PRESIDENTIAL ADDRESS DELIVERED AT THE AHNIVEESAEY MEETING OF THE ROYAL ANTHROPOLOGICAL INSTITUTE OF GREAT BRITAIN AND IRELAND. PROFESSOR WILLIAM GOWLAND, A.R.S.M., F.S.A., F.I.C., Professor of Metallurgy at the Royal School of Mines. THE BURIAL MOUUDS AND DOLMENS OF THE EARLY EMPERORS OF JAPAN. [WITH PLATES I -VIII.] PUBLISHED BY THE (gogaf (^nfDrojjofogtcaf Jneftfufe of (Breaf an^ Jrefanb* 3, HANOVER SQUARE, LONDON, W. PRESIDENTIAL ADDRESS DELIVERED AT THE AHHIVERSARV MEETZHD OF THE ROYAL ANTHROPOLOGICAL INSTITUTE or GREAT BRITAIN AND IRELAND, BY PHOFESSOR WILLIAM GOWLAND, A.R.S.M., F.S.A, F.I.C., Professor of Metallurgy at the Royal School of Mines, THE BURIAL MOUNDS AND DOLMENS OF THE EARLY EMPERORS OF JAPAN. [WITH PLATES I-VIIT.J PUBLISHED BY THE (Rogaf (^nfOropofogtcaf Jnsft'fufe of (Brcaf (J^nfatn an^ 3tefan^, 3, HANOVER SQUARE, LONDON, W. Digitized by the Internet Archive in 2016 https://archive.org/details/burialmoundsdolmOOgowl [ 10 ] ( 5 ) PKESIDENT’S ADDRESS. THE BURIAL MOUXDS AND DOLMENS OF THE EARLY EMPERORS OF JAPAN. By Professor William Gowland, A.R.S.M., F.S.A., F.I.C. [With Plates I-VIII.] We have but little exact knowledge of the mode in which the Japanese disposed of the bodies of their famous dead in the very earliest times. The somewhat vague statements of their ancient traditionary records would seem to point to burial or mere deposition on the summits of natural hills as their earliest practice, but the most ancient remains yet discovered have not been found in such localities, hut on the lower grounds bordering the plains, and on the plains themselves. These remains, which consist of bronze swords and arrowheads, personal ornaments of steatite, jasper, rock crystal, and other stones, and along with which no objects of iron occur, are generally found at but slight depths below the surface of the ground. It is impossible to say with absolute certainty whether they had or had not been originally covered with mounds of earth. If they had been so covered and the mounds were of only small dimensions, the action of long weathering or the agricultural operations of bygone ages would have amply sufficed to level and destroy them. The mass of evidence is in favour of the belief that low mounds had been erected over them and that the Japanese were a race of mound builders in very early times indeed. In China, mound-burial was practised at a very remote period, and, althougli implicit credence cannot be altogether given to the specific statements of the early writers relating to this matter, yet broadly considered their testimony doubtless contains some elements of truth. The first burial mound of which they give a record is the tomb of Hia How Kao, the date assigned to it being 1848 R.c.^ Several others, which I need not specify, of later centuries B.c. are also mentioned ; but, apart from these records, we have very weighty evidence in favour of the extreme antiquity of mound-burial in that country in the use of the ideograpli Jap. reading ryd, from very remote times to denote a burial mound, its original proper meaning being a high mount or peak. As the civilisation of China, even during these times, was probably not without influence on the tribes beyond ■ “On the Stone Figures at Chinese Tombs and the Offering of Living Sacrifices.” By W. F. Mayers. 1‘roc. N. China Branch of the Asiatic Society. March, 1878. G Presidential Address. [ 11 ] its frontiers, it is not impossible that the Japanese may have been mound builders before they migrated from their old home on the mainland. Whether this FIG. 1. — INTERIOR OF DOLMEN AT TSUK.VHARA (.SETTyU). (Reiirolhiccd Uii the covriesy of the Council of the Society of Antiquaries.) sup])nsitioii may be cori'ect or not, it is certain that the race practised mound Inirial, especially in the western parts of the islands they now occupy, several ceiitiuies before our era. [ 12 ] Presidential Address. 7 That the simple mounds preceded those that contain a rude stone chamber which we call a dolmen/ is also not open to doubt, for associated with them we find tfie rudest hand-made pottery, and neither this pottery nor the swords of bronze previously mentioned have ever been discovered in dolmens. Stone beads and ornaments and sometimes bronze arrow-heads are, however, found in dolmens, but then they occur along with weapons and objects of iron and beads of glass. The period of the dolmens is thus a continuation of that of the simple mounds. During the dolmen period, and certainly after it, the building of simple mounds still survived, but sarcophagi of wood, stone or terra-cotta, of which there are no traces in those of the earliest date, were then used in the burials. Excepting the earliest mounds, all others enclose either dolmens or sarco- phagi. Those containing dolmens are generally older than those containing sarcophagi ; there are, however, several examples in which both classes are undoubtedly contemporaneous. Burial mounds containing dolmens are very nuurerous in Japan, many hundreds are known to me. Of these I have carefully examined over four hundred. A detailed account of their various forms, distriljution and contents will be fomid in Archceologia, vol. Iv, p. 439 et seq? All dolmens, with one or two exceptions of late date, are constructed of rude unhewn blocks, often weathered boulders just as taken from the mountain sides; but in some localities where there is an outcrop of suitable rock, some of the stones seem to have been roughly quarried. A view of the interior of a typical dolmen is given in Eig. 1, and the exterior of a mound in Plate I. Usually the dolmen is covered by a simple conical, circular, or somewhat elongated, mound ; the mounds, however, with which we are specially concerned in this paper are of an entirely different form. They are generally known as misasagi or imperial burial mounds, and are of more than usual interest, as, so far as my knowledge goes, they are peculiar to Japan. From their form they may be not inaccurately termed “ double ” mounds although they never contain more than one dolmen. Fig. 2, which is drawn fi'om my surveys, represents a typical one in the neighbourhood of Nara (Yamato). Although it is of considerable size it is not one of the largest, yet I have selected it for description as it is in a better state of preservation tlian any others I have seen, and besides I was able to go upon it and make careful measurements — as it had not, until I called attention to it, received official recognition as an imperial toml), whereas in other cases this was prohibited, and I had then to make my observations from outside the moats. As seen in plan. Fig, 2, it appears to consist of a circular mound combined with ' The term “dolmen” is used in this paper in its broad or generic sense, and signifies a stone burial chamber, generally of rude megalithic stnictuie, larger than a cist, covered by a mound. ^ Gowland, The Dolmens and Burial Mounds in Japan. 8 Presidential Address. [13] another intermediate in form between a triangle and a square. But as there are no triangular mounds in Japan, and no symbolic use of the triangle until compara- tively late times, I think, the form may be regarded as a combination of a circular with a square mound. This curious shape is doubtless not without symbolic meaning, yet Japanese archaeologists have not been able to give any satisfactory interpretation of it. Elevation Scale of Peet O 50 100 200 3P0 roduced hy the courtesy oj the Council of the Society of Antiq^iaries.) The circular end rises in the form of a truncated cone forming a flat peak 113 feet in diameter at its summit, and this is always the highest part of these mounds. The square end, on the other hand, has no distinct peak, its summit lieing an inclined plane, also flat, about 215 feet long, rising at a gentle angle from tlie inner slope of the conical peak, and terminating in a long, straight edge, 90 feet long, at riglit angles to the middle line of the mound. [14] Presidential Address. 9 Some of these mounds, especially when seen from a distance, appear to have two peaks, and from this feature the name “ futa-go-yama ” or twin hills^ has been applied to them, but on examining them closely I have always found that there was only one original peak, and that the other had been formed by the excessive weathering of the narrower part of the square end. The word ^ Misasagi or Teiryo, is often applied to them as a specific name, but this is not strictly correct, as its meaning is merely “ imperial mausoleum,” and in that sense is used for all imperial tombs of whatever form, whether they are double or simple mounds.^ The burial, whether in a dolmen or sarcophagus, invariably took place in the circular end of the mound. In the square end no remains of any interment have ever been found, but on its surface fragments of ceremonial vessels sometimes occur, indicating that some of tlie funeral or subsequent rites were celebrated there. The chief dimensions of this mound are : — Total length of base ... Extreme length of square end Diameter of round end Height of conical peak Height of terminal edge of square end 674 feet. 425 „ 420 „ 65 „ 52| „ The exact relative proportions which these measurements bear one to another differ in nearly every one of these double mounds, but the proportions of length and breadth generally range from 1‘42 : 1 to f.’58 : 1. These mounds have usually terraced sides. In the example shown in the diagram there are two well-formed terraces completely encircling it. In some smaller mounds there is only one terrace or none, but in those of the largest size there are often three. At the re-entering angle on each side a projection about 55 feet broad, nov/ generally of irregular form, but probably originally semicircular, extends from the lowest terrace into the moat, and upon this there is generally a low, circular mound. The moat has an average breadth of 100 feet, and completely surrounds the mound. Nearly all these double mounds possess a wide moat, although some, from their position on sloping ground, are not surrounded by one. Others, as the huge mound of the Emperor Nintoku, in Izunii, and another, also of vast size, near Fujiidera (Kawachi), had two moats encircling them. Around the outer embank- ment of the moat of several of the larger mounds, small conical mounds are ranged at varying distances apart. A curious feature which they all possess is the rows of terra-cotta tubes termed “ hanivja” with which the borders of their summits, terraces and moats, are ' Other names by which they are popularly known are : Hyotan-yama =Hill resembling a bottle gourd. Samisen-dznka=^lom\d of the shape of a Japanese lute. Cha-usu-yania =Hill of the shape of a mill for grinding tea. 10 Presidential Address. [15] lined. One of these tubes is represented in Fig. 3. It is 1 foot 1 inch to 1 foot 3 inches in diameter, 1 foot 5 inches long, and to f inch thick, and is strengthened by three horizontal ridges encircling it. Two holes 2 inches in diameter are pierced in it opposite each other near its middle. In each row these tubes are set upright from 3 to 6 inches apart, and are almost completely buried in the earth, about an inch or so only being exposed. On this mound the row which encircles the entire summit is 8 feet, that on the upper terrace only 4 feet from the edge. On the lower terrace the tubes are exposed, and are being washed away by the water of the moat (Fig. 4). The total length of the rows on this mound, if placed in a straight line, would exceed 1^ miles, whilst the number of tubes at the lowest computation is not less than 4,740, exclusive of those on the embankment of the moat. It is difficult to determine, with absolute certainty, the exact intention of the early Japanese in using these rows of terra-cotta tubes. It may be that they w^ere placed in the positions we have seen for structural reasons, to aid in preserviug the form of the summit and terraces of the mound and FIG. 4 . — TERRA-COTTA TUIJES ON THE LOWER TERRACE OF AN IMPERIAL MOUND. NARA (YAMATO). FIG. 3 . — HANIWA. TERRA-COTTA TUBE. [16] Presidential Address. 11 the embankment of the moat from being destroyed by weathering, but, if so, it is not obvious why they were ever buried as far as 8 feet from tlie edge they were intended to protect. On the other hand they may liave been intended to represent the retainers, who, in earlier times, were immolated on the mound, but to this it may be objected that they would then have borne at least some rough resemblance to the human form, or some indications that they represented it. They occur, too, on mounds upon which rude terra-cotta human figures have been foimd along with them. Possibly there may be some truth in both suppositions. Opposite the square end and on the outer embankment of the moats of these imperial burial mounds is the “ kakusaku ” or sacred enclosure, where offerings are made to the manes of the deceased emperor on the anniversary of Ins death. This will be described later when describing the misasagi of the Emperor Iveitai. Frequent mention occurs in Japanese literature of the ancient custom of burying human beings and horses at the tombs of meniliers of the imperial family and of chieftains. The most important passages are in the JSfihongi.^ (2 B.o.) “ 28th year. Winter, 10th month, 5th year. Yaniato-hiko no Mikoto, the Emperor’s younger brother by the mother’s side, died. 11th month, 2nd day. Yamato-hiko was Iniried at Tsukizaka in Musa. Thereupon his personal attendants were assembled, and were all buried alive upright in the precinct of the misasagi. For several days they died not, but wept and wailed day and night. At last they died and rotted. Dogs and crows gathered and ate them.” The Emperor Suinin, hearing the sound of their weeping and wailing, was grieved in heart, and commanded his high officers, saying : ‘ It is a very painful thing to force those whom one has loved in life to follow him in death. Though it be an ancient custom, why follow it if it is bad ? From this time forward take council, so as to put a stop to the following of the dead.” (a.d. 3.) “ 32nd jmar. Autumn, 7th month, 6th day. The Emq)ress Hibasu- hime no IMikoto died. Some time before the burial, the Emperor commanded his ministers, saying: ‘ We have already recognised that the practice of following the dead is not good. M^liat should now be done in performing this burial ? ’ There- upon Xomi no Sukune came forward and said : ‘ It is not good to bury living men upright at the tumulus of a prince. How can such a practice l)e handed down to posterity ? I beg leave to propose an expedient which 1 will sul)mit to your iMajesty.’ So he sent messengers to summon up from the Land of Idzumo a hundred men of the clay workers Be. Fie himself directed the men of the clay workers Be to take clay and form therewith sha}>es of men, horses, and various objects which he presented to the Emperor, saying: ‘Henceforward let it be the law for future ages to substitute things of clay for living men, and to set them up at tumuli.’ Then the Emperor was greatly rejoiced, and commended Nomi no ’ Nihoiigi. Chronicles of Japan from the Earliest Times to a.d. G97. Completed, a.d. 720. Translation by W. G. Aston, C.M.G., vol. i, pp. 178-18], 12 Presidential Address. [17] Sukune, saying : ‘ Thy expedient hath greatly pleased our heart.’ So the things of clay were first set up at the tomb of Hib asu-hime no Mikoto. And a name was given to those clay objects. They were called Hani-wa (clay rings). Another name is Tatemono (things set up). “ Then a decree was issued, saying : ‘ Henceforth these clay figures must be set up at tumuli : let not men be harmed.’ “ The Emperor bountifully rewarded Nomi-no-Sukune for this service, and also bestowed on him a kneading place, and appointed him to the official cimrge of the clay workers Be.”^ The following examples of these customs as practised in China, are given by Mayers in the paper already referred to. 678 B.c. Human beings were first slain at tlie grave of the deceased Kung. The 621 B.c. 210 B.c. Wu number was 66. At the death of Emperor Muh Kung 177 were slain. At the death of Emperor She Kwang-ti, concubines who had borne no children, and others were put to death. No other later instances are given, but it is recorded that “ at the tomb of Hob Kii-ping (117 B.c.), stone figures of men and horses were arrayed.” In the province of Yamato, after these sacrifices had ceased, there was for some time a pretence of immolating victims. They were shut up in tlie chaml:)er of the mound with the dead, but an opening was left through which they might escape. These persons (termed om/w, smoke vanishing ”) were, however, considered to be dead, and liad to live in districts specially set apart for them. The custodians of the l)urial mounds formed another grade of men who were similarly compelled to live apart from the ordinary people. They were termed “ sliiku.” Both these grades usually carried on farming operations. The figures set up on the ancient burial mounds are called l)y the Japanese tsuchi ningyo,” a term merely signifying clay images. They are, with rare FIG. 5. — TERRA-COTTA FEMALE FIGURE. {Tsuchi-ningyo). f linear. {Reproduced hy the courtesy of the Council of the Society of Antiqiiarics.) ' According to Mr. W. G. Aston, “ tlie date ascribed to this incident cannot be depended on. At least, Chinese accounts speak of the custom of human sacrifices at the burial of a sovereign as in force in Japan so late as a.d. 247.” [18] Presidential Address. 13 exceptions, made of only lightly burnt terra-cotta, generally red in colour. Owing to the perishable nature of this material when exposed to the action of the weather, they would be rapidly destroyed as long as they stood above the ground, and only when by chance they were overturned and became covered with earth, was there any possibility of their preservation, hence but few have survived, and most of these are in a fragmen- tary condition. Unfortunately no records have been kept of the positions in wliich the existing specimens were found, but there is not the least doubt, judging from the forms of tlieir pedestals, that they were set up above the surface of the mound and not buried within it. My own opinion, which is based on the position in which I found a pedestal on a lai’ge mound of imperial form, and numerous fragments of terra-cotta, not pieces of ordinary tubular haniiva on others, is that they were so set up around the level summit of circular mounds and of the round peak of the double mounds. Fig. 5 represents one of these archaic figures from a mound in the province of Kozuke, which I was fortunate in being able to secure, and it is now in the British Museum. From the mode in which the hair is arranged, it is evidently intended to represent a woman ; around the neck is a necklace of round beads. FIG. 6. — TERRA-COTTA MALE FIGURE {TsucM-ningyo). The pedestal is in the form of a tube 5 LINEAR. resembling the haniiva, previously Antvluanes.) mentioned, and like them is pierced with two holes through wJiicli, it is said, a bar of wood was fixed to assist in keeping the figure in an upright position on the mound. In Fig. 6 is illustrated a male figure wearing a helmet, also a curious necklace of the curved ornaments, magatama, cylindrical beads alternating, which seems to have been worn only by men. Terra-cotta figures of horses were also frei^uently set up along wdth the human figures. 14 Presidential Address. [19] Stone figures, called hayato, are even rarer than those of terra-cotta ; it would hence seem that they had never been extensively used One from the province of Chikugo, now in the Imperial Museum, Tokyo, is shown in Fig. 7. It is a flat slab 3 feet in height (including the pedestal), 6|- to 7 inches thick, roughly hewn to represent a man wearing a short sword. On the back are perpendicular incised lines which are supposed by some to represent arrows. The mound from whicli it was taken is one of the double form, and formerly contained a dolmen hut now all the stones are gone.^ A curioTis form in which these guardians of the tomb are represented is shown in Fig. 8, which is a print of a rubbing of one of three rude stones, each of which has incised on its surface, in boldly cut lines, a rude drawing of a human figure with the head of a hare. This stone measures 4'1 x 1'6 x I’O feet, and its sides are not hewn hut weathered. The others are much smaller. They were found on the top of a burial mound, near Nara, attributed to the Empress Genimyo, who died in A.D. 721, and near them was also unearthed a hewn stone slal) bearing an inscri[)tion and the above date. When I saw them they were in tlie temple Todaiji (Nara), and were laljelled Hayato, which signifies “ Imperial guard.” The largest doulde mounds are situated in tlie provinces of Izumi, Kawachi, and Yamato, hut many others of imposing size I have also found in the provinces of Kozuke, Settsu, Hoki, Izunio, Yamashiro, Harima, Ilizen, and Hyuga. They vary in dimensions from a diminutive exanqde in Ilyuga, only 125 feet long and 18 feet high, to the stu]iendous piles officially recognised as the toinljs of the Emperors Nintoku and Eicliu in Izumi, and ( )jin in Kawachi, none of which are less than 1,200 feet in length and GO feet in height. That of Nintoku is specially noteworthy for its vast extent, l)eing about 90 feet high, and with its two nuAats covering about 80 acres of ground. Tlie manner in which the dead were disposed in tliese double mounds is by no means uniform. Some do not contain a megalithic dolmen, but only a sarcophagus of stone or wood not very deeply Inuied in the round peak. This I have found in some cases surrounded with a low wall of stones, over which larger slaljs were laid; in otliers these walls are wanting, and huge boulders then seem to have lieen simply placed over the col'lii'. ' In the Sliaku Niliougi (written in the thii’teentli century), it is stated tliat there were many other tignres of men and animals on tins mound. FIG. 7. — STONE FIGURE FROM A DOLMEN MOUND. {Reproihircd hy the courtesy of the Council of the Siocieiy of Antiquaries.) [ 20 ] Presidential Address. 15 FIG. 8. — HATATO. From a Burial mound attributed to the Empress Gemmyo. {From a rvbhhifi.) One of the largest and most noteworthy of these double mounds containing a dolmen, is situated in the village of Mise, in the most classic region of the province of Yamato. The vast proportions of this mound, and the magnitude of the dolmen within it, exemplify well the importance which the ancient Japanese attached to the sepulture of their illustrious dead. With the exception of the circular mound which forms its eastern peak, it is covered with terraced fields and part of the village. When first erected it cannot have been less than 1,000 feet long and 600 feet broad, although now it is slightly smaller, and the peak is 84 feet high. A little to the south of Unebiyama, around the villages Mise, Myohoji and Koshi, is the most important dolmen region in the province. Altliough now the number of dolmens is but few, yet three of these are amono'st the most remarkable O in Japan, whilst everywhere ruined mounds and piles of lu’oken stones mark the sites of scores of others, some of which were destroyed to furnisli stones for the modern mausoleum of the Emperor Jimmu. Fig. 9 represents the mound in longitudinal and transverse sections, showing the position of the dolmen within it. These lu’ing very forcilJy before us Presidential Address. 16 [ 21 ] the comparative insignificance of the burial chamber when contrasted with the vast dimensions of the mound. The mound which is also shown in Plate II is much dilapidated. Its sides and summit have long been under cultivation, and are clothed with terraced barley fields, excepting a portion at its eastern end, where a grove covers an irregularly rounded mound, which originally formed its peak. Yet on account of its vast size agricultural operations have failed to destroy the chief features of its original form —a double mound of the imperial type with four terraces. The moat has been almost completely absorbed by the surrounding fields so that its width cannot be ascertained. The dolmen, it will be seen, is situated below the round peak, and lies within the mound at right angles to its long axis ; its entrance, which faces south 10° west, being almost on a level with the second terrace. The gallery leading to the •f -e 120 C/- ^ Scale of Feet O 50 100 200 300 fOO O 50 100 200 300 4g0 FIG. 9. — DOUBLE MOUND AT MISE (YAMATO) In the elevation the entrance of the dolmen is seen in tl e side of the mound below the peak. The transverse section shows the position of the dolmen within the mound. {Rejirodurcd hit the courtesy of the Couacil of the Socifiy of Antiquarus.) chamber is about 60 feet long, about 8 to 10 feet high, and varies in breadth from 4 to 8 feet. Its roof consists of six huge undressed stones, one of which is 16 feet in length. Its walls are built of similar cyclopean blocks of somewhat smaller size, and all are of tlie rudest irregular forms without any trace of tool marks. Its fioor slopes gradually inwards. I was, unfortunately, able to penetrate only about 40 feet into this dolmen, when I was stopped by water and mud, which liad accumulated to a depth of about 4 feet further in, so that I did not reach the chamber, but, so far as I could see it, its structure is the same as the gallery. It contains two stone sarcophagi. Part of the cover of one of these — that placed longitudinally near the middle of the cliamber — was just visible above the water. It is of the usual form and size, well liewn, and with projecting lugs. The other sarcophagus, wliicli is placed transversely near the back wall, was not seen, being covered with water. I made two other special journeys to this dolmen during periods of drouglit, but on neither occasion had the water diminished. [ 22 ] Presidential Address. 17 I am hence compelled to rely for the dimensions of the chamber on the figures given in the Sei Seki Dzii Shi, a Japanese archceological work, dealing chiefly with the imperial burial mounds, and published in 1853. In this hook it is stated that the length of this chamber is 24 feet, breadth 18 feet, and that its roof consists of three stones. These measurements must be received with some reserve, but the chamber is certainly a large one. According to the author of this hook, the dolmen is the tomb of the Emperor Mommu (died a.d. 686) and the Empress Jito (died A.D. 702), but there are no grounds whatever for such an attribution. It is, undoubtedly, an imperial mound, although it is not officially recognised as one by the Imperial Board of Ceremonies at the present time, and the extensive cultivated fields which cover it show that this non - recognition dates from at least a century or two ago. The reasons for its rejection from the official list of the burial mounds of emperors or princes is impossible to conjecture, and especially so when we find so many puny and insignificant mounds are included in it. When we consider that tliis mound is exceeded in magnitude by only two others — and these are those of the famous Emperors Nintoku and Ojin — that the dolmen it contains is the largest in the country, and is, besides, unequalled as a specimen of megalithic structure, I think we cannot avoid arriving at the conclusion that it is, without doubt, the tomb of an einperor, and from the rude, undressed blocks of its dolmen — a very early one — one of the many of whose lives the ancient books, the “ Kojiki ” and “ Nihongi,” contain no record ; whose very names are unknown. The smallest double mound which I have found containing a large chamber is one of a rather extensive group of dolmens with simple mounds which is scattered over the lower slopes of Mount Kazuraki, near the village of Teraguchi (Yamato). The mound is 167 feet long and 32 feet higli. The dolmen^ is tlie largest in the group, and both gallery and chamber are of rude megalithic structure. That these large double mounds are the tombs of men of imperial rank of pre-eminent power is, I think, not open to doubt. Their vast bulk implies the labour of many hundreds of men for a considerable time for their construction, and this only a chief or supreme ruler could command. According to Japanese archteologists the earliest is the tomb of the Emperor Annei (c. fourth century A.C.), and the latest that of the Emperor Bidatsu (died a.d. 585). Whilst not accepting the strict accuracy of these dates, there seems to be no reason to doubt that several are as early as one or two centuries or more before our era, and that they continued to be built for five or six centuries after it. During this range in time nearly all the emperors whose names are recorded in tlie Kojiki, and many whose names and existence have been forgotten, were buried in these double mounds in the central provinces, but I have also found these mounds of imperial form in the important dolmen districts of Izumo, Hoki, Bizeu, Kozuke and Hiuga which are remote from the central provinces, the seats of the above recognised emperors. This would seem to indicate that these regions were once independent centres or were governed by chiefs who were regarded as ecprals with the central ruling family. Archceologia, Iv. 520, Tab. 1, No. 123. h 18 Presidential Address. [ 23 ] According to the statements of Japanese records the care of the imperial tomhs was entrusted to special resident officers from very early times, a custom surviving at the present day. But the appointment of custodians was frequently discontinued, sometimes for a considerable time ; the mounds were then neglected and their sanctity was disregarded, and it is probably owing to this that many are now under FIG. 10. — MISASAGI OF THE EMPEROR JIMMU. {From a Japaiiesc drawing.) cultivation, and with numerous others, are not recognised as imperial tomhs. Yet, notwithstanding these breaks in the regular succession of custodians, it is not unreasonable to suppose that the most important mounds, those which from their imposing size form such conspicuous olijects on the plains, would not altogether lose the early traditions attached to them, and would at least retain the names of the [24] Presidential Address. 19 emperors whose tombs these traditions indicated them to be. If any reliance may be placed on this supposition, then the enormous mounds of Nintoku, Eichu and Ojin may be considered to be the tombs of the emperors whose names they bear, although in the great majority of lesser magnitude the attribution will be doubtful. From this it follows that the building of , double mounds reached its zenith about the fourth century of our era. Before proceeding to deal further with typical misasagi of the double mound form, I will enter into a brief consideration of the misasagi or imperial mausoleum of the Emperor Jimmu, the founder of the Imperial dynasty.^ At the foot of tlie northern side of Mount Unebi, in the province of Yamato, a mound of insignificant size but a few feet in height and diameter, one of a small scattered group, had been long pointed out by tradition as the misasagi or liurial mound of the Emperor Jimmu, the founder of the Imperial dynasty. Shortly after the fall of the Shogunate it was recognised as such by the govern- ment, and was surrounded with a moat and embankment together with a large tract of the ground around it. The present emperor visited the misasagi on April 3rd, 1877, the anniversary of the death of Jimmu, and was present at the celebration of the annual ceremonies held before it. It is curious to note that the mound attributed to the first Emperor Jimmu is situated not on Mount Unebi but on the plain at its foot ; yet, according to the old traditionary records, the earliest emperors were all interred on the summits or lirows of natural hills. The misasagi is of little archaeological importance as its construction differs entirely from any of the ancient mausolea. It is difficult to conjecture the grounds on which its form was decided, yet it is worthy of description as showing what is regarded by the government to be a fitting mausoleum for the first of the imperial line. The misasagi is represented in Fig. 10, which is copied from a Japanese drawing. It consists of an inner square of level ground with sides 358 feet long, containing two low mounds, each about 18 feet in diameter and 2 feet to 3 feet high. Tins is enclosed by a low emlrankment and surrounded l>y a moat containing water. The mound attributed to the Emperor Jimmu is neai’er the middle of the square than is shown in the figure. ISTeither mound is visilile, even from the neighbouring slope of Moimt Unebi, as the enclosure is thickly planted with trees. Outside the moat, on the north and also on the south side, a broad roadway runs from the east to the west sides of the misasagi, and the whole is enclosed by a turfed embankment surmounted by a fence of stone rails. The entire enclosure is rectangular, and measured along the stone fence is 871 yards in circuit. ' It is recorded in the Kojiki that in the “ 76th year of his reign (585 B.C.), on the 11th day of the .3rd mouth, the Emperor .Jimmu died in the palace of Kashi wa-bai’a. His age was then 127. The following year, autumn, the 12th day of the 9th month, he was buried in the misasagi, north-east of Mount Unebi.” I 2 20 Presidential Address. [25] The outer gate of the form usually found at all imperial tombs is shown by a broad line on the outer side of the embankment which crosses the moat. On the other side of the moat is a wooden torii 18 feet high, and behind this another torii 12 feet high, constructed of stems of the tree hinold (Thuya obtusa) from which the bark has not been removed. This kind of torii is said to be peculiar to this misasagi. On each side of the space between the torii is a stone lantern such as are seen at all misasagi. On one of my visits (2nd April, 1888), 1 found a new wooden shed had been erected in the space between the two torii just mentioned for the annual ceremony of the ord, the date of the death of the emperor, when a representative of the ]\Iikado' visits the tomb to make the customary offerings. This officer is called Cliohishi or imperial ambassador, and, in addition to the offerings mentioned below presents a special offering from the Mikado, the nature of which I was unable to ascertain. The offerings are made on eleven sambo (ceremonial stands of white wood), and consist of products of the sea, river, and mountain such as tai (serranus marginalis), carp, sea weed, salt, water, sahe (rice beer), mochi (rice bread), horseradish, oranges, pheasants and wild ducks. A ceremonial ottering of raoehi and saJc,e at the outer gate by Shinto priests is illustrated in Plate III. Three sambo, bearing the offerings, are shown on the table in front of the otticiating priest. The public are allowed to go only as far as the gate on the outer side of the moat where they pay reverence to the tomb. The most ancient Japanese pottery yet found are the rude vessels which wer e unearthed near the small tumuli of (Miproauced ly the cov.rusy of the . Council of the Society of Antiquo.ries.) this vi'isasagi whilst tlie present moat was being excavated. Fig. 11 represents one of these, now in the Gowland collection in the British Museum ; others, not much different from it, are figured and described in the Kwanlfo Dzusetsu, by the late Japanese archieologist Ninagawa. The misasagi which I now propose to describe is that of the Emperor Keitai who died in A.u. 531, as it is one of the best preserved of all the misasagi recognised by the Imperial Board of Ceremonies ; we will then consider the most important ‘ Mikado. Though tliis is the name by which the whole outer world knows the sovereign of Japan, it is not tliat now used in Japan itself, except in poetry and on great occasions. The word is one indicative of the highest respect, as it is but natural that the name used by the Japanese of old to designate their heaven -descended sovereign should he. Tlie Japanese have got into the habit of calling their sovereign by such alien Chinese titles as Senshi, “ the son of Heaven ” ; Sen-0 or SennO, “ the Heavenly Emperor ” ; Shujo, “ the Supreme Master.” His designation in the official translation of modern public documents into English is “ Emjieror.” Things Japanese. Basil Hall Chamberlain, 1891, p. 291. FIG. 11. — TERRA-COTTA VASE (hand made), height, 7| INCHES. [•26] Presidential AdAress. 21 of those of the earlier emperors in the central provinces ; and afterwards some typical examples in other centres. Keitai Tcnno . — Born A.D. 450, reigned from 507-531. Eesidence at Tama ho (Yamato). The misasar/i of this emperor is an excellent example of a well-preserved doulde mound, situated on the plain between Takatsuki and Ota in the province of Settsu. Yo natural eminence has been taken advantage of in building it, and this, together with the extensive embankments which still bound the moat on two sides, show that its erection must have been a work of great magnitude, although in this respect it falls far behind the enormous piles of Nintoku and Ojin. It possesses for us, however, more than usual interest, as it is the last great example of the double mound period, which, beginning about one or two centuries u.c., reached its zenith during the reigns of Ujin and Kintoku and ended at the death of Yoinei a.d. 587. Four emperors succeeding Keitai were interred in doulde mounds it is true ; yet these are all of insignificant dimensions. During later times the building of large double mounds appears to have l)ee]i resumed, otherwise it is difficult to explain the occurrence of several both in 22 Presidential Address. [27] Yamato and Kawachi with sides, peaks and terraces so little weathered that they seem as if made Imt a century or two ago, yet no imperial name is attached to them. The tumulus of the Emperor Keitai, Plate III, is a long terraced momid, the upper portion rising from the third terrace ; this terrace, however, is insignificant in size and somewhat ill-defined compared with the two lower terraces. The Ijase of tlie mound measures 729 feet in length, 497 feet in breadth at its southern end which is straight, and 465 feet at its opposite and circular end. Fig. 12, wliich is drawn from my survey, sliows the mound in plan and section. The moat varies in breadth from 80 feet to 130 feet. The height of the momid could not he satisfactorily measured, hut its straight end is not less than GO feet high above the level of the water in the moat. The top of the round end cannot he well seen as it is covered closely with trees. From the top of the straight end to the base of the round end is 235 feet. At the south end, the lowest terrace is about 34 feet broad and the succeeding one 42 feet. There is a projection from the lower terrace in the re-entering curve of each long side. These projections are not distinct mounds although they are somewhat higher at the middle than at the sides. The east hank of the moat has been much cut away, and cultivated fields extend to within a few yards of the moat. On the west side the ground has not been disuulied, and here fragments of lutnitm occur, although none of the tulies can be seen in situ. On the north summit there are said to be three huge stones similar to those of dolmens. A short distance from the embankment on the west side, nearly opposite the re-entering curve, there is a small circular mound with haniwa fragments scattered over it, and two other similar mounds occur near the north-west corner, and remains of several others are seen in the fields on the east. The Irikuscdcu, or sacred eiiclosure, shown in Plate IV is of the usual form, On its right is a stone granite pillar, modern, hearing the name of the emperor, and on the left the usual official notice hoard.^ Its dimensions are as follows : — Breadth, 42 feet 2 indies ; length, 30 feet 2 inches. The torii are each 1 foot in diameter. The front torii is 10 feet 0^ inch high from the stone lintel to the bottom of the top bar. Both torii are of round hewn timber, both uprights and cross bars. ' Official inscription on the notice board at a nikasagi. August Tomb of the Emperor East to West Ken Nortli to kSouth Ken (In some cases only the circuit is given.) Regulations. It is not permitted to enter here or to climb the mound. It is not permitted to catch the fish or birds. It is not peimitted to cut down the trees or bamboos. Date. The above notifieation is to he strictlg obeyed. [28] Presidential Address. 23 The fence consists of carefully-hewn wooden posts rising from a horizontal beam which rest on a foundation of one row of squai'ed stones. Within the enclosure there are two stone lanterns, one on each side of the inner torii. The space within the enclosure and also in front of the gate is covered with sand or fine gravel and is kept carefully swept. Outside the gate the sand is carefully heaped into two conical momids, one on each side, and when offerings are made here on the appointed days the sand is scattered over this space. The same small mounds of sand may be seen occasionally at the entrances of cremation grounds in the country. Chuai Tennd’s Misasagi. PI. V. Chaai Tenno, born a.d. 149, accession 192, died 200.2.6. Age 52. Eesideiice, Kehi. The misasagi,^ Plate V, is situated at Fujudera in Kawachi. It is a double mound surrounded by a moat. The embankment and the tumulus also have been recently repaired and put in order, so that the original shape of the former may FIG. 13. — PLAN OF EAST HALF OF THE BURIAL JIOUND OF THE EMPEROR CHUAI WITH ITS MOAT. have been slightly altered. The east side of the mound (Fig. 13) is more perfectly preserved than the west, the projection on the latter side being now almost shapeless. The measurements were hence made on the east side only. The kalcusalcu is placed on the embankment of the moat, almost opposite the middle of the square end of the end. It faces S. 20° W. The dimensions of the Imhisaku enclosure are:— length, 30 feet 2 inches, and breadth, 42 feet 7 inches ; gate, 8 feet 6 inches broad. On one visit the moat was almost empty of water, fragments of haniwa., of both red and a bluish colour, were then seen to be strewn along the bottom, both at the liase of the tumulus and of the embankment. They were more numerous in the former position. According to the official notice board, the misasagi measures 651 ken = 1,294 ' This misasagi was only determined to be that of Chuai about foi ty-four years ago wlien it was repaired. Presidential Address. 24 [29] yards in circuit, along the embankment of the moat. The principal dimensions of the mound are : — Feet. Length of base ... ... ... ... ... ... 803 top Breadth of base, square end „ „ round „ Height of round end about The moat varies in breadth from 109-172 feet. 381 637i 560 70 The summit rises from the fourth terrace, which is unusual, most misasagi having not more than three terraces. The fourth terrace is, however, very indistinct in places, and may be of later date than the mound, and this, T think, extremely probable, as the summit both at the roimd and the square end is much less liroad than we should expect to find in a misasagi of such large dimensions. The other terraces have also been much cut up for some reason wliich is not apparent. The projections from the lower terrace, especially that on the east side, are seen to he not of the form of ordinary small I’ound tumuli, but merely prolongations of the lowest terrace, rising about two feet or so from their sides to the middle, with a low rounded surface, Near it are the following mounds A small round mound almost in front of the kakusahu, 40 yards distant, and two others. A similar mound, on east side, in line with the moat at the square base and 150 feet from the embankment. „ „ east side, opposite the curve, and 150 feet from the emlrankment. „ „ west side, in line with the moat at the square base and 50 yards from the embankment. A double ,, with a moat, very small, lying east and west, about 200 yards north of the nortli end of the misasagi. There may haA'^e been several other small mounds arranged around the outer embankment which liave been levelled l)y the farmers, as cultivation extends quite lip to the moat. None of tliese small mounds, which in Kawachi are ranged around the imperial tumuli, contain dolmens. Tliey are generally simple mounds of earth. In one example near the misasagi of the Emperor Ingio, about lh~2 miles distant, an earthen or clay coffin, containing iveapons, ornaments, and vermilion, was found liuried a few feet lielow its top. Around or near a misasagi in ancient times, there were, doubtless, Imried the wives or conculfines and the chief retainers of the emperor. When the misasagi was constructed the small mounds were perhaps made with it, during the lives of those wlio were, at tlieir deatli, to be liuried in them. And as the coffin is always l)ut shallowly Imried in the top of these mounds there would lie no difficulty in subsequently using them for interments. Many which have been opened in tlie neighljourliood of the Kawachi misasagi have been found to contain nothing, others to contain weapons only and no bones, others to contain [30] Presidential Address. 25 bones only and no weapons. The absence of bones in those cases where the sarco- phagi were found to contain only weapons is not certain, as, in several cases known to me, the bones have disintegrated and fallen to powder, the enamel of the teeth alone being preserved, so that, unless they were very carefully looked for, they would escape observation. The misasagi of the recognised emperors earlier than Chuai are so much weathered and dilapidated that they afford no useful measurements for the determination of their original size, I will therefore enter into a brief consideration of the most important of those of liis successors. They are all of the type shown in Figs. 2, 12 and 13, so that only a short notice of each will suffice. The first is the burial mound of the Empress Jingo to whom the conquest of Korea in the early years of the third century a.d., is attributed. The Empress died in A.n. 269. The mound, winch is a large one, is 660 feet long, 418 feet broad, and about 60 to 70 feet high. Its circuit as measured along the outer embankment of the moat is stated on the official notice board to be 563 ken = 1,119 yards. The misasagi of the Emperor Ojin^ (died A.D. 310), the son of the Empress Jingo, is situated in the province of Kawachi. It is one of the largest of the recognised imperial mounds, being 2,312 yards or nearly a mile and a third in circuit, as measured along the outer embankment of the moat. Its height at the circular end is said to be aliout 60 or 70 feet. The misasagi of the next Emperor Kintoku (died a.d. 399) is shown in Plate V. It is situated near Sakai in the province of Izumi, and is the largest of all the imperial burial mounds. It possesses two moats and three terraces, and is 2,475 yards or nearly a mile and a-half in circuit. The north or circular end is is about 94 feet high. This stupendous pile was erected by the people, doubtless in recognition of the beneficent reign of the emperor.^ The misasagi of the Emperor liichu (died A.D. 405), Plate VI, is not far dista7it from the last. It also is a large mound, but somewhat smaller than those of his two predecessors. Its base is 1,224 feet in length, and the circuit of its moat 871 ken, or 1,742 yards, i.e., about a mile. The misasagi of the Emperor Ingyo (died a.d. 453), the successor of Eichu, is in the province of Kawachi near the village of Domioji. Plate VI. * This emperor some centuries after his death was deified as the War God Hachiman. - It is recorded in tlie JVihongi, that the emperor ascended a higli tower and looked far and wide, but no smoke arose in the land ; from this he inferred that the people were so poor that none in the houses were cooking rice. He then decreed that for the space of three years no taxes and no forced labour should be imposed. His own palace, for want of funds to repair it, was allowed to become so dilapidated that the roof admitted the rain. Three yeais later he again ascended the tower and beheld smoke rising from every dwelling. The people were now rich enough to bear taxation wfithout feeling the burden, and voluntarily offered their labour and contributions towards the rebuilding of the palace. Nintoku is said to have reached the advanced age of 122, but, it must be remembered, that it is not until the next i-eign that the miraculous details which characterise the early portion of .Japanese history cease. In this connection too, it must be noted that the dates I have given, which ai-e those of the Nihongi, befox’e the reign of Richu, should be accepted with reserve. 26 Presidential Address. [31] The mound is 750 feet long and 528 feet broad, and about 60 feet high. According to the official notice board its circuit is 559 ken or 1,111 yards. The burial mounds of the succeeding emperors up to Keitai are all of comparatively small size and much dilapidated so that none requires any special notice. Passing now from the central provinces, the recognised sites of the capitals of the early emperors, and crossing over to the opposite side of the island, we find in the province of Izumo burial mounds of imperial form precisely similar to those with which I have already dealt. The first which I shall describe is a large double mound behind the temple Dainenji, on the borders of the town of Imaichi. It contains a double-chambered dolmen. The mound is very much weathered, but from parts of it which have FIG. 14. — DOLVEN CONTAINING TWO STONE SARCOPHAGI AT IMAICHI (iZUMO.) With elevation and transverse section of largest sarcophagus. i nduced h}/ the courtesy of the Council of the Society of Antifiuarics.) retained tlieir form tlu; original shape and some of its dimensions can be well deter- mined. It possessors two terraces, the floor of the dolmen being on a level with the top of the lowest. In direction tlie mound runs E.N.E. and W.S.W., the round end, which contains the dolmen, being dir nted to the latter quarter. The dolmen ha.s the same orientation as the round peak, lying longitudinally in the mound, a. ]iosii 1 11 whicli I have not found elsewhere in Japan except at Teraguchi. In all other mounds of this form the direction of the dolmen is at right angles to their lenglh. As the breadth '-i the mound is more than sufficiently great to permit the dolmen to have been built in tbe normal position there must have been some syiecial reason for its unusual position. [32] Presidential Address. 27 Dimensions. Height of round end above the plain ... ... 42 feet. Length of base of upper terrace ... ... ... 280 „ „ „ suniniit ... ... ... ... ... 145 „ Tire dolmen (Fig. 14) is constructed of rude stones, some of which have natural flat faces, hut others are much rounded by weathering. It has two chambers, both of which contain stone sarcophagi. The inner, which is the most capacious, has the following dimensions : — Average Length ... ... ... ... 18 feet 10 inches. „ Breadth ... ... ... ... 9 „ 9 „ „ Height 11 „ 2 „ Placed longitudinally in it is a huge sarcophagus hewn out of a single block of hard volcanic tuff, measuring internally at the top 9 feet long by 8 feet, 74 inches broad ; at the bottom 9 feet 4 inches long, 4 feet 5 inches broad, and 3 feet, 6 inches in depth. Its cover is of the usual roof-shaped form with projecting lugs, and is 1 foot 11 inches thick. This sarcophagus is one of the largest I have found, and is remarkable also for the curious opening hewn in its front side. The opening is 4 feet 4 inches long by 2 feet 4 inches high and is recesseil to receive a slab by which it was closed. Below it the bottom of the sarcophagus projects in the form of a step, upon which the slab rested. This peculiar feature is seen in three other sarcophagi in dolmens not far distant, and seems to be confined to the province of Izumo, as it has not yet been found elsewhere. As to the purpose it can have served I am unable to offer any explanation. It is too large to be intended for the introduction of offerings of food, or for the egress or ingress of the spirit of the dead. The other chamber is much smaller than the inner, being only about 10 feet long, 9 feet high, and tapering from 8 feet at one end to 6 feet 4 inches at the other. Its floor is nearly 18 inches lower than that of the inner. The sarcophagus which it contains is much smaller, and is constructed of slabs somewhat roughly hewn. Its cover is broken up and also its front side. Its internal dimensions are : — length, 5 feet 1 inch ; breadth, 2 feet 5 inches ; depth, 1 foot 10| inches. The inner chamber is partially separated from the outer by large stones which project like rude lintels from the walls on each side and also by a huge capstone resting upon them. The outer chamber is also marked off from the entrance gallery by similar side stones and a depression of the roof. The total length of the dolmen from the back wall to the entrance of the gallery is 43 feet, and its breadth diminishes from 10 feet 1 inch at its inner extremity to 3 feet at its entrance. This dolmen, I was informed, was opened in 1825, its floor was then covered with small round stones : since that date it has stood open. A large quantity of 28 Presidential Address. metallic vemains, and many vessels of pottery, were taken out, lint unfortunately all have been lost excepting the following, which are now kept in the temple : — One straight ,?word blade, part only 28| inches long, inches liroad. Several iron arrow heads. Cheek piece of a horse hit, iron plated with copper. Several metal ornaments for the trappings of a horse. Part of spear head, iron. (_)ne socket piece, iron. Xone of tlie ornaments are gilt, all are simply of iron coated with copper. On the exterior of the mound I found many fragments of red terra-cotta haniwa. The inner .sarcophagus is doubtless that of an emperor or ruler of this important centre in the early history of Japan, and the outer, that of his empress or consort. On a low upland near the village Enya inura, about a mile to the south of the Imaichi mound is another douljle mound containing a large dolmen of even greater importance than that just described, liy reason of the structure of the dolmen chamber, its two well-hewn sarcophagi, and more esj^ecially for tlie numerous objects, weapons, armour, etc., which were found in it when it was opened in 1886. The mound has, unfortunately, been reduced by the needs of agriculture and by weathering to a shapeless heap, hut from its great length, compared with its breadth, and from the position of the dolmen within it, it was certainly originally a double mound of the imperial type. The dolmen, with its sarcophagi in situ, is represented in plan and sections in PlateVII. It is contained in the west-north-west end of the mound, and its mouth is directed towards the west-south-west (W.S.W 3° S.). It possesses a single chaml^er only, and in this the two sarcophagi are placed ; the larger longitudinally against the west-north-west side, and the other transversely against the l)ack wall. The chamber has an average measurement of 21 feet in length, 8 feet 5 inches in ljreadth,and 9 feet 9 inches in height, and is separated from the entrance gallery by one of the wall stones on each side being set forward, and by a huge capstone which rests upon them as shown in the figure. When the dolmen was opened the entrance to the chaml)er was closed by hewn blocks built up across the gallery from side to side, some of which are still in position. The floor of the chamber is covered with large and small rounded pebbles. T1 le.total length of the dolmen from the end of the chamber to the mouth of the gallery is -16 feet 8 inches, and its lireadth diminishes from 8 feet 8 inches at its inner extremity to 3 feet 5 inches at its entrance. Botl) the chamber and gallery are constructed of moderately well-hewn blocks of volcanic rock. . Tlie sarcophagi which are more carefully hewn than tlie walls of the dolmen have each a large well-cut aperture in the front side, that of the smaller being recessed [ 34 ] Presidential Address. 29 so that it might be closed with a slab. The apertiu'e in the lai'gest sarcophagus measures 4 feet 1 inch by 1 foot 9 inches aiicl in the smaller 3 feet 5 inches by 1 foot 44 inches. At the bottom of the former is an angular groove leading into the interior, probably intended to allow any water which might find its way into the sarcophagus to drain away. Their internal dimensions are as follows : — Length. Breadth. Depth. Tliickness of sides. Thickness of bottom. Longitudinal sarcopha- gus ... J 7' 4" top I 7' 9" bottom ... 2' 10" top \ 3' 3'' bottom / 3' 2" 9" to 10" 1' 0" Transverse „ 5' 10" 3' 1" 2' 3" 6" to 8" 9" Tire capacity of both is increased by the lower sides of the covers being hewn out to a depth of 11 inches and 7 inches, respectively. The chief objects which were found in this dolmen are now in the Government offices at Matsuye. The iron objects are nearly all converted into iron oxide, and, as little care seems to have been exercised in removing them, most are more or less imperfect. I7o record was made of the position of any in the dolmen, or of the shapeless masses of rust into which many otlrers were oxidized which were thrown away. List of articles now at the Government ojficcs. 24 Iron arrow heads. 1 Straight iron one-edged sword, with the point broken off. Length of remaining part of blade, 2 feet 7^ inches ; breadth, IJ inch. Wooden scabbard of the above, mounted with a single encircling band of copper, coated with silver, which bears a simple line pattern of punched dots. 1 Iron sword, straight blade, 16-^ inches long. Total length, including grip, 234 inches. 1 Iron sword, straight blade, 17 inches long. Tang imperfect. 1 n ^ 7 „ spear heads, socketed. Triangular blades. . 4 „ halberd-shaped ornaments for horse trappings, coated with copper gilt. 8 Iron ornaments for liorse trappings, plated with gilt copper. 4 ^ yy 9) yy yy ?> yy yy yy 99 2 „ „ coated with silver. 1 „ horse-bit, with check pieces of iron open work, plated with gilt copper. 1 Iron buckle. 30 Presidential Address. [35] 1 Bronze bell. 2 Covered pots of ordinary dolmen form. Besides the above there were several magatama ("curved beads), and I'v.datama (l)ugles). Respectiog the position of these objects in the dolmen, the statements of tlie officials at the Government office, and of the head man of tlie village in which the dolmen is situated, were very conflicting ; but as he was present at the opening, and assisted in taking out the objects, I am inclined to accept his version. According to him, the longitudinal sarcophagus contained the short swords, bronze bell, some of the arrow heads, and some of the spear heads. The transverse sarcophagus contained large masses of iron rust resemlding plates of armour, magatama, the long sword, some arrow heads, and the silvered ornaments. Some spear heads were found outside this sarcopliagus near its W.N.W. end, and the horsel)it and horse ornaments on its cover. < 96 - - 46 -' - -> No leones or vermilion were seen, and no pottei-y excepting the covers of the two shallow dishes, which were found just within the entrance of the chamber. Leaving Izumo for a' tract of country to the north-west of Tokyo, where the provinces of Kozuke, Musashi, and Shimotsuke meet, we find the most remarkable and extensive series of Imrial mounds which occurs outside the central provinces. Here, within an area about twenty-five miles long, and about the same breadth, are scattered either singly or in groups about 300 mounds, nearly all containing dolmens. The district to which I gave special attention was that in the neighlioiirhood of the villages of Oya and Omuro in Kozuke, as one of its noteworthy features is the occurrence of six large double mounds of the imperial type, containing^ dolmens and one a cistj and two of the former had yielded some interesting vessels of pottery as well as metallic remains. 1’he occuri ence of these double mounds, and the highly ornamented metal work which was found in the dolmens which both tliese and several of the simple mounds enclose, also the nundjer of ts'udiiningijo (terra-cotta figures) which [36] Presidential Address. 31 have been unearthed from these and others, indicate clearly that the region was one of the leading centres during the flourishing part of the mound-building period. The most important of those I examined were two double mounds near the village Omuro. The mounds are situated on the north and south of a much larger double mound with two moats, which is said to have never been opened. The sectional sketch (Fig. 15), made from my own measurements, illustrates the north mound, winch is called “ Futago yaina ” or “ twin hill.” The mound is of the ordinary imperial type, with two terraces, and surrounded l>y a moat. It is, however, very much weathered, so that these features are partially obliterated Its direction is E. 20° N., to W. 20° S., the rounded peak lieing at the eastern end. Numerous lianivm, (terra-cotta tubes) are embedded in the usual manner near the edges of the terraces and summit of both this and the next mound. The dolmen, which is contained in the round end, is 48 feet in length, and is placed approximately at right angles to the long axis of the mound, with its entrance a little higher than the lower terrace, and facing S. 12° E. It consists of a chamber separated from the outer gallery by two rude slabs fixed vertically against the side walls with a capstone resting upon them, leaving a doorway of about 4 feet by 1 foot 9 inches. The floor of the hinder part of the chamber for about 6 feet from the hack wall is raised about 9 to 10 inches higher than the other part of the floor. Both were paved with large, flat, rough slabs, some of which are still in situ. The dimensions of the chamber are : — Length of Chamber ... ... 17 feet. Gallery ... ... ... 26 ,, Total length 43 „ Chamber : — Breadth at back ... ... ... 6 feet 6 inches. „ at front ... ... ... 5 „ Height ... ...5 feet 6 inches to 6 ,, Gallery : — Breadth ... ... ... ... 4 „ Height ... ... ... ... 5 „ 6 inches. A diagram (Fig. 16) only is given, as it was impossible to make a complete drawing of the interior, owing to the quantity of earth which it contained, and, to obtain the measurements, this had to be dug into at several points. The dolmen is constructed of unhewn, irregular blocks, none of which are specially remarkable for their size, the largest only measuring about 6 feet by 4 feet 3 inches. The objects which were found in it when it was opened are now in the possession of Mr. Negishi, of Omuro, and have been already described Presidential Address. 32 [37] by Sir Ernest M. Satow in a paper read before the Asiatic Society of Tokyo, Japan 2 The most important are the following : — From the raised part of the chamber : — 4 horse ornaments of iron, plated with gilt copper. 1 small bronze mirror. 1 iron spear head. Nnmerons beads of blue glass. 1 penannnlar ring, plated with gold. From lower part of the chamber : — 1 horsebit, with cheek-]3ieces of iron, plated with gilt copper. 1 stirrup iron. 1 iron spear head. Several iron arrow heads. 17 vessels of pottery, chiefly of types b, d, e, k, f (Fig. 22), but several of softer clay than ordinary dolmen pottery. FIG. 16. — PLAN AND SECTIONS OF DOLMEN IN THE DOUBLE MOUND AT OMURO. The head of a tsuehi-ningyd (terra-cotta figure) was found buried in the earth on the outside of the mound. The horse ornaments and cheek-pieces of a bit are practically identical with tliose in the British Musenm, which I obtained from a dolmen in Tamba.^ ■ A notalJe feature of the mound, which, however, as we have already seen, is not peculiar to it, is its imposing size when compared with that of the dolmen it contains. This teaches us how very careful we should be in our explorations of sepulchral mounds in pronouncing any to be without a chamber, although we may have sunk shafts and driven tunnels in them without finding one. In this example we might have sunk a shaft from the middle of its summit to its base, and tunnelled through it from end to end along its median line, and never found the comparatively large dolmen which it contains. Tlie southern mound, whicli is called “ JJehihori tsuka,” is slightly smaller than ' Trans, of the Asiatic Society of Japan, vol. vii, 313, et seq. - Archceologia, vol. 55, Figs. 29, 30a, pp. 487, 488. [38] Presidential Address. 33 that just described, and has only one terrace, hut otherwise it is similar to it. The dolmen it contains is similarly placed, but is only 27 feet 6 inches in total length, and faces S. 25° W. The chamber is large, 21 feet 6 inches long, 8 feet 4 inches wide at the hack, and 5 feet at the front, and 7 to 8 feet high. At a distance of 9 feet from its back wall it is divided into two by two slabs of stone placed transversely across the floor and rising about 13 inches above it. The inner portion of the floor is strewn with round pebbles. It is merely a form of cdUe converte, its walls gradually converging to the entrance to the short gallery. The floor is about on a level with the top of the terrace. FIG.' 17 . — FLAX AXD SECTION OF DOUBLE MOUND WITH, A SARCOFHAGUS EXPOSED ON ITS SUMMIT. The remains fountl in this mound were as follows: — Four iron swords, Several iron rings „ human teeth „ iron arrow-heads. Six vessels of pottery of ordinary dolmen shapes. A large double mound about four miles to the south of the above is w'orthy of note, as it contains no dolmen, but merely a stone sarcophagus, wdiich is placed at the summit of the round peak. It is the only example I have found of a mound of this form containing a sarcophagus not placed in a dolmen chamber although such is by no means uncommon in simple conical mounds. The mound e 34 Presidential Address. m with the sarcophagus in sitn is shown in Fig. 17. The mound lies E. 20° N., W 20° S., and the sarcophagus has the same direction. Its base measures approxi- mately, lengtli, 405‘feet, extreme breadth, 225 feet, height, 32 feet, and it has been surrounded by a double moat. The sarcophagus consists of two longitudinal side slabs rather roughly hewn, which are channelled to receive the transverse slabs forming the ends. The bases of these slabs rest on a ledge cut in the large slab forming the bottom. The interior dimensions are, length, 6 feet 7^ inches, breadth, 2 feet 2 1 inches, depth, 2 feet 9 inches. The sarcophagus now projects about 6 inches aboye the ground, but originally it was covered with earth, which was removed when the summit was levelled long ago to form a site for a Shinto shrine. Fragments of hanivja are found on the mound, and these and its double form are the only clues to its approximate age. Another important and, in fact, the most ancient, centre of the early rulers is the island of Kyushu. Many of the very numerous mounds there have been opened and a great number of ol>jects have been obtained from their chambers. The FIG. 18. — SWORDS FROM THE HIGO DOLMEN. mounds are chiefly of the simple conical type, some being of imposing size, and most contain a dolmen. Eonble mounds of imperial form also occur. Some of the most important objects which have yet been found in any burial mound were taken out from the dolmen chamber of a double mound in the province of Higo, which has been previously mentioned as having had many stone figures of men set up upon it. The objects taken from this mound are now in the Imperial Museum, Tokyo. They comprise, amongst others of more common occurrence : — Fourteen sword blades, three of which are illustrated in Fig. 18. The longest blade is 3 feet 0'5 inch long and its tang inches, the total length being 3 feet 9 inches. Tlie shortest blade is 2 feet 6 inches long with a tang of 6 inches. Some pieces of armour and a helmet of special interest were found together witli the swords, also several spear heads. Tlie armour consisted of two cuirasses, one of which is illustrated in Fig. 19. It is formed of iron plates very skilfully forged and riveted together. The helmet is of similar construction. Both the helmet and the cuirasses are entirely different [40] Presidential Address. oO in form from those of historical times, but the latter agree very closely with the armour represented on the terra-cotta figure (tsuchi ningyd). Fig. 6. Thin plates or bands of gilt copper seem to have played an important part in the decoration of the robes of the dead. They are of frequent occurrence in the dolmen chamber of important mounds, and are always found along with the fragments of bones, when there are any, or in that part of the chamber where the body had lain. The largest piece from this mound is a broad hand of copper gilt foil, orna- mented with a hexagonal net-like pattern, the decorative effect being increased by a small circular pendant of gilt copper foil suspended by wires from the angles of the hexagons. But more important than these by reason of its elaborate decoration is a tiara of gilt copper having in addition to the punched dot decoration, rich scroll designs in pierced work. FIG. 19. — IROX CUIRA.SS AND HELIIET. ^ LINEAR. {Hcpi'oduccd hu the conrUsy of the Council of the lSoci ' • « / / ■•<•1 i ? I •n % Journal of the Royal Anthropological Institute, Vol. XXXVII, 1907, Plate II. DOUBLE MOUND CONTAINING A DOLMEN AT MISE (YAMATO). (Reproduced by the courtesy of the Council of the Japan Society.) THE BURIAL MOUNDS AND DOLMENS OF THE EARLY EMPERORS OF JAPAN. 9 m V e- I •I . < 1 J >;■ C < / ni Journal of the Royal Anthropological Institute, J^ol. XXXVII, 1907, Plate V. FIG. 2. — MISASAGI OF NINTOKU. THE DOLMENS AND BURIAL MOUNDS OF THE EARLY EMPERORS OF JAPAN. FIG. 1. — MISASAGI OF CHUAI. I ./tl Journal of the Royal Anthropological Institute, Vol. XXXVII, 1907, Plate VI. I t FIG. 1. — MISASAGI OF THE EMPEROR RICHU. FIG. 2. — MISASAGI OF THE EMPEROR INGYO. THE BURIAL MOUA'DS AND DOLMENS OF THE EARLY EMl’ERORS OF .JAPAN. Jo'urnal of the Roiial Anthropological Institute, Yol. XXXVII, 1907, Plate VII. DOLMEN AT ENYA. THE BURIAL MOUNDS AND DOLMENS OF THE EARLY EMFERORS OF JAPAN. <1 t. ■' Al' I <1 •iir*’** Journal of the. Boi/al Anthropological Inxtitute, T~ol. XXXT^II, 1907, Plate I'lll. BURIAL MOUND OF THE LAST EMPEROR { from a Japanese Print). THE BURIAL MOUNDS AND DOLMENS OF THE EARLY EMPERORS OF JAPAN.