 1st century C.E. to the 3rd century C.E. It is therefore safe to say that foxtail millet from this site belongs to the Yayoi period. In addition, stable isotope analysis of carbon
and
nitrogen
support
our
identification of the seeds. Values of rice fall into an area of C3 plant, while result of foxtail millet was plotted in an area of C4 plant (Slide 30). In conclusion, use of foxtail millet began in the Japanse archipelago from the Early Yayoi. And specimens from Tagoyama site also provide reliable data for the Late Yayoi. Foxtail millet of Yayoi period provided a basis for intensive utilization after the Kofun period and in classical times, although there is still room for investigating the situation during the Middle Yayoi.
Broomcorn Millet
The oldest reliable data for broomcorn millet was found at the Nakayashiki site—dating
from
the
Early
Yayoi—mentioned above. In addition, other
reliable
data
comes
from
RyugasakiA site in the Kinki region. A lump of charred seeds adhered to the inside wall of a pottery vessel dating from the end of the Final Jomon or from the Early Yayoi. Prof. Akiko Matsutani 13
identified these seeds as broomcorn millet by tissue observation using SEM (Slide 31). Results of AMS radiocarbon dating are consistent to chronological estimates of the pottery type. These specimens suggest that the end of the Final Jomon or the Early Yayoi is an epoch of utilization of broomcorn millet, as was the case for rice and foxtail millet. After the beginning of the Kofun period,
considerable
numbers
of
broomcorn millet seeds appear, especially at settlements on hilly areas. However, little is known about broomcorn millet in the Middle and the Late Yayoi at this time.
Barnyard Millet
Echinochloa, types of barnyard millet should be examined in detail because some archaeologists have presented a hypothesis
that
domesticated
it
in
was the
firstly Japanese
archipelago.
Echinochloa species of Asia fall into three main groups of biosystematics as shown in Slide 32. The first is the phylogenesis
Echinochloa
utilis,
or
domesticated barnyard millet, which has been mainly cultivated in East Asia. Botanical constitution
study has
of
its
genome
demonstrated
that
domesticated barnyard millet originated from Echinochloa crusgalli. Meanwhile 14
archaeology in East Asia has revealed that millets in the first group came into use more than several thousand years ago. I would like to focus here on the use of
Echinochloa crusgalli and barnyard millet (Echinochloa utilis). In the 1980’s, Prof. Gary Crawford revealed
by
using
the
waterflotation
technique that Echinochloa was used in the Early and the Middle Jomon of northern Japan. After that, use of Echinochloa in Jomon has been confirmed in many sites by researchers
conducting largescale and
systematic archaeobotanical studies in the Japanese archipelago. Recent AMS radiocarbon dating has also demonstrated that Echinochloa use extends back to the Middle Jomon as shown in this table (Slide 33). The results suggest that these specimens go back at least to the 3rd millennium B.C.E. In addition, hundreds of Echinochloa seeds assigned to the Initial and the Early Jomon have been reported (ex. Slide 34). However, their dates had not yet been crosschecked by radiocarbon dating. Let
me
Echinochloa
examine seed
distribution from
pre
of and
protohistoric sites. This map shows the frequency of charred seed of Echinochloa (Slide 35). Black sections in the circle graphs indicate the ratio of Echinochloa seed to all grass plant seeds. The figure at left is based on data that pertains to artifacts before the Middle Jomon. The 15
calibrated radiocarbon age is estimated as earlier than about 2,400 B.C.E. Almost all the specimens of Echinochloa in this period are concentrated in northern Japan. The figure at the right shows the situation in the Late and the Final Jomon. Although the ratio of Echinochloa decreased sharply in the Tohoku region, it remained the dominant cereal in Hokkaido. This map (Slide 36) shows the case of Yayoi and the EpiJomon. As with the previous periods,
Echinochloa was a common millet on Hokkaido. In short, current archaeological data on Echinochloa from archaeological sites is mostly concentrated in northern Japan. I believe that Echinochloa has been utilized for a long time, since the Jomon period. In particular, Hokkaido is one of the regions where Echinochloa was used intensively and continuously. The morphology of seeds also provides an effective clue for examining past utilization of
Echinochloa. In 1983, Prof. Gary Crawford noted that the mean size of Echinochloa caryopses from the Middle Jomon is about 20 percent larger than specimens from the Early Jomon. In the 1990’s, Prof. Masakazu Yoshizaki recognized three types of Echinochloa caryopsis from the Jomon sites as shown in this figure (Slide 37). The first is a relatively small and slender type similar to Echinochloa crusgalli. The second is a relatively large and round type similar to domesticated barnyard millet. The last is an intermediate type with characteristics between the first and second types. Prof. Yoshizaki gave the nickname ‘Jomonhie’ to the second and the third types. Echinochloa seeds were found in the hearths of dwelling pits—they were not accompanied by traces of wild weeds and husks had been already removed. He also presumed that ‘Jomonhie’ increased after the Early and the Middle Jomon as a result of human husbandry. In Slide 38 we see SEM photographs of excavated barnyard millet in Hokkaido. From a macro perspective, it is possible to trace a morphological change in the seeds from Echinochloa crusgalli to barnyard millet via ‘Jomonhie’ in northern Japan. In 1988 Prof. Sadao Sakamoto presented his hypothesis that barnyard millet was domesticated in the northern part of the Japanese archipelago and diffused to a wide area of East Asia from this region. Nevertheless the domestication center of barnyard millet has not yet been thoroughly 16
explained. This graph (Slide 39) represents the length and breadth of Echinochloa caryopses from northern Japan. It is difficult to detect distinct morphological changes in specimens
from
Jomon
to
Yayoi
or
EpiJomon. Additionally seeds from the Satsumon period (dated from the 7th to the 12th century) tend to be rather smaller than those of previous periods. In contrast, significant changes occurred after the 13th century, signifying the appearance of real domesticated barnyard millet. If seed size became larger gradually, a possible interpretation is that barnyard millet was domesticated as a result of longterm use in this region. However, it changed drastically in a short period from the 12th to the 13th century. Therefore there is some possibility that barnyard millet was domesticated elsewhere and then imported to northern Japan after the 13th century. Notably Echinochloa
crusgalli is a ubiquitous wild weed in temperate regions. The wide natural geographical range of Echinochloa crusgalli makes it a suitable candidate for multiple domestication events in East Asia. Thus the relationship between ‘Jomonhie’ and barnyard millet should be carefully examined not only in the Japanese archipelago but also in the broader region of East Asia. In that regard, in Korea, however, there is currently no reliable data of Echinochloa before ThreeKingdom period, according to Prof. Ahn SungMo. But in recent years, Echinochloa seeds were found in Neolithic sites in the Russian Far East (Slide 40). Ms. Elena Sergusheva has reported that nonwild type Echinochloa seeds have been found in the specimens from Zaisanovka1, dated about 2,500 B.C.E. I think that this data indicates that further investigation of Echinochloa across East Asia is needed. For now, it is still difficult to determine the domestication center of barnyard millet.
Barley and Wheat
Until the last century, some archeologists argued that wheat was used in Yayoi period. However, many researchers now think that there is no reliable data in terms of seed remains for wheat before the Yayoi period. As for barley, Prof. Hiromich Ando reacted skeptically to views that it was used in Jomon and Yayoi periods because information on it is rare. However, an impression of barley was found by ReplicaSEM Method on a fragment of pottery from the end of the Final Jomon to the Early Yayoi. Slide 41 shows a SEM photograph of an impression of barley from Nakamichi site in the 17
KantoKoshjin’etsu region. I am therefore sure that barley was used in the Final Jomon or the Early Yayoi, but it is still difficult to estimate its significance as a major crop.
Adzuki Beans
Until the first half of the 1990’s, many archaeologists believed that the mug bean was used from the Early Jomon, based on result of excavation at the Torihama site, a famous Jomon site in the TokaiKinki region. However, it is difficult to identify mug beans by morphological observation because their shape is very similar to adzuki beans. They are both species classified in the genus Vigna. In 1992, Prof. Masakazu Yoshizaki and Ms. Yasuyo Tsubakisaka
developed
a
method
to
distinguish the mug bean from the adzuki bean by studying their primary leaves. These pictures (Slide 42, 43) show morphological differences between the primary leaves of the modern mug and adzuki bean. The primary leaf of a mug bean
is
relatively
large
and
hangs
vertically from the top of seed, while the small primary leaf comes down on a slant in the case of an adzuki bean. According to close examination using such a method, all Vigna beans from pre and
protohistory
of
the
Japanese
archipelago are adzuki beans. It is still difficult to conclude whether these adzuki 18
beans were domesticated or not; but at least, archaeology has begun to clarify longterm adzuki bean utilization in the Japanese archipelago.
Soybeans
Soybeans are attracting a great deal of attention in recent years. In 2007 Mr. Seiji Nakayama
and
others
reported
an
impression made by a bean on the jagged edge
of
a
Middle
Jomon
pot
from
Sakenomiba site, KantoKoshin’etsu region (Slide 44). This pot was found broken when it was excavated in 1997. Although its fragments
were
reconstructed
with
adhesive, a curator of the Yamanashi Prefectural Museum found that they did not adhere because secular changes of adhesive agent. At that time, a hole was discovered on broken surface of protruding part of the pot. On the opposite side of the broken fragment, the same impression was confirmed.
An
investigation
using
ReplicaSEM Method suggests that this hole was an impression made by a soybean. The soybean belongs to the genus
Glycine. Beans of the genus Glycine can be identifiable by existence of a hilum. Slide 45 shows a cross section of these beans. The hilum of beans in the genus Vigna and
Phaseplus, which include the species of kidney and navy beans, is covered with spongelike tissue known as the epihilum, seen here. Therefore it is impossible to see 19
the hilar groove in the center of the hilum. On the other hand, the hilar groove can be observed easily in the case of the genus
Glycine because there is no epihilum. Therefore, the exposed hilar groove of a replica recovered from Sakenomiba site makes it identifiable as a seed of the genus
Glycine (Slide 46). At present, Glycine seeds have been discovered from the Kyushu region to the Tohoku region from the Middle Jomon. Were these soybeans domesticated or not? Concerning this problem, further examination is needed. However, what is important is that the size of the Glycine seed is very large. Slide 47 shows a SEM photograph of two replica attached together to restore the entire shape of soybeans found at Sakenomiba. The length is 11.8 mm, longer than its original dry condition because it was mixed in pottery clay and contained water. A modern domesticated soybean also increases in size up to 11 or 12 mm after soaking in water. In contrast, a modern wild soybean expands to 8 or 9 mm after soaking. It is likely that Jomon soybeans were not wild soybeans. Another morphological characteristic of soybeans is thickness. The value of thickness of a Jomon soybean is usually less than 4mm. The thickness of a modern soybean soaked in water for six hours is 4 mm or greater. As for length, a Jomon soybean tends to be relatively flat. Such a flatshaped soybean is unique to the Japanese archipelago, and these are the oldest soybeans in East Asia. It is likely that the soybean was a highly controlled plant used by habitants in the Japanese archipelago in the Jomon period. While it is hard to say whether soybeans were domesticated in the Japanese archipelago, some archaeologists have begun to think that soybeans were domesticated in multiple places in East Asia, one of which was the Japanese archipelago.
Conclusion
In the Japanese archipelago, the occurrence of rice and major millets can be dated from the second half of the Final Jomon to the Early Yayoi. Although some scholars have insisted that rice and the major millets existed before the Late Jomon, current reliable data does not support this argument. At the same time, we can confirm that varieties of barnyard millet were used from the 3rd 20
millennium B.C.E., while husbandry of adzuki and soybeans seems to have been practiced to some degree at least after the 4th millennium B.C.E. In this phase, other useful plants like gourds, beefsteak plant/perilla, hemp plant, peach, burdock, oil seed, and rape were also used. Some archaeologists also think that forests of chestnut and horse chestnut were highly controlled around settlements in the Jomon period. And finally, the occurrence of wheat and buckwheat remains obscure, although they were popular crops in classical Japan.
References Akasaka, M. and Yamaguchi, H. 2006 Characterization of microsatellite of wild and domesticated E. oryzicola, Ikushugaku Kenkyu (Bleeding Science), 8, special issue1, p.241 (in Japanese). An SungMo 2006 Investigation of existence of the domestication of Echinochloa in the Ancient Korean Peninsula, Obata, H. ed.
Cultivated cereals in prehistoric and ancient Far East Asia 2, pp.3143, University of Kumamoto (in Japanese). Ando, H. 2006 A note on prehistoric plant remains and their impressions of potsherds, Nishisagami Koko, pp.112122 (in Japanese). Crawford, G.W. 1983 Paleoethnobotany of the Kameda Peninsula Jomon, University of Michigan. Crawford, G. W. 1992 Prehistoric Plant Domestication in East Asia, Cowan, C. W. and Watson, P. J. eds. The Origins of Agriculture:
International Perspective, pp.738, Smithsonian Institution. D’Andrea, A. C. 1992 Paleoethnobotany of later Jomon and Yayoi cultures of northeastern Japan: Northeastern Aomori and southwestern Hokkaido, Ph.D. thesis, University of Toronto. D’Andrea, A. C., Crawford, G. W., Yoshizaki, M. and Kudo, T. 1995 Late Jomon Cultigens in Northeastern Japan, Antiquity, 69, pp.146152. Gould, F. W., Ali, M. A. and Fairbrothers, D. E. 1972 A Revision of Echinochloa in the United States, The American Midland
Naturalist, 871, pp.3659. Hirai, Y. and Watanabe, T. et al. 1995 Minamimizote site I, Okayama Bunkazai Hogo Kyokai (in Japanese) Hirose, K. 1997 New Historical Images of the Jomon and the Yayoi, Kadokawashoten (in Japanese). Holm, L. G., Plucknett, D. L. Pancho, J. V. and Herberger, J. P. 1977 The World’s Worst Weeds, The University Press of Hawaii. Kimura, S. 1996 Cultivating Fields and the Japanese, Chuokoronsha (in Japanese). Kobayashi, K. 2008 Calendar year of the Jomon period, Archaeology of Jomon Period 2, pp.257269, Doseisha (in Japanese). Kodainomorikenkyusha 2006 Analysis of macro plant remains, Chikano Iseki IX, The Education Board of Aomori Prefecture, pp.297321 (in Japanese). Matsuda, O. 1980 Dictionary of Plants in Classical Literatures, Kodansha (in Japanese). Matsui, A. and Kanehara, M. 2006 The question of prehistoric plant husbandry during the Jomon Period in Japan, World
Archaeology, 382, pp.259273. Matsutani, A. 2006 SEM identification of charred material in a potsherd from Ryugasaki A site, Ryugasaki A site, pp.1839， Education board of Shiga Prefecture and Association for Protection of Culture Property of Shiga Prefecture (in Japanese).
21
Michael, P. 1994 Distribution and taxonomy of Echinochloa: a world view with a key to the species occurring in China, Proc. 5th
Weed Science Conf. of China, pp.161166. Miyamoto, K. 2000 Agriculture in Jomon and Jomon society, Sahara, M and H. Tsude eds. Issues of the Ancient History:
Environmental and Food Production, Shogakukan (in Japanese). Miyamoto, K. 2003 The emergence of agriculture on the Neolithic Korean Peninsula and Jomon Agriculture, Cultura Antiqua, 557, pp.116 (in Japanese). Miyamoto, K. 2005 History of China 01, Kodansha (in Japanese). Miyamoto, K. 2007 Rice cultivation culture in China and the Korean Peninsula and origin of Yayoi, Hirose, K. ed. Yayoi Jidai Wa
Dou Kawaruka, pp.7792, Gakuseisha (in Japanese). Morita, H. 2002 Rice cultivation and millet in Japan, Yamaguchi, H. ed. Hie To Iu Shokubutsu, pp.4968, Zenkoku Noson Kyoiku Kyokai (in Japanese). Nakayama, Y. 2009 The Adzuki beans (Vigna angularis) in the Jomon period, Abstracts Volume of 24th Annual Meeting, Japanese Association of Historical Botany, pp.4445 (in Japanese). Nakayama, Y., Umemoto, S., Yamaguchi, H. 1999 Identification of polyploidy groups in the Genus Echinochloa by izozyme analysis,
Zassou Kenkyu (Journal of Weed Science and Technology), 44, pp.205217 (in Japanese with English summary). Nakazawa, M. 2009 On arguments of Jomon agriculture, Archaeology of Yayoi Period 6, pp.228246, Doseisha (in Japanese). Nakazawa, M. and Ushino, T. 1998 Observation of ricelike impressions on potsherds of the Final Jomon using replication method,
Jomon Jidai, 9, pp.128 (in Japanese). Nakazawa, M. Ushino, T. and Matsutani, A. 2002 On impression of barley on potsherd from Nakamichi site Narasaki city, Yamanashi Prefecture: observation of ricelike impression of the Final Jomon (2), Kodai, 111, pp.6383 (in Japanese). Nozawa, S., Nakai, H. Sato, Y. 2004 Characterization of microsatellite and ISSR polymorphisms among Echinochloa (L.) Beauv. spp. in Japan, Ikushugaku Kenkyu, 6, pp.187193 (in Japanese with English Summary). Nishi, K. 2002 Natural history of Himetainubie (Echinochloa crusgalli Beauv. var. formosensis Ohwi), Yamaguchi, H. ed. Hie To Iu
Shokubutsu, pp.7996, Zenkoku Noson Kyoiku Kyokai (in Japanese). Nishimoto, T., Mira, K., Sumita, M. and Miyata, Y. 2007 On the cultivated barnyardgrass (Echinochloa crusgalli) in the Middle Period of Jomon Culture, Dobutsu Kokogaku, 24, pp.8588 (in Japanese). Nishiaki, T. 2000 Jomon Cold Frames?, Hokkaido Kokogaku, 36, pp.2946 (in Japanese). Obata, H. 2008 Agriculture and foods in NE Asia as seen from seeds, Prehistoric and Ancient Cultigens in the Far East 3, pp.183211, Kumamoto University (in Japanese). Sakamoto, S. 1988 Road of millets, NHK Press (in Japanese). Sergysheva, E. A. 2005a Domesticated plants from archaeological sites of the Primorye on basis of paleoethnobotanical data,
Prehistoric and Ancient Cultigens in the Far East, Kumamoto University (in Russian). Sergysheva, E. A. 2005b Paleoethnobotanical remains, Zaisanovka 7, pp.6568, Kumamoto University (in Russian). Sasaki, Y., M. Nakazawa, HI. Nasu, K. Yoneda, R. Koizumi 2009 Identification of Setaria impressions in potsherds of the end of the final Jomon to the early Yayoi periods recovered from the Nakayashiki site, Kanagawa, and the Ishigo site, Nagano, Abstracts
22
Volume of 24th Annual Meeting, Japanese Association of Historical Botany, pp.4849 (in Japanese). Showa women’s university excavation team of Nakayashiki site 2008 Report of archaeological investigation at Suginodo site, Ohyi
town, Kanagawa Prefecture, Rokuichishobo (in Japanese). Takase, K. 2004 Charred seeds from Suginodo site, Mizusawa city, Iwate Pref., Nakamura , N. ed. Report of archaeological
investigation at Suginodo site, Research Institute for Cultural Promotion of Iwate Prefecture, pp.161170 (in Japanese). Takase, K. 2005 Charred seeds from Tanakateramae site, Komae city, Tokyo, Report of archaeological investigation at
Tanakateramae site, Komae city, Tokyo, pp.117119, Josaishoji and Kato Kensetsu (in Japanese). Takase, K. 2006a Changes and characteristics of domesticated plant use in northeastern Japan, (A Report of GrantsinAid for Scientific Research). Takase, K. 2006b Charred seeds from the 25th investigation at KumadoB site, Morioka city, Iwate Prefecture, Report of the 25th
archaeological investigation at KumadoB site, Morioka city, Research Institute for Cultural Promotion of Iwate Prefecture (in Japanese). Takase, K. 2007 Seeds from Jomon and Yayoi sites in northeastern Honshu, Japanese Archaeological Association Meeting 2007
Kumamoto Abstracts, pp.398408 (in Japanese). Takase, K. 2009 Examination of millet farming and utilization of nuts in Yayoi period, Archaeology of Yayoi Period 6, pp.213227, Doseisha. (in Japanese). Takase, K, and Endo, E. n.d. Analysis of charred rice and foxtail millet from Tagoyama site, Shiki city, Saitama Prefecture,
KodaigakuKenkyujo Kiyo , 12, Meiji University. Takahashi, M. 1992 Pottery with impression of rice in Jomon, Kokogaku Journal, 355, pp.1517 (in Japanese). Takahashi, O. 2000 Subsistence in Hokkaido during Holocene, Hokkaido Kokogaku, 36, pp.113 (in Japanese). Tsubakisaka, Y. 1993 Identification of foxtail, barnyard and bloomcorn millet, Senshigaku to Kanren Kagaku, pp.261281 (in Japanese). Yabuno, T., 1962, Cytotaxonomic Studies on the Two Cultivated Species and the Wild Relatives in the Genus Echinochloa, Cytologia, 27, pp.296305. Yabuno, T. 1975 The classification and geographical distribution of the genus Echinochloa, Zassou Kenkyu, 20, pp.97104 (in Japanese). Yabuno, T. 1981 The classification and geographical distribution of the genus Echinochloa, Shuseibutsugaku Kenkyu, 5, pp.97104 (in Japanese). Yabuno, T. 1983 Biology of Echinochloa Species, Proceedings of the Conference on Weed Control in Rice, pp.307318. Yamaguchi, H. and Ohe, M. 2001 Basic morphological characteristics and scientific names of Echinochloa, Yamaguchi, H. ed., Hie To
Iu Shokubutsu, pp.3148, Zenkoku Noson Kyoiku Kyokai (in Japanese). Yamaguchi, H. et al. 2007 Paleoethnobotanical study of Hiezake of the Ainu, Report of study aided by the foundation for research and promotion of Ainu culture (in Japanese). Yamazaki, S. 2007 The origin of Yayoi culture, Hirose, K. ed. Yayoi Jidai Wa Dou Kawaruka, pp.97110, Gakuseisha (in Japanese). Yamada, G. 1994 Cultivated plants occurred from the Primorye region: especially on buckwheat and barley, Preliminary Reports on
23
“Research Project of the Historical and Cultural Exchange of the North” in 1993，pp.2950，Historical Museum of Hokkaido (in Japanese).． Yamada, G. and Tsubakisaka, Y. 2006 Barnyard millet, foxtail millet and Japanese barnyard millet from archaeological sites in Hokkaido, Obata, H. ed. Cultivated cereals in prehistoric and ancient Far East Asia 2, pp.1526, University of Kumamoto (in Japanese). Yokoyama, E. 2003 Traces of barnyard in Hokkaido, Bussitsu Bunka, 75, pp.113 (in Japanese). Yoshizaki, M. 1992 Detecting ancient millets: an archaeobotanical investigation, Kokogaku Journal, 355, pp.214 (in Japanese). Yoshizaki, M. 1993 A study on diffusion of technology of ancient agriculture in Tohoku and Hokkaido using flotation technique, University of Hokkaido (in Japanese). Yoshizaki, M. 1997 Cultivation plants of the Jomon Period, Dai Yonki Kenkyu (The Quaternary Research), 36, pp.343346 (in Japanese). Yoshizaki, M. 2003 Prehistoric millets: archaeobotany of Echinochloa and adzuki bean, Yamaguchi, H. and Kawasaki, M. eds.,
Natural history of millets: in pursuit of the origin and the culture, pp.5270, Hokkaido University Press (in Japanese). Yoshizaki, M. and Tsubakisaka, Y. 1993 Charred plant remains of the middle stage of Jomon from Tominosawa (2) site in Aomori Prefecture, Tominosawa (2) IV, pp.10971107, The education board of Aomori Prefecture (in Japanese). Yoshizaki, M. and Tsubakisaka, Y. 1996 Plant remains of the initial stage of Jomon from Nakano B site in Hokkaido, Nakano B site
(Part 4), pp.304313, Hokkaido Maizobunkazai Center (in Japanese). Yoshizaki, M. and Tsubakisaka, Y. 1997 Plant remains from K39 site, loc. Ohki, K39 site, pp.106110, loc. Ohki, Educational Board of Sapporo city. Yoshizaki, M. and Tsubakisaka, Y. 1998 Problems on plant remains and Genus Echinochloa seeds from Fugoppe Shell Mound,
Fugoppe Kailua, pp.535547, Hokkaido Maizobunkazai Center (in Japanese). Yoshizaki, M. and Tsubakisaka, Y. 2001 On beans from prehistoric sites, Journal of Bean, 24, pp.19 (in Japanese).
24
Prehistoric and Protohistoric Plant Use in the Japanese Archipelago Katsunori Takase （Faculty of Letters, Meiji University）
Introduction
This lecture outlines prehistoric and protohistoric plant use in the Japanese archipelago based on the latest information of plant archaeology. Slide 1 pictures useful plants which are examined today. I am focusing on crops of grass plants and beans including rice, broomcorn millet, foxtail millet, barnyard
millet,
barley,
wheat,
adzuki
and
soybeans. Slide 2 summarizes chronological and regional settings. I cannot talk about regional differences in detail due to time constraints. However, I would like to focus on diachronic change and periods of occurrence of useful plants. Please refer to the chronological table (Slide 2) as needed. Calendar years on this table are based on results of recent AMS radiocarbon dating and dendrochronology. In a framework of Japanese archaeology, prehistory means before the Jomon period, and protohistory means the Yayoi and the first half of the Kofun periods. Agriculture was the core economy of ancient Japan and paddy rice was the most important grain. Rice cultivation was the basis of the taxation system and rice has been continuously a major crop down to modern times in Japan. At the same time, however, it is notable that the ancient polity tried to increase 1
productivity
of
other
grains.
For
example, Prof. Shigemitsu Kimura has pointed out that there were three phases of policy concerning millets in classical Japan (Slide 3). In the first phase, until 820 C.E., foxtail millet was recommended. And in the second phase, barley
and
wheat
were
strongly
recommended. Then, after 920 C.E., various millets were recommended for cultivation. As is well known, various goods were paid as tribute to the ancient court. Millets and other useful plants were included. This table (Slide 4) shows various millets and useful plants that were cultivated in classical Japan. Cultivation not only of rice but also of some kinds of millet is evidenced by literary texts. According to Prof. Osamu Matsuda, about 160 species of plants are referred to in the Manyoshu, Kojiki,
Nihonshoki,
Fudoki,
Kokinwakashu,
Makuranososhi, and the Tale of Genji. Matsuda has noted that in total there are 38 useful plants mentioned in these texts,
with
representative
grains
including rice, barley, wheat, broomcorn millet, foxtail millet, and barnyard millet. Archaeological
records
further
confirm his argument. Many charred seeds of domesticated plants have been excavated
from
sites.
These
are
2
examples
of
carbonized
seeds
from
settlements dating from the 7th to the 9th centuries in eastern and northern Japan. Rice is usually dominant in a complex of domesticated plant seeds, but in the case of sites located in inland and hilly areas, seeds of millets and beans occasionally exceed rice seeds, as shown in Slide 5. There the picture on the left shows charred seeds from a settlement of the 7th century and located in the mountainous locale of western Tokyo. Foxtail millet is dominant at this site. The picture on the right shows seeds from Suginodo site in northern Japan. This site is a large settlement of the 8th to the 9th centuries. In this site, rice was the dominant crop; however, many kinds of millets have also been found there. Slide 6 shows additional seeds from the
Suginodo
site.
Charred
acorns
indicate that wild tree nuts were also used with domesticated plants. And I could also identify adzuki beans by observing an organ known as the primary leaf, a point on which I will elaborate later. Slide
7
also
shows
seeds
from
settlements of the 9th to the 10th centuries, and therein barley, beans, and acorns could be identified along with rice. To sum up, not only rice but also various millets and beans have been uncovered at ancient sites, even in 3
northern and eastern Japan.
Current
Status
of
Study
on
Prehistoric Plant Use in the Japanese archipelago
It is still difficult to know when and how these grains came to be used in the Japanese archipelago. There are two current hypotheses as to when domesticated plants such as rice and major millets came into use. The first hypothesis is that they were introduced before the Late Jomon. For example, authors of these highquality overviews (Slide 8) often argue that rice and millets were utilized before the Late Jomon, although farming at that time may not have played an important role in the economy. These books are very influential because they are useful to readers desirous of catching up on the latest knowledge of archaeology or paleoethnobotany. Therefore, this hypothesis has rapidly taken root in Japan during the last fifteen years. A second hypothesis is that rice and major millets were utilized after the second half of the Final Jomon or by the beginning of the Yayoi period. Its advocates take a cautious stance on the previous hypothesis. For example, they use AMS radiocarbon dating of charred seeds proactively to evaluate the reliability of plant remains. In addition, a new methodology known as the Replica SEM Method, which permits observation of seed impressions on potsherds, has afforded important results (Slide 9). In this method, replica of impressions on fragments of pottery are made using silicone rubber. Then observation of the replica with a scanning electron microscope enables much more accurate recognition of the character of the impression than does general microscopic observation. I will elaborate further on these achievements later. But first I would like to present an abstract of the diffusion process of farming in East Asia. Recent archaeological investigations in China have demonstrated that wild rice was first utilized in the Yangtze River basin prior to 9,000 or 10,000 years ago. After 7,000 years ago, domesticated rice became dominant, as shown by charred rice caryopses from archaeological sites in the Yangtze River basin. Slide 10 indicates the distribution of rice cultivation in each period. In contrast, millet cultivation also occurred before 8,000 or 7,000 years ago in the Yellow River basin. Much later, rice cultivation had diffused to the Shandon Peninsula by 2,500 B.C.E. from the south. In this region, wet rice cultivation was combined with millet farming. And still later the 4
technology of wet rice cultivation on the Korean
Peninsula
was
strongly
influenced by that on the Shandong Peninsula,
with
the
archipelago
also
being
Japanese affected
indirectly. Specifically, on the Korean Peninsula
millet
farming
and
rice
cultivation had already taken set form by 1,500 B.C.E. That makes it probable, as some archaeologists have expected, that rice and millet cultivation was introduced to the Japanese archipelago before the Late Jomon, dated to before 1,200 B.C.E. However to gain a more specific sense of when various grains were introduced to the Japanese archipelago, it is important to examine the archaeological data exactly. I turn to that task now.
Rice
In
last
decade
or
two
some
archaeologists have argued that rice cultivation goes back before the Middle Jomon. For example, Prof. Kazuo Hirose, Prof.
Kazuo
Miyamoto,
and
Prof.
Takehiko Matsuki have all stated in their books
that habitants
of the
Japanese archipelago started to use rice before the Middle or the Late Jomon. They
emphasize
the
existence
of
phytolith data and impressions of rice on pottery from before the Late Jomon. And Prof. Kazuo Miyamoto has opined 5
as well that charred seeds of rice indicate cultivation
before
the
Late
Jomon,
although the number of rice seeds found remains very small. Nevertheless there is always danger of contamination by tiny phytolith like plant opal in sediments of archaeological sites (Slide 11). To avoid this problem, fragments of pottery have been used as samples for plant opal analysis. If plant opal can be found in the pottery, its date is thought attributable to the date of the pottery. And Prof. Mamoru Takahashi has in fact reported that the plant opal of rice has been found in pottery dating from before the Middle Jomon. However, it is also important to recognize that phytolith is enough small to fly from the Chinese continent to the Japanese archipelago with yellow dust. As shown in Slide 11, a wide area of East Asia
is
covered
with
yellow
dust,
especially in spring. Yellow dust reaches as far as the Hawaiian Islands and sometimes can fly as far as Alaska. So it is likely that a small amount of rice opal is
found
in
Jomonperiod
sediment
simply because wild and domesticated rice existed on the Chinese continent at that time. Phytolith, of course, provides efficient data
when
they
are
quantitativelydetected. If there are more than five thousand plant opals of rice in 1g sediment from a site, it is possible to 6
estimate that there was a paddy rice field there.
However,
small
numbers
of
phytolith cannot be reliable evidence for proving the cultivation of rice in the Japanese archipelago. Then, what about charred seeds from archaeological sites? This table (Slide 12) shows the number of carbonized grass plant seeds found in sites from the Initial Jomon to the Middle Jomon. In this phase, only the genus Echinochloa—kinds of barnyard millet—was found in northern Japan. Data for barley is not reliable because the number of specimens is very small. Slide 13 shows a table summarizing the number of seeds from the Late Jomon to the Final Jomon. In this phase, the number of rice seeds increases. However, these
cases
still
seem
accidental,
considering the number of sites excavated in a year in Japan. In addition, it is notable that there is no example definitely dated to this period by physicochemical dating. Finally, Slide 14 includes a table showing the number of charred seeds from the first half of the Yayoi period. From this phase seeds of rice and millets can be found repeatedly in various sites in all areas except Hokkaido and Ryukyu. As these tables show, the number of rice seeds increase rapidly after the Yayoi period begins. Although more than two hundred charred seeds of rice have been 7
reported from the Late and Final Jomon, it is still hard to affirm the existence of rice cultivation on the basis of seeds. There is also danger of contamination by carbonized seeds because they move quite easily and naturally between sedimentary levels. And paddy rice fields have not yet been found before the Final Jomon. At present, carbonized seeds indicate that
fullscale
rice
cultivation
was
introduced after the end of the Final Jomon or during the Early Yayoi. This perspective is supported by paddy rice fields and implements of farming. Kazahari site in northern Japan is the only site where the radiocarbon date of carbonized seeds of rice was measured. The results were 2540±240BP and 2810± 270BP (Slide 15). These specimens were measured
in
the
early
1990’s
in
a
laboratory at the University of Toronto, and they represent one of the earliest examples of AMS radiocarbon dating of charred
seed
in
Japan.
Therefore,
calibrated ages have a wide range— from the 17th century B.C.E. to the 1st century B.C.E. These results indicate that charred seeds of rice from Kazahari cannot be indisputable evidence showing the existence of rice cultivation before the Late Jomon, because they might be assigned to the Yayoi period although more accurate periodization is not possible now. Present technology of AMS radiocarbon dating can provide results with a higher degree of accuracy. In recent years, the National Museum of Japanese History tried to determine the date of one specimen of charred rice from the same Kazahari site. Unfortunately, the result was 173±35BP, indicating a distinct case of contamination by modern rice. At present, there are no charred seeds of rice dated with certainty before the Late Jomon by physicochemical analysis. Researchers must also be quite careful in dating seed impressions on pottery because it is not easy to determine the chronology of the pottery. Slide 16 shows a fragment of a vessel with rice 8
impressions from Minamimizote site, in the ChugokuShikoku region. In popular books of Japanese archaeology, many authors have stated that this find constitutes reliable evidence of rice cultivation in the Late Jomon. But lately the fragment has been joined with others and restored as a large pot of the second half of the Final Jomon phase. It is not therefore evidence demonstrating rice cultivation in the Late Jomon. Next, let me introduce recent achievements of ReplicaSEM Method. Mr. Michihiko Nakazawa and Prof. Tsuyoshi Ushino have examined riceshaped impressions on pottery intensively using this method since 1998 (Slides 17, 18). As a result, it was revealed that ten of the eleven impressions were not made by rice, although formerly all impressions had been thought to be those of rice. Almost all of the impressions were made by small fragments of wood or unknown seeds of weed. It was an amazing result. In fact there is only a single reliable impression made by rice: that on a fragment of pottery from Ishigo site in the KantoKoshin’etsu region, which has been assigned to the second half of the Final Jomon (Slide 19). Using this method, researchers have continued to detect reliable data of the oldest rice in each region. This table (Slide 20) shows a chronological sequence of pottery; a black circular mark means the occurrence of reliable rice impressions. At present, such impressions never predate the first half of the Final Jomon in the western part of the Japanese archipelago. And in the Kanto and Tohoku regions, the number of impressions of rice suddenly increases after the Early Yayoi. And while I have searched intensively for impressions of seeds in the Kanto and Tohoku regions, I have found no reliable rice impressions before the Final Jomon. On the other hand I have already found more than 60 impressions of rice on pottery vessels dating from the Early and Middle Yayoi (Slide 21). As expected, the Yayoi period has proven to be an important epoch for rice cultivation in the eastern part of the archipelago. To sum up, there is no archaeological record showing the existence of rice before the Late Jomon. And according to seed impressions on pottery, the oldest rice seeds date from after the second half of the Final Jomon, or during the Early Yayoi.
An Insect Pest, the Maize Weevil
Another important problem highlighted by ReplicaSEM Method is that of an insect pest. Mr. Sumio Yamazaki has reported that 9
impressions of the maize weevil go back to at least the Late Jomon (Slide 22). He has also opined that cultivation of rice must also date back to the Late Jomon, since the maize weevil has a close relationship with rice. On the other hand, Prof. Hiromichi Ando asserts that this theory needs cautious examination because no one knows how the maize weevil lived before the introduction of domesticated crops in Japan, even if it is now seen as a rice
pest.
Meanwhile
pestology
has
studied the insect to learn how to prevent its destructive activities, but we do not yet know its longterm life history. Pestological study (Slide 23) has already clarified that the food of the maize weevil is restricted to rice，barley，wheat, and maize. They grow from egg to pula in one grain and therefore they need grains longer than their body length to become adult. Their reproduction also requires threshing because the tube for bearing eggs cannot penetrate the husks of the grain. Pestology shows that the maize weevil might be related to rice. Therefore, the presence of this insect may suggest utilization of rice before the Late Jomon. However, there is no entomological and pestological data as to whether they eat other kinds of nuts or grains, or as to conditions of reproduction. Archaeologists have to investigate this problem. Slide 24 shows results of a breeding experiment for maize weevils conducted by Prof. Hiroki Obata and Ms. Yasuko Senba. It looks like a crazy experiment, but I think it is important. According to this study, the maize weevil eats threshed rice, barley, wheat, broomcorn millet and acorn. But survival into a second generation occurred only in the case of threshed rice, barley, and wheat. If so, the existence of the maize weevil may well suggest the use of rice. Nevertheless, one more update needs noting: Prof. Obata recently stated that maize weevils fed with acorn were still alive although he once thought that they had all died. Our studies of the maize weevil and its significance for rice are 10
right back where they started.
Foxtail Millet
Until the 1970’s, some archaeologists believed that foxtail millet was cultivated during
the
Middle
KantoKoshin’etsu
Jomon region.
in
the Tiny
roundshaped charred seeds from sites of this period had been regarded as foxtail millet for a long time. However, Prof. Akiko Matsutani revealed that they were seeds of the beefsteak plant, also known as perilla. At present, the reliable data for foxtail millet is still very small. However, positive
evidence
has
been
reported
recently. This table (Slide 25) shows the number of seed from the Nakayashiki site of Kanto region, which is dated as the Early Yayoi. At this site some pit features including a lot of charred seeds of rice, millets, and horse chestnut were excavated. AMS radiocarbon dates of rice are consistent with remarks of investigation (Slide 26). Foxtail millet should therefore be assigned to the Early Yayoi. Furthermore last month Ms. Yuka Sasaki and others reported that plural impressions of foxtail millet could be seen on fragments of pottery from this site. In addition, many charred seeds of foxtail millet were found in a field cultivated during the Early Yayoi at ShoKuramoto site located on a campus of Tokushima 11
University,
ChugokuShikoku
region.
Current data seems to suggest that the Early Yayoi was the time when foxtail millet came to be utilized. Another reliable example of foxtail millet was excavated at Tagoyama site, Kanto region (Slide 27). This is a dwelling pit of the Late Yayoi. Carbonized wood and heataltered soil suggest that this house was burned down. In areas A and B, many carbonized seeds of rice, millet, and beans were found around the floor of the dwelling pit. The picture on the right strongly suggests that ears of rice were charred for some reason because rice husks face in the same direction in area A. In area B, on the other hand, many seeds of foxtail millet and beans were recovered by the waterflotation technique as shown in this table (Slide 28). In total, about two hundred thousand seeds of foxtail millet were found in this dwelling pit (Slide 28). This is the largest number of foxtail millet seeds from any single pre or protohistoric site in Japan, and one of the most reliable examples of foxtail millet use in the Yayoi period. Recently,
we
also
examined
AMS
radiocarbon dating of rice and foxtail millet from this site. Specimens are one caryopsis
rice
for
AMS
dating,
five
caryopses of rice for stable isotope analysis of
carbon
and
nitrogen,
and
sixty
caryopses of foxtail millet for AMS dating and stable isotope analysis. The results of 12
radiocarbon dating, seen in Slide 29, demonstrate that the rice and foxtail millet date from the 1st century C.E. to the 3rd century C.E. It is therefore safe to say that foxtail millet from this site belongs to the Yayoi period. In addition, stable isotope analysis of carbon
and
nitrogen
support
our
identification of the seeds. Values of rice fall into an area of C3 plant, while result of foxtail millet was plotted in an area of C4 plant (Slide 30). In conclusion, use of foxtail millet began in the Japanse archipelago from the Early Yayoi. And specimens from Tagoyama site also provide reliable data for the Late Yayoi. Foxtail millet of Yayoi period provided a basis for intensive utilization after the Kofun period and in classical times, although there is still room for investigating the situation during the Middle Yayoi.
Broomcorn Millet
The oldest reliable data for broomcorn millet was found at the Nakayashiki site—dating
from
the
Early
Yayoi—mentioned above. In addition, other
reliable
data
comes
from
RyugasakiA site in the Kinki region. A lump of charred seeds adhered to the inside wall of a pottery vessel dating from the end of the Final Jomon or from the Early Yayoi. Prof. Akiko Matsutani 13
identified these seeds as broomcorn millet by tissue observation using SEM (Slide 31). Results of AMS radiocarbon dating are consistent to chronological estimates of the pottery type. These specimens suggest that the end of the Final Jomon or the Early Yayoi is an epoch of utilization of broomcorn millet, as was the case for rice and foxtail millet. After the beginning of the Kofun period,
considerable
numbers
of
broomcorn millet seeds appear, especially at settlements on hilly areas. However, little is known about broomcorn millet in the Middle and the Late Yayoi at this time.
Barnyard Millet
Echinochloa, types of barnyard millet should be examined in detail because some archaeologists have presented a hypothesis
that
domesticated
it
in
was the
firstly Japanese
archipelago.
Echinochloa species of Asia fall into three main groups of biosystematics as shown in Slide 32. The first is the phylogenesis
Echinochloa
utilis,
or
domesticated barnyard millet, which has been mainly cultivated in East Asia. Botanical constitution
study has
of
its
genome
demonstrated
that
domesticated barnyard millet originated from Echinochloa crusgalli. Meanwhile 14
archaeology in East Asia has revealed that millets in the first group came into use more than several thousand years ago. I would like to focus here on the use of
Echinochloa crusgalli and barnyard millet (Echinochloa utilis). In the 1980’s, Prof. Gary Crawford revealed
by
using
the
waterflotation
technique that Echinochloa was used in the Early and the Middle Jomon of northern Japan. After that, use of Echinochloa in Jomon has been confirmed in many sites by researchers
conducting largescale and
systematic archaeobotanical studies in the Japanese archipelago. Recent AMS radiocarbon dating has also demonstrated that Echinochloa use extends back to the Middle Jomon as shown in this table (Slide 33). The results suggest that these specimens go back at least to the 3rd millennium B.C.E. In addition, hundreds of Echinochloa seeds assigned to the Initial and the Early Jomon have been reported (ex. Slide 34). However, their dates had not yet been crosschecked by radiocarbon dating. Let
me
Echinochloa
examine seed
distribution from
pre
of and
protohistoric sites. This map shows the frequency of charred seed of Echinochloa (Slide 35). Black sections in the circle graphs indicate the ratio of Echinochloa seed to all grass plant seeds. The figure at left is based on data that pertains to artifacts before the Middle Jomon. The 15
calibrated radiocarbon age is estimated as earlier than about 2,400 B.C.E. Almost all the specimens of Echinochloa in this period are concentrated in northern Japan. The figure at the right shows the situation in the Late and the Final Jomon. Although the ratio of Echinochloa decreased sharply in the Tohoku region, it remained the dominant cereal in Hokkaido. This map (Slide 36) shows the case of Yayoi and the EpiJomon. As with the previous periods,
Echinochloa was a common millet on Hokkaido. In short, current archaeological data on Echinochloa from archaeological sites is mostly concentrated in northern Japan. I believe that Echinochloa has been utilized for a long time, since the Jomon period. In particular, Hokkaido is one of the regions where Echinochloa was used intensively and continuously. The morphology of seeds also provides an effective clue for examining past utilization of
Echinochloa. In 1983, Prof. Gary Crawford noted that the mean size of Echinochloa caryopses from the Middle Jomon is about 20 percent larger than specimens from the Early Jomon. In the 1990’s, Prof. Masakazu Yoshizaki recognized three types of Echinochloa caryopsis from the Jomon sites as shown in this figure (Slide 37). The first is a relatively small and slender type similar to Echinochloa crusgalli. The second is a relatively large and round type similar to domesticated barnyard millet. The last is an intermediate type with characteristics between the first and second types. Prof. Yoshizaki gave the nickname ‘Jomonhie’ to the second and the third types. Echinochloa seeds were found in the hearths of dwelling pits—they were not accompanied by traces of wild weeds and husks had been already removed. He also presumed that ‘Jomonhie’ increased after the Early and the Middle Jomon as a result of human husbandry. In Slide 38 we see SEM photographs of excavated barnyard millet in Hokkaido. From a macro perspective, it is possible to trace a morphological change in the seeds from Echinochloa crusgalli to barnyard millet via ‘Jomonhie’ in northern Japan. In 1988 Prof. Sadao Sakamoto presented his hypothesis that barnyard millet was domesticated in the northern part of the Japanese archipelago and diffused to a wide area of East Asia from this region. Nevertheless the domestication center of barnyard millet has not yet been thoroughly 16
explained. This graph (Slide 39) represents the length and breadth of Echinochloa caryopses from northern Japan. It is difficult to detect distinct morphological changes in specimens
from
Jomon
to
Yayoi
or
EpiJomon. Additionally seeds from the Satsumon period (dated from the 7th to the 12th century) tend to be rather smaller than those of previous periods. In contrast, significant changes occurred after the 13th century, signifying the appearance of real domesticated barnyard millet. If seed size became larger gradually, a possible interpretation is that barnyard millet was domesticated as a result of longterm use in this region. However, it changed drastically in a short period from the 12th to the 13th century. Therefore there is some possibility that barnyard millet was domesticated elsewhere and then imported to northern Japan after the 13th century. Notably Echinochloa
crusgalli is a ubiquitous wild weed in temperate regions. The wide natural geographical range of Echinochloa crusgalli makes it a suitable candidate for multiple domestication events in East Asia. Thus the relationship between ‘Jomonhie’ and barnyard millet should be carefully examined not only in the Japanese archipelago but also in the broader region of East Asia. In that regard, in Korea, however, there is currently no reliable data of Echinochloa before ThreeKingdom period, according to Prof. Ahn SungMo. But in recent years, Echinochloa seeds were found in Neolithic sites in the Russian Far East (Slide 40). Ms. Elena Sergusheva has reported that nonwild type Echinochloa seeds have been found in the specimens from Zaisanovka1, dated about 2,500 B.C.E. I think that this data indicates that further investigation of Echinochloa across East Asia is needed. For now, it is still difficult to determine the domestication center of barnyard millet.
Barley and Wheat
Until the last century, some archeologists argued that wheat was used in Yayoi period. However, many researchers now think that there is no reliable data in terms of seed remains for wheat before the Yayoi period. As for barley, Prof. Hiromich Ando reacted skeptically to views that it was used in Jomon and Yayoi periods because information on it is rare. However, an impression of barley was found by ReplicaSEM Method on a fragment of pottery from the end of the Final Jomon to the Early Yayoi. Slide 41 shows a SEM photograph of an impression of barley from Nakamichi site in the 17
KantoKoshjin’etsu region. I am therefore sure that barley was used in the Final Jomon or the Early Yayoi, but it is still difficult to estimate its significance as a major crop.
Adzuki Beans
Until the first half of the 1990’s, many archaeologists believed that the mug bean was used from the Early Jomon, based on result of excavation at the Torihama site, a famous Jomon site in the TokaiKinki region. However, it is difficult to identify mug beans by morphological observation because their shape is very similar to adzuki beans. They are both species classified in the genus Vigna. In 1992, Prof. Masakazu Yoshizaki and Ms. Yasuyo Tsubakisaka
developed
a
method
to
distinguish the mug bean from the adzuki bean by studying their primary leaves. These pictures (Slide 42, 43) show morphological differences between the primary leaves of the modern mug and adzuki bean. The primary leaf of a mug bean
is
relatively
large
and
hangs
vertically from the top of seed, while the small primary leaf comes down on a slant in the case of an adzuki bean. According to close examination using such a method, all Vigna beans from pre and
protohistory
of
the
Japanese
archipelago are adzuki beans. It is still difficult to conclude whether these adzuki 18
beans were domesticated or not; but at least, archaeology has begun to clarify longterm adzuki bean utilization in the Japanese archipelago.
Soybeans
Soybeans are attracting a great deal of attention in recent years. In 2007 Mr. Seiji Nakayama
and
others
reported
an
impression made by a bean on the jagged edge
of
a
Middle
Jomon
pot
from
Sakenomiba site, KantoKoshin’etsu region (Slide 44). This pot was found broken when it was excavated in 1997. Although its fragments
were
reconstructed
with
adhesive, a curator of the Yamanashi Prefectural Museum found that they did not adhere because secular changes of adhesive agent. At that time, a hole was discovered on broken surface of protruding part of the pot. On the opposite side of the broken fragment, the same impression was confirmed.
An
investigation
using
ReplicaSEM Method suggests that this hole was an impression made by a soybean. The soybean belongs to the genus
Glycine. Beans of the genus Glycine can be identifiable by existence of a hilum. Slide 45 shows a cross section of these beans. The hilum of beans in the genus Vigna and
Phaseplus, which include the species of kidney and navy beans, is covered with spongelike tissue known as the epihilum, seen here. Therefore it is impossible to see 19
the hilar groove in the center of the hilum. On the other hand, the hilar groove can be observed easily in the case of the genus
Glycine because there is no epihilum. Therefore, the exposed hilar groove of a replica recovered from Sakenomiba site makes it identifiable as a seed of the genus
Glycine (Slide 46). At present, Glycine seeds have been discovered from the Kyushu region to the Tohoku region from the Middle Jomon. Were these soybeans domesticated or not? Concerning this problem, further examination is needed. However, what is important is that the size of the Glycine seed is very large. Slide 47 shows a SEM photograph of two replica attached together to restore the entire shape of soybeans found at Sakenomiba. The length is 11.8 mm, longer than its original dry condition because it was mixed in pottery clay and contained water. A modern domesticated soybean also increases in size up to 11 or 12 mm after soaking in water. In contrast, a modern wild soybean expands to 8 or 9 mm after soaking. It is likely that Jomon soybeans were not wild soybeans. Another morphological characteristic of soybeans is thickness. The value of thickness of a Jomon soybean is usually less than 4mm. The thickness of a modern soybean soaked in water for six hours is 4 mm or greater. As for length, a Jomon soybean tends to be relatively flat. Such a flatshaped soybean is unique to the Japanese archipelago, and these are the oldest soybeans in East Asia. It is likely that the soybean was a highly controlled plant used by habitants in the Japanese archipelago in the Jomon period. While it is hard to say whether soybeans were domesticated in the Japanese archipelago, some archaeologists have begun to think that soybeans were domesticated in multiple places in East Asia, one of which was the Japanese archipelago.
Conclusion
In the Japanese archipelago, the occurrence of rice and major millets can be dated from the second half of the Final Jomon to the Early Yayoi. Although some scholars have insisted that rice and the major millets existed before the Late Jomon, current reliable data does not support this argument. At the same time, we can confirm that varieties of barnyard millet were used from the 3rd 20
millennium B.C.E., while husbandry of adzuki and soybeans seems to have been practiced to some degree at least after the 4th millennium B.C.E. In this phase, other useful plants like gourds, beefsteak plant/perilla, hemp plant, peach, burdock, oil seed, and rape were also used. Some archaeologists also think that forests of chestnut and horse chestnut were highly controlled around settlements in the Jomon period. And finally, the occurrence of wheat and buckwheat remains obscure, although they were popular crops in classical Japan.
References Akasaka, M. and Yamaguchi, H. 2006 Characterization of microsatellite of wild and domesticated E. oryzicola, Ikushugaku Kenkyu (Bleeding Science), 8, special issue1, p.241 (in Japanese). An SungMo 2006 Investigation of existence of the domestication of Echinochloa in the Ancient Korean Peninsula, Obata, H. ed.
Cultivated cereals in prehistoric and ancient Far East Asia 2, pp.3143, University of Kumamoto (in Japanese). Ando, H. 2006 A note on prehistoric plant remains and their impressions of potsherds, Nishisagami Koko, pp.112122 (in Japanese). Crawford, G.W. 1983 Paleoethnobotany of the Kameda Peninsula Jomon, University of Michigan. Crawford, G. W. 1992 Prehistoric Plant Domestication in East Asia, Cowan, C. W. and Watson, P. J. eds. The Origins of Agriculture:
International Perspective, pp.738, Smithsonian Institution. D’Andrea, A. C. 1992 Paleoethnobotany of later Jomon and Yayoi cultures of northeastern Japan: Northeastern Aomori and southwestern Hokkaido, Ph.D. thesis, University of Toronto. D’Andrea, A. C., Crawford, G. W., Yoshizaki, M. and Kudo, T. 1995 Late Jomon Cultigens in Northeastern Japan, Antiquity, 69, pp.146152. Gould, F. W., Ali, M. A. and Fairbrothers, D. E. 1972 A Revision of Echinochloa in the United States, The American Midland
Naturalist, 871, pp.3659. Hirai, Y. and Watanabe, T. et al. 1995 Minamimizote site I, Okayama Bunkazai Hogo Kyokai (in Japanese) Hirose, K. 1997 New Historical Images of the Jomon and the Yayoi, Kadokawashoten (in Japanese). Holm, L. G., Plucknett, D. L. Pancho, J. V. and Herberger, J. P. 1977 The World’s Worst Weeds, The University Press of Hawaii. Kimura, S. 1996 Cultivating Fields and the Japanese, Chuokoronsha (in Japanese). Kobayashi, K. 2008 Calendar year of the Jomon period, Archaeology of Jomon Period 2, pp.257269, Doseisha (in Japanese). Kodainomorikenkyusha 2006 Analysis of macro plant remains, Chikano Iseki IX, The Education Board of Aomori Prefecture, pp.297321 (in Japanese). Matsuda, O. 1980 Dictionary of Plants in Classical Literatures, Kodansha (in Japanese). Matsui, A. and Kanehara, M. 2006 The question of prehistoric plant husbandry during the Jomon Period in Japan, World
Archaeology, 382, pp.259273. Matsutani, A. 2006 SEM identification of charred material in a potsherd from Ryugasaki A site, Ryugasaki A site, pp.1839， Education board of Shiga Prefecture and Association for Protection of Culture Property of Shiga Prefecture (in Japanese).
21
Michael, P. 1994 Distribution and taxonomy of Echinochloa: a world view with a key to the species occurring in China, Proc. 5th
Weed Science Conf. of China, pp.161166. Miyamoto, K. 2000 Agriculture in Jomon and Jomon society, Sahara, M and H. Tsude eds. Issues of the Ancient History:
Environmental and Food Production, Shogakukan (in Japanese). Miyamoto, K. 2003 The emergence of agriculture on the Neolithic Korean Peninsula and Jomon Agriculture, Cultura Antiqua, 557, pp.116 (in Japanese). Miyamoto, K. 2005 History of China 01, Kodansha (in Japanese). Miyamoto, K. 2007 Rice cultivation culture in China and the Korean Peninsula and origin of Yayoi, Hirose, K. ed. Yayoi Jidai Wa
Dou Kawaruka, pp.7792, Gakuseisha (in Japanese). Morita, H. 2002 Rice cultivation and millet in Japan, Yamaguchi, H. ed. Hie To Iu Shokubutsu, pp.4968, Zenkoku Noson Kyoiku Kyokai (in Japanese). Nakayama, Y. 2009 The Adzuki beans (Vigna angularis) in the Jomon period, Abstracts Volume of 24th Annual Meeting, Japanese Association of Historical Botany, pp.4445 (in Japanese). Nakayama, Y., Umemoto, S., Yamaguchi, H. 1999 Identification of polyploidy groups in the Genus Echinochloa by izozyme analysis,
Zassou Kenkyu (Journal of Weed Science and Technology), 44, pp.205217 (in Japanese with English summary). Nakazawa, M. 2009 On arguments of Jomon agriculture, Archaeology of Yayoi Period 6, pp.228246, Doseisha (in Japanese). Nakazawa, M. and Ushino, T. 1998 Observation of ricelike impressions on potsherds of the Final Jomon using replication method,
Jomon Jidai, 9, pp.128 (in Japanese). Nakazawa, M. Ushino, T. and Matsutani, A. 2002 On impression of barley on potsherd from Nakamichi site Narasaki city, Yamanashi Prefecture: observation of ricelike impression of the Final Jomon (2), Kodai, 111, pp.6383 (in Japanese). Nozawa, S., Nakai, H. Sato, Y. 2004 Characterization of microsatellite and ISSR polymorphisms among Echinochloa (L.) Beauv. spp. in Japan, Ikushugaku Kenkyu, 6, pp.187193 (in Japanese with English Summary). Nishi, K. 2002 Natural history of Himetainubie (Echinochloa crusgalli Beauv. var. formosensis Ohwi), Yamaguchi, H. ed. Hie To Iu
Shokubutsu, pp.7996, Zenkoku Noson Kyoiku Kyokai (in Japanese). Nishimoto, T., Mira, K., Sumita, M. and Miyata, Y. 2007 On the cultivated barnyardgrass (Echinochloa crusgalli) in the Middle Period of Jomon Culture, Dobutsu Kokogaku, 24, pp.8588 (in Japanese). Nishiaki, T. 2000 Jomon Cold Frames?, Hokkaido Kokogaku, 36, pp.2946 (in Japanese). Obata, H. 2008 Agriculture and foods in NE Asia as seen from seeds, Prehistoric and Ancient Cultigens in the Far East 3, pp.183211, Kumamoto University (in Japanese). Sakamoto, S. 1988 Road of millets, NHK Press (in Japanese). Sergysheva, E. A. 2005a Domesticated plants from archaeological sites of the Primorye on basis of paleoethnobotanical data,
Prehistoric and Ancient Cultigens in the Far East, Kumamoto University (in Russian). Sergysheva, E. A. 2005b Paleoethnobotanical remains, Zaisanovka 7, pp.6568, Kumamoto University (in Russian). Sasaki, Y., M. Nakazawa, HI. Nasu, K. Yoneda, R. Koizumi 2009 Identification of Setaria impressions in potsherds of the end of the final Jomon to the early Yayoi periods recovered from the Nakayashiki site, Kanagawa, and the Ishigo site, Nagano, Abstracts
22
Volume of 24th Annual Meeting, Japanese Association of Historical Botany, pp.4849 (in Japanese). Showa women’s university excavation team of Nakayashiki site 2008 Report of archaeological investigation at Suginodo site, Ohyi
town, Kanagawa Prefecture, Rokuichishobo (in Japanese). Takase, K. 2004 Charred seeds from Suginodo site, Mizusawa city, Iwate Pref., Nakamura , N. ed. Report of archaeological
investigation at Suginodo site, Research Institute for Cultural Promotion of Iwate Prefecture, pp.161170 (in Japanese). Takase, K. 2005 Charred seeds from Tanakateramae site, Komae city, Tokyo, Report of archaeological investigation at
Tanakateramae site, Komae city, Tokyo, pp.117119, Josaishoji and Kato Kensetsu (in Japanese). Takase, K. 2006a Changes and characteristics of domesticated plant use in northeastern Japan, (A Report of GrantsinAid for Scientific Research). Takase, K. 2006b Charred seeds from the 25th investigation at KumadoB site, Morioka city, Iwate Prefecture, Report of the 25th
archaeological investigation at KumadoB site, Morioka city, Research Institute for Cultural Promotion of Iwate Prefecture (in Japanese). Takase, K. 2007 Seeds from Jomon and Yayoi sites in northeastern Honshu, Japanese Archaeological Association Meeting 2007
Kumamoto Abstracts, pp.398408 (in Japanese). Takase, K. 2009 Examination of millet farming and utilization of nuts in Yayoi period, Archaeology of Yayoi Period 6, pp.213227, Doseisha. (in Japanese). Takase, K, and Endo, E. n.d. Analysis of charred rice and foxtail millet from Tagoyama site, Shiki city, Saitama Prefecture,
KodaigakuKenkyujo Kiyo , 12, Meiji University. Takahashi, M. 1992 Pottery with impression of rice in Jomon, Kokogaku Journal, 355, pp.1517 (in Japanese). Takahashi, O. 2000 Subsistence in Hokkaido during Holocene, Hokkaido Kokogaku, 36, pp.113 (in Japanese). Tsubakisaka, Y. 1993 Identification of foxtail, barnyard and bloomcorn millet, Senshigaku to Kanren Kagaku, pp.261281 (in Japanese). Yabuno, T., 1962, Cytotaxonomic Studies on the Two Cultivated Species and the Wild Relatives in the Genus Echinochloa, Cytologia, 27, pp.296305. Yabuno, T. 1975 The classification and geographical distribution of the genus Echinochloa, Zassou Kenkyu, 20, pp.97104 (in Japanese). Yabuno, T. 1981 The classification and geographical distribution of the genus Echinochloa, Shuseibutsugaku Kenkyu, 5, pp.97104 (in Japanese). Yabuno, T. 1983 Biology of Echinochloa Species, Proceedings of the Conference on Weed Control in Rice, pp.307318. Yamaguchi, H. and Ohe, M. 2001 Basic morphological characteristics and scientific names of Echinochloa, Yamaguchi, H. ed., Hie To
Iu Shokubutsu, pp.3148, Zenkoku Noson Kyoiku Kyokai (in Japanese). Yamaguchi, H. et al. 2007 Paleoethnobotanical study of Hiezake of the Ainu, Report of study aided by the foundation for research and promotion of Ainu culture (in Japanese). Yamazaki, S. 2007 The origin of Yayoi culture, Hirose, K. ed. Yayoi Jidai Wa Dou Kawaruka, pp.97110, Gakuseisha (in Japanese). Yamada, G. 1994 Cultivated plants occurred from the Primorye region: especially on buckwheat and barley, Preliminary Reports on
23
“Research Project of the Historical and Cultural Exchange of the North” in 1993，pp.2950，Historical Museum of Hokkaido (in Japanese).． Yamada, G. and Tsubakisaka, Y. 2006 Barnyard millet, foxtail millet and Japanese barnyard millet from archaeological sites in Hokkaido, Obata, H. ed. Cultivated cereals in prehistoric and ancient Far East Asia 2, pp.1526, University of Kumamoto (in Japanese). Yokoyama, E. 2003 Traces of barnyard in Hokkaido, Bussitsu Bunka, 75, pp.113 (in Japanese). Yoshizaki, M. 1992 Detecting ancient millets: an archaeobotanical investigation, Kokogaku Journal, 355, pp.214 (in Japanese). Yoshizaki, M. 1993 A study on diffusion of technology of ancient agriculture in Tohoku and Hokkaido using flotation technique, University of Hokkaido (in Japanese). Yoshizaki, M. 1997 Cultivation plants of the Jomon Period, Dai Yonki Kenkyu (The Quaternary Research), 36, pp.343346 (in Japanese). Yoshizaki, M. 2003 Prehistoric millets: archaeobotany of Echinochloa and adzuki bean, Yamaguchi, H. and Kawasaki, M. eds.,
Natural history of millets: in pursuit of the origin and the culture, pp.5270, Hokkaido University Press (in Japanese). Yoshizaki, M. and Tsubakisaka, Y. 1993 Charred plant remains of the middle stage of Jomon from Tominosawa (2) site in Aomori Prefecture, Tominosawa (2) IV, pp.10971107, The education board of Aomori Prefecture (in Japanese). Yoshizaki, M. and Tsubakisaka, Y. 1996 Plant remains of the initial stage of Jomon from Nakano B site in Hokkaido, Nakano B site
(Part 4), pp.304313, Hokkaido Maizobunkazai Center (in Japanese). Yoshizaki, M. and Tsubakisaka, Y. 1997 Plant remains from K39 site, loc. Ohki, K39 site, pp.106110, loc. Ohki, Educational Board of Sapporo city. Yoshizaki, M. and Tsubakisaka, Y. 1998 Problems on plant remains and Genus Echinochloa seeds from Fugoppe Shell Mound,
Fugoppe Kailua, pp.535547, Hokkaido Maizobunkazai Center (in Japanese). Yoshizaki, M. and Tsubakisaka, Y. 2001 On beans from prehistoric sites, Journal of Bean, 24, pp.19 (in Japanese).
24
