J'^M \r?l = 0) »CNJ I CO tcD m M ^^^ ^*' CO 1- t-lL^ iHL H F, R / \RGIVE A ;^Sk(!'v:-^ il W^ i« $^ p &^ wr. %1 f5^ i^lHKHUHHjaBMlil' CHAR i^g>/\ei;^ig6Hg>g5S>Ai^S5 STEIN ■Tiai-tft^/ftVri.s-ai^'fa i^y^or^ »0(r 1 1* "K. Kr 1 i 'ihr ^'" '" 4IU- ■• -■ . Digitized by the Internet Archive in 2007 with funding from Microsoft Corporation http://www.archive.org/details/argiveheraeum02walduoft THE ARGIVE HERAEUM ^> ^rc{)aeological 3nstitutE of 9linerica American ^ci^ool of Clascal ^tudiejsj at ati^enjs THE ARGIYE HEEAEUM BY CHARLES WALDSTEIN Ph. D., L. H. D., Liti. D. sometime director of the american school of classical studies at athens slade professor of fine art, university reader in classical archaeology, and fellow of icing's college, cambridge sometime director of the fitz william museum, cambridge, etc., etc. WITH THE COOPERATION OF GEORGE HENRY CHASE, HERBERT FLETCHER DE COU, THEODORE WOOLSEY HEERMANCE, JOSEPH CLARK HOPPIN, ALBERT MORTON LYTHGOE, RICHARD NORTON, RUFUS BYAM RICHARDSON, EDWARD LIPPINCOTT TILTON HENRY STEPHENS WASHINGTON, AND JAMES RIGNALL WHEELER / IN TWO VOLUMES VOLUME II TERRA-COTTA FIGURINES, TERRA-COTTA RELIEFS, VASES AND VASE FRAGMENTS, BRONZES, ENGRAVED STONES, GEMS, AND IVORIES, COINS, EGYPTIAN, OR GRAECO-EGYPTIAN, OBJECTS U^jl ^i- n BOSTON AND NEW YORK HOUGHTON, MIFFLIN AND COMPANY tlt^e HibewtDe press;, CambriDge 1905 MICROFILMED BY UNIVERSITY OF TORONTO LIBRARY MASTER NEGATIVE NO.; 9^30/77 • • . . . / 1 . M I . I r T t • I t I f • t t * r T • ? f • COPYRIGHT, 1905 BY THE TRUSTEES OF THE AMERICAN SCHOOL OF CLASSICAL STUDIES AT ATHENS ALL RIGHTS RESERVED Published April, igos NOTE This work is issued by the authority and under the auspices of the Archaeological Institute of America and the American School of Classical Studies at Athens, which share the financial responsibility for it. The supervision of its publication has been intrusted to a Committee which consists of representatives of both bodies. Each contributor has been left entirely free to express his opinions and sentiments regarding the subjects treated by him — even where in these he differed from his asso- ciates— and is therefore solely responsible for the statements made in the articles written by him. Committee on Publication : — JOHN WILLIAMS WHITE, HAROLD N. FOWLER, EDWARD ROBINSON, On behalf of the Institute. THOMAS D. SEYMOUR, --.,^ JAMES R. WHEELER, JOHN H. WRIGHT, On behalf of the School at Athens. PREFATORY NOTE TO VOLUME 11. In the preface to the first volume of this publication I expressed my regret at the delay in bringing- before the world the results of our excavations which were completed in 1895. I also regretted that we were obliged to issue the first volume before the completion of the second. I then thought myself justified in announcing that the second volume would " follow within a few months." Circumstances over which 1 have had no control have extended this period of a few months to what will now in all probability exceed a twelvemonth. As in the case of the first volume I deplore this, not only because it has kept from the archaeological world the material which we have to ofEer for study, but also because full justice cannot be done to my collaborators, especially to Dr. Hoppin and Mr. Chase, whose work has been completed and in type for several years. Quite apart from the question of priority with regard to the antiquities we have excavated as well as the deductions we may have drawn from them, the work of my collaborators may not be jnstly appreciated when considered in the light of more recent discoveries, — I mean those that have been made since our excavations were completed. I have in mind especially the important discoveries made in Crete, Melos, and the more recent excavations of Thera. In any case, the complete and final publication of the work on those sites was not available for our purposes ; while in the case of Crete, notably of Cnossos, the exca- vations themselves have not been brought to a termination, and the material furnished does not present itself in a form admitting of final conclusions and general deductions. Still, it may not seem right that, covering as we do similar fields of inquiry to that of recent work on other Aegean sites, we should not have considered at least the suggestions which they offer in the elaboration of our own material ; and it may not be unfair to criticise us for not having redistributed all the type set years ago and rewritten all that had been written. But I venture to think that the scientific value of the archaeological material which we here offer in our publication, and even of the conclusions which we have drawn from the study of them, is increased by the fact that in our study we have confined ourselves to what was actually before us in our own material, and to the defi- nitely published results of the work that had preceded us. Whether our results as here published are confirmed, rejected, or modified by those of future excavators and inquirers, our exposition can but gain in its claim to trustworthiness from not having been biased by premature side-glances at the foreshadowed results of contemporary exca- vation ; while those who must elaborate the discoveries made since our own will be able to make the better use of ours in their independent presentation, even if this use should be to controvert our conclusions. A similar explanation may be called for with regard to the fact that we have not viii PREFATORY NOTE considered the work done since our excavations were completed on wider questions of historic and prehistoric archaeology, in which our own material must find a place as evidence. These include especially the wide and complicated prohlems of the earliest ethnology of the ancient world, together with the greatly expanded study of Hellenic and primitive mythology. The whole question of the Mycenaean age and the periods that preceded and followed it has entered a new phase, owing to the remarkahle work done hy such writers as Professor Ridgeway and Dr. J. G. Frazer, and to the cognate discoveries and inquiries of explorers and scholars like Professor Flinders Petrie. I have been du-ectly criticised for not having, in my General Introduction in the first volume, clearly taken sides on the main issues raised by such important inquiries. Dr. Rouse, in his review of our first volume,* says : " Dr. Waldstein has not thought well to examine the bearing of this book (Professor Ridgeway's Early Age of Greece) on his Introduction, part of which might have been modified by it. He does indeed mention Professor Ridgeway in a footnote, not without approval ; but there was time to have formed a more definite opinion than he gives on the new theories, whether to accept them or to reject." Now, all that I have urged above regarding the undesirability of entering into wider inquiry beyond the range of our own material in this publication applies still more emphatically to the questions raised by Professor Ridgeway's bold and far-reaching work. I wish in no way to discourage this stirring expansion of the study of classical antiquity into wider spheres, where it establishes its immediate relationship to general ethnology, — nay, to anthropology. The day may come when all our views of the early history of man, including the peoples of Ancient Greece and Rome, will present a picture containing a full and organic composition instead of the smaller groups and single portraits which we now have before us ; and when the whole image of the past in the eyes of our poster- ity may gain in truth and even in clearness. But to my mind this day has not yet come ; because the material for such wider generalization, which we may reasonably hope for in the future, is not yet to hand. Excavations, not only in Egypt, in Assyria, in the whole of Asia Minor, in the northern portions of Ancient Greece, in the islands of the Aegean, in Sicily and in Italy, but also in many other centres of Europe and other parts of the world, will have to yield their treasures of the past, before a great genius, happily arising, is prepared to fuse the disjointed ingots of evidence together, at the slow and steady fire of sober yet imaginative induction, into the great and malleable bar of precious metal out of which the monumental forms of truth are forged and cast. In the mean while, how- ever, truth is advanced by a select few who form legitimate scientific hypotheses in presenting theories by means of which, for the time being, we can in some way grapple with the vast mass of material before us. But in presenting the results of our excava- tions of one site, and in drawing our conclusions from the material before us, we are emphatically not called upon to take sides on problems of much wider issue, the solution of which may ultimately be furnished by our facts and our exposition.^ ' Classical Review, October, 1903, p. 368. opinions tended to support those of Professor Ridgeway, * I think it right to state that wherever my results and they were on my part arrived at quite independently PREFATORY NOTE IX Still, since the first volume was published, the final and official publication of Troja and Hlon has been issued . by Dr. Diirpfeld and his collaborators. Furthermore, the work done and the wonderful results attained by Dr. A. J. Evans and his colleagues in Cnossos of Crete, as well as that of Professor Halbherr and his colleagues at Phaestos, though the excavations are not completed, have led to preliminary publications in which final conclusions have in some instances been drawn which bear directly upon essential points of our discoveries at the Argive Heraeum. As the conclusions thus arrived at by the excavators of Cnossos seem to militate against my main contention with regard to the Argive Heraeum, I feel that I must in a few words touch upon this question, even at the risk of being premature. I claimed preeminent importance for the Argive Heraeum, because, as compared with sites like Troy and other early " classical sites " including Crete, it was one about which there could be no doubt in regard to its intrinsic association from the earliest times with what is to us Hellenic life ; while, on the other hand, in contradistinction to Tiryns and Mycenae, which only present definite periods in the development of Hellenism, the Argive Heraeum was continuously, in all ages about which we know anything, an important centre of that life. I maintained, and I still maintain, that our excavations have yielded material giving undeniable proof that this site was occupied in what is commonly called the Neolithic period. By an oversight the first volume contained no illustrations of stone implements found in our excavations. I wish here to supplement this deficiency by inserting specimens of such Neohthic articles of use. Stone Implements from the Argive Heraeum. From the Neolithic age onward there is abundant evidence of a continuous occupation of that site, be it as a citadel or as a temple, down to Roman times. I wish here to emphasize more strongly than I have done in the first volume that the earUest walls and dwellings, together with the stone unplements, confirm the tradition which I maintained * from him. Any suggestion I received from him has been duly acknowledged ; just as he has acknowledged, both in his article in the Hellenic Journal in 1896, as well as in bis book, the cases in which I enabled him to consider some of our material before publication. We are nei- ther of us responsible for the other's views and conclu- sions. 1 Classical Review, December, 1900, pp. 473 S. X PREFATORY NOTE is given in Bacchylides (xi. 43-84), that the site of the Temple of Hera was also that of the Pre-Proetean citadel of Argos. On every side there is thus evidence of a Pre- Mycenaeau civilization at the Heraeum ; and in tradition, monuments, and minor works we have there an illustration of the systematic origin and organic growth and develop- ment of the Mycenaean civilization and the periods following it. No doubt there is evi- dence (which moreover we should a priori expect) of waves of foreign influence. But the main current of civilization leading up to the Mycenaean age, through it and beyond it, is central and continuous on the site of the Argive Heraeum. As far as our evidence goes, there is no reason for assuming that the Mycenaean civihzation came to the Argolid from without, and that it was not continuously developed in that centre of ancient life. Now, the conclusions which Dr. Evans has drawn from his Cnossian discoveries tend to make Crete the earliest focus of Mycenaean civilization from which it radiated to other parts of the world, including Greece Proper. Mr. Mackenzie, moreover, in a very ingen- ious and able article* has endeavored to prove this definitely by the evidence of the vases foiuid at Cnossos. He maintains that the evolution of the Mycenaean ceramic style out of the Kamares ware, which again is evolved from Cretan Neolithic ware, is to be found in Crete. He further maintains that from Crete it found its way into the Argolid and the other parts of the ancient world. In short, it is maintained that Crete is '• the cradle of the Mycenaean, and hence of Hellenic, civilization." This seems to me improbable. Moreover, if it were true, my own conclusions with regard to the position of the Argive Heraeum woidd be unfounded. In spite of the prematureness of any attempt to arrive at a final decision on the nature of the Cnossian excavations while they are not yet completed, and before the objects there found have been subjected to a thorough examination in their completeness, I wish, in view of the partial publication of such conclusions on the part of the excavators, to enumerate briefly my doubts, and to commend them to the consideration of the Cretan explorers. On the one hand, I do not doubt of the ample evidence as to the early settlement of the Cnossian and other Cretan sites in the Neolithic period, and I view favorably Dr. Evans's suggestion of a chronology reaching back much farther than we have hitherto been wont to conceive, namely, well into the third and even the fourth milleunia b. c. As Dr. Diirpfeld has admitted, the earliest layers of Hissarlik distinctly point to such remote ages ; and I have always expressed my belief that we should have to go much farther back when looking for the beginnings of early Greek civilization. But, on the other hand, we are informed that the objects on the Cnossian site end with the Mycenaean period, that there are no objects belonging to a later date found there, and that therefore the life they reflect antedates that of the sites in Greece Proper, and that the Mycenaean civilization found in the latter is derived from Crete. It is now some years ago ^ since I expressed the view that it is a mistake to believe that the Mycenaean and Geometric fonns of art production ceased in all parts of Greece at the same time, and were superseded in the same period by the archaic forms of > J. H, S. XXIII. 1903, pp. 157 ff. <■ At a public meeting of the Americau School of Athens in 1892. PREFATORY NOTE xi distinctly " Hellenic " character. It then appeared to me that, especially in ceramic art, the familiar polychrome, black-figured and red-figured vases never became the indigenous manufacture of certain portions of the ancient world ; but that the Mycenaean and the Geometric forms of ornamentation as well as the so-called " Proto-Corinthian " (our later Argive-Linear) maintained themselves long into the histoiic period. Of this survival, this persistency of earlier styles, Crete seems to me a most probable instance. Though the Cnossian city reach far back into the remotest antiquity, and nothing later than Mycenaean ware has as yet been found there, we must at least await the completion of those excava- tions and those of other Cretan sites before we exclude the possibility, nay, the probability, that the latest of the " Mycenaean " remains at Cnossos reach far down to the gates of the historic Greek periods when elsewhere in the Hellenic world other Hellenic forms had been developed. The grounds for my doubt in the face of the mass of evidence which is even now adduced by the Cnossian excavators are based upon the intrinsic char- acter of this evidence itself. To begin with the more general point of view (though I cannot enter upon a minute critical estimate of the ethnological " evidence "), the confused tangle of ancient tradition concerning the early inhabitants of ancient Crete, — the Eteocretans, Pelasgians, Achaeans, Dorians, the Carians, Leleges, and their relation to the Lycians, — does not point to a unity of organization and civilization which could have produced the Mycenaean style foreshadowing the typically Hellenic art. Moreover, I cannot feel that the literary traditions concerning the Minoan Cycle in any way warrant us in assigning a central position to it in the formation of Hellenic tradi- tion. In its relation to the established and predominant body of earliest tradition it par- takes of what I should hke to call an " episodical " character. We can conceive of the organism of Hellenic myth and heroology retaining its organic character if this cycle were expunged. An intimate connection with Greece Proper, so as to suggest an essen- tial influence in the form of civilization, is not manifest, certainly not as regards the Argolid. The only real point of union in this sphere is with Attica in the Thesean story ; and here it is of a passing nature, indicating a war which implied no conquest or lasting occupation of the country. On the contrary, the balance of evidence in the tradition of Daidalos points, as regards art and style, rather to an influence from the mainland (in this case Attica) upon Crete, than from Crete on Greece Proper. This is quite different from the wave of influence which we can clearly trace in sculpture in the Daedalids, the so-called Ionian period, when we are at the very gates of the Historic period.^ In fact, I cannot help feeling that the Minoan and Thesean tradition is comparatively late, as the complex elaboration of the tradition concerning the life and deeds of Theseus and his worship bears traces of contamination and points to a late date compared with the body of Greek myths and heroology. There was one period when the influence of Crete upon the mainland of Greece was ' See my article on ' D^dale et I'ArWmis de D^los,' = Cf. Essays on the Art of Pheidias, pp. 254 ff. Rev. Archeologique, December, 1881, pp. 321 S. xii PREFATORY NOTE paramoiuit, when accordingly the life in Crete, especially at Cnossos, must have been most thriving and have manifested itself in active art-production. This is the period which in the history of sculpture we call the Ionian period : when the craft of making teraple-stiitues, the profession of the dyaX/xaTo;roio'? was really estiiblished and introduced from these centres into Greece Proper. According to the literary traditions the chief centres whence this art sprung were Crete, Chios, and Samos. The activity of these schools of sculptors, immediately derived from Daidalos (in my opinion the historical Daidalos), and the most effective transmitters of this skill, were the Cretan Daidalidae, Dipoinos and Skyllis. These we can trace in their journeys through the Greek continent engrafting their craft at Sicyon, through the Peloponnesus, and as far north as the Aetolian Ambracia. Is it not startling to find that of this period, corresponding to the orientalizing wave in ceramic art in Greece, concerning which we have undoubted evi- dence as to the predominant position of Crete, not a single trace should have been found, especially in such centres as Cnossos ? Is this conceivable ? Are we not justified in ask- ing for some reasonable explanation of such a singidar phenomenon ? And may I not suggest that, however early some of the Cnossian remains and the earliest building may be, some parts of the palace, especially its plastic decoration in stucco as well as some of the wall-painting, belong to this later historical period ? We need not expect to find "archaic Greek" objects there, unless they were imported; but the native orientalizing style, which presents a continuous survival of the Cretan-Mycenaean art down to these later historical times, we can expect.^ But let us turn to the more specifically archaeological evidence. (1) The architecture of Cnossos in its " Minoan " periods appears to me to be of a distinctly later date than the Cyclopean structures of Tiryns and of those parts of the Heraeum of the " Tirynthian " period — a fortiori the earlier walls of more primitive settlements. If the Cnossian walls correspond most closely to the sixth layer of Hissarlik- Troy, it stands to reason that the earliest Heraeum walls, which correspond to the first, second, and third layers of Hissarlik, are distinctly earlier. The walls of the Heraeum present a continuous development from these primitive beginnings through all ages to the later Roman period, and thus admit of no break or hiatus.** ' Since I wrote the above, my attention has been drawn fact ? The date of the " later Greek alphabet " we do to a passage which I overlooked in Mr. Evans's Report know. Do not these later Greek objects speak rather on the Excavations of the Palace of Knossos in the Annual in favor of my view, that the Mycenaean style survived of the Brilish School at Athens, vii. (1900-1901), p. 118 ff. in the Palace of Cnossos down to historical Greek times ? In it is an account of tlie find of " bone ' fish ' " upon '' Quite recently Professor Noack (Homerische Paliiste, which were incised signs. " Out of twenty-one varieties, 1903) has pointed out that there is a " contrast " (ein ten are practically identical, both in shape and positioq, Gegensatz) between the elevation and ground-plan of the with later Greek alphabetic forms, wliile four more are " Greek " (Troy, Tiryns, Mycenae, and Arne) and the the i«ame, though in a different position. Thus we have Cretan palaces. It seems strange to me that he should A, A, H, A^, I, n , (*, Y, + and a form approaching the nevertheless consider the Greek palaces to have been in- digamma, etc. . . . Yet the Mycenaean date of these flnenoed by tlie Cretan form, especially when he (pp. bone pieces is as well ascertained as anything found 27 ff.) refers the principle of the hreitstimige Front which within the walls of the Palace." T must ask : What is distinguishes Cretan from " Greek " palaces to the East, the " Mycenaean date " here referred to as an established Syria and Egypt, for its origin. This " oriental " ele- PREFATORY NOTE xm (2) Though the Terra-Cotta figurines from Cnossos have not yet been published sys- tematically, occasional publications have given enough to show that they do not present that full and organic development which Mr. Chase and I have endeavored to demon- strate in this publication. None of the specimens from Crete which I have since seen represent the earliest beginnings as illustrated by our first and second class ; nor do they show the continuity of development of our Tirynthian-Argive class out of its primitive beginnings. The causes for regret at the delay in our publication are exemplified by the fact that in the excellent publication of ancient Terra-Cottas just issued by Dr. Winter * he should have had to rely for the choice of his terra-cottas from the Heraeum upon my preliminary publication of our first year's work pubHshed in 1892, and that he has thus been unable to use for his general exposition the richest find of such objects, represent- ing the most varied classes and presenting instances of such continuous development. In view of what I had put in the first volume ^ on the nature of the typically Mycenaean terra-cottas and their " ceramic " style, he could hardly have given to them a position earlier than our Tirynthian-Argive, still less than our primitive class.^ (3) As regards the pottery from Cnossos, we are already in possession of much fuller publication ; and quite recently Mr. Duncan Mackenzie * in a very thoughtful and able article has made an attempt to give a systematic account of the evolution of Cretan pottery. The discovery of Mr. J. L. Myres established the fact that the pottery to which he gave the name Kamares ware is peculiar to, and characteristic of, Crete. Mr. Mac- kenzie has in a conclusive manner shown how the earliest form of this Kamares ware is immediately developed out of the primitive Neolithic pottery with incised patterns. Upon this follows a class of " early Minoan " ware which, according to him, is the precursor of the Mycenaean ware, out of which, in fact, he maintains the Mycenaean style grew. But this Minoan class presents two different, if not antagonistic, groups : the one monochrome, the other polychrome in its decoration. On his own showing (p. 171) the polychrome was the more favorite style of the two. Now he maintains that these two groups contend with one another, as it were, mingle their influence, the monochrome ultimately winning the day, and that the final result is the establishment of the Mycenaean style. It is on this most important point with regard to the evolution of the Mycenaean style that Mr. Mackenzie's arguments have failed to convince me. For the whole nature of this polychrome style in the Kamares as well as in the Cretan-Mycenaean ware is to my ment (contrasted with the "Greek" element) in the by Mr. Chase, containing the objects from a tholes tomb Mycenaean style of Crete would correspond exactly to at Mycenae in which a Heraeum terra-cotta of our second what I maintain below characterizes the Cretan pottery (Argive-Tirynthian) class was found, to prove that even of the Mycenaean age. our second class can at least not be later than the Myce- 1 Die Antiken Terrakotten, etc., by R. Kekule von naean period. Cf. p. 10^. Stradonitz, vol. III. 1903. * J. Hellen. Stud. XXIII. (1903) pp. 157 ff. Other ^ Vol. I. p. 45. articles dealing with Cnossian pottery have appeared by * When Dr. Winter, /. c. (p. XXX.), says of these earliest Mr. D. G. Hogarth, BrUish School Annual, VI. 1899- classes : Die Entstehungszeit auch der primitiv gebildeten 1900, p. 74 ; by Dr. A. J. Evans, B. S. A. VII. 1900- Stiicke dilrfte kaum sehr weit uber das achte Jahrhundert 1901; by the same writer, B. S: A. VIII. 1901-1902, pp. zuriickreichen, it is enough to point to the tray, examined 88 S. xiv PREFATORY NOTE mind a most characteristic ilifferentiation of Cretan pottery ; one, moreover, which also differentiates it from the essential characteristics of the Mycenaean style as presented to ns in the representjitive vases from the mainland of Greece and other sites. The very first specimens of Mycenaean vases foimd in Crete which were shown to me some time ago gave me the impression that they had local characteristics mixed up with the Mycenaean style which distinguished them from the general run of Mycenaean ware, and which I con- sider foreign to the main feature of Mycenaean ware, in some sense antagonistic to it. For want of a better word I should describe this differential characteristic as " orientalizing." I mean by that term those featiu-es of decoration which are found m a much later age, namely, the " orientalizing " period as illustrated in the Corinthian vases preceding the development of the archaic black-figured vases which are more purely Greek in style. The distinctive decorative features of this " orientalizing " style are a tendency towards the polychrome or piebald in color and towards a redundancy of ornaments in design. Plate XII. vol. XXII. of the Journal of Hellenic Studies will illustrate my meaning. The bronze age vase (No. 1 in the middle) from Zakro is in shape and general ornamen- tation distinctly Mycenaean. But the manner in which the partly naturalistic and other ornaments have been scattered about the whole body and neck of the vase (with a redun- dancy which has led German archaeologists to talk of such as " Filllornamente ") is foreign to the essential characteristics of the pure Mycenaean vase and style. Yet we must not be astonished to find this characteristic in a Mycenaean vase from Crete when it is manifestly already the leading characteristic of the " early Minoan " style, the precursor of the Mycenaean period, in that island. The primitive Cretan painted vases published by Messrs. Hogarth and Welsh in 1901 ^ with the polychrome and characteristic Kamares vases given on Plates VI. and VII. will illustrate this more fully ; while the polychrome cups Nos. 1 and 2 on Plate VI. of Mr. Mackenzie's article will most strikingly illustrate the principle of the " Fiillornamente" at the early stage of Cretan ware preceding the Mycenaean period. Thus, through a careful and sober study of the style in the Cretan remains themselves, without the intrusion of any further theory, I cannot feel on the one hand that this polychrome style could ever lead to the Mycenaean style, and, on the other hand, that the distinctive characteristics which it embodies could ever be entirely expunged from the ware manufactured in Crete at any period of its production. It is to be found in the Mycenaean ware of distinctly Cretan origin, and, in some form or other, the devel- oped Kamares ware manifests a similar character. It is a significant and interesting fact that the place where these characteristics of Cretan style manifest themselves in a strong survival most clearly in a far later period is Naucratis, where we find the main features of the Kamares ware in the ornamentation of the interior of the characteristic Naucratite pottery. With the evidence furnished up to this moment before us, it seems to me more probable that the active influence which won the day in Crete for the monochrome decoration contending with the polychrome style was the introduction of pure Mycenaean patterns 1 /. H. S. 1901, pp. 78 S. PREFATORY NOTE xv which probably came into Crete from elsewhere, and which were never assimilated in their complete purity of character. This course in the history of ceramic decoration seems to me more probable than that the elements of the Mycenaean style were engrafted from Crete upon the ware of those centres, notably the Argohd, where the Mycenaean style has hitherto been supposed to have had its native home. As far as the Heraeum is concerned, there appears to me to be no room for Mr. Mackenzie's hypothesis. For we have there, as 1 have endeavored to trace it,' a continuous and organic development from the incised and early painted ornamentation of the Neolithic ware, through the dull-colored linear ornament with free-hand drawing of the early Mycenaean ware, to the fully developed forms of that style ; and it appears to me that, out of the principle of free-hand drawing with spirals and curved lines and waved patterns in the dull-colored vases, the natural- istic ornament of the period of advanced ceramic art in the Mycenaean age naturally follows. There is no room in the chain of development for the intrusion of the peculiar early Minoan vase. One important point Mr. Mackenzie has succeeded in making highly probable, namely, that the use of glaze-color in vase decoration, which is to be found at the very earliest period in Crete, may have originated there, and may have spread thence to other parts of the Mycenaean area. This is most probable in view of the large nimibers of such specimens found in Crete ; though all sites of the ancient world have not yet been exca- vated. But the introduction of such a technical innovation is readily made from one part of the world into another ; not so the introduction of a characteristic style. The isolated specimens of distinctively Cretan Mycenaean ware which have been found in other parts of the Mycenaean world do not justify the conclusion of such a predominant influence on the part of the Cretan style, when we consider the paucity of these isolated specimens in their proportion to the pure Mycenaean ware found in the Argolid and elsewhere. We must wait for a careful differentiation between the peculiar Mycenaean ware found in Crete and the ware which is purely Mycenaean and corresponds to the predominant class of Mycenaean vases found in the Argohd, and, comparing their relative nimibers, we must then ask whether there is not actual evidence which makes it probable that a great quantity of Mycenaean ware may have been imported into Crete from the Argolid. I may finally add that the evidence from Troy confirms me in this supposition ; for we there find a pronounced and distinctive form of primitive ware with its characteristic material, shape, and ornaments ; while the advent of the Mycenaean style in the sixth layer is foreshadowed in the adoption of spiral ornaments in the previous layer, which, though incised in the predominant system of Trojan vases, manifests a decorative feeling intro- duced and adopted from elsewhere. Dr. Dorpfeld says of these ornaments : ^ " Ebenso neu, aber nicht eigentlich troisch, sind die eingetieften Spiralen, welche in der Zeit der dritten Periode aufkommen." These designs correspond to the usual ornaments as we find them on the earliest dull-colored vases in the beginning of the Mycenaean period. These remarks on the Cretan question in its relation to our discoveries may, as I have 1 Vol. I. pp. 53 ff. ' ^'■"y" """^ ■^"'^' ^^^> P- 2'^^- x^n PREFATORY NOTE siiid brfore, be premature, and I am quite ready to retract the statements here made when fuller evidence warrants such a change. But as it was not proper to ignore the brilliant results and the excellent work of the Cretan excavators while presenting this volume to the public, it was not possible to consider them without giving such a critical estimate of the bearings of their work upon ours. In the preface to the first volume I announced that " the principle which I followed in this publication was to allow each member intrusted with a department as much inde- pendence as possible ; " and that " I have not stood in the way of exjjression of well- founded individual opinion for which each collaborator deserves the credit and retains the responsibility." This clearly applies to Dr. Hoppin's conscientious publication of the Vases. It will be seen from his own statement that there is a point on which we differ, namely, the question of the Argive-Linear ware. But upon examination it will be found that the difference is but slight and not of essential importance. Dr. Hoppin accepts what is, after all, the most important point, which I raised some years ago, and which to my knowledge had not been put before, namely, that the linear ornament (we might call it the geometric system of ornamentation — not to be confused with the later geometric period of Greek ceramics) preceded the naturalistic ornamentation of the Mycenaean vases. As far as I can ascertain, the more recent excavations in Crete and Melos and other ancient sites have strongly supported the view I urged in 1892. I have not yet come across any specimen of early dull-colored Mycenaean vases, nor of Neolithic vases from classical sites, which can be held to contain ornamentation that is not purely linear in character. Moreover, Dr. Hoppin admits that in the more advanced and wheel-turned vases of the Mycenaean period an important part of the ornamentation always remains linear ; while in many of the small vases which he attributes to the Mycenaean period this linear system alone appears. Furthermore, in some small vases which he attributes to the Geometric period, the same principle applies ; while, of course, in the succeeding period his own Argive-Linear, there is nothing but linear ornamentation. He can thus not deny my contention that in the early vases from the Heraeum the linear principle of ornamenta- tion was never absent, from its beginnings in the incised Neolithic ware to the end of the Prehistoric period. Our difference of view thus narrows itself down to what is almost a difference of words. In view of the continuous presence of a certain system of ornament, its predominance at the earUest period, as well as at the latest, I have maintained that there was thus in the vase decoration a continuous Argive-Linear system at the Heraeum, the pedigree of which goes back to the Neolithic period, and the flourishing day of which is to be found in the later Argive-Linear period, formerly called Proto-Corinthian. Dr. Hoppin believes that my later Argive-Linear ware marks the degeneration of the Mycenaean system, and he proposes to restrict the term Argive-Linear to that later development. I am quite willing and quite ready to grant the main point of Dr. Hoppin's observation, namely, that the linear principle could not have asserted itself so strongly in the later period if the naturalistic element, predominant in the Mycenaean period, had not lost its vitality. But in view of the fact that in none of the previous periods had the linear PREFATORY NOTE xvu principle of decoration been eliminated, it cannot be correct to imply that there is any new departure in its more exclusive assertion when the Mycenaean style lost its vitality. Nor can it be correct to maintain that in this later period it originated out of the decline of the Mycenaean system immediately preceding it, when it was the predominant principle in the period preceding the Mycenaean, out of which this grew. The conscientious work that Mr. DeCou has bestowed upon the publication of the Bronzes will be manifest to all readers. But they cannot realize the infinite care and persevering labor which, before he wrote, he bestowed for several years upon the cleaning and sorting of the confused mass of shapeless material with which he had to deal. He has chosen to avoid drawing any wider conclusions from his facts, — a course which is thoroughly justifiable and has much in its favor. But we should have also been grate- ful to him if he had given us the fruits of his experience in further historic and archaeo- logical appreciation of his material, seeing that there is hardly any archaeologist alive who has had more experience in dealing with ancient bronzes than he. Professor R. Norton's work is of high interest. If he is right in his classification, there are two distinct classes of engraved stones preceding the distinctly Mycenaean stones. This would furnish very strong evidence of artistic activity on our site before the developed Mycenaean period. Moreover, the numerical preponderance of the " new type Argive " stones over the Mycenaean stones is significant. I woidd suggest as an explanation of the comparative paucity of Mycenaean stones on this site, that this class of gem was not commonly used as a votive offering. They are thus generally found in graves, and were used in life as ornaments or amulets. At the same time, the presence of comparatively large numbers of stones on the Heraeum site antedating the Mycenaean period might be a confirmation of our view that in these earlier times the site was a citadel and not merely a sanctuary, as it became in the later period. Mr. Lythgoe's publication of the Egyptian objects confirms the views he previously expressed, and which are embodied in the General Introduction in the first volume.* King's College, Cambridge, February 28, 1904. CHARLES WALDSTEIN. * Since this has been in type, and nearly two years and his school. . . . The head of Hera and most of the after Volume I. of this publication appeared, a review sculptures from the Heraeum known to me are certainly is published in the Berliner Philologische Wochenschrift Attic." While denying all my views in his review, he (June 25, 1904, pp. 811 ff.) by Professor Furtwangler inserts the phrase that two of the metope heads " may Je which is a caricature of all that a fair review ought to be. admitted to have something of Polycleitan style," What This I answered in No. 39, September 24, of the same does this retreat from his former position mean ? That journal. To all who are acquainted with the nature of two heads of this chief Argive temple, in which Polyclei- the controversies carried on at present between us his tus, the leader of the Argive school of sculpture, made action will be intelligible. In this long review he passes his famous statue, are admitted by him to be Polycleitan over the most important question of the Polycleitan char- in character. If any of the metopes show Polycleitan acter of the Heraeum sculptures — the main subject of style, the leading art of the place was present in the controversy between us — with a few lines. Readers of sculptured decorations. As a matter of fact, with minor Volume I. (pp. 184 ff., and elsewhere) will remember differences among each other (such as the Parthenon how fully 1 dealt with the question of the Polycleitan marbles and all other temple-sculptures show), all these character of the sculptures, and how I controverted in Argive sculptures from this temple are of the same style, detail Professor Furtwiingler's dogmatic statement that C. W. " all these sculptures haue not the least relation to Polycleitus January 30, 1905. CONTENTS THE TERRA-COTTA FIGURINES BY CHARLES WALDSTEIN AND GEORGE HENRY CHASE PAGE Introduction „ 3 I. Pbimitive Argive 16 II. TiBTNTHIAN ARGIVE 17 Group 1 : Sliglitly Developed Plastic Ornamentation 17 Group 2 : Considerably Developed Plastic Ornamentation 18 Group 3 : Elaborate Plastic Ornamentation 19 III. Mycenaean 22 IV. Geometric 23 V. Advanced Argive 24 VI. Figures under Oriental Influence 28 VII. Early Archaic 29 Group A : Hand-made Figures 29 Group B : Mould-made Figures 31^ VIII. Advanced Archaic 38 IX. Free Style 39 X. Animals 39 XI. Various Objects 42 THE TERRA-COTTA RELIEFS BY CHARLES WALDSTEIN AND JOSEPH CLARK HOPPIN Introduction 47 Reliefs Nos. 1-10 49^ THE VASES AND VASE FRAGMENTS BY JOSEPH CLARK HOPPIN Preface "' Abbreviations Employed •'^ Introduction "" Chapter I. Primitive Vases ^^ Chapter II. The Mycenaean Style 71 Vases with Dull Decoration 72 Fragments '^ Vases ' Vases w^ith Lustrous Decoration <4 Class I., Not represented '4 Class II., Division 1 78 Class II., Division 2 79 Class III., Division 1 ^1 Class III., Division 2 82 Flower Buds ^^ Circles, Diamonds Spiral Forms and their Development °4 XX CONTENTS Flower Patterns 85 Flower Branches 85 Murex, or Purple Fish Patterns 86 Linear and Textile Patterns 86 Class IV 89 Birds, Animals, and Human Figures 90 The Bee-Hive Tombs near the Hekaeum 91 Jugs with Handle 92 Aniphoras 93 Three-handled Vases 94 Teapot-shaped Jugs 94 Pseud-amphora 95 Kylixes 95 Cups 95 Miscellaneous Vases 96 Saucers 96 Bowls 98 Jugs 99 I. Without Decoration 99 One-handled Jugs 99 Th?ee-handled Jugs 100 II. With Decoration 100 Two-handled Vase 101 CuAPTEB III. The Geometric Style 101 Class I. Transitional Stage 104 Class II. Purely Geometrical Types 106 Class III. The Introduction of Human and Animal Motives 107 A. Birds 108 B. Horses 109 C. Men Ill D. Women 114 E. Miscellaneous 115 Other Miscellaneous Types 115 Chapter IV. The Argive Style 119 Class I. Early Argive 124 Lekythoi 124 Kalathoi 124 Saucers 125 Class II. Linear Argive 126 Lekythoi 126 Oinochoai 128 Skyphoi 132 Pyxides 136 Covers 139 Knobs 140 Plates 141 Miscellaneous Forms 143 Kothons 143 Ring-form Aryballos 143 Jugs 143 Openwork Vases 144 Class III. Oriental Argive 144 Lekythoi 145 Oinochoai 148 Skyphoi 151 Pyxides 153 Jugs 163 CONTENTS xxi Plates J54 Alabastra ]^55 Aryballoi I55 Note on the Use of Applied Color in Certain Styles 156 Mycenaean 156 Dull Finish 156 Lustrous Finish, Class III 156 Geometric I57 Argive 157 Polychrome Vases, or Decoration on a White Ground 158 Chapter V. Miscellaneous Types 159 Chapter VI. The Corinthian Style 165 Old Corinthian Style 167 Skyphoi 167 Oinochoai 168 Pyxides 168 Covers 169 Sugar-bowl Form 170 Tripod Bowls 170 Alabastra 171 Amphora 171 Aryballoi 171 Askos 171 Kylikes 172 Plates 172 New Corinthian Style 172 Foreign Types 173 Early Attic 173 Cyrenean 173 Chapter VII. Black- and Red-Figure Styles 174 Black-Figure Style 176 Kylikes 176 Skyphoi 177 Tripod Bowl 177 Amphora 178 Lekythos 178 Plate 178 Polychrome Ware 178 Eed-Fiqure Style 178 Severe Style 17" Later Style 179 Polychrome Ware 1'" Chapter VIII. Vases in Relief and Later Vases 180 Red Ware 180 Megarean Ware ^"^ Lamps 18^ Glass 184 Byzantine 1°'* I. Painted Inscriptions II. Incised Inscriptions . INSCRIPTIONS ON VASES BY THEODORE WOOLSEY HEERMANCE 186 185 xxii CONTENTS ' THE BRONZES OF THE AKGIVE HERAEUM BY HERBERT FLETCHER DE COU IXTRODUCTION 191 CATALOdUK 19"^ I. Fkaumexts of Statues, Nos. 1, 2 194 II. Statuettes and Fkaumexts of Statuettes, Nos. 3-7 194 III. Animals (free and attached), Nos. 8-i)l 197 1. Horses, Nos. 8-18 197 2. Deer, Nos. 19-21 200 3. Sheep, No. 22 201 4. Cattle, Nos. 23-28 201 5. Lions, Nos. 29, 720, 946, 2204 203, 235, 249, 293 6. Rodent, No. 30 203 7. Frog, No. 31 203 8. Serpents, Nos. 32-35 204 9. Birds, Nos. 36-51 204 IV. Ornament and Toilet, Nos. 52-1849 207 1. Pins, Nos. 52-948 207 A. Straight, Nos. 52-807 207 B. Safety, Nos. 808-946 240 C. Pseudo-safety, Nos. 947, 948 249 2. Needle, No. 949 250 3. Rings, Nos. 950-1546 250 A. Finger-rings, Nos. 950-970 h 250 B. Bracelets, Nos. 971-974 251 C. Decorative, Nos. 975-1524 251 D. Structural, Nos. 1525-1540 263 E. Pieces of wire twisted into the form of rings, Nos. 1541-1546 264 4. Beads, Nos. 1547-1552 264 5. Pendants, Nos. 1553-1559 264 6. Mirrors and Mirror-handles, Nos. 1560-1589 a 264 7. Objects in Sheet Bronze, Nos. 1590-1846 266 A. Diadems, Nos. 1590-1599 266 B. Disks, Nos. 1600-1718 b 267 C. Binding strips, Nos. 1719-1747 269 D. Ornamented bands, Nos. 1748-1751 270 E. Coatings, Nos. 1752-1796 270 F. Rims, Nos. 1797-1800 272 G. Plates, Nos. 1801-1827 272 H. Cut ornaments, Nos. 1828-1832 a 274 I. Figures, Nos. 1833-1838 274 J. Uncertein objects, Nos. 1839-1846 274 8. Leaves, Nos. 1847, 1848 274 9. Leather gear, No. 1849 275 V. Vessels, Nos. 1850-2251 275 1. Vases and Cauldrons, Nos. 1850-2217 275 A. Without shoulder, Nos. 1850-2019 275 B. With shoulder, Nos. 2020, 2021 286 C. With shoulder and neck, Nos. 2022-2047 286 D. Handles, Nos. 2048-2207 288 E. Connecting rods, Nos. 2208-2214 294 F. Ornaments, Nos. 2215-2217 294 2. Tripods and Standards, Nos. 2218-2234 294 A. Fragments of large tripods, Nos. 2218-2224 294 B. Low trii)od standards, Nos. 2225-2230 295 0. Feet of standards, Nos. 2231-2234 296 CONTENTS xxiii 3. Miscellaneous Forms, Nos. 2235-2251 296 A. Containel-s, Nos. 2235-2244 296 B. Handles, Nos. 2245-2251 297 VI. Implemknts, Utknsils, Stkuctubal Pieces and Materials, Nos. 2252-2841 .... 298 1. Parts of Vehicles, Nos. 2252-2255 298 A. Fragment of antyx, No. 2252 298 B. SmaU wheels, Nos. 2253-2255 298 2. Strigil, No. 2256 299 3. Bell, No. 2257 299 4. Cymbals, Nos. 2258-2261 299 5. Key, No. 2262 299 6. Knives, Nos. 2263-2272 299 7. Spits, Nos. 2273-2711 e 300 8. Spear-butt, No. 2712 323 9. Pestle or Plektron, No. 2713 324 10. Hooks, Nos. 2714-2728 324 A. Free, Nos. 2714-2722 324 B. Attached, Nos. 2723-2728 324 11. Nails, Nos. 2729-2764 324 12. Button, No. 2765 327 13. Fasteners, Nos. 2766-2772 327 A. Wire, Nos. 2766-2769 327 B. Chain, Nos. 2770, 2771 327 C. Eyelet, No. 2722 328 14. Clamps, Nos. 2773-2776 328 15. Caps, Nos. 2777-2782 328 16. Centre-piece, No. 2783 328 17. Binding, No. 2784 328 18. Decorative Braces and Strips, Nos. 2785-2808 328 A. Straight applied braces, Nos. 2785-2787 328 B. Curved applied brace, No. 2788 329 C. Strips and rods, Nos. 2789-2808 329 19. Inserted Disks, Nos. 2809-2812 a 330 20. Other Decorative Attachments and Fragments, Nos. 2813-2816 330 21. Supports, Nos. 2817-2818 a 330 22. Rods, Nos. 2819-2821 330 23. Bars, Nos. 2822-2827 330 24. Ball, No. 2828 331 25. Castings, Nos. 2829-2841 331 A. Partially formed objects, Nos. 2829-2835 331 B. With accidental resemblances to various objects, Nos. 2836-2839 331 C. Formless, Nos. 2840, 2841 331 Appendix: Insckiptions on the Bronzes 332 ENGRAVED STONES, GEMS, AND IVORIES BY RICHARD NORTON The Stones and Gems 343 Class I. Prehistoric or Primitive 345 CLA.SS II. New Type — 'Argivb' 346 Class III. 'Geometric' ^^ Class IV. Mycenaean ^^^ Class V. Nondescript The Ivories Articles of Apparel or of Household Use 353 Human or Animal Figures Miscellaneous Objects xxiv CONTENTS f COINS FROM THE AKGIVE HERAEUM BY HERBERT FLETCHER DE COU IxTRODUcnoN 357 Cataloguk 357 A. Greece (Autonomous and Municipal) 367 Addenda to Coins of Argos 359 B. Rome (Empire) 361 C. Rome (Eastern Empire) 362 D. Venice 363 E. Uncertain 363 EGYPTIAN, OR GRAECO-EGYPTIAN, ORTECTS BY ALBERT MORTON LYTHGOE Introduction ■ 367 Catalogue 370 Index 375 ILLUSTRATIONS IN THE TEXT PAOE SiojfE Implements ix Tekra-Cotta Figukines. — Primitive Argive Type 16, 17 Terba-Cotta Figurines. — Tiryntliian Argive Type, Group 1 17, 18 Terba-Cotta Figurines. — Tiryntliian Argive Type, Group 2 19 Tebba-Cotta Figubines. — Tiryntliian Argive Type, Group 3 19-22 Terra-Cotta Figurines. — Mycenaean Types 22, 23 Tebra-Cotta Figurines. — Geometric Types 23 Tebba-Cotta Figurines. — Advanced Argive Types 24-27 Teeba-Cotta Figubines. — Under Oriental Influence 28, 29 Terra-Cotta Figurines. — Early Archaic : Hand-made 29-31 Tebra-Cotta Figubines. — Early Archaic : Mould-made 31-38 Terra-Cotta Figurines. — Advanced Archaic 38, 39 Tebba-Cotta Figurines. — Free Style 39 Terra-Cotta Figurines. — Animals 40, 41 Terba-Cotta Figubines. — Various Objects 42-44 Vases and Vase Fbagments. — Primitive Vases 70, 71 Vases aud Vase Fbagments. — Mycenaean Style : Dull Decoration 73, 74 Vases and Vase Fbagments. — Mycenaean Style : Lustrous Decoration, Class II. 1 ... . 79 Vases and Vase Fbagments. — Mycenaean Style : Lustrous Decoration, Class II. 2 ... . 80 Vases and Vase Fbagments. — Mycenaean Style : Lustrous Decoration, Class III. 2 . . 83, 84, 87, 89 Vases and Vase Fragments. — Mycenaean Style : Lustrous Decoration, Class IV 90, 91 Vases and Vase Fragments. — Mycenaean Style : Contents of Bee-Hive Tomb 92-95 Vases and Vase Fragments. — Mycenaean Style : Miscellaneous Vases 96-101 Vases and Vase Fragments. — Geometric Style : Amphora, etc 117, 118 Vases and Vase Fragments. — Early Argive 124, 125 Vases and Vase Fragments. — Linear Argive 126-143 Vases and Vase Fbagments. — Oriental Argive 146,147,151,155 Vases and Vase Fbagments. — Miscellaneous Types 159, 161 Vases and Vase Fbagments. — Old Corintliian Style 171,172 Vases and Vase Fragments. — New Corinthian Style 172 Vases and Vase Fragments. — Local Style 176 Vases and Vase Fragments. — Glass Vase 184 Vases and Vase Fragments. — Byzantine Ware 184 Inscriptions on Vases. — Painted Inscriptions 18>> Inscriptions on Bronzes "^^^i •^•^' > ''•'8 Engraved Stones. — Argive Style "^8, 349 Engraved Stones. — Geometric Style "^^ Engraved Stones. — Mycenaean "'''" Engraved Stones. — Nondescript ''•*" Ivories 351,352,354 Archaistic Scarab of Thothmes III. from Eleusis 369 ScARABEOiD from Eleusis. Probably of Phoenician Origin 372 Porcelain Whorl '^"' PLATES •»• The descriptions of the Plates begin at the pages indicated, but the Plates themselves are grouped and placed as follows : Plates XLII.-XLViri, after page 44 ; Plate XLIX., after page 54 ; Plates L.-LXIX., after page 188 ; Plates LXX.-CXXXVIl, after page ^40 ; Plates CXXXVIII.-CXL., after page 354 ; Plates CXLL, CXLIL, after page 3tU ; and Plates CXLIII. and CXLIV., after page 374. . *«* Plates I.-XLI. are in Volume I. Plate XLII. XLIII. XLIV. XLV. XLVI. XLVII. XLVIII. XLIX. L. LI. LII. LIII. LIV. LV. LVI. LVII. LVIII. LIX. LX. LXI. LXIL LXIII. LXIV. LXV. LXVI. Lxvn. LXVIIL LXIX. LXX. LXXI. LXXII. LXXIII. LXXIV. LXXV. LXXVI. LXXVII. LXXVIII. LXXIX. LXXX. LXXXI. LXXXII. Paoe Terra-Cotta Figurines. — Primitive and Tirynthian Argive Types : Standing Figures 1 6 Terra-Cotta Figurines. — Tirynthian Argive Type : Seated Figures 17 Terra-Cotta Figurines. — Advanced Argive Type 24 Terra-Cotta Figurines. — Advanced Argive Type : Heads 26 Terra-Cotta Figurines. — Early Archaic Type 33 Terra-Cotta Figurines. — Early Archaic, Advanced Archaic, and Free Style Types : Heads 31, 37 Terra-Cotta Figurines. — Animals 23, 29, 39 Terra-Cotta Reliefs 47 Vase Fragments. — Primitive Vases 69 Vase Fragments. — Mycenaean Style : Dull Finish, and Lustrous Finish, Class II. 1 72 Vase Fragments. — Mycenaean Style : Classes II. 2, and III. 1 79 Vase Fragments. — Mycenaean Style : Class III. 2 83 Vase Fragments. — Mycenaean Style : Class HI. 2 85 Vase Fragments. — Mycenaean Style : Classes III. 2, and TV 87 Vase Fragments. — Geometric Style : Linear Motives, Horses, Birds, etc 105 Vase Fragments. — Geometric Style : Human and Animal Figures Ill Vase Fragments. — Geometric Style : Dipylon and Aeginetan 115 Vase Fragments. — Argive Style 128, 130, 134, 138, 141, 148, 151, 152 Vase Fragments. — Local Style 156, 159, 164 Vase Fragments. — Corinthian Style 167 Vase Fragments. — Early Attic, Cyrenean, Red-Figure, Black-Figure, and Megara Ware, etc 173, 177, 183 Vase Fragments. — Vases in Relief 180 Vase Fragments. — Various Argive Types (in color) 91, 134, 145, 147 Vase Fragments. — Argive Style : Class III. (in color) 135, 150, 153, 158 Vase Fragments Geometric, Argive, and Local Types (in color) . 153, 154, 157, 165 Vase Fragments. — Local Type (in color) 161 Vase Fragments. — Fragments of a Polychrome Kylix (in color) 179 Vase Fragments. — Inscriptions Incised on Vases 185 Bronzes. — Fragments of Statues, Statuettes 194 Bronzes. — Statuette Bronzes. — Animal Figurines Bronzes. — Animal Figurines Bronzes. — Animal Figurines Bronzes. — Animal Figurines Bronzes. — Animal Figurines Bronzes. — Animal Figurines 205, 207 Bronzes. — Animal Figurines, Straight Pins 205, 206 Bronzes. — Straight Pins Bronzes. — Straight Pins Bronzes. — Straight Pins Bronzes. — Straight Pins .• * • 194 197 199 200 201 203 210 215 220 224 XXVIJI » Lxxxni. LXXXIV. LXXXV. LXXXVI. LXXXVII. LXXXVIII. LXXXIX. XC. XCI. XCII. XCIII. XCIV. xcv. XCVI. XCVII. XCVIII. XCIX. c. CI. CII. cm. CIV. CV. CVI. evil. CVIII. CIX. ex. CXI. CXII. CXIII. CXIV. cxv. CXVI. CXVII. CXVIII. CXIX. CXX. CXXI. CXXII. CXXIII. CXXIV. cxxv. CXXVI. CXXVII. CXXVIII. CXXIX. CXXX. CXXXI. CXXXII. cxxxni. CXXXIV. CXXXV. CXXXVI. CXXXVII. CXXXVIII. PLATES Broxzes. — Straight Pins ,• ""^ Bkoxzes. — Straight Pins, Safety Pins 235 Bkonzks. — Safety Pins 240 Bkoxzes. — Safety Pins 242 Bkoxzes. — Safety Pins •^^^ Bkoxzes. — Safety Pins, etc. ; Needle, Finger-Rings 247 Bkoxzes. — Finger-Rings and Bracelets 250 Bkoxzes. — Decorative Rings 2i)2 Bronzes. — Decorative and Structural Rings 259 Beoxzes. — Structural Rings, etc. ; Beads, Pendants, Mirrors 263 Bkoxzes. — Mirrors 265 Bronzes. — Mirrors 265 Bronzes. — Mirrors 265 Bkoxzes. — Mirrors 265 Bkoxzes. — Mirror 266 Broxzes. — Mirrors 266 Broxzes. — Diadems, etc. ; Disk 266 Broxzes. — Disks 268 Broxzes. — Disks, Binding-Strips 269 Bronzes. — Binding-Strips, Ornamented Bands, Coatings 270 Bronzes. — Coatings 271 Bronzes. — Coatings, Rims, Plate 271 Bronzes. — Plate, Cut Ornaments 272 Bronzes. — Plate with Inscription. From Photogi-aph 273 Bronzes. — Plate with Inscription. From Drawing 273 Bronzes. — Figures, etc. ; Leaves, Thongs, Plates 274 Bronzes. — Plates, Platters 275 Bronzes. — Plates with Inscriptions 277 Bronzes. — Plates, Platters 277 Bronzes. — Platters, Saucers 278 Bronzes. — Saucers 279 Broxzes. — Saucers 281 Bronzes. — Saucers, Bowls 283 Bronzes. — Bowls 284 Bronzes. — Bowls, Suspension Vase, etc. ; Pitcher, Jars 285 Bronzes. — Jars, Basin 286 Bronzes. — Cauldron, Bosses, Horizontal Handles 287 BRONZE.S Horizontal Handles 288 Bronzes. — Horizontal Handles 289 Bronzes. — Horizontal and Vertical Handles 292 Bronzes. — Vertical Handles, Various Ornaments, Fragments of Large Tripods . . 293 Bronzes. — Fragments of Large Tripods, Small Tripod-Standards 294 Bronzes. — Feet of Standards, Miscellaneous Vessels 296 Bronzes. — Miscellaneous Vessels and Handles, Various Implements and Utensils . . 297 Bronzes. — Various Implements, Sjjits 299 Bronzes. — Spits 302 Bronzes. — Spits 309 Bronzes. — Spits 313 Bronzes. — Spits 315 Bronzes — Spits 319 Bronzes. — Spits, Various Implements and Utensils 321 Bronzes. — Various Utensils and Constructive Pieces 326 Bronzes. — Constructive and Decorative Pieces 329 Bronzes. — Castings 331 Bronze-s. — Miscellaneous Addenda . . . 199, 211, 215, 217 n., 251, 263. 266. 268, 269 Engraved Stones. — Classes I., II 345, 347 PLATES xxix CXXXIX. Engraved Stones. — Classes III.- v. — Ivories 349,361 CXL. Ivories. — Articles of Ajjparel or of Household Use, Human or Animal Figures, Miscel- laneous 352 CXLI. Coins. — Mainly Greek 357 CXLII. Coins. — Mainly Roman 360 CXLIII. Egyptian, or Graeoo-Ecjyptian, Objects 370 CXLIV. Egyptian Objects. — From the Heraeum and from Aegina 373 THE TERRACOTTA FIGURINES THE TEREA-COTTA FIGURINES By CHARLES WALDSTEIN and GEORGE HENRY CHASE The great bulk of the Terra-cotta Figurines found at the Heraeum belong to what must be called pre-archaic Greek types. When we consider the find as a whole, we dis- cover the following significant facts : — I. The number of terra-cottas dating from the clearly historical periods of Greek civilization, i. e. the eighth century b. c. and later times, is very small. Out of 2865 figures preserved by us, not counting the ordinary figures of animals and small objects, which cannot be accurately classified, there are 2557 of pre-archaic style to 308 of archaic and later types.^ II. Among the terra-cottas of the archaic and later periods, a strikingly small pro- portion belong to the periods of advanced archaism and technical freedom ; in fact, there are only fifty-two specimens of the former and two specimens of the latter. Thus it will be seen that the closest parallelism exists between the main conditions of the find in terra-cottas and those of the vases, bronzes, and other objects ; namely, the absolute preponderance of " prehistoric " material, the absence of works of the later Greek periods, and the fact that in the layers which we excavated, hardly any object is syn- chronous with the erection of the Second Temple or of a later period.^ The bulk of the find in terra-cottas was made in the black layer about and below the foundation walls of the Second Temple, as well as in the deposits on the southeast and southwest slopes of the Second Temple terrace.^ But this general statement must be qualified by the following significant facts : — 1. We can assert as practically established the fact that no terra-cotta figurines were found beneath the surface of the upper or Old Temple platform. 2. Nearly all the specimens of the archaic and subsequent periods were found on the southern slope and the southeast corner and slope of the Second Temple (not in the black layer about the foundations of the Second Temple). 3. No prehistoric Greek specimens were found within the large West Stoa (one speci- men of a draped female figure, belonging to the archaic period, was found here), and none at all on the site of the Roman Building. The fact that no terra-cottas were found on the site of the Old Temple, as it jyre- sented itself to us for excavation, may lead us to any one of three conclusions : (1) that our terra-cottas belong to a later period than that marked by the site ; or (2) that the custom of offering such figurines did not exist at the time when the temple was built — ' Cf . the Table, p. 9. elsewhere (cf. Homolle, ' Donarium ' in Daremberg et 2 Cf. General Introd. vol. I. pp. 37 fE. Saglio, Diet, des Ant. Gr. et Rom. II. pp. 363 ff., esp. 8 The custom of burying the smaller objects dedicated p. 371, note 130). In the case of the Heraeum, these in temples, often apparently with great care (cf. the find offerings seem to have been used as "dry rubbish" in of terra-cottas made by Orsi at Terravecchia near Gran- leveling up about the Second Temple, just as the offer- michele in the Province of Catania, Mon. Ant. VII. pp. ings upon the Acropolis which were demolished by the 201-274, esp. pp. 212 ff.), is one which has been illustrated Persians were later used in leveling that site, by a large number of excavations upon Greek soil and 4 THE TERRA-COTTA FIGURINES that itfwas an aniconic period ; or (3) that the site in question was not the natural reposi- tory for such votive offerings. We are inehned to one of the latter views.* The two other points, while they show that the main sites excavated about the Second Tenijjle were pre-archaic Greek, and that the Argive period of the Heraeuni marks the change in the entrance from the west side to the south slope, also show that the main bulk of our terra-cottas belong to the pre-archaic period. With such a mass of material before us, even in the selected specimens which we brought to Athens, our first care was to find some principle of classification. Two courses seemed open to us : We might classify our figures either according to their f(nm and style, (u- accorduig to their decoration. On mature consideration, the former method has appeared preferable, especially as a classification by decoration must confine itself almost entirely to plastic decoration — the paint having disaj)peared entirely from many of our figures. Moreover, no systematic chronological development could be estiiblished on the decora- tive pruiciple only. At the sjime time, the principle of classification by decoration has seemed to us a sound one, and we have employe^ it as a secondary principle for dis- tinguishing a few classes," and for drawing distinctions within one dass.^ Class I. — Such, then, being our j^rinciple, our first class, which we have called " Prim- itive Argive," consists of a great number of very rudely modeled figures, to which, in fact, the term " modeled " can hardly be applied. They are really nothing but a small liunp of clay, elongated, then pinched in the middle to indicate a waist, the residting projection above forming two stumpy aims, while another pinch at the top indicates the head. This head soon develops into the well-known bird-like head, in which the beak-like nose is formed by a compression of the clay between the two fingers of the " modeler." * These are the characteristics of the earhest specimens.^ Later, two disks are plastered to the sides of the " beak " to represent the eyes," a lump of clay is added to form a kind of pilos or cap,' or a strip of clay wound about the head fonns a stephane,* while in many figures one or more short strips of clay are added to the back of the head to represent the hair * (exceptionally the hair is indicated in a few early specimens by grooves at the back of the head). So one sometimes finds — though this is quite exceptional — a fig- ure in which the artist has tried to imitate nature more closely, in which two lumps of clay are added to the front of the figure to represent the breasts ; '" the fall of the dra- pery in front is rudely reproduced by one or more grooves," or a cutting across the " beak " forms a rude mouth. '^ But even in the most advanced figures of this class, the development is exceedingly slight, and the body always remams essentially a rude lump of clay." Class II. — A distinct advance is made when the rude standing figure is converted into a seated one.'* The technical change, indeed, is not great ; the standing body is simply ' Cf. General Introd. vol. I. pp. .39 ff. ' Cf. No. 3. '^ Such, for instance, as the Mycenaean and the Bipy- * Cf. Nos. 4, 6, etc. Ion classes, in which the influence of vase painting is ' Cf. No. 4. very apparent. lo Cf. No. 8. " Cf. p. 11. 11 Cf. No. 9. * It should be noted that this lack of modeling is '^ Cf. Nos. 9 and 10. partially accounted for by the fact that nearly all our '' We might have subdivided this first " Primitive " figures are female (cf. p. 13, note 6). The drapery of class into three minor classes, according to the differences Buch figures in the early stages does not admit of fine dis- enumerated above. But we find it desirable at this tinctions in modeling, such .as the indication of the legs, stage not to confuse by too minute classification, leaving the torso, and its articulations. it to future inquiry to accentuate these, or even to estab- Cf. No. 1. lisli further distinctions. • Cf. Nos. 2, 3, 5, etc. " Cf. Nos. 12 ff. PRINCIPLES OF CLASSIFICATION 6 bent at the middle and supported by two stumpy legs,' a process whieli prevails down to the latest period of this class. Even in the most advanced specimens, the employment of a separately made chair as a support is rare. But although this change is of little impor- tance technically, it is of great importance historically. It marks the change in the con- ception of the goddess which is represented in the traditions of Argos by the story of Pirasus and his seated figure of Hera dedicated at Tiryns." Now it is a striking fact that, aside from the Heraeum and Tegea (of which we shall speak later), Tiryns is the place where these seated figures have been found in greatest nimibers. We have there- fore ventured to call the figures of this class " Tirynthian Argive," and we believe that they represent roughly the period of the supremacy of Tiryns in the Argive plain.^ In these " Tirynthian Argive " figures, we have a long and complicated development of the drapery ; while the figures themselves, although constantly increasing in size, remain, with very few exceptions, a mere oblong piece of clay, bent forward into the seated posture, and the heads still have the " bird face." This development of the drapery even affects the standing type, and we find in our later standing figures a tend- ency to elaboration which corresponds closely to the development in the seated types. The various steps in this development are exceedingly interesting, and we shall revert to them later. Suffice it here to say that, for practical purposes, we have divided our " Tirynthian Argive " class into three subordinate groups, basing our distinctions upon the greater or less elaborateness of the ornamentation.'' The next step in the development of this type of terra-cotta figurines is the change from the " bird face " to the human face, with all the features carefully distinguished. But before we take up this class of " Advanced Argive " figures, we must speak of two intermediate classes, which we have set apart by reason of their decoration as well as their form and style. We mean the Mycenaean and Geometric classes. Class III. — The " Mycenaean " terra-cottas differ from those of the Primitive and Tirynthian Argive classes in that they show a quite different technique, both in regard to the fineness of the clay and in regard to ornamentation. A careful comparison of the Mycenaean figures with the preceding classes, moreover, brings out the following facts : (1) They are numerically very weak — there are only sixty-three fragments of Myce- naean types among our selected terra-cottas, and of these ten are fragments of animals.'^ (2) Frima facie, the schematic outline of these figures, ending below like the foot of a vase, and above with the ;;o/o.s-, like the neck of a vase," the whole crossed by round or pelta-shaped body and arms,' is much less naturalistic than the figures of the primi- tive and Tirynthian Argive classes. On the other hand, the detail workings of the body, such as the indication of the breasts," show far greater knowledge and skill in modeling than we find in Classes I. and II. Both these facts would seem to indicate that we are dealing with a later class of figures. Moreover, the color decoration shows the same use of glaze color, applied in a similar manner, that we find in Mycenaean vases. We thus consider this class of terra-cottas an intrusion into the general and con- tinuous scheme of Argive terra-cottas, which began and ceased during the period of our first and second classes, but which, strangely enough, was but very little influenced by the Primitive and Tirynthian Argive styles, and exercised Httle influence on them." 1 As it happens, the most primitive specimen of the » Cf. the Table, p. 9. seated type which we found (No. 12) is provided with a » Cf. No. 67. separate four-legged chair, but this is exceptional. ' Cf. Nos. 65, 67, 70. a Cf Pans II. 17. 5, and General Introd. vol. I. p. 44. « Cf. especially No. Go. « Cf. General Introd. vol. I. p. 44. ' See, however, our remarks on No. 70. * Cf. p. 11. 6 THE tp:rra-cotta figukines The peculiarities of this Myeeuaeau style, we believe, are due to the fact that it was esseutially a developuient from vase pauitiug, perhaps even the work of the Mycenaean Ke/ja/icu's, not of the coroplast proper.' We believe this for these three reasons : — These fijjiires are never found, to our knowledge, except where there are also Myce- naean vases (whereas our " Primitive " figures were found together with all classes of vases). The conventionalization of the figure, notiibly at head and foot, points to ceramic art. The use of glaze color is certainly borrowed from vase painting. In this matter of the use of glaze color, we have come to a question which requires much greater amplification, and which Avill lead us much further afield. In the figures of Classes I. and II. (as also, to a less extent, in the later classes), the whole figure was usually covered with a white, chalky slip, and upon this was applied a simple line deco- ration in red and black.^ This white slip, owing to its flaky character, has in many cases been entirely lost,^ and is often preserved only in cracks and holes.* The isolated instances in which figures of Classes I. and II. have a glaze color applied directly to the clay ® seem to us to point to an introduction of this specifically Mycenaean technique to contemporary specimens of those classes. Indeed, we have a number of specimens in which the white coat is used for the upper part of the body, while the lower part is given a solid color, applied directly to the clay," and this is common in the later, " Ad- vanced Argive " specimens.' The use of the white slip, however, appears to have been the specifically coroplastic method, which obtained throughout all periods down to the latest times ** — a fact which strongly confirms our contention that these terra-cottsis (in which the color is applied directly) were made imder the influence of the /cepa/ieu?-" Class IV. — The single female figure (No. 75), which we have assigned to a distinct " Geometric " class, seems to us to warrant this distinction by reason of its peculiar decora- tion, which, with its bands of geometric ornaments, divided by straight lines, is very similar in plan to the decoration of the large Geometric vases which form this class. Its position in the series is determined by the fact that the face was evidently of the primitive " bird- faced " type. With this figure we have grouped another distinctly Geometric specimen (No. 76), a horse of the type so common upon the covers of Geometric vases.'" The two specimens do not represent the entire influence of the Geometric style upon our terra- cottas. Traces of that influence are to be seen in a number of other figures, especially among the animals and small objects," but the two which we have placed together are the only ones which are so distinctly Geometric in character as to warrant a place by themselves. In fact, though at Olympia a large number of Geometric animals (espe- cially horses) have been found, it is a noteworthy fact that we can hardly identify any plastic representations of the human figure with this period. It appears to us that this » Cf. General Introd. vol. I. pp. 45 ff. s Q{ Martha, in tlie introduction to his Calalogue ries ' Cf. Xos. 7, 11, 12, 13, 14, etc. Figurines en Terre Cuite du Musee de la Societe Archio- « Cf. Nos. 1, 2, •3, 5, 9, etc. logique d'Athenes. * It is this perishability of the white slip, doubtless, » This matter of the interaction of vase painting and that led Schliemann to speak so often of " nupainted " coroplastic art has never been carefully worked out. terra-cottas from Tiryns (cf. Tiryns, pp. 150 if.). In the The early Tanagra "TrairoSfs," for instance, show distinct light of our discoveries, there can be no doubt that nearly traces of the Geometric influence. The white Athenian all, if not all, the figures from Tiryns were originally lekythi, on the other hand, seem to be a result of the ap- painted. plication of coroplastic technique to vase painting. = Cf. Nos. 10, 21, 25. 10 Cf. Rayet et Collignon, //i.sV. de la Ceramirjue Grecque, « Cf. Nos. 4, 24. p. 33, fig. 21. ' t;f- P- 7. n Cf. Nos. 248, 272. PRINCIPLES OF CLASSIFICATION 7 arises out of the constructive nature of that style, which is essentially decorative and conventional — in so far un-naturaHstic and opposed to plastic art. Class V. — A very marked advance in the development of the terra-cotta figurines found at the Heraeum from the earliest primitive type onward is made when once the " bird face," which obtains in all the classes hitherto mentioned, is superseded by the human head.' This type we have called "Advanced Argive." The figures of this class are plainly a development from our second, Tirynthian Argive, class. Even in the most advanced specimens, with very few exceptions,' the bodies of these figures are essentially the same flat oblong which we found in the figures of the Tirynthian Argive class. The heads, on the other hand, are of archaic type, with wide, staring eyes, large ears placed high, the mouth often having the well-known archaic smile.'* Most importjint of all, these heads, almost Avithout exception, were made in moulds. Here, then, we see the beginnings of that use of moulds which later became the prevailing mode, and revolu- tionized the manufacture of terra-cottas. Another characteristic of this class of figures is the tendency to elaborate ornamentation, of which we shall have more to say later.* In tlie colored decoration of the figures of this class, also, we find again the two pro- cesses which we have before noted. Thus, while the use of the white slip as a basis for decoration in color is always retained for the upper parts of the figure, and in many cases is used for the entire figure as well,^ in the large specimens the color is often applied directly to the chiton, usually a red stripe at the top, and dark brown for the rest of the garment." It is also a noteworthy fact that in these figures the seated type prevails almost exclu- sively. Indeed, the only standing figure which we have ventured to place among the advanced Argive specimens differs from them so widely that we have even doubted whether it ought not to be placed in a category by itself. This is No. 107, a figure which again illustrates the influence of ceramic upon coroplastic art. This figure has traces of a human head. In the upper part of the figure there is a certain degree of naturalness, but the lower part of the body is emphatically conventionalized, without any attemjit at naturalism. The roundness is so complete that it at once suggests mechanical work, such as the turning of the potter's wheel. The use of the wheel is still further confirmed by the application of the brush in making the accurate parallel hues in groujjs of three, and the broad line at the waist around the cylindrical body. The ornamentation, as well as the structure of the figure, thus points to the influence of the Kepa^evs, belonging to the period of the fully developed Argive-Linear,^ as the Myce- naean terra-cottas showed the influence of the Mycenaean Kepafievs- We have placed it among the advanced Argive figures on the ground that it has a human head and a conventionalized body,* and that the decoration would mark it as belonging to about the period to which we must assign our advanced Argive figures. Class VI. — Our next class, " Figures under Oriental Influence," consists of a smaU number of figures which are marked out as a separate category by their subjects — two lions, heraldicaUy grouped, winged female figures like the so-called Nike of Archermus,'' I Cf. Nos. 77 £f. ' Cf. Nos. 86, 90, 91, 92. - Figures like Nos. 90 and 91, in which the bodies ' Cf. pp. 126 ff. have a considerable thickness and there is some attempt ' It should be noted, however, that the conveutionali- to represent the waist, are extremely rare. zation in this figure is quite different from that which we 8 Cf. Nos. 93 ff. have seen in the Mycenaean figurines. 4 Cf pp iQ £f_ ^ Cf. Collignon, Hist, de la Sculpture Grecque, I. p. 136, 5 Cf. Nos. 79 and 80. %• 68. 8 THE TERRA-COTTA FIGURINES severul figures of the Egyptian Bes, and a niiniher of representations of the Sphinx.' With these we have grouped a small figure of a horseman,'- which is certiiinly of Egyp- tian manitfacture. It would of course have been jjossible to catalogue these figures under our next class of archaic figures, as they are all archaic in style ; but it has seemed to us more desirable to collect in one category the specimens which distinctly show the influence of the E;ist, just as in vases this influence is recognized in the so-called Corin- thian vases. It is the analogy of the history of vases, moreover, in which the Corinthian style foUows the Argive style, that has led us to place this category immediately after the class of advanced Argive figures.^ Class VII. — With the beginning of the historical Greek period, the art of terra-cotta making becomes more dependent upon sciUjjture, and from this time on, its history is bet- ter known and more easily traceable. To the beginning of this archaic period we must assign a number of figures rudely made by hand,* which, although they still preserve a high degree of conventionalism, manifest an endeavor on the part of the coroplast to imi- tate more closely the forms of the human body. Their conventionalism, moreover, is the conventionalism with which we are familiar in the early statues of Greek art (the ^da^a), beginning with the Nicandra statue from Delos.° We have therefore classed these figures, and those immediately following them, as " Early Archaic," but have subdivided them into two groups : — Group a. — Hand-made Figures. These we regard as in a certain sense a transitional type from the earlier, hand-made Argive figures to the moulded figures of the archaic period proper. Group 6. — Mould-made Figures. The archaic period proper begins with the intro- duction of the use of moulds for the whole figure, the second step — the modeling of the heads of the advanced Argive figures being the first — toward the developed terra- cotta figurine of later centuries. Here we have, at the beginning of the class, a very interesting transitional specimen," in which we have a moidd-made torso, decked out at the shoulders with two disks,' such as are common in the primitive and advanced Argive tyi)es.* After this we have a long series of both seated and standing figures, hi which we can trace, as in the work of the sculptor, the struggle of the artist with his material, the gradual advance in the representation of the folds of the chiton," the development of the arms from mere stumps to well-modeled members of the body,'" which are raised to the breast," and carry attributes.'- The final period of this de- velopment corresponds to the period of the well-known female figures on the Athenian Acropolis.'^ As to color technique, we find in the archaic terra-cottas the same two opposing influ- ences which we have noted in our earlier classes. The great majority of our archaic 1 Cf. Nos. 108-116. 8 It shoultl be noted, liowever, that this figure is almost Cf. No. 117. without a parallel in our finds. For tlie most part, these » Here, too, might be placed Nos. 207-213, which archaic terra-cottas were very little iiifliiencea by the Ar- also show traces of Oriental influence ; but it seemed give type, although the two classes must have existed better to catalogue them among the archaic heads, for side by side for a considerable period, in them it is the archaic, rather than the Oriental ele- ^ Cf. Nos. 153, 157, 163, etc. ment, which is most characteristic. lo (jf ^o. 157 with No. 148 and No. 154. * Cf. Nos. 118-131. a cf. Nos. 153, 155. » Cf. Collignon, Hist, de la Sculpture Grecque, I. p. 120, '2 qj jJj, jgg g ^S- "^- " Cf. Collignon, Hist, de la Sculpture Grecque, I. pp. •Cf.No.l32. 341 ff..flgs. 17011. / ' IP ' Cf. p. 10. NUMBER OF kSPECIMENS IN EACH CLASS 9 figures show luimeious tiaees of the white shp, which we have described as the typically coroplastic! method of decoration ; ' but some have paint applied directly to them, at least in parts.^ Class VIII. — Of terra-cottas belonging- to a later age than this early archaic period we have but few specimens, and these are almost exclusively heads of figures and votive busts, in Avhich, owing to the very gradual development of the archaic style, no sharp line of demarcation can be drawn between early and later specimens. We believe, however, that the forty-two specimens ^ which we have included in an " Advanced Archaic " class will be found to mark a higher stage of development than is attained by any specimens of Class VII. Class IX. — Finally, we come to the last group, which, as has been siiid above, is re- presented by a remarkably small nundier of specimens. In regard to Nos. 23J) and 240, there can be no doubt that they belong to the best period of Greek terra-cotta work, and we have not hesitated to place them as " Figures of Free Style," in a separate class. The figures of animals and of varicms small objects are in general so rudely modeled that they do not lend themselves to any classification based uj^on style."* We have therefore been led to adopt the manifest principle of classification according to subject, and thus to disthiguish two further classes, — " Animals " (under which we include fig- ures of men and women seated upon animals) and '• Various Objects." So much for the principles upon which we have based our classification, and the gen- eral development of the manufacture of terra-cottas at Argos, as represented by our finds. But, as we have already suggested, these classes are by no means of ecpial importance. Nothing, indeed, is more instructive than the table of the numerical strength of the classes, which is as follows : — NUMBKK OK SPECIMENS IN THE VARIOUS CLASSES. I. Primitive Argive .... 122 VII. Early Archaic .... 249 II. Tirynthian Argive . . 19G1 VIII. Advanced Archaic ... 42 III. Mycenaean 63 IX. Free Style IV. Geometric ..... 2 V. Advanced Argive . . . .409 Animals 715 VI. Figures under Oriental Influence 15 Various Objects 433 That is, of 2865 specimens, not counting animals and various objects, 2492, or over 85 per cent., are figures of the flat-bodied type, with either the " bird face " (2083 speci- mens) or with the archaic head which is its successor (409 specimens). It is the pre- ponderance of this type which justifies us in calling these specimens distinctively Argive. They have been found, it is true, in other places, nottibly at Tiryns^(as we have recognized by our qualification of a certain class of them as Tirynthian), at Haghios Sostis (Tegea),« 1 Ct. Nos. 132 135 140, etc. glo"- ^"o"* ^^«"'- f^dl' Ins. II. pp. 72-70, pi. vi.; Martha, -' Cf. Nos. 137 138 163* etc. C"'' ''«•' ^'ff- ''" ^^'"- d'Alhbxes, Nos. 541 ff. These Tegeaii 8 Cf Nos! 227-238.' ' figures present the closest parallels to our Argive types. * The onlyTxceptioii whicli we have made to this rule They occur with human as well as witli " bird faces," aiid is in the case of Mycenaean and Txeometric animals, the later specimens are often very elaborately decorated. where the decoration leaves no possible doubt as to their They were found in such numbers as to preclude the , theory that they were imported from Argos. vV e must '' "^"cf Sehliemann, Tirym, pp. 149 ff., figs. 76-90. couch.de, therefore, that the art of the coroplast passed « Cf . Lenormant, Gaz. A rch. 1878, pp. 42-48 ; Pervano- through much the same development .at Tegea as at Argos, 0 10 THE TERRA-COTTA FIGURINES and spoiaflifallv elsewhere.' But nowliere have they been found in such nund)eis or exhil)itini; sueli a dearly marked development as at Argos. They are hardly found at all at Mycenae." We feel justified, therefore, in giving them the distinctive name of " Argive." But if these Argive figures form the great bulk of our find in terra-cottas, they also present the most serious and ditticult problems of all our figures. How, for insfcvnce, shall we designate the great mass of plastic ornament which is so characteristic of the later " advanced " type, and also, to some extent, of the earlier specimens ? In the pre- liminary })ublication of the Heraeum finds, it was suggested that the elaborate orna- ments upon the shoulder of the Argive figures might he bunches of flowers, having some reference to Hera Antheia.^ But further consideration has convinced us that, for the great mass of our figures, this view is untenable. In the first place, the large orna- ments for which this interjiretation was suggested * are ])lainly developed from the small roiuid bosses of earlier figures."^ The beginning of this development is very plain in one specimen," where the wedge-shaped ornament at the shoidder is no more than an elongated boss. Now in many later specimens ^ this boss develops into a regidar fibida of the " f oui'-leaved clover " type,* which Helbig '' proposes to identify with the Homeric ikt^.^" Moreover, in one fragment " we have the most evident attempt to represent a bronze fibida of the usual "bow" shape ; '^ while with the elaborate ornaments like or (wliat seems to us more probable) that the coroplasts Primitive standing figures, no ornamentation .2 of Tegea were strongly influenced by those of Argos, and Primitive standing figures, one necklace •....•) that there was an intimate connection between these two ''ri'"it>ve sitting figures, no ornamentation .... 1 „i / c t>. 1 „ r u f vxTT rt of\i*-\ n^ xi. Primitive sitting figures, one necklace .... 2 places (cf. Kidgeway, J. H. b. XVI. 118961, p. 99, on the „ • . ■ -,,. /. . , ,., .„ ,7 ,■ ^ r, . 1 nmitive sittmg figures, two necklaces 8 close connection between legea and Argolis); Curtius, as Primitive heads o Ridge way remarks, pointed out {Gr. Gesch. I. p. 156) Advanced sitting figures . . . . . . . .1 that Nauplius is represented as the servitor of the king Advanced heads 2 of Tegea. We may also poiut to the fact that the Warriors 11 priestess Chrysis sought and found refuge at Tegea after Animals 6 leaving the Heraeum (Pans. III. o. 6). ^"""*' 7*^ ^'f, **«* 'P^^^^^'y * """t*"^- ,.„, ,,, T ^..T., , r ragment of a Mycenaean vase. ' At liathos, on a sj)ur of Mt. Lycaon (cf. Bather and Yorke, J. H. S. XIII. [1893], p. 228;; in Cyrenaiea (cf. Several of the figures have pins of tlie "dumb-bell" Heuzey, Fig. Ant. pi. xl. 1); at Thisbe in Boeotia (cf. shape (cf. No. 84), while the "clover-leaf" type (cf. stir- Arch. Anz. 1895, p. 220, 4) ; one or two specimens were pro) also occurs. One of the advanced heads has a cala- foiind by the English School at Phylakopi in Melos. thos and a double taenia, closely resembling Nos. 101- There are a few specimens very similar to ours (all with 104. We thiuk it right to draw atteution to the fact human faces) in the museum at Syracuse, but the Sicilian that this beehive tomb, even if it was disturbed, contained terra-cottas are for the most part of a far less primitive no object later than the Mycenaean perio<l. In so far, character than those of Argos ; cf. the figures from the presence of the " Tirynthian Argive " terra-cottas Agrigentum (cf. Kekule, Ant. Terrakotten, II. pp. 17, 18, would place them not later than the Mycenaean period, figs. 21-27), from Gela {ibid. p. 22, fig. 47), and the '' Cf. Waldstein, Excavations of the American School at specimens from the vicinity of Terravecehia near Gran- the Heraeum of Argos, 1892, p. 20. Cf. No. 52, where raichele, published by Orsi (^Mon. Ant. VII. pp. 239 ff., the ornament certainly suggests a mass of flowers, figs. 31, 32). Finally, a number of vases from Cyprus * Cf. Nos. 83 and 93. in the British Museum, of Geometric style, are decorated ^ Cf. Nos. 27, 28, 40. with figures very similar to our advanced Argive type ; * Cf. No. 43. they are placed on the neck of the vase, and hold in ' Cf. Nos. 64, 90, 91. one hand a small jug which served as a spout to the « cf. Montelius, La Civilization Primitive en Italic, pi. ^**«- xxi. 286-289 ; Furtwangler, Olympia ; Ergehnisse, IV. ^ A tray in the Central Museum at Athens (numbered pi. xxi. 360. 3071) contains a number of figures of the type of our » Cf. Das Horn. Epos aus den Denkmalern erlautert, 2te primitive and advanced Argive specimens, which came Aufl. pp. 280 ff. from a tholos tomb near Mycenae. Dr. Tsountas in- "> But et StuAniczka., Beit, zur Geschichte d. Gr. Tracht, forms us, however, that the roof of this tomb had fallen Ahh. d. Arch.-Epig. Seminars d. Universitiit Wien, VI. p. in and the contents had been disturbed, so that no sure 114, footnote. conclusions can be drawn from them. The tray con- " No. 82. **"•* '- Cf. Montelius, he. cit. pis. i., iv., v., viii., ix. THE ARGIVE TYPE 11 No. 84, we may compare fibulae found in our own excavation,' and others published by Montelius.^ It thus appears evident that these ornaments are intended to represent simply the pins by which the Doric chiton was fastened at the shoulder, whether this be the straight pin {TrepovT}), or the fibula of the clover leaf or bow form ; and we have therefore had no hesitation in so designating them in our descriptions. There is room for doubt, perhajjs, in the specimens which have only the simple round boss, whether this boss is meant to represent a fibida of the eXtf type or a simple straight pin. We have used the term " pin," therefore, to include both fibulae and wepovai. The bands across the breasts of our figures* are in many cases evidently meant to represent necklaces, and we can perhaps distinguish in some cases between the laOfiiov, or close-fitting necklace, of Homer ^ and the 6pp.o<;,^ or long necklace. Certamly the many cases in which we find bands ornamented with pendants " can be meant for nothing but necklaces. But very often our band extends only from shoidder to shoiUder, and appears to have some intimate connection with the pins which we have just discussed.^ Schliemann * called such ornaments simply " bands," whUe Perrot," in describing a figure very simdar to No. 34, calls the ornament " a sort of scarf." In the later development,"* when the figures are adorned with a large number of these bands, the lower one is usually so wide, and is stretched so straight from shoulder to shoidder, that there can be no dovibt that it is the fold of the Doric chiton." Moreover, in specimens like No. 30, the band is plainly the fold of a garment arranged like the ifiaTtov m later works.'" It must be said, however, that even in fairly early specimens the " bands " often become subordinated to a general principle of elaborate ornamentation, so that it is impossible to assign a given ornament to one class or the other, to say categorically, it is a neck- lace or it is a fold. In the most advanced Argive specimens '* we find a further com- plication in the elaborate plastic ornaments across the breast, which seem to be made in imitation of lace or metal adornments attached to the fold itself, or possibly of long chains, such as the women of modern Greece wear on feast days. But even if they are such, they are plainly developed from the earUer necklace and simple fold, and these are the most characteristic forms of decoration of the Argive type. It is the constant recurrence of these bands which has led us to adopt, as convenient subdivisions of our Tirynthian Argive class, the following categories : '* — 1. Figures with slightly developed plastic ornamentation at shoulder (no decorative bands). 2. Figures with considerably developed plastic ornamentation (one band). * Cf . p. 242, No. 852. ^Pl^<" '' *H>' i^oAf Sdpfi irepiKoXX^«j ^aav 2 Loc. cit. pis. xvii. and xviii. '"'^'''' xpi'^^o,, Tc^^oimXoi is ii „,K^^ ^ 8 Of "W 97 ff tTT^Bfaiv ci/xi^' aTraKoiai iXait-Ktro, Savfia iS(ff6ai. « Cf. Nos. 27 and 36. Compare also Od. xviii. 300 : — " Ct. Nos. 40, 42, 80, etc. ia0iJ.iov fjueiKfy Sepiwav, irepi/caXAfS SyaA/ua. ' Cf. Nos. 28, 34, 37, 40, 41, etc. And Schol.:- ^ C(. Tiryns, -p. 156. Ue^iov he^h. 6 rpdxv^os. foflM'o. »5- Tep.Tpaxi)X.ov «<(<rMO. ^ Cf. Perrot et Chipiez, Hust. de I Art, VI. p. <50. ■„,piwtl^\(y^iiyo•', ov jk^Vto. K,>af,i,fiard nva iKKpifii/Mfva. The figure is No. 344, taken from Schliemann, Ttryns, (tot SAAojj- irepiTpoX'iX""', ^"^f ""^ irap((r8^ia- Siafepei Je tov p. 359, fig. 159. ipfiov. rh fify yap irpoaix""-^ ''<? Tpax^^V) ^ '' optios KexiiAa- l" Cf. Nos. 86, 93. ffTai. " Cf. Studniczka, loc. cit. pp. 6 ff. 6 Cf. Hym. H(m. v. 10 (Gemoll) : — '^ cf. Nos. 163, 164, 175. Sfipp 5' i.ijL<p' awaXv "a) rrT^fltirii', ipyvpiotaiy " Cf. Nos. 90 and 91. gpMO.T. xpv<r^oic/iK6,T,.,oy, ohi T.p aBra. " For the advanced Argive terra-cottas no such dis- ^npai Koaixfi<xBvy xp^^^MT"""- tinction seems necessary, as they are practically all of . ,.,.,... „o the elaborately ornamented type. And tbid. in. 88 : — •" 12 THE TERRA-COTTA FIGURINES 3. Piguies with elaborate plastic oinamentation (two or more bands). It will be found, in general, that increasing elaborateness of ornamentation at the shoulders of these figures corresponds to the multiplication of bands, and that both advance with the growth of perfection in technique.' It remains to note the other decorative elements, which, although they are not of sufficient imi)ortjince to form a basis for classification, yet display a very considerable development in our series of figures, — namely, the earrings, the hair, and the head- dress. Of the earrings not much need be Siiid. They begin comparatively late in the series (the small size of the earlier figures naturally ])recludes the use of such ornaments) with much-decorated " bird-faced " figures, and develop from the sinqile disk to the disk with a central boss," the rosette,' and even the disk with pendant.* The development of the hair is more complex. In the earliest figures, as we have before stated," the hair is not represented at all, or at best it is marked by a few par- allel grooves at the back of the head or by two or three strips of clay attached to the back of the head and falling to the shoidders, most usually notched or twisted. Then one or two strands are brought forward over the shoulders," while a corkscrew curl is added to each side of the forehead.^ The use of the simple mass to represent the hair as it falls to the shoulders ** seems to be a later development. It is employed almost exclusively in the most advanced heads," Avhere the front hair is usually represented by two notched bands,'" or by large curls which leave between them a wide parting." In a few cases we find over the forehead a row of small, separate curls, similar to those commonly given to archaic heads m stone.'" In the headdress we find a development very similar to that which we have noted for the hair. Very early in our series, as we have already pointed out,'^ we find a smiple band of clay woiuid about the head to represent the stephane. This is often so broad and high that it might more properly be called a polos or Stephanos.'* To this is often applied a single boss,'° or a double boss,'" or even a number of bosses in a row, with perhaps a double boss in the middle.'' In one case ''^ the edge of the stephane is pierced with holes, apparently to receive metal ornaments. But the stephane (stephanos, polos) is not the exclusive headdress of our figures. As the coroplast came to model larger and taller figures, he evidently began to feel the need of a more stately headdress to correspond to the elaborate adornment of the body. Thus we find that in the later figures the most common headdress is the high, bell-shaped calathos.'" This is usually bound about the bottom Avith a taenia, and is often decorated at the top with bosses.'''" Most remarkable, of all, however, are the horn-like adornments which are the final deco- ' It is not always possible to use the number of bands as an absolute criterion, for there are cases in which other decorative elements, such as applied layers of clay, representing the garments (cf. Kos. 56, 60, 61), or the in- fants which the KoupoTp6<pos figures hold (cf. Nos. 37, 57, 58, etc.), take the place of tlie bauds. But, for the most part, the distinction of the classes according to the num- ber of bands upon the figures will be found to have been carried out with considerable strictness. - Cf. No. 53. » Cf. Nos. 54, 104. * Cf. No. 94. ■• Cf. p. 4. « Cf. No. 26. ' Cf. Nos. 25, 45, etc. 8 Cf. Nos. 42 and 51 (which arc of the Tirynthian Argive class), where the mass is crossed by grooves. « Cf. Nos. 93, 94, 97, 98, 99, etc. "> Cf. Nos. 94, 97. " Cf. Nos. 101-104. '■^ Cf. No. 105. IS Cf. p. 4. "« Cf. Nos. 40, 53. '•' Cf. Nos. 7, 33, 53, 96. '« Cf. Nos. 44, 97. " Cf. Nos. 40, 42, 98, 99. '» Cf. No. 54. i» Cf. No. 94. 2« Cf. Nos. 98, 99. WHOM DO THE FIGURES REPRESENT? 13 ration of this later type.' These consist of a strip of clay, l)nl<;in<>- slightly at the extremities, j)laced about the calathos in such a way that the two ends rise above it in front. A boss is sometimes added to each end for further effect," and between the " horns " a disk or rosette is quite usual.^ We are inclined to regard these " horns " as an attempt to represent plumes of some sort, although we know of no parallel cases.* Another point which naturally arises in connection with these " Argive " figures is the question. Whom do the figures represent ? Now the most striking fact about them (as, indeed, about the human figures in general) is the great preponderance of female forms. In the Avhole find of terra-cottas, there are but sixty-six figures which are male beyond a doul)t, and of these forty-eight belong to the Avell-known type of the mounted warrior, and eight to the type of the Egyptian Bes.' This fact would naturally lead us to the conclusion that our female figures represent the chief goddess of Argos, Hera, and many arguments can be brought to support this interprefcition. Thus, the great majority of our Argive figures are seated, and we know from the Pirasus story and from Pausanias's account of an early image of Hera, as well as from his account of the statue of Polycleitus, that Hera was conceived at Argos as a seated divinity. The head- dresses which we have noted upon our figures, stephane, polos, Stephanos, and calathos, belong to the regular wardrobe of Hera ; " while as a goddess of childbirth ' she could be represented with an infant in her arms. But such arguments as these are extremely fallible ; and for the early time, especially, we certainly cannot postulate any such fixity in the functions and attributes of the goddess as such statements imply. The proof of this fact is not far to seek ; for in Tegea, where, as we have already stated,* we find figures exactly similar to all classes of our Argive terra-cottas, these figures were dedi- cated to Demeter." The headdresses that we have mentioned are found in the Tegean figures, as in those of Argos, and they are by no means the exclusive property of Hera. The KovpoTp6(f)o<; figures are much better explained as human mothers,'" especially as Hera (and the divinities of the Argive plain in general) is rarely represented with an infant in her arms,*' and the epithet KovpoTpo^o'i for Hera rests upon very doubtful authority.'- Again, the groups of two seated females," while they might be taken as Hera and Hebe 1 Cf Nos. 100-104. '" t)ii the analogy of the woman vvitli an infant on her 2 Q^ j^o 203 back (No. 39), the woman kneading bread (No. 24), and ' Cf Nos. 100 104. the performers upon niiisical instruments (Xos. 21, 22, ■• Mr. De Cou suggests that these adornments may be 23). This is the interpretation which we are inclined to borrowed from the headdress of Isis, and that the disk or favor for these KovpoTpi^o'i figures. A third possibility, rosette may then represent the sun. If this theory be tliat these figures are nu)diacations of the Oriental As- correct, we should then find in these figures another tarte, seems to be excluded by the fact that they bear trace of Oriental influence. such close resemblance to the Argive type, and are plainly '- It may be said that the early numbers of our scries only a development of it. are so rude as to be practically sexless. But the early " Cf. O. Miiller, Handhnch d. Arch. d. Kuiist, § 3i>3, 1; development of the hair and the ornaments, which points and Tsountas, -Efvf^fpU 'Apxaio\oyiK-li, 1888, p. 170. distinctly to female figures, as well as the great pre- "^ Suidas, s. v. -O^ripo,, gives the verses : — ponderance of ft'male figures in the later types, make it KKvOi iioi fvxotiLfi"f. Kovporpitpt. ihs 5* yvva'Ka, practically certain that our early figures, also, are meant r-hvif fftuv im^yaadai (piKirriTa Kal (vfrii', to represent females. ,^^^^ .^jj^ . '• Cf. for the polos, the well-known head from Olym- ' ' _ /^ «. , pia, Bcitticher, Olympia, p. 237, fig. 44 ; for the stephane, "^ ^ _ . Overbeek, Kitnstmythnlogie, pis. ix., x.; for the Stephanos, ibid. vol. III. Hera, Miinztafel ii.; for the calathos, tW. Bernhardy (cf. Suid. ed. Bern. II.' p. 1101) refers Miinztafel i. Nos. 1-9. Kovporpitptp to Apollo. Athenaeus (XIII. 592) refers the T Cf. on Hera Eileithyia, Oeneral lutrod. vol. I. p. 8. verses to Sophocles, and says the poet calls on Aphro- 8 Cf. p. 9. note 6. '''*"■ » Cf. Lenormant, Gaz. Arch. 1878, pp. 44 ff. " Cf. Nos. 59-02. 14 THE TERRA-COTTA FIGURINES or Hera anti Eileithyia, seem to us, to judge from the analogy of similar groups in later times, to be better t<iken as representiitions of the Earth-mother and her daughter, the later Demeter and Kore.' In the case of these Argive terra-cottas, therefore, we can only siiy that they prove to us the existence at the Heraeum, in very early times, of a cult of the primitive female divinity, the real " great mother of the gods," Avhether we call her Ge, or Demeter, or Aphrodite, or Cybele, or Hera, or perhaps even Athena or Artemis. She was early conceived as a seated figure, and this conception strongly influ- enced the ideas of later times, when men began to call her Hera, and to reproduce her image in wood and stone. Finally, the use of exactly the form that is employed to repre- sent the goddess for figures that are certainly human '" proves conclusively that we are here deaUng with one of those early types which are already familiar to us in the seated figures from Branchidae, the " Apollo " figures from all over the Greek world, the standing types of the Acropolis, and many other works. We have dwelt thus at length upon the " Argive " figures both because they form the most importiint and characteristic part of our find, and because this type has never heretofore been subjected to the carefid study which it deserves, and Avhich, fortunately, our numerous finds of early figures in so unbroken a series enabled us to make. The remainder of our finds belong, for the most part, to well-known types, and can be dismissed in a few words. Of the Mycenaean figures, the greater part belong to the common type with round or pelta-shaped bodies. The large bidl's or cow's head is, we believe, unique in terra-cotta, although this subject is common enough in Mycenaean art in other materials.* The paucity of figures of this class of terra-cottas at the Heraeum is very striking, in view of the large numbers of them that were found at Mycenae and at Tiryns, and suggests the explanation that these figures were not regarded as appropriate offerings at this particular temple. The Argive style, in fact, had become invested with a hieratic character, and was regarded as the proper object of votive dedication at the Heraeum.* The same remark holds true of figures of Geometric style, of which the paucity is very striking, in marked contrast to the early figures of Boeotia, which are distinctly Geometric.'^ So, too, even our 283 figures of the Archaic class foi-m a strikingly small proportion of the find, when we consider the prevalence of this class on other early sites." Here, indeed, the figures themselves seem to show that these archaic terra-cottas are not a natural Argive product. We do not mean to say that they were not made at Argos. > Cf. the argument, pp. 22 f. 6 Cf. Heuzey, Fig. Ant. pi. xvii. 3 ; Jamot, B. C. H. « Cf. Xos. 21-24, 39. XIV. (1890), pp. 207 ff., flgs. 1, 2, 6, and pi. xiv. ; Mar- « Cf. the parallels cited on p. 23. tha, Cat. de.i Fig. du Mus. d'Athmes, Nos. 213 S. These * So at Athens, in early times, the fashion seems to figures are interesting as examples of coroplastic art have been for figures of Athena of archaic type (cf. developing on lines parallel to those which it followed at Winter, Arch. Am. 189,3, pp. 141 ft.). At Corcyra the Argos, yet differing from Argive art in many particulars, popular offering was apparently a figure of Artemis (cf. In Boeotia, as at Argos, the flat, conventional body of the Lechat, B. C. //. XV. [1891], pp. 1 ff.). At Tegea the " bird-faced " type was retained long after the advent of Argive type was later superseded by an archaic hydro- the archaic style of head. In marked contrast to the Ar- phoros type (cf. Pervanoglou, Nuove Mem. deW Ins. II. give figures, the staiuliug type is, in Boeotia, the prevail- pp. 74 f.). In Sicily archaic figures carrying a pig are ing one, and the ornamentation is almost entirely painted, found in great numbers (perhaps connected with the not plastic. The teclinique, too, is ceramic, rather than cult of Demeter and Kord ; cf. Li^nard, Gaz. Arch. 1880, coroplastic, i. e. the paint is applied directly to the clay, pp. ir, ff.; Caylus, Rec. d'A,it. vol. VI. pi. xxxvii.). On « Cf. the excavations cited in note 4, and especially the subject of "fashions" in terra-cottas, cf. the re- those conducted by Orsi in Sicily — at Megara Hyblaea marks of Paris upon the terra-cottas of Klateia, B. C. (cf. Mon. Ant. I. p. 689, esp. pp. 913 fl^.), and at Terra- H. XI. (1887), pp. 405-444. veechia (ibid. VII. p. 201, and e.sp. pp. 21(i ff.). MYCENAEAN AND AECHAIC FIGURES. ANIMALS. VARIOUS OBJECTS 15 The large number of replicas ' would prove that most of them were made there, if such proof were necessary ; but they seem to be the product of outside influences, rather than the result of native development.^ So we find a number which show traces of the schools of Rhodes and of Cyprus.^ So, too, the great majority of these archaic figures are standing types, whereas the conception of Hera which prevailed at the Hemeum was, as we have seen, that of a seated figure. Moreover, a number of these archaic figures are shown by their attributes to be representations of goddesses in no way connected with Hera, or even hostile to her. Such are the figui-es of Artemis and Aphrodite.* For the rest we can only say that the archaic figures without attributes may be meant for sfcitues of Hera, and in many cases, doubtless, were so thought of by the donors ; but in many others they may have been meant for priestesses or even persons unconnected with the cidt of the goddess, who offer their own image to her. We are agaui deaUng, that is to say, with a type, and all attempts to dogmatize at this stage on the subjects of these figures are futile." The animals display all the variety usually found in temple offerings. They include horses (both with and without riders), bidls, dogs, pigs, bears, cocks, and birds, and even some less common ty|)es, such as the centaur, the serpent, the tortoise, the monkey, and the bidl attacked by a lion. It does not seem possible that even the hvely imagina- tion of the Greeks can have thought of any very definite association of all these animals with the goddess. But few of the animals represented are sacrificial, so that the idea that they are " substitutions " ® for actual sacrifices is in most cases excluded. At best this explanation can hardly apply to more than the cows,' the rams, and the pigs. In most of our figures, therefore, we are inclined to see only examples of the cheap kind of offerings which were sold at the entrance to the temple, — offerings whose value depended not so much upon their intrinsic worth as upon the spirit of the donor. The same must be said of most of the " various objects " which we have catalogued in our last class. The rough pear-shaped weights * and the cones," — which were probably meant for use upon the loom,'" — the spools," and the rude oven ^- — probably from a baking scene — are not inappropriate offerings to a female divinity like Hera. So, too, the dish of cakes '^ and the numerous fragments of tables " are doubtless to be regarded in the light of banquets offered to the divinity. The flowei-like disks,^^ which occurred in ' Cf. Nos. 141, 144, 148, 149, 177, 188, 200, 231. Myth. p. 410 : " Doch ist der Typus eiu so allgemeiiier, 2 Cf. the analogy in the proportion of black-figured dass er audi fiir andere Gottheiten, ja aueh sterbliche, vases to those of the Argive-Linear style, pp. 60, 174. die Votivgabeii darhringen, benutzt wurde." ^ Cf. Nos. 207-211, 212, 21.3. ° Cf. Perrot et Chipiez, Hist, de I'Art, VI. p. 818. * So we interpret the figure holding a bow and a ' It is interesting to note that tlie large Mycenaean deer (No. 179), and the figures which have in their hands bull's or cow's head (No. 72) has a hole in the top, which a hare (Nos. 176-178), or a flower (Nos. 166, 174), or a may have been meant to receive a bronze axe (cf. the fruit (Nos. 172, 173, 197), or a dove (Nos. 166-171, 199, references given in the catalogue, p. 23). 200). Some of these, to be sure, may be connected with 8 pf, jjog, 290, 291. the cult of Hera, as the figure with the wreath (No. » Cf. Nos. 283-285. 175) probably is. Thus we know that Hera as Eilei- i" Pottier and Reinaeh (La Necrople de Myrina, pp. thyia was represented with bow (cf. vol. I. p. 8) ; and 248 ff.) explain these cones as cakes "by substitution," the fruit of Nos. 172, 173, and 197 may be a pome- and give the literature of the subject. The older inter- granate (cf. Philost. Apoll. Tyan. TV. 28 : v l>oa S^ fi.6ini pretation of them as loom-weights seems to us to be pre- <t,VT5iv ry 'Hp? <pifrai). But such attempts are certainly ferable. far-fetched, and if we consider the wide distribution " Cf. Nos. 286-289. of these types (see the references under the separate '^ Cf. No. 279. numbers), the connection with Hera does not seem prob- ^» Cf. No. 267. Cf. also the mould, No. 277. ^We. '^ Cf. Nos. 268-270. 6 Cf. Furtwangler, 'Aphrodite,' in R«scher, Lex. <l. '^ Cf. Nos. 281, 282. 10 THE TKRRA-COTTA FIGURINES siitli muuHers, have reference, peiliaj)s, to Hera Antheia. But the aijpiopiiateness of the foot with a siindal ^ (which seems too elaborate to have been part of a human fig- ure), of the object which we have called a tree," of the whorls ^ and the rings,* it is impossible to see. Here, then, we doubtless again have examples of small, inexpensive offerings, which were turned out by the score by the coroplasts, and served to Scitisfy those among the poorer classes whose means did not permit them to jjresent more elabo- rate olferings. But these, after all, are points of minor importance. The chief interest and value of our find of terra-cottas lies in the fact that tliey reveal to us — in the figures of the " Argive " classes — the existence of an artistic tradition wliicli began long before the period of the distinctly Mycenaean civilization, and continued unbroken for centuries after it. Tliey thus strengthen us in the position which we have been led to take from a general survey of the whole result of our excavations. A detailed catalogue of the Terra-cotfci Figurines follows. Upon tlie Plates (XLII.- XLVIII.) the figures iu parentheses refer to the numbers of this catalogue. I. PRIMITIVE ARGIVE. Fig. 2. 1. (Plate XLII. 1.) Rude standing figure, modeled by hand, — a simple strip of clay, pinched together at the middle to form a waist and a stumpy pair of arms (now broken), and pinched at top into a pointed head. No traces of color. Height, 5 cm. Dark red clay. Cf. Seldiemann, T'lrynn, p. 150, No. 78. i2. (Fig. 1.) Similar figure, arms broken. To each side of the nose is fixed a Inni]) of clay, to represent the eyes. The bottom is rounded to J^*' form a base. No trace of color. Height, 4.5 cm. Light yellow 11 clay. ^ 3. (Fig. 2.) Similar figure with a lump of clay attached to eacli side of tlie nose to represent the eyes, and another lump on top of head, to form a sort of i)ilos or cap. The figui-e is attached Fig. 1. *° ^ hAne and possibly formed part of a group. No traces of color. Height, 5 cm. Dark red clay. Cf. Schliemann, Tiri/ns, p. 153, No. 82. 4. (Figs. 3 and 4.) Similar figure, with lumps for eyes and a strip of clay wound about the head, forming a stephane. Below this stephane, at back, three strips of clay are applied to represent the hair. The figure .stands on a plinth and may liave formed part of a group. Tlie body is painted brown ; the face is covered with a white •slip. Height, 5.5 cm. Clay, reddish yellow. 5. (Fig. 5.) Similar figure, wearing stephane. Tlie body is pierced by a round hole, apparently for the Fig. 4. purpose of suspending the figure. No traces of color. Height, 6.5 cm. Red clay. 6. Fragments of two similar figures with holes for suspension. 7. (Plate XLll. 6.) Primitive figure, right arm broken. Eyes, the usual lumps. The stephane has a boss added in front. Traces of white slip. Height, 6 cm. Clay, light yellow. a (Fig. 6.) Primitive figure, which originally had two lumps for eyes, now lost. The figure is remarkable for the two lumps of clay wliich are evidently meant for the breasts, although " Cf. No. 276. » Cf. No. 280. Cf. No. 294. * Cf. No. 292. Fig. 3. TIRYNTHIAN ARGIVE 17 tliey are i)lacecl very low. Color, red, shading to black (apparently from burning). Yellow clay. 9. (Fig. 7.) Similar figure, with the usual lumps for eyes. The represented by a cutting across the " beak." The ver- tical grooves in the lower part of the figure seem to represent the folds of the chiton. No trace of color. Height, 5.5 cm. Light yellow clay. . 10. (Fig. 8.) Standing male figure with stephane. The eyes are not indicated, but the lower part of the "beak" is drawn out to form a beard, and the mouth is represented by a sliort cutting. Body and stephane are painted red. Height, 5 cm. Yellow clay. 11. One hundred and eleven fragments of figures similar to Nos. 1-9, showing the same development of hair and stephane, but no devel- opment of drapery. They are all of small dimensions, njade by hand, of rather generally show traces of a white coat, with simple line patterns applied in red and Height, 6.6 cm. mouth is rudely Fig. 6. Fig. 7. Fig. 8. coarse clay, and black. II. TIRYNTHIAN ARGIVE. GROUP 1: SLIGHTLY DEVELOPED PLASTIC ORNAMENTATION (NO DECORATIVE BANDS). 12. (Plate XLIII. 1.) Seated female figure, simplest form. The figure is made in the same way as No. 1, but it is slightly bent in the middle, and to the back is applied a four-legged support (cf. p. 5, note 1). There is a simple stephane on the head, and the feet are represented by two lumps applied to the bottom of the figure. White slip, traces of red on stephane. Height, 4.5 cm. Dark yellow clay. 13. Seventy-five fragments of figures similar to No. 12, exhibiting no adornment other than stephane and slight development of hair. They are usually covered with a white coat, and decorated with red and black lines. 14. Fragment of figure similar to No. 12, broken at waist, and at neck and left shoulder. Above the waist are two holes for inserted arms, with fragments of arms still in place. Traces of white. Height, 6.5 cm. Clay, reddish yellow. 15. (Fig. 9.) Bearded seated figure, broken at waist and lacking arms. Eyes, the usual lumps. Round pilos-like cap. Mouth, a deep cut, and beard marked by four vertical grooves. White slip. Height, 6.5 cm. YeUow clay. 16. (Fig. 10.) Seated male figure in high pilos, with strap passing under chin. The figure has a band across the mouth, plainly for holding a flute, which is now broken away. The ai8<Ha are represented by a roinid lump of clay. Traces of white. Height, 8.5 cm. Red clay. Cf. for headdress, Schliemann, Tiryns, p. 159, No. 93: Furtwiingler, Jiihrhir/i, 1887, p. 202, No. 8: Fnrtwiingler, 0/i/nipi(i, Erriebnis»e : JJic Bronzen, pi. xvii. 288. 17. (Fig. 11.) Seated figure with feet extended. Part of face, both arms, feet, and legs of chair missing. The fio-ure has a stephane with boss, but otherwise is unadorned. A strip of clay over each temple forms the hair. White slip. Height, 8 cm. Length, 6.5 cm. Yellow clay. Cf. Schliemann, Tiryns, ]>. 159, No. 91. 18. Two similar figures, badly broken. 19. Similar figure, with band over mouth, api)arently for holding flute. Traces of white coat. Height, 5 cm. Length, 6 cm. Yellow clay. Fig. 9. Fig. 11. Fig. 10. 18 TERRA-COTTA FIGURINES 20. Simifer figure, broken at waist, legs of chair also broken. At left side, near feet, are traces ^ of a smaller figure, also broken at waist. No traces of color. Height, 5 cm. Length, 6.5 cm. Yellow clay. 21. (Fig. 12.) Flute plaj'er of uncertain sex. Right arm and right half of , flute broken away. With the exception of the face and the flutes, the figure is entirely covered with dark red paint. Height, 7 cm. Light yellow clay. .i 22. (Fig. 13.) Performer on the syrinx — the " bird-faced ' Fig. 13. t i^H type, with simjile stephane. The nature of the Pan's pipe is clearly J^B indicated by a cross strip at top and bottom of the instrument. |9| Broken at bottom. Height, 4.5 cm. Red clay. Fig 12 ^3. (Fig. 14.) Similar figure with double flute. Stephane adorned with a boss. Red stripe indicates the flute strap. Height, 5.5 cm. Red clay. 24. (Fig. 15.) Woman kneading bread. She wears a stephane, and has a lump of clay attached to each shoulder, to represent dress-pins (Trcpoi-ai). Small disks of clay represent the dough in the pan. Traces of white on arms of woman and on pan, of red on stephane, of brown on woman's body. Height, 6 cm. Clay, reddish yellow. Cf. Schliemann, Tiryns, p. 149, No. 76. The motive is very ancient — it occurs in the oldest sculpture of Egypt ; cf. Perrot et Chipiez, Hist, de PArt, I. p. 74, fig. 48, and p. 662, fig. 448. 25. (Plate XLH. 3.) Standing female figure, showing an attempt at more careful re- presentation of the hair. Under the stephane there is a large curl over each eye, and an oblong strip of clay at the back of the head, now mostly broken away. Stephane, red ; hair and body black. Height, 8 cm. Red clay. Cf. Schliemann, Tiryjis, pi. xxv. k. 26. (Fig. 16.) Flute player (flutes now missing), showing further development of hair into two strands in front of each shoulder, and four curls across forehead. There is also an attempt to represent the feet. Traces of white slip. Height, Fig. 16. 8.5 cm. Clay, dark yellow. Fig. 15. GROUP 2 : CONSIDERABLY DEVELOPED PLASTIC ORNAMENTATION (ONE BAND). 27. (Plate XLII. 4.) Standing figure, with close-fitting necklace (cf. p. 11) extending to nape of neck, and bosses at shoulders, to represent TrepoVai. The figure has a stephane. Traces of white. Height, 7.5 cm. Red clay. 28. (Plate XLII. 12.) Similar figure, with wide band stretched from shoulder to shoulder, surmounted by round bosses (cf. pp. 10 f.). The hair is represented by curls across the fore- head, bound with a taenia. White slip, traces of red lines on band and down left side. Height, 11 cm. Light yellow clay. 29. Forty-three fragments of figures similar to Nos. 27 and 28, adorned with one necklace only. Almost all have the stephane, and show some develo])ment of the hair. They also have very considerable traces of white coat and ornamentation in red and black. 30. (Plate XLII. 2.) Standing figure, arms, head, and feet broken away, with narrow band which passes over left shoulder and under right arm. About the neck are traces of a close-fitting necklace. Slight traces of white. Height, 6 cm. Yellow clay. Here the band seems clearly to represent the fold of an outer garment, an-anged in the manner of the later himation (cf. the Acropolis statues, Mvnees iTAtheiiex, pis. ii. and iii. ; Heuzey, Fif/. Ant. pi. xii. 4 ; and our own Nos. 163, 164, and 175). The oblique band is very similar to that of the primitive vase in human form from Hissarlik, Schliemann, Ilios, p. 343, No. 235. TIRYNTHIAN ARGIVE 19 31. Two fragments of similar figures, with l)ainls passing from left shoulder to right side. 32. (Fig. 17.) Standing figure without stepliane. The hair is formed by curls arranged about a centre. The arms and all the upper part of the body were enveloped in a thin layer of clay, which formed a sort of shawl (now preserved only on right side). No traces of color. Height, 7 cm. Gi'eenish yellow clay. 33. (Plate XLIII. 2.) Seated female figure with stephane, similar to No. 12, but decorated with a necklace from shoulder to shoulder. The eyes are not indicated plastically. The stephane has a boss. The feet are indicated by two projections. Covered with white slip, marked with red and black horizontal lines on chiton and chair. Height, 9 cm. Red clay. Cf. Heuzey, Fig. Ant. pi. xl. 1 (from Cyrenaica). 34. (Plate XLIII. 3.) Similar figure, with ends of necklace enlarged into ■n-epovai. The figure has no separate support, but is held upright by two legs attached directly to the back. Covered with white slip, traces of red lines at neck and waist. Height, 9.5 cm. Red clay. Cf . Schliemann, Tiryns, p. 157, Nos. 87 and 88 ; and on the whole series of seated figures, the figures from Tegea (Martha, Cat. des Fig. du Mus. d'Athhie.% Nos. 541, 542 ; and Lenormant, Gaz. Ardi. 1878, pp. 44 ff.). 35. Four hundred and ninety-one fragments of figures similar to Nos. 33 and 34, with stephane and single necklace, both with and without separately made chairs. They all show the same sys- tem of decoration in red and black lines on a white ground, with occasional use of color applied directly (for the bodies). Clay, usually red or yellow. GROUP 3: ELABORATE PLASTIC ORNAMENTATION (TWO OR MORE BANDS). 36. (Plate XLII. 5.) Standing figure with two necklaces. The hair was formed by four notched strands at back of head, and shows traces of a stephane. Hair, eyes, and body painted black, face and breast natural color of the clay. Height, 6.5 cm. Dark yellow clay. 37. (Plate XLII. 11.) Similar figure with two necklaces and ir^pofaL as KovporpotfxK (with an infant in her arms). The head of the child is broken away. White slip, traces of red on neck- lace, on infant, and on lower edge of chiton. Height, 7.5 cm. Red clay. Cf. p. 13. 38. Six almost identical figures carrying infants, both with and without necklaces. 39. (Fig. 18.) Standing female figure, carrying a child on her back; he sits on her shoulders and grasps her around the head. She wears the customary stephane with boss. Hair, two strands over eyes. General traces of wliite. Height, 5.5 cm. Gray clay. Cf. the figure given by Heuzey, Fig. Ant. pi. iv. 4, which has a child both at front and at back. J^ ^^^ ^- (Plate XLII. 10.) Standing female figure of usual type in stephane, ^^H ^m with one double and (originally) two single bosses. She wears a close-fitting ^^%-Mf necklace with pendant, and long double necklace across shoulders, between two Fig 18 round pins. Plentiful traces of white slip, and of red lines on stephane and neck- laces. Height, 8 cm. Dark yellow clay. 41. (Plate XLII. 13.) Similar figure, but without stephane. Tlie hair was originally long curls, now broken away. She has three necklaces, consisting of a twisted band between two plain ones, and double bosses represent the dress-pins. The fold of the chiton between the legs seems to be represented by a slight indentation, but this might be due to a defect of the clay. The body is painted black. Height, 7.5 cm. Gray clay. 42. (Plate XLII. 7.) Similar figure in high stephane with three bosses. Hair, simple mass, crossed by horizontal grooves. Long necklace with three pendants across breast. Large round pin on left shoulder ; the one on the right is broken away. The feet are carefully indicated by wedge-shaped projections, and the figure stands on a plinth. White slip, with traces of red on 20 •** TERRA-COTTA FIGURINES necklace, pii», liair, and stephane ; body, from waist down, black. Height, 10 cm. Dark yellow clay. 43. (Fig. 19.) Fragment of figure with two necklaces and pecnliar form of pin on riglit shoulder ; tliis consists of a wedge-shajjed i)iece of clay, and is plainly only an elongated form of the round i)in of earlier numbers ; it forms a transition, however, to the elaborate pins of later numbers. Ilair two large curls, surmounted by stepliane. Traces of white. Height, 5 cm. Vellow clay. 44. (Fig. 20.) Similar female figure, with elab- orate applied ornaments. She has a stephane with double boss, bound about the bottom witii a taenia. The hair is represented by four spiral curls over Fig. 19. forehead, three long curls at back. Four neck- laces (one twisted, one with pendant), and under them (appearing at left .side) two more bands, which can hardly be anything but an attem])t to rej)resent the fold of the chiton. On tlie right shoulder is an elaborate j)in, consisting of two cross-bars and two bosses, with traces of a third cross-bar (tyi)e of No. 84). The feet are marked with three gi-ooves each. A wide band of clay, broken at both extremities, appears under right arm, and there are traces of a similar band under left arm. These may have served to form a loop for suspension, for the figure could never have been intended to stand. White slip, with red lines on face and breast ; stephane, taenia, and body red : Fig. 20. hair black, black band at waist and at bottom of chiton. Height, 9 cm. Clay, dark yellow. 45. (Plate XLII. 8.) Similar figure, with deep indentation in beak, which brings out nose and chin. The figure wears a simple stephane. Hair, two large curls over forehead, six notched strands at back. There are three necklaces (simple band between two notched bands), with flower- shaped irtfrnvr} ou left shoulder. Face and necklaces show thick white slip; the body has horizontal red lines. Height, 13.5 cm. Clay, greenish yellow. 46. (PivATK XLII. 9.) Similar figure with the same indentation for mouth. The hair con- sisted originally of thick twisted braids (two pieces only are preserved). Four necklaces — twisted band between two plain ones, toppe<l by waved band. The fibulae were elaborate ; each consisted originally of two large bosses of three concentric circles, with small bosses between. Thick white slip on face and necklaces, narrow red line at waist. Height, 11.5 cm. Clay, reddish yellow. 47. Thirty fragments of figures similar to Nos. 40-46, with two or more necklaces. They are all crudely made, often without trace of color, although all ])robably had at least the usual white coat with line decoration. 48. Fifteen fragments of bodies, broken below waist, so that it is impossible to determine scheme of decoration. 49. (Plate XLII I. 6.) Seated figure with stephane (broken at front) and two necklaces ; also originally with pins at the shoulders. The hair is treated as a single mass over each eye. Cus- tomary white slip, some traces of red horizontal lines on neck (to represent another necklace?), necklace, waist, and lower part of chiton. Height, 12.5 cm. Clay, red. 50. (Plate XLIII. 7.) Similar figure with four necklaces (three plain bands and one twisted band), seated in wide chair. Hair is treated as one single, large curl on top of head. White slip. Height, 9.5 cm. Dark red clay. Cf. Schliemann, TlrynH, ]>. 156, No. 86. 51. (Plate XLIII. 10.) Fragment of similar figure, broken at waist and right shoulder, with three necklaces (a i)lain band between two twisted bands), ending on each shoulder in two round bosses. Tiie mouth is distinctly marked by a straight groove across beak. The figure has earrings consisting of one boss applied to another, and stejjhane. Tiie hair is treated as a single strand over each eye, notched in front, as a mass at back, criss-crossed with vertical and liorizontal grooves, and notched at sides of head. Traces of wliite. Height, 5.5 cm. Dark yellow clay. TIRYNTIIIAN ARGIVE 21 52. (Fig-. 21.) Upper part of female figure, with a remarkable brooch on right shoulder — a large spiral, adorneil with rosettes and bosses. (In this single case, it seems possible that this ornament is of a floral nature, and has some connec- tion with Hera Antheia; ef. j). 10.) Hair, a large mass, which covers head and shoulders, hatched with horizontal and vertical grooves. The earrings have tiie form of rosettes, witli raised bosses in the centre. Traces of white. Height, 5 cm. Red clay. 53. (Plate. XLIII. 8.) Similar figure, elaborate decoration. It has six necklaces, four plain and two twisted bands ; the lower one passes around the back. Earrings in the forms of double bosses. The stephane also has a large boss. The hair is treated as a long notched curl over each eye, with five spirals at back of head. Traces of white, with red on stephane. Height, 7 cm. Reddish yellow clay. 54. (Plate XLIII. 9.) Elaborate figure, broken at waist ; left arm and right forearm missing. The figure has no necklaces, but a deep groove runs across the breast and around the shoulders anil back, to mark the edge of the chiton. The stephane is pierced by three holes, i)robably intended for inserted ornaments. Earrings in the form of rosettes. The hair consists of two bands above forehead, filled with irregular lines of small holes, — a not altogether unsuccessful attempt to indicate the texture of the hair. Below these bands the hair falls in spiral curls (originally there were two wiiich fell in front of shoulders, and four at back), which remind one strongly of metal work. The treatment of the body at the waist marks a style which is not constructively that of terra-cotta figures, but is more like the technique of beaten metal. On the shoulders are fibulae in the form of four-leaved clovers (cf. p. 10). Some traces of red on chiton and on stephane. Height, 8.5 cm. Gray clay, now black from burning. 55. Five hundred and seventy-seven more or less broken seated figures of the type of Nos. 49-53, with two or more necklaces, usually with stephane and pins. Some traces of decoration in red and black lines on a white ground, with occasional application of paint directly to the clay. 56. (Fig. 22.) Upper part of "■ bird-faced" figure, broken at waist. Hair, single wavy curl above forehead. The figure is entirely wrai)ped in a sort of • shawl, which rises to a point over the head. Thick white slip ; traces of red and black on fragment of back of chair. Height, 6.5 cm. Yellow clay. Cf. No. 32 and the figure from Tegea, Martha, Cat. des Fig. du Mtis. / 'A fr^<^e«es, Nos. 554-558. t^ « 57. (Plate XLIII. 4.) Seated female figure, holding infant (heads of both ^^ ^, figures missing). The seated figure is of the regular Tirynthian type, with a necklace from shoulder to shoulder. The child is an oblong lumj) of clay, with a groove at bottom to mark the feet. Some traces of white. Height, Fig 22. ^ '""• •^^'^ clay. Cf. p. 13. Sa Fragments of seven similar figures, all of the regular " bird-faced " type, to which a lump of clay is added to represent a child. Some traces of white. Clay, red to yellow. 59. (Plate XLIII. 11.) Fragment of female figure, broken at neck and waist, on wide seat, which was evidently intended for two figures. The fragment preserved wears a waved necklace of two bands, and was doubtless of the " bird-faced " tyjie. In her laj) she holds an object like a broad roll or band, marked with five grooves, which evidently extended to the other figure of the group. Slight traces of white, with red on taenia and black on the body of the woman. Height, 6.5 cm. Length, 8.5 cm. Reddish yellow clay. 60. (Plate XLIII. 5.) Female figure from a similar group (the seat is hei-e broken close to the fio-ure, but was plainly intended for two figures, as it shows no traces of legs at the point of breakao-e. The face is of the usual "Tirynthian Argive" sort, with a single mass of hair over each eye. The entire figure is wrapped in a mantle (cf. No. 56) meeting in front, which rises to a point above the head. In the lap are traces of a roll or band, similar to that of No. 59. Traces of brown on hair and lower part of chiton. Height, 8.5 cm. Reddish yellow clay. 00 TERRA-COTTA FIGURINES Fig. 23. 61. (Fig. 83.) Similar female figure from right hand side of group. The mantle is here broken at the back, while in front it leaves the neck and shonlders bare. The figure has a simple stephane and a single necklace, adorned with three bosses. Wliite slip, traces of red on necklace. Height, 6 cm. Clay, reddish yellow. 62. Fragment of a similar group of large size (only left hand figure preserved, broken at shoulder and at lower right hand corner). In the lap is a fragment of a broad band, similar to the bands of Nos. 59 and 60. General traces of white. Height, 11 cm. Red clay. The interpretation of these groups is an interesting problem. So far as we can judge, both the figures were feminine, for in Nos. 59, GO, and 62 we have the left hand figure of the group, and in No. 61 the right hand figure, and these are all feminine. Moreover, the traces of the second figure, preserved on the seat of No. 59, point to a female rather than to a male figure. This fact at once declares against the theory that we are here dealing with a representation of Zeus and Hera, a sort of [c^jos ya/ios (such as the group given by Overbeck, Kunstmythologie, Hera, fig. 4 a, from Gerhard, Ant. Bildwerke, pi. i.), which we should most naturally expect to find at a sanctuary of Hera.^ The association of Hebe with Hera, as she was later associated with her in the group of Polycleitos and Naukydes, seems hardly probable at the early date to which we must assign these groups. We are reduced, then, to the view that we have here an early form of two female divinities like the later Demeter and Kore. (For another very primitive group of these goddesses, cf. Heuzey, Fiff. Ant. pi. xiii. 3 ; later examples are pis. xviii. bis. 3 ; xxiv. 1. Cf. the same author's article, ' Groupe de Demeter et Kore,' in Mo?i. grecs 2)ub. par V Assoc, pour V encouragement des etudes grecques en France, 1876.) The roU may then be the symbol of the bond between the two goddesses. 63. Four hundred and thirteen fragments of bodies of the early seated type — lower parts only preserved. They usually show some traces of the white coat, and often are decorated at the lower edge and at the waist with simple line patterns in applied red and black. 64. Two hundred and fifty-nine detached heads of the " bird-faced " type, with round lumps for eyes, and (usually) simple stephane. Traces of the white coat are common, and the stephane is often red. III. MYCENAEAN. 65. (Fig. 24.) Fragment of female idol of the usual Mycenaean form round body — broken at waist and neck. Ornamenta- tion in wavy lines, brown to black. Height, 4.5 cm. Fine yellow clay. Cf. Schliemann, Mycenae and Tiryns, pi. C, fig. m. 66. Nine fragments of similar figures, same clay, same ornamentation. 67. (Fig. 25.) Female figure with arms raised (pelta- shaped body.) Ornamentation in red lines. Height, 8 cm. Fine yellow clay. Cf. Schliemann, 3Iycenac and Tiryns, pi. B, figs, e and f. 68. Thirteen fragments of similar figures, same clay, orna- mentation in red, changing to brown and black. 69. (Fig. 26.) Female idol with arms (two applied strips of clay) crossed below breasts, which are also represented plastically, disks. Broken at top and bottom. Ornamentation in red lines. Hei Fine yellow clay. Cf. Schliemann, Mycenae and Tiryns, pi. C, fig. 1. 1 The Heraea mentioned by Pans. (II. 24. 2) probably whole subject, cf. ihid. pp. 177-181 ; Roscher, Lex. der had some relation to the Up),! yd/ios (ef. Daremberg et Myth. I^ p. 2008; Fiirster, Die Hochzelt des Zfm u. der Saglio, Did. des Ant. ' Hi^ros gamos,' p. 179). On the Hera, Breslau, 1867. Fig. 25. by applied ght, 5.5 cm. ftG. 26. Fig. 27. GEOMETRIC 23 70. (Fig. 27.) Body of Mycenaean figure with arms raised, broken at top and bottom. The breasts are connected by a crescent-shaped band (meant to represent a neck- lace ? cf. No. 36, and the Tirynthian Argive figures generally). Usual wavy hues, shading from light to dark brown. Height, 3.5 cm. Reddish yellow clay. 71. Twenty-six fragments of Mycenaean ware, so broken that the exact form cannot be determined ; same fine clay, with ornamentation in wavy lines. 72. (Fig. 28.) Large bull's or cow's head, broken at neck and minus horns. Ornamentation in light red, straight lines on fore- head and nose, circle around eyes and end of nose, hatched lines ^^ on horns, filling of double squares on cheeks, and y|k^^ peculiar ornament at back of neck (Fig. 28 b). In £l^^L the middle of the forehead is a vertical hole, possibly ^^JF^^ meant to contain a bronze axe or some other orna- ^ ment (cf. Schliemann, Mycenue, p. 218, figs. 329, 330 ; and the large silver head, Fro. 28 b. i^id- P- 216 f., figs. 327, 328). The theory of Perrot (Mstoire tie V Art, VI. p. 822), that this axe symbolized the axe used to slaughter the victim, seems to us probable. Height, 8 cm. Width, 10.5 cm. Fine yellow clay. 73. (Fig. 29.) Body of horse, broken at legs and neck. Line orna- mentation in brown, shading to black. Length, 7.5 cm. Fine yellow clay. 74. Nine fragments of other animals, much broken, same clay, same ornamentation. Fio. 28. Fig. 29. IV. GEOMETRIC. 75. (Fig. 30.) Rude round-bodied figure, overladen with ornamentation. It presents many analogies to the elaborate Tirynthian Argive figures. The nose was originally of the " beak " type. The eyes are incised triangles. Ear very large, witli large pendant. The hair was a large mass, now broken away. A single band passes from the neck wider each arm, and there are two simple necklaces, which carry a large brooch, broken at lower edge. On the shoulders are " clover-leaf " pins, with five bosses each. The decoration in color is Geometric (Dipylon), in red paint, sliading to black. It consists of horizontal lines, zigzags, and dots below waist, of dots alone on necklace, pins, and brooch. Height, 12.5 cm. Reddish yellow clay. Cf. pp. 6 f. 76. (Plate XLVIII. 13.) Geometric horse, legs broken. The deco- ration is entirely in straight lines, except at shoulder, where there are two curved lines. This horse is very similar to the horses which are so com- mon upon the covers of Geometric vases (cf. Rayet et Collignon, Histoire de la Ceramique Grecque, p. 33, fig. 21), and perhaps came from a vase. Color, lustrous black. Height, 9 cm. Yellow clay. Among the other animals of Class IX. several show traces of Geometric influence in their deco- ration, but in none are these traces so pronounced as to justify their transference to the Geometric class. Cf. p. 6. Fig. 30. 24 TERRA-COTTA FIGURINES V. ADVANCED ARGIVE. eyes wide and staring, traces of two necklaces. Till" body General i.s still a traces of Face of Fig. 31. 31.) Seated female figure, varying from primitive type only in that the head has eyes, nose, and month carefully modeled. The face is of pronounced archaic type — mouth turned up at corners, mere oblong piece of clay ; it shows white. Height, 7.5 cm. Clay, red. C'f. Schliemann, Tiryiix, p. 160, No. 95. 78. (Fig. 32.) Similar figure, broken at waist, and lacking right ai-m. marked archaic type — eyes, protruding masses. Ilair, single long curl over each shoulder, topped by stephane with (originally) seven small bosses. The breasts were modeled separately and ap])lied, as is evident from depression to receive right breast. In the hand, the thumb and the fingers are distinguished, and a hole is left between them (for holding an object ?). Traces of white, of red on stephane. Height, 8 cm. Red clay. 79. (Plate XLIV. 2.) Seated figure, which shows some attempt at model- ing the body. The swell of the breast is quite well reproduced, the hair is moulded with the face, the ear is fairly well given, although placed too high. The stephane has long ends behind, and thus forms a sort of taenia. The dress- pins are of the usual " boss " type. The band is here plainly a necklace, for the fold of the chiton is reproduced at the neck. There is no chair, but only two legs attached to the back of the figure. Color, white, with red on necklace, at waist, and on lower edge of chiton. Height, 12 cm. Light yellow clay. 80. (Plate XLIV. 1.) Flat-bodied figure (head missing), seated in elaborate chair. She wears three necklaces — the second has three pendants, and the third ends in small round pins. On the back and arms of the chair are round bosses. Ground color, white ; the shorter necklaces are red ; the long one has oblique lines ; back of chair, vertical lines ; arms of chair, horizontal lines ; bosses, radiating lines ; broad band at waist and below knees — all red. Height, 8 cm. Yellow clay. Cf. Martha. Cat. ties Fuj. du Mus. d'Athenes, No. 731 ; and the statuette from Thisbe, now in Dresden, Arch. Anz. 1895, p. 220, 4. 81. (Fig. 33.) Fragment of seated figure, broken at neck and below waist. Simple necklace with pendant, followed by twisted necklace, and wide band, which is probably here the fold of the chiton. The fragment on the right shoulder is part of a large fibula. The figure is interesting as showing the method of attach- ing the head in these advanced Argive figures; the body was drawn out into a wedge-shaped projection at the top, upon which the head was fitted ; the joint was then covered by a necklace in . front and by the mass of the hair behind. This figure also shows the legs of chair attached directly to tlie figure. General traces of white ; slight traces of red on orna- ments. Height, 12 em. Clay, red. 82. (Fig. 34.) Fragment of right t'iG. 33. shoulder, with fibula of the usual bronze type (cf. p. 10). The figure has two twi.sted necklaces, and the fold of the chiton is distinctly marked by an applied band of clay ; to this is affixed the fibula in form of a semicircle, with two pairs of rings for decoration. Slight traces of white, and of red on fold. Height, 4 cm. Reddish yellow clay. ADVANCED ARGIVE 25 Fig. 3C. Fig. 35. 85. (Fig. 37.) 83. (Fig. 35.) Body of figure, which shows greatest development of fibula. This here consists of a long bar, decorated with rosettes, and with cross-bai's of dumb-bell shape. Four bars and three rosettes are preserved, and this is probably the original number, as the pin begins to narrow considerably at the points of breakage. The ends undoubtedly curved forward, as in No. 84. The figure has two necklaces (one twisted) and distinctly marked fold. Traces of white on neck, neck- laces, fibula, and back, of red on fold and on cross-bars of fibula. Height, 12.5 cm. Clay, red. 84. (Fig. 36.) Fibula of most elaborate type, broken from its figure. It has three cross-bars and two rosettes, and curves for- ward at the ends. Ground color white, cross-bars marked with red and black criss-cross lines, rosettes with radiating red and black lines ; lower tip red. Height, 7 cm. Clay, yellow. Cf. pp. 11, 242. Head and left arm of seated, flat-bodied figure, with an infant on her arm (only legs and feet of infant preserved). Of the hair, only a single mass, with horizontal grooves, is preserved over the left temple. The figure wears a necklace with three pendants. General traces of white, red lines below neck- lace, red band at infant's knees. Height, 8 cm. Red clay. Cf. p. 13. 86. (Plate XLIV. 3.) Large seated female figure, broken at neck and at left lower edge. She held an infant in her arms, of which traces are pre- served only in her lap. Elaborate adornment with double necklace, which carries two pendants at the sides, double waved band, double straiglit band, and wide fold. The pins, strangely enough, are only small disks, as in the earlier types. Left arm disproportionately short. The chair was made Traces of white, with applied red on upper body and child. Lower body is painted red horizontal lines. Height, 18 cm. Yellow clay. 87. (Fig. 38.) Similar KovpoTp6<j)Oi figure with large infant on left arm. The infant's head is missing. Both woman and child are entirely enveloped in a thin layer of clay, which includes the back of the chair as well (cf. Nos. 56 and 60) ; but a break on the right side shows that the figures were com- plete before the appli- cation of this layer. The woman has a necklace with pen- dant, which was entirely hidden. The child wears the chiton, with small, round pins at the shoulders. The woman's face is round, witli protrud- ing eyes. The hair is treated in notched strands. Traces of white, of red on shawl. Height, 7.5 cm. Yel- low clay. 88. (Fig. 39.) Bust of very large seated figure, with three necklaces Fig. 37. separately, brown with Fig. 39. •26 TERRA-COTTA FIGURINES (plain necklac* between two twisted ones), double waved band and wide fold. On the back are traces of hair treated as a mass, and of back of chair. General traces of white, with red on fold. Height, 12 cm. The entire figure must have been fully 40 cm. tall. Red clay. Similar large fragment, with three necklaces and wide fold, topped by two narrow bands with round pendants. Traces of white, fold red. Height, 10 cm. Yellow clay. 90. ( Plate XLIV. 4.) Torso of seated figure with very elaborate adornment. About the neck runs a simple necklace decorated with a line of holes, followed by a twisted chain with large pen- dants, wliich are decorated alternately with grooves |E*'^f ** ^1 and rows of holes. The chiton is a distinct layer '^ T '^ ^ , . ^ ^ > '^ ^B of clay. From shoulder to shoulder, over the chiton, runs a waved band, adorned with fine holes ; between the turns of this band are rosettes (an imitation of lace ? or should we find here a |.j^. ^y trace of the metal ornaments found by Schlie- mann, which were intended to be sewn to the garment? cf. p. 11). The pins ai-e of the "clover-leaf" type, with double bosses in the centre, and single bosses on the leaves. At the back are traces of curls, and two notched bands are stretched from shoulder to shoulder over the chiton. General traces of white on neck and neck- laces, body brown. Height, 8 cm. Clay, light yellow. 91. (Plate XLIV. 5.) Similar elaborate torso. The chiton is a separate layer of clay. Two twisted necklaces. Over the chiton in front is stretched a waved stripe, decorated with incuse disks (such as are found in ceramic ware) followed by a line of pendants, with similar incuse disks, bosses at top and notches at bottom (cf. p. 11). The pin is similar to that of No. 90, but larger. Behind, there are traces of curls and a line of pendants from shoulder to shoulder. Traces of brown paint on chiton. Height, 8 cm. Clay, yellow. 92. One hundred and eighty fragments of figures similar to Nos. 77-91. The plastic decora- tion is throughout very elaborate ; the painted decoration consists sometimes of red and black lines on a white ground. Sometimes this treatment is applied only to the upper part of the body, and the lower part is given a solid brown color, applied directly to the clay. 93. (Plate XLV. 11.) Head and shoulder of advanced Argive type. The headdress is broken away. The hair was a row of curls over the forehead, topped by a notched mass, with the usual simple mass at the back, now broken away. The figure has two close-fitting necklaces with three pendants, followed by three larger ones (one of them twisted), decorated at the shoulder with a pin of the round boss type. The fold of the chiton is represented by two narrow bands ; it was held by a large fibula of tlie usual elaborate kind, of which only the upper part remains. White slip, red on ornaments. Height, 12 cm. Yellow clay. 94. (Plate XLV. 13.) Similar head and shoulder, color well preserved. Close-fitting necklace with pendant, followed by three others (simple band between two twisted bands) and double waved band. The pin is placed altogether too high, so that it can only be intended for ornament, — possibly as a pendant to the necklace. Disk earrings with pendants. Hair, two masses above forehead, with horizontal grooves, large mass at back, notched at sides of neck, The headdress consists of a calathus, bound by a wide taenia. The ground color is white ; black is found on hair (except at sides of neck), above the eyebrows, and inside of eyes; in radiating lines on ear- rings, in dots on pin and waved band ; red appears in radiating lines on earrings, in dots on pin and waved band, and in two wide streaks above and below close-fitting necklace. All the colors are carelessly applied, and seem to be meant simply to enliven the figure, without much regard to its anatomy. Height, 12 cm. Clay, yellow. 95. (Plate XLV. 2.) Head of advanced Argive type, with double stephane. The hair consists of four loops over forehead, single long curl at each side. Ground color, white ; traces of red on hair. Height, 5.5 cm. Gray clay. ADVANCED ARGTVE 27 96. (Plate XLV. 4.) Similar head, with stepiiane and boss. Hair, two notched strands over forehead, topped by two masses with horizontal grooves ; single strand over shoulder. Simple disk earrings. Traces of white. Height, 5.5 cm. Yellow clay. 97. (Plate XLV. 8.) Similar head ; has stephane with double boss, bound by a taenia. Hair, two notched strands over forehead, and mass at back. Double boss earrings. Traces of two necklaces. Ground color, white ; stephane, red ; hair, brown. Height, 6 cm. Yellow clay. 98. (Plate XLV. 9.) Head of advanced Argive type, in high calathus with three bosses and double taenia. Hair, two masses over forehead, wedge-shaped mass at back. One earring is pre- served — a simjjle disk. Traces of one plastic necklace, with pendant and boss, and of two painted necklaces, red. Ground color, white ; hair, black ; red stripes on calathus, necklace, and neck. Height, 9 cm. Yellow clay. 99. (Plate XLV. 5.) Similar head. The calathus has three bosses and is bound by double taenia. Hair, two notched strands, topped by two masses over forehead, large mass at back. Plain disk earrings. White slip ; taenia, red ; hair, black. Height, 7 cm. Clay, yellow. 100. (Plate XLV. 7.) Head of advanced Argive type with high calathus, fronted by two plumes (cf. p. 13) with a disk at the base. Hair, notched mass over forehead, topped by wide strands, which extend to shoulders. Earrings, large disks. A wide band represents the fold of the chiton. White coat ; traces of red on brows, earrings, side hair, and fold. Height, 8.5 cm. Dark yellow clay. 101. (Plate XLV. 10.) Similar head, plumes broken away. Double taenia. Hair, two notched masses over forehead, topped by large curls, mass at back. Plain round earrings. Traces of necklace. White slip ; hair, black ; red on taenia. Height, 9.5 cm. Red clay. 102. (Plate XLV. 1.) Similar head, calathus small, plumes large. Double taenia. Hair, two large curls over forehead, mass at back. Kosette earrings. General traces of white. Height, 7.5 cm. Clay, gray. 103. (Plate XLV. 6.) Similar head. The plumes are here connected behind the calathus, and the one that is preserved is adorned with a disk. Double taenia. Round earrings. Hair, large curl over each temple, mass at back. Traces of necklace with pendants. Slight traces of white. Height, 11 cm. Clay, dark yellow, shading to red. 104. (Plate XLV. 12.) Similar head, with large rosette between plumes, which here, also, are connected behind calathus. Double taenia. Rosette earrings. Hair, mass over forehead and at back, topped by two large curls. Slight traces of necklace with pendant. White slip ; hair black ; red on necklace. Height, 9 cm. Dark yellow clay. 105. (Plate XLV. 3.) Head of advanced Argive type. The features are very sharp and dis- tinctly archaic, — indeed, it may be a question whether this head ought not to be placed in the " Archaic " class. The eyes are wide and staring, mouth accentuated, in that the region separating the mouth from the cheeks and the chin is here defined by a round hollowing, which at the same time gives hardness. Hair, tight curls over forehead, treated in a conventional manner, like inverted wave pattern, spirals at back, now broken. Simple band about hair, traces of another ornament (perhaps calathus) on crown of head. Simple earrings. Traces of white on face ; hair, black. Height, 4 cm. Clay, yellow. 106. Two hundred archaic heads, similar to Nos. 93-105, always showing the same tendency to elaborate headdresses. _^ — ^_ Here, too, we have placed (see p. 7) — - I ^B 107. (Fig. 41.) Body of standing female figure, broken at neck. The body is ^^^m ^ perfect cylinder ; in fact, it seems to have been made on the wheel. The hair was a simple mass at back, red. The ornamentation consists of a broad stripe at waist, and four series of three fine red lines below waist — very similar to Argive-Linear work (see p. 3) in vases. Height, 8.5 cm. Reddish yellow clay. Fig. 4L Cf. for round body, Heuzey, Fig. Ant. pi. ix. No. 3 (from Cyprus). 28 TERKA-COTTA FIGURINES VI. FIGURES UNDER ORIENTAL INFLUENCE. Fig. 42. 108. (Fig. i'2.) Plaque with heraldic lions, broken at right lower corner. The position of the _ lions is very schematic ; each rests one fore paw on a pedestal, and raises the other liorizontally, so that the two raised legs meet for their wliole length. The tail is curved over the hack. Tiie heads turn and look backward. The surface is treated ratlier superficially, and the muscles are but little developed. No trace of color. Height, 10 cm. Chiy, red, in many places burned black. The Lion gate at Mycenae is strikingly similar to this plaque, as well as some Mycenaean gems (for instance, the one published by Tsountas, 'E(^. 'Apx- 1888, p. 175, 2 ; pi. x. 2). Between the two lions of our plaque probably stood a pedestal or altar (the upper edge can still be made out) similar to those of the gate and the gem. Tlie plaque would seem to belong to the end of the Mycenaean civilization. On the whole subject of heraldic animals, see Curtius, ' Uber Wappengebrauch n. Wappenstil im gr. Alterthum,' in his Gesammelte Abhamllungeii, II. pp. 77-115, pi. i. (cf. with our plaque especially Nos. 8 and 11). 109. (Fig. 43.) Upper part of winged female figure, broken at waist. The face is badly worn, but plainly archaic. Three notelied strands of hair fall to each shoulder. The head is surmounted by a stephane. Arms akimbo, — the hands held sickle-shaped objects, badly worn. Hair, red ; red bands on stephane, body, and wings. Height, 7 cm. Yellow clay. This figure is very similar to the Delos statue (^B. C. 11. III. [1879], pp. 393-399, pis. vi. and vii.), and, like it, was undoubtedly a running figure with one knee touching the ground. In terra-cotta, we have similar figures in Martha, Cat. des Fifj. dii Mus. d'Athhies, No. 9, and in Dumont et Chaplain, Les Ceramiques de la Grece Propre, II. p. 229, No. 46. These are both explained as goi'gons, and that is the interpreta- tion we are inclined to give to our figure. On the subject of running figures, cf. Curtius, ' Die knieenden Figuren d. altgr. Kunst,' in Gesam- melte Ahhandlnngen, II. pp. 116 ff. pi. ii., and Gerhardt, 'Uber die Fliigelgestalten der alten Kunst,' in his Gesammelte Akad. Ahhandlungen, I. pp. 157 ff. pis. ix.-xii. 110. (Fig. 44.) Fragment of a similar figure, running to riglit, badly broken. All that appears is the body, with arms akimbo, a small part of right wing, right leg to knee, and left leg to middle of thigh. But the figure is plainly an archaic running figure. Traces of white ; some red on chiton. Height, 5.5 cm. Reddish yellow clay. Cf. No. 109. lU. (Fig. 45.) Squatting male figure, type of the Egyptian Bes, broken at knees. The figure is that of a dwarf with hands placed on the abdomen, which is abnormally developed. Hair, a mass with horizontal grooves. No trace of color. Height, 6 cm. Red clay. The wide distribution of this class of figures may be seen from Orsi's note on the find at Megara Hyblaea, Mon. Ant. I. p. 838, note 1. Cf. the figure there given, pi. vi. 4. The development of this type is traced by Heuzey, 'Sur quelques representations du dieu grotesque appele Bes par les Egyptiens,' Comptes liendus de l' Academic des Inscriptions et Belles-Lettres, 1879, p. 140. 112. Six replicas of No. Ill, all more or less broken. Same clay, some traces of white coat. 113. Similar figure of smaller Heiglit, 5 em. Dark yellow clay. Fig. 43. size, feet broken away. No trace of color. EARLY ARCHAIC 29 114. (Plate XLVIII. 1G.) Fragment of large sphinx, broken at middle of body. The figure is very rude, consisting of no more than a head and neck ( in full face), set upon a winged body (whether of a lion or of some other animal it is impossible to determine). The technique is most interesting, for all the surfaces of the face ai'e plane surfaces, and appear to have been made with a knife or some similar tool ; this is true, too, of the wing. Moreover, there are no traces of paint on the head or on the wing, whereas they are plentiful on the other parts of the figure. It appears, then, that the figure was finished and painted, and at some later time the surfaces which show the knife, marks were formed by cutting away the original contours. The paint, which appears on legs and body, is I'ed. Height, 11 cm. Yellow clay. An interesting head for comparison is published by Tsountas, '1C<)!). 'A.px^. 1892, p. 13, pi. iv. 4 and 4" ; cf. also the poros heads from the Acropolis {Bev. Arch. XVII. [1891], pp. 304 ff. j)l. x.), which, some have held, were made by a workman during the noonday rest, exactly as our figure may have been remodeled. 115. (Fig. 46.) Small sphinx, head missing. The head was apparently feminine, for two locks of hair appear in front of each shoulder. No traces of color. Height, 4.5 cm. Clay, light yellow. 116. Fragment of similar sphinx, only front leg, body, and beginning of wing preserved. Traces of white. Height, 4.5 cm. Clay, brown. 117. (Fig. 47.) Small figure on a horse, gal- loping to left — plaque, stamped on both sides. Rude modeling, but with some attempt at repre- senting muscles of horse by means of grooves. Surface, green enamel. Height, 6.5 cm. Light, sandy material. This figure is stamped as Egyptian by its enamel. Moreover, Professor Petrie, in his Naukratis (p. 14), speaks of " figures of light, friable, sandy ware " — a description which applies exactly to our plaque ; and Professor Ernest Gardner informs us tliat some of these figures were exactly similar to ours. There can be no doubt, then, that we have here an imported article, as is also the case with a sniall fragment of similar material and glaze — the lower portion of a squatting figure, very much like the type of the Egyptian Bes, Nos. 111-113. Fig. 40. I'u,. VIL EARLY ARCHAIC. 118. (Fig. 48.) Fig. 48. GROUP A: HAND-MADE FIGURES. Standing male figure, both legs, left arm, and right forearm missing. The figure presents many analogies to primitive Argive figures — the eyes are round lumps, the nose is almost a beak, and the hair consists of four strands falling to the shoulders behind and a wavy curl on top of the head. Yet the mouth and chin are plainly indicated, and there is some attempt at modeling in the body, which produces something the effect of the early Apollo figures. The body is covered with a thick white slip, with applied red on hair, red lines at breast and waist, and red hatchings on legs (meant to represent pattern on chiton ?). Height, 11 cm. Clay, red. 119. Fragment of seated male figure, broken at hips. The upper part is a mere wedge, to which was attached a face in front, and a mass of hair behind. The aiSoIa are roughly represented, but far too high. Slight traces of dark brown on left arm, back of head and lower abdo- men. Height, 9 cm. Clay, straw. 120. (Fig. 49.) Standing female figure, rudely modeled by hand, but showing an attempt to distinguish the features of the face. The so TERRA-COTTA FIGURINES Fig. 49. Fig. 50. nose, now broken, was di.stinguisheil from the chin ; the eyes are incised circles ; the hair is a series of notched strands falling to shoulders. The arms are raised to the breasts, and three grooves at the end of each mark the fingers, (ieneral traces of white. Height, 8 cm. Light yellow clay. 121. Standing female figure with remarkably long neck, broken at waist, arms missing. Face very rude, nose very much like a beak, but distinguished from chin, mouth not indicated. The breasts are rudely applied lumps. Eyes, two incised circles. Hair consisted of notched mass at top and strands down the back (now broken off). White color is preserved in circles of eyes. Height, 10.5 cm. Clay, reddish yellow. 122. (Fig. 50.) Standing female figure with nose and mouth care- fully distinguished. Eyes, however, are lumps — a reminiscence of the earlier types. Hair, nine notched strands, spreading over shoul- ders. A slight groove from shoulder to shoulder marks the fold of the chiton. The forearms (now broken off) were extended forward. Three grooves extend from waist to bottom of chiton on each side, doubtless to represent folds. The ground color is black, covered with yellow dots, even on the parts of the neck above fold of chiton ; between the grooves, on the lower part of the figure, are hatchings of light yellow. Height, 11.5 cm. Clay, grayish yellow. 123. (Fig. 51.) Rudely made figure with round body ; head, right arm, and left forearm miss- ing. Traces of white. Height, 8.5 cm. Red clay. This figure stands in the same relation to the two preceding figures that the statue of Cheramyes (cf. Collignon, Hist, de la SculjHure Grecque, I. p. 163, fig. 73) does to the Nicandra statue {ibid. p. 120, fig. 59). It is an attemjit to vary the monotony of the square, board-shaped body by another form, which, however, is equally summary and untrue to nature. 124. (Fig. 52.) Rude round figure of a pregnant woman, head, arms, and feet broken away. The feet were attached to two stumps inserted in the lower part of the figure. The edge of the chiton below the neck is carefully indicated by a deep cutting. White slip. Height, 9.5 cm. Yellow clay. Representations of pregnant women are not uncommon in all periods of Greek terra-cottas ; cf. Schliemann, Bericht iiher die hungen in Troia im Jahre 1890, pi. i. 3 ; Dorpfeld, Troia, 1893, Schone, Gr. Eeliefs, p. 67, No. 142, pi. xxxvi. { = Bull. 1868, p. 54, No. 20) ; Stephani, Compte Rendu, 1865, pp. 193, 194, pi. vi. 6 ; Ant. du Bosj^hore Cimmenen, II. p. 91, pi. Ixix. 7. But all these figures are plainly intended for caricatures, whereas our figure, as a temple gift, can hardly be anything but a thank offering for a successful childbirth. It is appropriately dedicated to Hera as 'ElKfidvia. (Cf. Hesych, s. v. ElXeiOvlar "Hpa iv 'Apyei, and on the whole sub- ject, Roscher, Lexicon der Mythologie, ¥. pp. 2087 ff. 125. Forty-four fragments of figures similar to Nos. 118-122, mostly bodies of the o-aits type, roughly made by hand. A few show traces of applied plastic ornamentation. The painted orna- mentation is simple ; it consists principally of lines and dots, usually in red and black, although there are a few cases of applied white (to mark the girdle), and one figure has two purple bands falling from the girdle in front. The only pattern is found on the right side of one figure, where we may suppose the fold of the Doric chiton to be : — Ausqra- p. 101 ; Fig. 52. :^3C^: EARLY ARCHAIC 31 126. (Fig. 53.) Torso of female figure of coarse clay, round-bodied type. On the riglit slionl- ^^^^^^^^^^ der is an elaborate fibula of the " clover-leaf " type, and the left shoulder I^^^^BJBJl^l^^ shows traces of a similar fibula. At the back are traces of strands of T^^^^^^^^^^^M hair, and also of an applied band wiiich was stretched from shoulder to ^K^^^^j^HBf shoulder. The ground color is red, covered by a black coat, with ^^^^^^^^^f^ applied yellow lines at waist and neck, and yellow dots on fibula. ^^^^H^ Height, 6 cm. Clay, yellow. ^^^^r^ 127. Similar torso, forearms (which were extended) broken off. Pie 53 Traces of four strands of hair at back, and of one strand in front of each shoulder. Fold of chiton represented plastically and decorated with " laufender Hund " pattern in light brown. The entire body is covered with a light brown coat, to which is applied dark brown on bust and arms. Dark brown stripe at waist. Height, 5.5 cm. Yellow clay. 128. Third torso of similar style, arms broken. Fold of chiton moulded in the clay, both at front and back. Traces of long necklace from shoulder to shoulder behind. Hair was six notched strands. Fragment of large fibula on left shoulder. Ground color red, changing to black (appar- ently from burning). Applied white on neck and fibula ; white line at waist. Height, 5 cm. Dark yellow clay. 129. Fragment of heavy, seated figure of coarse clay (similar to the clay of Nos. 126-128), broken at neck and below knees. The woman leans forward and rests her elbows on her knees. The forearms are broken off. Color, black, with red lines at u])per edge of chiton and at waist. Traces of large fibula at left shoulder. Height, 8.5 cm. Yellow clay. Here, too, we may place, as a transitional type — 130. (Plate XLVII. 3.) Bust and head of rudely made figure of archaic style, with a mod- eled head. This figure presents striking analogies to the advanced Argive figures ; it has a low stephane, six large curls, large earrings, and a close-fitting necklace. Yet it is a real archaic figure ; the body was of the same type as Nos. 126-129. Paint applied directly : pupils of eyes, hair, and body, black ; brows, brown. Height, 6.5 cm. Yellow clay. 131. Fragments of four replicas of No. 130, with varying amounts of plastic ornament. Three have necklaces, the fourth is broken close to the neck. One has marks of a long necklace from shoulder to shoulder at the back. Two have applied earrings, while of the others, one has the ear- rings represented by circles sunk in the clay, the other has none at all. The hair consists in three cases of curls ; in the other it is a mass. All show marks of having had a stephane. The system of color is the same in all — body, solid color, in one case black, shading to red ; in the others, black ; necklace, a row of dots ; pupils of eyes, brows, and hair, same color as the body — in one case red, in the others, black. One figure has a black dot in the centre of each earring. Height, 3.5 cm. to 6.5 cm. Clay, yellow, except one figure, which is of gray clay. GROUP B: MOULD-MADE FIGURES. 132. (Fig. 54.) Fragment of standing female figure, broken at neck and waist, arms missing. The modeling is careless. The fig- ure is interesting only because it has at the shoulders round pins similar to those of the early and advanced Argive styles (cf. p. 8). General traces of white. Height, 4 cm. Clay, red. 133. Fragment of similar figure, without pins. The lower body is no more than a squared lump of clay ; the arms are stumps. Fold of chiton represented by a slight groove. Traces of white. Height, 7.5 cm. Reddish yellow clay. 134. (Fig. 55.) Moulded female figure, broken at neck and above knees. She wears a sleeveless chiton— the upper edge plainly marked by a depres- FiG. 54. Fig. 55. 32 TERRA-COTTA FIGURINES sion ' Fig. 56. Fig. 57. Fig. 58. ■and necUace with pendant. At each side of neck, one or two curls of a long lock of hair. The arms fall close to the sides. The relief is very flat, but there is an attempt to represent the breasts. No trace of color. Height, 7.5 cm. Clay, dark red. 135. (Fig. 56.) Badly worn figure of similar make, no orna- mentation. At height of shoulders, two supports were fixed to the back to hold the figure upright (cf. the Argive seated fig- ures). Traces of white. Height, 8.5 cm. Yellow clay. 136. Three fragments of similar figures, all showing remains of supports at level of shoulders. They all show traces of white coat. Heights, 4.5 cm. to 8 cm. Yellow clay. 137. (Fig. 57.) Lower part of large moulded figure, in very high relief (broken at waist). The chiton hangs in a long straight fold between the legs (cf. Miisees d'Athhies, pi. v.). Anns are pressed close to sides. Color, bright red ; arms, pink. Height, 10 cm. Reddish yellow clay. 138. (Fig. 58.) Similar fragment in lower relief, with line ornamentation in pur- ple (simple vertical lines, joined by horizontal and criss-cross lines ; at sides, rude horizontal lines only). Height, 6 cm. Clay, reddish yellow. 139. Fragments of three similar figures in low relief, arms always at sides. In two cases, the toes are modeled rather carefully. 140. (Fig. 59.) Moulded figure, broken at neck. Hands still at sides. From the waist down, the figure is a mere case, in which all the forms disappear, but there is an attempt at representing the feet. Above the waist, the modeling is summary, but fairly accurate. The breasts are clearly brought out, as are also the lower edge of the chiton and the indentation of the girdle (cf. Musees d'Athenes, pi. x.). The figure stands on a narrow plinth, which was included in the mould. White slip, with red bands around lower part of chiton. Height, 7 cm. Clay, gray. 141. Exact replica of No. 140, similarly broken, decorated with red bands. Height, 7 cm. Clay, gray. 142. Very similar figure, but not replica. The proportions are slightly smaller than those of No. 140 and No. 141 ; the figure may have been made from a mould which was itself made from No. 140 or No. 141 and shrank in baking.^ It shows only traces of white. Height, 7 cm. Red clay. 143. (Fig. 60.) Fragment of similar figure, broken at waist. The hair is represented by notches and falls to shoulders. The ear is too high, as is usual in archaic work. The edges of the chiton at neck and waist are dis- tinctly marked. Traces of white, of red on chiton. Height, 4.5 cm. Red clay. 144. (Fig. 61.) Fragment from the same mould as No. 143, minus face and broken at knees. This figure shows not only the lower edge of the fold of the chiton, but also the indentation of the girdle. Arms held close to sides. Slight traces of red. Height, 7.5 cm. Red clay. 145. Upper part of similar figure in high calathus (broken at waist). Fold of chiton and indentation of girdle well marked. Color, white ; traces of red on calathus. Height, 5 cm. Clay, red. 146. Small fragment of similar figure, broken at waist and below knees. Behind each hand (arms still hang close to sides) a small hole, possibly for suspending the figure. Traces of white. Height, 4.5 cm. Red clay. 147. (Fig. 62.) Similar figure, showing wide border all around, where the clay ran over the ' Cf . Martba, Cat. des Fig. du Mus. d'Athines, p. xxiii, on the subject of " sunnoulages." Fig. 59. Fig. 62. EARLY ARCHAIC 33 Fig. ()4. Fig. 63. edges of the mould. Arms hang close to sides. Edges of chiton marked by grooves at neck and waist, also at sides. The hair is represented by grooves, worked after the figure left the mould. Slight traces of red. Height, 8.5 cm. Dark yellow clay. 148. (Plate XL VI. 5.) Figure from the same mould, with stephane added. Traces of white on face, of red on chiton. Height, 8.5 cm. Yellow clay. 149. Fragment from the same mould as Nos. 147 and 148, broken at neck and knees. 150. (Plate XLVI. 14.) Bust of large standing figure broken at waist. High round stephane. The hair is carefully represented by curls in low relief over forehead, and three long curls in front of each shoulder. The eyes are remarkably long and triangular. Traces of white, of red on left side. Height, 10 cm. Gray clay. 151. (Fig. 63.) Somewhat similar fragment, broken at waist, face badly broken. Edges of chiton very plainly marked, also edge of girdle. Hair, large mass at back. Ears, remarkably large. No traces of color. Height, 10 cm. Red clay. 152. Fragment of similar figure, broken at waist and neck. General traces of white. Height, 7 cm. Red clay. 153. (Plate XLVI. 1.) Rudely modeled figure, with arms bent at elbows, hands touching shoulders. No attempt at representing details. Traces of white. Height, 7 cm. Yellow clay. 154. (Fig. 64.) Fragment of figure with arms simi- larly bent (broken at neck and above knees). Very careless modeling. Traces of white. Two red lines at waist, one at neck. Height, 7 cm. Greenish yellow clay. 155. (Plate XLVI. 6.) Standing figure in stephane, with hands raised to breast. All the lines are remarkably deep ; they seem to have been made with a pointed instrument. Hair treated in notches. Two grooves mark the lower border of chiton. No trace of color. Height, 8 cm. Clay, dark red, shading to black. 156. Two fragments of very similar figures, with hands raised to breasts (both broken at neck). One has white slip, with red horizontal lines, one at upper edge of chiton, three at waist, two at lov/er edge of chiton. 157. (Plate XLVI. 2.) Somewhat more advanced archaic figure, showing an attempt to represent the folds of the chiton. The figure wears the stephane, and the arms are represented as bent at the elbow, although the forearms are mere stumps. Traces of white coat. Height, 8.5 cm. Red clay. 158. Fourteen fragments of similar figures, mostly lower parts with folds in low relief. One is an exact replica of No. 157. 159l Figure very similar to No. 167 — folds of chiton fairly well given (broken at neck). The arms hang slightly forward, and are entirely given in the moidd. Traces of white. Height, 6 cm. Clay, reddish yellow. / ? l^^H ^^- Fragment of left side of archaic figure — forearm and hand k^K S 4.^^^^! "•rasping edge of chiton preserved — with fairly careful reproduction of ^B 1 ■ T^^m folds. This figure is not flat at back, but hollow, with thin walls, corre- ^ 1 ■ -^^m sponding to the later technique. No traces of color. Height, 8 cm. Red clay. 161. (Fig. 65.) Lower part of archaic figure, showing still greater development of folds of chiton. Here, not only the vertical folds which fall from the arm are brought out, but tlie folds between the legs are also fairly well reproduced. The modeling of these folds is softer than in most of the other figures of this class, giving an impression of gi-eater Fig. 65. freedom. This may partly be due to the fact that this was perhaps made 34 TERRA-COTTA FIGURINES when the mould itself was somewhat worn, and the lines, therefore, not so sharp. No traces of color, lleiglit, 8 cm. Light yellow clay. 162. Two fragments of similarly drajjed figures. One of them is painted a bright red ; the other shows a veiT plain thumb-mark at the back. 163. (Plate XLVI. 3.) Figure of the " Spes " type, broken at neck and at knees. The hitnatiou falls from right shoulder and j)asses under right arm. The chiton was doubtless indi- cated in color. Right hand raised to waist, and pierced by a hole for an attribute (perhaps a flower?). Left hand holds edge of himation. The folds are very carefully indicated. Himation, red — color applied directly to clay. Height, 8.5 cm. Yellow clay. Cf. Winter, 'Die Terrakotten von der Akropolis,' Arch. Am. 1898, pp. 140 if., esp. figs. 5, 11, 15 ; Lechat, ' Terres Cuites de Corcyre,' B. C. H. XV. (1891), pp. 1-112, esp. p. 79, No. 74, pi. viii. 2 (the specimen there mentioned had a hole in the hand, exactly as No. 163 lias) ; Heuzey, Fig. Ant. pi. xl. 2 (from Cyrenaica) ; Martha, Cat. des Fig. du 3fus. d'Athhies, No. 733 (provenance unknown, but probably Greek). The best short discussion of the " Spes " figures is that of Orsi, in his publication of the Megara Hyblaea terra-cottas ; see Moti. Ant. I. pp. 924 ff. 164. (Plate XLVI. 10.) Fragment of figure similar to No. 163 — broken at neck and waist. Same arrangement of himation, same careful adjustment of folds. Right hand raised to waist, left doubtless grasped edge of garment. Slight traces of white. Height, 5 cm. Light yellow clay. 165l Fragment of archaic figure, broken above waist and below knees. The folds are hastily indicated. The left hand hangs somewhat forward and holds an apple. At the back is a very plain thumb-mark. Traces of white. Height, 6.5 cm. Dark red clay. 166. (Plate XLVI. 9.) Aphrodite of archaic type, broken at neck and knees. Folds of chiton barely indicated. Both hands are raised, the right, carrying a dove, to breast ; the left, with flower, below it. Traces of white. Height, 5.5 cm. Clay, reddish yellow and very fine. 167. Fragment of similar figure, only left hand with dove and right hand side of drapeiy pre- served. Traces of white. Height, 4.5 cm. Fine yellow clay. 168. (Fig. 66.) Similar Aphrodite, broken at neck and above knees. Right hand, raised to breast, carries dove ; left grasps edge of sleeve of himation. Traces of white. Height, 7 cm. Dark red clay. 169. Similar figure, badly broken. Right holds dove, left raised edge of hima- tion. The folds of the himation are carefully given. On the shoulders, traces of three locks of hair. Traces of white. Height, 7 cm. Clay, reddish yellow. 170. Fragment of similar figure, broken at neck and waist. Right hand holds dove. In front of shoulders, four notched strands form the hair. Traces of white. Height, 6 cm. Reddish yellow clay. 171. Second fragment, broken at neck and just below breasts. Right holds dove. Traces of white. Height, 2.5 cm. Fine yellow clay. Cf. on all these figures, Nos. 166-171, the marble statue in Lyons (^Gaz. Arch. 1876, p. 133, pi. xxxi.) ; also the terracottas, Heuzey, Fig. Ant. pi. xii. 5 ; Orsi, ' Megara Hyblaea,' 3Ion. Ant. I. p. 925, pi. vii. 11; Michaelis, Arch. Zeit. 1864, p. 137, pi. clxxxii. 1; Martha, Cat. des Fig. du Mus. d'Athines, No. 433. 172. (Plate XLVI. 4.) Figure of usual archaic type, broken at neck and at knees. The figure wears the chiton, with a thick fold between the legs, and over it another garment (xitwi'iVkos ?) reaching to the knees (cf. the statue, Musees d'Athenes, pi. v.). Both hands are raised in front of body, the left below the right, and each holds a fruit. Traces of white. Height, 7.5 cm. Clay, light yellow and very fine. Cf. p. 15 ; Heuzey, Fig. Ant. pi. xl. 2 ; Martha, Cat. des Fig. du Mus. d'Athenes, No. 736. 173. Very similar fragment, broken at neck and thighs. Left hand holds a fruit ; attribute of right uncertain, but probably also a fruit. Traces of white. Height, 5 cm. Fine yellow clay. 174. Fragment of archaic figure, broken at neck and waist. Left hand at breasts, holds flower. Traces of white. Height, 4 cm. Red clay. This type is very common; cf. Orsi, 'Megara Hyblaea,' J/on. Ant. I. p. 924, pi. vii. 4, pi. viii. 3 ; idem, ' D' Una Citta Greca a Terravecchia presso Granmichele in Provincia di Catania,' EARLY ARCHAIC 35 ibid. VII. pp. 230 ff., pi. iv. ; Clarae, Miixee de ScuIpUire, pi. dcxxxii. J. 1449 H ; Lechat, 'Terres Cuites de Corcyie,' B. C. H. XV. (1891), p. 30, No. 7, pi. i. 2 ; p. 33, No. 14, pi. i. 8. 175. (Plate XLVI. 8.) Fragment of archaic figure, broiien at neck and at waist. She wears the himation falling from right shoulder to left side, and holds in front of her, with her right hand, a wreath. Traces of red on himation. Height, 4.5 cm. Clay, yellow and fine. Cf. the statue from the Acropolis (Collignon, Hht. de la Sculpture Grevque, I. p. 353, fig. 178), which holds a wreath in the right hand and a jar in the left; Orsi, ' D' Una Citttl Greea a Terravecchia presso Granmichele in Provincia di Catania,' 3Ion. Aiit. VII. p. 234, fig. 24 ; Kekule, Ant. Terrakotten, II. p. 9, fig. 4. It seems not unlikely that this figure has reference to the cult of Hera Antheia. 176. (Plate XLVI. 7.) Standing figure in chiton (edge at neck and waist, with folds at shoulder and below girdle carefully rendered). Broken at neck and at knees. Right hand holds a small animal (probably a hare) ; left grasps edge of sleeve of chiton at breast. At back are marks of a thumb and two fingers. Traces of white. Height, 11 cm. Dark red clay. Cf . the fragment of a pores statuette, Salzmann, La Necrojmle de Camiros, pi. ix. ; Bernouilli, Aphrodite, p. 38. 177. Fragments of two replicas of No. 176, showing the same finger-marks at back. Traces of white. Height, 7.5 cm. and 8.5 cm. respectively. Same clay as No. 176. 178. Twenty fragments of figures similar to Nos. 176 and 177, principally small bits of drapery. Dark red clay. 179. (Plate XLVI. 11.) Similar archaic figure, broken at neck. She wears chiton with thick fold between legs, and long overgarment like that of No. 172. Both hands hold attributes — right, an animal (stag?), left, bow (cf. p. 15). The break at the left shoulder is interesting, as it shows the method of manufacture by layers — the outer layer is of much finer clay than the inner (filling) layer. Traces of white. Height, 11.5 cm. Fine, light yellow clay. Cf. Heuzey, Fig. Ant. pi. xviii. bis. 1 (from Thespiae) ; Stephani, Compte Rendu, 1872, p. 161, pi. iii. 4 (from Kertsch ; the figure looks as if it had come from the same mould as No. 179) ; Lechat, ' Terres Cuites de Corcyre,' B. C. H. XV. (1891), p. 48, No. 37, pi. iii. 2 ; p. 51, No. 40, pi. iii. 1 ; Berlin Museum, 6831 (Attica) and 6262 (Thespiae) ; Syracuse, large terra-cotta room, case III. (Megara Hyblaea). 180. Four fragments of figures similar to Nos. 166-179, attributes indistinguishable on account of breakage. Traces of white. Clay, fine, varying from yellow to red. 181. Eight fragments of drapery veiy similar to that of Nos. 166-179, but belonging to hollow figures with thin walls. Traces of white slip ; slight traces of red. Heights, 4 em. to 13 cm. Clay, light yellow to red. 182. Basis of large statue of archaic type, with fragment of the figure from the knees down. The feet are carefully modeled and fairly correct. The bottom is pierced by a round vent. The chiton was red. Height, 7 cm. Height of plinth, 2 em. Length, 6 cm. Width, 5 cm. Clay, reddish yellow. 183. Six fragments of similar bases, usually with traces of feet only. General traces of white. The dimensions vary from 3 cm. X 3.5 cm. to 5.5 cm. X 7.5 cm. Clay, light red to yellow. 184. (Fig. 67.) Very rude seated figure, broken at bottom. The figure varies little from the standing type, except that it is slightly bent at the middle and has two supports behind (cf. No. 135). The arms, which are resting on the knees, are little better than two stumps. On the breast is a pendant. The upper part forms a rude stephane. The figure hardly appears to be modeled at all, yet the surface shows plainly that it was. Traces of white. Height, 5.5 cm. Red clay. 185. Similar figure with supports behind. She wears stephane and necklace with pendant (these made in the mould), and long veil falling to shoulders (a separate piece added after the figure was complete). Traces of white ; three red lines on bottom of chiton. Height, 10.5 cm. Reddish yellow clay. i'lG. 07. jgQ Third fi<>ure of the same sort with arms bent at elbows, and hands raised to shoulders. Traces of white. Height, 8 cm. Light yellow clay. 36 TERRA-COTTA FIGURINES 68*) Seated figure, very summary modeling — only feet, hands, and features brought out at all. The figure rests on a high plinth, whicli was moulded with it. The hair is a simple baud above forehead. Slie wears a stephane, from which a veil falls over shoulders. Solid, pierced only by air-vent. Traces of white. Height, 8 cm. Dark red clay. 188. (Plate XLVI. 12.) Replica of No. 187, only that plinth was made lower. Traces of wliite, of red on lower part of chiton. Height, 7 cm. Dark red clay. 189. Similar figure, but not from same mould. Modeling very summary. Traces of white. Height, 6.5 cm. Red clay. 190. Similar figure, but with liigli polos instead of stephane. Traces of red. lleiglit, 8 cm. Red clay. 191. Similar figure, seated on wide chair (broken at waist). Here, also, the veil covers the hair and falls to shoulders ; cf. No. 187). White slip. Height, 6 cm. Red claj'. For all these carelessly modeled figures, cf. Heuzey, Fig. Ant. pi. xi et Chipiez, Hist, de F Art, III. p. 425, fig. 299 ; Kekule, Ant. Termhottcn, II. p. 9, fig. 3 ; Fig. 68. Perrot Martha, Cat. des Fig. du 3£iis. d'Athenes, Nos. 227 £E. ; Panofka, Terrakotten des A". Mus. zu Berlin, p. 12, pi. ii. 192. (Fig. 69.) Hastily modeled seated figure, wrapped in flowing chiton, under which, how- ever, the outlines of arms, hands, and breasts are carefully brought out. The figure is pierced by an air-vent. Traces of white. Height, 7.5 cm. Reddish yellow clay. Cf. Gerhard, Ant. Bild. pp. 338 ff., pi. xcv. 1, 2 (Sicily); also in stone, the seated figures, discovered at Branchidae by Newton (^Discoveries at Ilalicarnassns, pp. 530 if., pis. Ixxiv., Ixxv.), and the archaic statue in the National Museum at Athens (Kavvadias, No. 6 ; 'E^. 'Apx- 1874, p. 480, pi. ixxi. Aa and A/3^. 193. Four fragments of similar figures, not, however, from the same mould. Slight ti-aces of white ; some red on chairs, feet, and chitons. Clay, yellow and red. 194. Similar figure showing an attempt at reproducing the folds of the chiton where it falls across the breast. Arms still at sides, hands resting on knees. Fragments of two supports at back. White slip with red lines Height, 6 cm. Yellow clay. figures, not from same mould. Traces of white ; one figure has red lines Fig. C9. on arms and breast. 195. Two similar (criss-cross pattern) on breast and above knees. Height, 6 cm. and 7.5 cm. Clay, yellow. 196. Lower part of large seated figure on high plinth, with considerable treatment of the folds at sides and between legs. Traces of white. Height, 7.5 cm. Dark red clay. 197. (Plate XLVI. 16.) Figure similar to No. 192, but with much better treatment of folds across breast and at waist. The outlines of the legs are also well brought out. The bauds rest on arms of chair, and both hold round objects (probably apples ; cf. Nos. 172, 173, and p. 15). The figure is pierced by an air- vent. Traces of white. Height, 7 cm. Dark red clay. 198. (Plate XLVI. 13.) Similar figure, with careful, though stylistic representation of folds by deep vertical lines, from breast to feet. Hands at knees ; fingers rudely indicated. The figure has an air-vent. Red line at neck. Height, 6.5 cm. Clay, light yellow and very fine. 199. (Plate XLVI. 15.) Similar figure without indication of folds, but with right hand raised, holding a bird to breast. Left hand rests on left knee. The throne on which she sits is carefully distinguished. Traces of red on chair. Height, 6 cm. Clay, light yellow and very fine. Cf. Nos. 166-171; KeTcule, Ant. Terrakotten, II. p. 10, fig. 10, p. 11, fig. 13 : also figures in Case VI., large terra-cotta room, Syracuse. 200. Fragments of five replicas of No. 199, all solid. Traces of white, with red on chair. Fine, liglit yellow clay. EARLY ARCHAIC 37 201. (Fig 70.) Hastily modeled figure of a child, broken at waist. On her right arm she holds a small doll. Traces of white. Height, 5.5 em. Red clay. 202. Thirteen small heads belonging to the type of Nos. 198 and 199. They are flat at back, of pronounced archaic style. The hair is usually a mass falling g_ ^I^H *'' shoulders, slightly notched over forehead, with polos or stephane. General ^^M^^^ traces of white. Heights, 2.5 cm. to 5 cm. Clay, fine, ranging from yellow to red. 203. Three very similar heads, not, however, flat at back. The features are archaic. Hair, mass. One head has the polos. Traces of white, of red on polos. liG. lO. Heights, 3 cm., 3.5 cm., and 4 cm. Clay, light yellow to red. 204. Lower part of crouching male figure with hands (disproportionately large) on knees. The figure corresponds very closely to one from Megara Hyblaea (cf. Mon. Ant. I. pi. vi. 5), and was undoubtedly a satyr or silenus. Traces of white. Height, 6 cm. Reddish yellow clay. 205. Lower part of squatting figure, probably a satyr, broken at waist. Hands rest on knees. Traces of white, sporadic bits of red. Height, 4.5 cm. Light yellow clay. There is a similar figure in Syracuse, large terra-cotta room, Case IH., from Megara Hyblaea. 206. Rude seated Pan — hollow figure, which probably served as a vase. Modeling very flat, only face, arms, and syrinx brought out. Slight traces of white. Height, 8.5 cm. Reddish yellow clay. 207. Head of archaic type, hollow, with vase orifice at the top. The hair seems to be held close to the skull by a veil. Traces of white. Height, 7.5 cm. Red clay. 208. (Plate XLVH. 13.) Very similar head, also with orifice at top, and casque-like hair- covering. Traces of white. Height, 6.5 cm. Red clay. 209. Eight fragments of similar heads, badly broken. General traces of white. Yellow clay, shading to red. These heads (Nos. 207-209) have the marks of the Rhodian ware as described by Heuzey ( Cat. des Fig. du Mus. du Louvre, p. 220), " la machoire inferieure osseuse et developpee a I'exces, ce que nous appelons le menton cfaloche ; de gros yeux triangulaires, dont la paupiere superieure est seule arquee." They are jirobably, therefore, imported, although a few may be local imitations. Cf. Heuzey, loc. cit. pp. 229, 230 ; Kekule, Ant. Terrakotten,U. p. 12, fig. 16, and p. 18, fig. 34 ; Winter, Arch. Am. 1893, p. 147, No. 28 ; Orsi, ' Megara Hyblaea,' 3Ion. Ant. I. ]). 805. 210. (Plate XLVIL 15.) Head of pronounced archaic type — eyes set obliquely to nose, mouth raised at corners. The headdress is very high, a sort of tiara (the kydaris ?), and from it a veil falls to the shoulders. Traces of white. Height, 7.5 cm. Clay, light yellow. 211. Three very similar heads, all with same headdress and veil. Slight traces of white. Heights, 5.5 em., 5.5 cm., and 7 cm. Yellow clay. These heads (Nos. 210 and 211) again resemble very closely a Rhodian type (cf. Heuzey, Fhj. du Mus. du Louvre, pi. xi. 2), which is found pretty generally distributed in the basin of the Mediterranean — in Greece, in Sicily, in Italy, and even in Phoenicia.^ Yet they show some di- vergences, for instance, the eyes are not set quite so obliquely as those of the Rhodian figures, and they are not so narrow ; so that they are probably of local manufacture, imitating Rhodian work, like the Syracusan figures mentioned by Heuzey.^ 212. (Plate XLVIL 14.) Archaic head, with elaborate treatment of the hair. This consists of two rows of curls over the forehead, moulded in the form, then a row of elaborate applied spiral curls, with eight curls at the back (now broken), the whole topped by a double, twisted taenia. The eyes are wide and very long, nose in two planes, mouth straight, ending in a pad of flesh at each corner ; a deep furrow extends from cheek-bone to chin on each side of nose. Traces of white. Height, 5 cm. Yellow clay. 213. (Plate XL VII. 12.) Large archaic head, very similar to No. 212. Same type of face. The hair here consisted of a row of scallops moulded in the form, with a row of applied curls above and large mass behind. The figure had large earrings, now broken. Traces of white. Height, 8.5 cm. Light yellow clay. These two heads (Nos. 212 and 213) differ greatly from the Argive type. Their nearest 1 Cf. Ilouzev, Cat. des Fig. du ^fus. du Louvre, p. 240. = Jbid. p. 227, n. 3. 38 TERRA-COTTA FIGURINES analogy in the HA'aeum finds is the bronze bust, 5, pi. II. 1, 2 ; and in terra-cotta, the head pub- lislied by Cesnola, Coll. of Cypriote Ant. II' pi. xv. No. 117 (cf. the sculptures, ibid. Ir pi. xciii. Nos. 622 and 623 ; pi. xcvii. Nos. 662 and 663). We are inclined, therefore, to regard these two heads as imported. 214. (Platk XLVII. 4.) Head of archaic style, with very sharp features. The hair is an applied mass at back, marked with notches and bound with a taenia. The ground color is greenish yellow, to which red was applied on eyes, cheeks, and brows. Hair, red, shading to black. Height, 4.5 em. Yellow clay. 215. Small, very rude bearded masque, slightly chijjped at sides. Eyes and mouth marked by wide grooves, beard by coarse vertical lines. No trace of color. Height, 5 cm. Red clay. 216. (Plate XLVII. 8.) Archaic female masque, original edge all around. The hair is moulded in masses. Hole for suspension above forehead. Traces of white. Height, 5.5 cm. Light yellow clay. 217. Large votive head, broken at right side and at neck. The hair is a row of round bosses, surmounted by stephane with suspension hole. Modeling sui^erficial. Traces of white. Height, 9 cm. Dark red clay. 218. Very similar head, almost replica, broken at neck, left side of face chipped away. Traces of white. Height, 7 cm. Clay, dark red. 219. Right side of large votive head, with treatment of hair in four lines of zigzags. Narrow stephane. Traces of white. Height, 6 cm. Dark yellow clay. 220. Similar fragment, with large round earring. Hair modeled as a mass, topped by stephane. Traces of white. Height, 8.5 cm. Light yellow clay. 221. (Plate XLVII. 9.) Similar masque with stephane. The hair is treated as short curls. Hole for suspension in front of stephane. Traces of white. Height, 5.5 cm. Red clay. 222. (Plate XLVII. 6.) Small masque with very high stephane, broken at neck. Ilair, a line of notches above forehead. Customary hole for suspension. No trace of color. Height, 4 cm. Light yellow clay. 223. Three replicas of No. 222, same clay. 224. Small bust of hasty workmanship, original edge preserved all around. The hair is a simple mass. Hole for suspension. No trace of color. Height, 8 cm. Greenish yellow clay. 225. (Plate XLVII. 10.) Archaic masque in high stephane, broken only at upper right hand corner. The hair is treated in two rows of loops. Traces of white. Height, 6.5 cm. Clay, straw. 226. Eight fragments of archaic masques, similar to Nos. 221-225, both with and without stephane. The suspension hole is common. General traces of white. Yellow clay. VIII. ADVANCED ARCHAIC. 227. (Fig. 71.) Upper part of standing figure with high stephane (broken at waist). Hair, a simple roll below stephane. At the shoulders were two projections (now broken off), which doubtless served to suspend the figure. Some traces of white coat. Height, 9.5 cm. Reddish yellow clay. 228. Very similar smaller figure, broken at waist. This figure also has stephane and projections at shoulders. Two strands of hair fall in front of left shoulder, one in front of right. Traces of white. Height, 6 cm. Yellow clay. 229. Lower part of seated figure with hands on knees. The figure is hollow, like the later terra-cottas of free style. The folds of the chiton, however, are still represented in a stiff, archaic fashion. The color is well preserved — white slip, with traces of applied red. Height, 5 cm. Clay, reddish yellow. 230. (Plate XLVII. 11.) Votive head, hair treated in waves above forehead, and in notches at sides. Stephane, pierced by suspension hole. Traces of white. Height, 5.5 cm. Dark red clay. Fig. 71 ANIMALS 39 231. Three exact replicas of the preceding, same traces of color ; same clay. 232. Seven heads similar to No. 230, but not replicas. Same treatment of hair ; stephane with hole for suspension. General traces of white. Height, 4 cm. to 5 cm. Clay, light yellow to red. 233. Twenty-three fragments of similar advanced archaic heads, much mutilated, but probably all votive heads. General trace of white. Clay, usually red ; one or two specimens are of yellow clay. f234. (Fig. 72.) Head of advanced archaic style, solid. Hair treated as a simple mass. Traces of white. Height, 3.5 cm. Light yellow clay. 235. (Plate XLVIL 1.) Similar head, solid. Hair, mass. Traces of white. Height, 4.5 cm. Clay, light yellow. ^^ ^^ 236. (Plate XLVH. 2.) Similar head, solid. Hair, mass, but treated in waves over forehead. Traces of white, of red on hair. Height, 4.5 cm. Dark yellow clay. 237. (Plate XLVH. 5.) Similar head, solid. High coiffure, treated as a mass. Traces of white. Height, 4 cm. Dark red clay. ^'<'* '^^• 238. (Fig. 78.) Head of similar style. Hair, wavy curls over forehead, and at sides of neck ; mass at back under veil. Traces of white. Height, 6 cm. Clay, reddish yellow. IX. FREE STYLE. 239. (Plate XLVIL 7.) Fragment of large votive head, broken at left side. The whole treatment is in large surfaces, free from archaic restraint ; the mouth has the natural curve, the outlines of the eye are correctly given, even to the lapping of the upper lid over the under. A ridge down the middle of the face seems to show that the head was made in two pieces. Traces of white. Height, 8.5 cm. Red clay. This head, in spite of its fragmentary condition, seems to us to represent the great style of the fifth century, and has inherent traces of the influence of some large statue. The peculiar marking of the hair, undercut and thus relieved from the face, and falling on either side in a mass of curls (here boldly sketched, though not elaborately modeled), is closely analogous to the Argive coin in which a copy of the Polycleitan Hera has been identified. It also suggests a comparison with the newly identified marble head in the British Museum (see General Introd. to Sculpture, p. 23). We are thus encouraged to consider this terra-cotta head as immediately related to the great Polycleitan Hera. Cf. Waldstein, ' The Argive Hera of Polycleitus,' J. H. S. XXI. (1901), pp. 30 ff., especially pp. 43, 44, and fig. 2. 240. (Fig. 74.) Votive bust, broken at neck, but with original edge at waist. The folds of the chiton and of the long veil which falls over the shoulders and the arms are all given with perfect freedom and naturalness. The figure is covered with a white coat. Height, 9 cm. Clay, reddish yellow. X. ANIMALS. 241. (Plate XLVIII. 7 and Fig. 75.) Group, representing a bull attacked by a lion. The lion grasps the bull around the shoulders and bites him in the neck. The bull is plainly exhausted ; his head sinks to dhe side, and one of his hind legs is curled up under him, while the other is stretched straight out behind, as in the well-known Acropolis group. The execution of the group is fair — the anatomy of the bull's head is quite well given, and the muscles are fairly well brought out. The fore legs of tlie lion are too long, otherwise he is well modeled ; his mane is formed of a number of round disks, close together. Technically, it may be noted that the bull seems to have been made complete (his tail curls over his back), and the lion was then " applied," so to speak, to him. White slip ; black lines on forehead of bull and front legs of lion ; red 40 ... TERRA-COTTA FIGURINES streak at lion's inotith, extending across the bull's neck (probably a representation of blood). Length, 10 em. Height, 5 cm. Clay, light yellow. The motive of a bull attacked by one or more lions is one of the oldest, and one of the com- monest in Greek art. In Mycenaean times, we find rejiresentations of this subject on gems (cf. Tsountas, 'E.^. 'Apx- 1888, p. 177, No. 25, pi. x. ; ibid. 1889, p. 166, No. 18, pi. x. ; Perrot et Cliipiez, Ifist. de VArty VI. p. 771, pi. xvi. 21) ; and on an ivory plaque from Sparta (Z?. C. II. II. [1878], p. 213, pi. xvi. 4). In archaic art it is very common ; cf. for example, the frieze of the temple of Assos (Clarac, Musee de Sculpture, II. pi. cxvi. B, No. 5) ; the Lycian relief (Clarac, op. cit. II. pi. ccxxiii. No. 189) ; the relief from Marsala {Arch. Zcit. 1872, pp. 133, 134, pi. li.) ; group from Myra (Fellows, Discoveries in Lyciu, p. 197) ; relief from Xanthus (^ibid. p. 174) ; frieze from Aezani (Le Bas et Waddington, Vpi/aye Arclieologique en Grece et en Asie J/iwcMre, atlas, under ' Asie Mineure,'pl. xiv.) ; and especially the Acropolis group (B. C IT. XIII. [1889], p. 139, and Eev. Arch. XVIII. [1891], p. 137, pi. xiv. bis.). The motive is also common in branches other than sculpture in stone ; cf . the shield from Amathus published by Cesnola (Cyprus, pi. xx.) ; the tripod from Vulci (Braun, Annali, 1842, pp. 62 ff. ; Mon. III. pi. xliii.) ; and the oenochoe (Gerhard, Avs. Vus. II. p. 134, pis. cxxii., cxxiii). In terra-cotta, I know of no examples except two reliefs — one in Berlin (Gerhard, Ant. Bild. p. 317, pi. Ixxviii. 2), the other in Palermo (Kekule, Aiit. Terrakotten, II. p. 82, pi. liv. 2). Of all these groups, the Acropolis one approaches most nearly to ours. In both, the bull is ,. plainly exhausted and overcome ; the hind legs even have the ^^^^^^^j w same position in both — that of the " split," and the decoration ^^^^^V J^^^ in red stripes is common to both. In fact, it seems probable ^^^^^^k '^^'^'^^^^^ ^^^^ ^^'' gi'oiip ^^s copied from a work of sculpture similar to WH^Mk ^^tr tlie Acropolis group, and dates from about the same time, i. e., ^^^ ^ the first half of the sixth century B. C. — 242. (Plate XLVIII. 11.) Fragment of a centaur, top of head, legs, and horse's body missing. The break at buttocks shows plainly that the figure was a centaur. He has a long, pointed beard, in which a triangular hole marks the mouth. The left hand is pressed against the left hip ; the right was apparently raised. White slip. Height, 6.5 cm. Light yellow clay. Cf. the figure from Liuiniti, Arch. Anz. 1889, p. 88, E. 243. (Plate XLVIII. 6.) Horse carrying a woman of primitive type. She faces the horse's right side, and no saddle is represented. Eyes and breasts of woman and eyes of horse are applied disks. White slip, with applied red lines on neck, waist, and knees of woman. Height, 6.5 cm. Red clay. Reinach, in a study of female figures associated with horses {La JVecropoIe de 3Iyrina, pp. 401 ff.), comes to the conclusion that " Ton pent hesiter entre Eos, Selene et Aphrodite, mais sans exclure la possibilite d'une autre hypothese." At Argos, a figure like No. 243 is much more likely to be associated with Hera iim-to? (cf. Paus. V. 15. 5), but this cannot be regarded as anything more than a conjecture. 244. (Plate XLVIII. 2.) Mounted warrior of the ordinary primitive type. Face, a beak, with disks for eyes; high helmet. He has no shield, but extends his arms to the sides of the horse's head. Traces of white. Height, 7.5 cm. Red clay. 245. (Plate XLVIII. 4.) Similar warrior, with large shield on left arm. Only neck and shoulders of horse are preserved. Traces of white. Height, 9 cm. Red clay. 246. (Plate XLVIII. 3.) Similar warrior, no part of horse preserved. The warrior wears a peculiar helmet, the effect of which resembles that of the American "cocked hat." White slip. Height, 7 cm. Light yellow clay. 247. Forty-five fragments of similar figures, all more or less broken. They usually show traces of a white slip. Clay, usually red. For this very common type, cf. Martha, Cat. des Fig. du Mus. d'Atkenes, Nos. 602, 603 (from Tegea) ; Jamot, B. C. II. XIV. (1890), pp. 217 ff. (Tanagra figures) ; Cesnola, Coll. of Cypriote Antiquities, IIP pi. Ixix. ANIMALS 41 248. (Plate XLVIII. 8.) Horse with pannier on each side, bound by straps which pass around neck, under belly, and under tail. Ornamented with black lines across breast, black dots on straps. Length, 7.5 cm. Yellow clay. Cf. the horse with a basket on his back found at the sanctuary of the Kabiri (Welters, Ath. Ifitt. XV. [1890], p. 358). 249. (Plate XLVIII. 10.) Horse's head of good style. The forelock is carefully brought out, and even the wrinkles over the eyes. Traces of white. Height, 5 cm. Clay, light yellow and fine. 250. (Plate XLVIII. 12.) Small animal (dog?), legs broken. The eyes are holes bordered by raised rims. Traces of white. Height, 3.5 cm. Yellow clay. 251. (Plate XLVIII. 1.) Bull's head of good style, complete except right horn. Traces of white. Height, 6.5 cm. Clay, red. 252. (Plate XLVIII. 5.) Similar head, horns broken. Traces of white. Height, 6.5 cm. Clay, red. 253. (Plate XLVIII. 17.) Bull's head in profile, good workmanship. Traces of white. Height, 5.5 cm. Dark yellov^ clay. Cf. the large finds of similar bulls at the sanctuary of the Kabiri (Wolters, Ath. 3fitt. XV. [1890], pp. 355 ff.). 254. (Plate XLVIII. 24.) Serpent's head of good style. Body, black ; head, green, bordered by maroon line ; eyes and nose touched up with black. Length, 5.5 cm. Yellow clay. 255. (Plate XLVIII. 9.) Small dove with young under her wings. Traces of red. Height, 2.5 cm. Red clay. 256. (Plate XLVIII. 23.) Cock, practically unbroken. White slip, traces of red on comb and wattles. Height, 7 cm. Yellow clay. 257. (Plate XLVIII. 14.) Small human-headed bird, with ring for suspension. The face is archaic, with high coiffure, two braids falling over right shoulder and one over left ; mass at back. Traces of white with applied red. Height, 5 cm. Dark yellow clay. Cf. Martha, Cat. des Fig. du Mus. d'At/ienes, Nos. 241, 242, 964 ; Kekulc, Ant. Terrakotten, II. p. 26, figs. 63 and 64 ; Heuzey, Fir/, du 3Ius. du Lonvre, pi. xiii. 6. We agree with Heuzey (loc. cit. p. 11), in regarding this figure as a Siren or Harpy, borrowed by the Greeks from the Egyptian representation of tlie soul. 258. (Fig. 76.) Cock, broken at neck. The background is cut away all around, and the plaque is pierced by two holes for suspending it or for fixing it upon a background. Traces of white. Height, 5.5 cm. Dark red clay. This type is very common. Cf. Martha, Cat. des Fig. du Mus. d'AtJienes, Nos. 11-13; Dumont et Chaplain, Ceramiques de la Grcce Propre, II. p. 229, No. 45; Girard, ' Sur Quelques Nccro- poles de la Grece du Nord,' B. C. H. III. (1879), p. 217, No. 2 ; Stephani, Compte lieiuhi, 1868, p. 71, pi. iii. 14 ; Rayet, Cat. de la Coll. d'Ant. Gr. de M. O. liayet, p. 8, No. 30 ; there are similar ^^' ' specimens in Syracuse. 259. (Plate XLVIII. 19.) Dog's head, broken at neck, which is decorated with a twisted collar. The modeling is very careless. Traces of white. Height, 6 cm. Clay, red. 260. (Plate XLVIII. 18.) Rude ram's head, broken at neck. Traces of red. Height, 5 cm. Light yellow clay. Cf. Schliemann, Tiryns, p. 106, fig. 22. 261. Bear's head, broken at neck, very rude eye and mouth marked by long grooves. Traces of red. Height, 5 cm. Gray clay. 262. (Plate XLVIII. 15.) Head of griffin, broken at neck. The mouth is wide open. The hair is represented by notches above the forehead and at sides of neck. Traces of white. Height, 6 cm. Red clay. 263. (Plate XLVIII. 20.) Squatting pig, fore legs (which were extended horizontally) broken. No trace of color. Height, 6 cm. Dark red clay. 42 TERRA-COTTA FIGURINES 264. (^1'latk XLYIII. 21.) Sitting monkey, arms broken. No trace of color. Height, 9.0 cm. Light yellow chiy. Cf. Martha, Cut. des Fig. du 3his. d'Athhies, No. 100; Frohner, Collection Burrc, No. 4G4. It is probal)le, from the position of the arms and from a remnant of some object on the left shoulder, that our figure held a spear in the right hand, and carried a shield on the left arm — the attitude of a similar figure from Megara Hyblaea in Syracuse (vase room, central case). 265. (Plate XLVIII. 22.) Tortoise, head missing. Traces of white. Length, 4 cm. Dark red clay. 266. Six hundred and forty-six fragments of animals, similar to the types here described, all more or less mutilated. XL VARIOUS OBJECTS. Clay, rwldi^h yellow. 267. (Fig. 77.) Fragment of dish with cakes. Length, 6 cm. ^^ ^^^^^ Undoubtedly an offering " by sub- ^^B^^^^^^ stitution " to the gods ; cf . Pottier and ^^^^^^^^^^^^^ Reinach, La Nicropole de Myrinu, mKKt^K PP-246ff. ^ '■ "' 1^^ 268. (Fig. 78.) Table, legs broken iQ iiiy-*"^ close to top. The edges are grooved *"■ to form an ornamental pattern. On liG. 77. ^jjg table, thin strips of clay, applied irregularly, seem to us to represent food (cf. No. 267). Traces of white and some red. Length. 7.5 cm. Yellow clay. 269. (Fig. 79.) Top of a table, legs broken. At each corner is a boss, meant, possibly, to represent a bronze nail-head. Top painted with cross-lines and round dots in maroon, shading to black. Length, 5.5 cm. "Width, 5 cm. Yellow clay. 270. Forty-eight fragments of similar tables, mostly small, flat pieces of tops, with incised patterns of parallel and criss-cross lines. They are usually painted solid red, but a few have the reg- ular white coat, with applied red and black lines. The lai'gest fragment measures 7.5 cm. X 6 cm. Yellow clay. 271. (Fig. 80.) Fragment of bed, legs broken. The sides are covered with incised patterns ; on one side, in the centre, meander, and at sides, criss-cross. No traces of color. Height, 3.5 cm. Length, 4.25 cm. Width, 2.75 cm. Yellow clay. 272. (Fig. 81.) Armchair, legs and back broken, a hole at the end (to receive a bronze nail-head ?). Traces of a seated figure in the chair. Decoration in straight and zigzag lines, brown to black. Height, 5 cm. Length, 6 cm. Width, 4.5 cm. Light yellow Fig. 81. "'^y. Cf. Schliemann, Tiryns, p. 148, pi. xxiii. fig. c. ; Furtwangler, Samm. Sahourqf, II. pi. cxliv. 3 ; Man. Ant. VI. p. 107, pi. iv. 9 ; ibid. p. 129, pi. V. 10. 273. Similar armchair, containing fragment of a female figure of the seated, flat-bodied type, legs and back broken. One arm has a round boss at the end. Decoration in brown criss-cross lines on back of chair. The body of the woman was painted black. Height. 7 cm. Length, 6.75 cm. Width, 5 cm. Dark yellow clay. Cf. No. 272. Fig. 80. The arm has Fig. 82. VARIOUS ORIECTS 43 274. (Fig. 82.) Leg from a siinihir chair or from a table, with ornamental nail-head, decorated by eight radiating grooves, with a boss at the centre. The end is painted red. Height, 4.5 cm. Clay, yellow. 275. Fifty-two fragments of similar legs, all decorated on the ends, usually with a boss or a rosette or radiating lines. They usually show traces of white or red color. Lengtii, 2 cm. to 8 cm. Clay, yellow or red. 276. (Fig. 83.) Foot with sandal. The sole is a separate piece, and there is an elaborate No trace of color. Length, 6 cm. Yel- ,^-^^ Fig. 83. I'iG. 84. Fig. 85. bake-shop scene. bow on the instep. low clay. 277. (Fig. 84.) Terra-cotta mould. The objects repre- sented come out more plainly in the impression (which is here given) than in the mould itself ; but they are difficult to determine with certainty. Two are plainly shells, one seems to be a pear, another a fig, another a nut. Length, 4 cm. Width, 2.6 cm. Clay, yellow. The objects represented in this mould make it appear probable that the mould was used for making offerings of food, " by substitution," to the goddess. Cf. No. "267, and tlie reference there given. 278. (Fig. 85.) Small set of three large-handled vases, set into one another. They served no practical purpose, but were simjily a votive offering. Traces of white. Height, 6.25 cm. Dark red clay. Orsi, in his article ' Megara Hyblaea ' (JMon. Ant. L p. 874), quotes a case of two stamni set into each other in a grave (No. cxciii. of his enumera- tion). 279. (Fig. 86.) Fragment of a All that is preserved is a portion of the long table, with two boards upon it, each car- rying a cake of dough. On the base, beside this table, traces of three figures of prim- itive type. Traces of black. 7.5 cm. Yellow clay. Cf. the bake-shop scenes in Athens, Martha, Cat. tZe.f Fkj. du 3hts. cVAtheucs, Nos. 39, 40, published by Kourouniotes, 'E.^. 'Kpx- 1896, pp. 201 ff., pi. xi. 280. (Fig. 87.) Small tree (?) with seven projections about main stalk (to represent branches?). White coat, ornamentation in red and black lines. Height, 7 cm. Clay, reddish yellow. 281. (Fig. 88.) Two flower-like disks, fastened together by the stems. One is painted solid red, the other has cross-lines in red, shading to black : stems, dark red. Length, 5 cm. Yellow clay. J- 282. Twenty-two fragments of similar disks. Tiiey are Blk^^B^^^ all painted, usually in maroon and black, sometimes in W^^^^^^^P solid color. Length, 1.5 cm. to 5 cm. Yellow clay. * ^^^^^H 283. (Fig. 89.) Four-sided cone, perforated at the top. ^^^^m Ornament, a wide stripe in dark red to black, on each face. '^^^r Height, 5 cm. Red clay. Fig. 88. Cf . p. 15, note 10. Fig. 86. Height, 4.5 cm. Length, 10 cm. Width, Fig. 87. L-^' .a-j^aK^' / 44 TERRA-COTTA FIGURINES / l-IG. 'Jl. 284. (Fig. 90.) Similar perforated cone ; on each side, six horizontal lines. Color, light red, shading to dark red, brown, and black. Height, 7 cm. Yellow clay. 285. Thirty-one similar cones. Decoration is rare, and is never more than a sinii)lt; scheme of line ornament. Height, 3.5 cm. to G.5 cm. Clay, yellow to red. 286. (Fig. 91.) Large spool. On one face, a stamp, consisting of curved lines, I'adiating from a centre. Slight traces of red. Height, 4 cm. Diameter, 5 cm. Clay, greenish yellow. 287. (Fig. 92.) Similar spool. On one face, an ani- mal (cat or dog), stamped in deep lines. Traces of white, lleiglit, 5.5 cm. Diameter, 5.75 cm. Yellow clay. 288. Forty-nine similar spools, with sim^jle designs, usually a single circle or simple rosette ; two have a Greek Fig. 90. cross. They almost never show traces of color. Heights, 3 cm. to 5 cm. Diameter, 3 cm. to 4.5 cm. Clay, light yellow to red. 289. One hundred and seventy-eight plain spools, sometimes rounded at the ends. Almost no traces of color. Heights, 2.25 cm. to 6 cm. O Diameters, 2.25 cm. to 6 cm. Clay, light yellow to red. 290. Large pear-shai^ed weight (probably loom-weight), perforated 4.5 cm. from top. The bottom, which is flattened, is pierced by three holes, ajjparently unconnected, and another hole, parallel to the axis, enters at the side. On one side, at bottom, is stamjied an anthemion. No trace of color. Height, 14.5 cm. Greatest diameter, 7 cm. Dark yellow clay. 291. Twenty fragments of similar large weights, perforated at top, P „2 ^^^ with one to five holes at bottom, but without decoration. No traces of color. Height, 7.5 cm. to 13 cm. Clay, usually red. 292. Eight whorls of the ordinary type, all plain except one, which has four radiating grooves on one side. No decoration. Diameters, 2.75 cm. to 5 cm. Clay, yellow to red. Cf. Schliemann, Tiryns, pp. 147 ff. ; Mycenae and Tiryns, p. 77. 293. Three objects in the form of a Greek cross with rounded ends, pierced by a hole at the centre. No color. Lengths, 4.75 cm., 5.5 cm., 6.5 cm. Clay, red. 294. Four small rings with notched edges. They are painted red. Diameters, 2.75 cm. to 4 cm. Red clay. Cf. Schliemann, Tiryns, p. 148, No. 73. ,NOTE. It is to be regretted that these pages were in type before the appearance of Hadaczek's two articles, ' Zur Sclimuckkuust des altgriecliisclien Mittelalters ' and ' Die Fibel des Odysseus, Helikes und Kalykes ' {Jahres- hefte d. oester. arch. Inst, in Wien, V. [1902], pp. 207 ff., and ibid. VI. [1903], pp. 108 ff.), so that it was impossible to give definite references to the Argive figures which he mentions. We are pleased to note that his conclusions as to the dress of these figures are substantially in agreement with our own, as expressed on pp. 10 ff. and passim. The Argivk Heraeum Plate XLII 10(40) 12(28) 1 1 (37) TERRA COTTA FIGURINES. - PRIMITIVE AND TIRYNTHIAN ARGIVE TYPE: STANDING FIGURES Thk Argive Hrrakum Plate XLIII 10(51) TERRA COTTA FIGURINES. -TIRYNTHIAN ARGIVE TYPE: SEATED FIGURES The Argivk Hkraeum Plate XLIV P:^ 4(90) 5(91) TERRA COTTA FIGURINES. -ADVANCED ARGIVE TYPE Thk AReiivF. Meraeum Plate XLV 11(93) 13(94) TERRA COTTA FIGURINES. -ADVANCED ARGIVE TYPE: HEADS The Argive Heraeum Plate XLVI 12(188 14(150) 15(199) TERRA COTTA FIGURINES. -EARLY ARCHAIC TYPE Thk Argive Hkraei'm Plate XLVII 12(213) 1 3(208) TERRA COTTA FIGURINES. -EARLY ARCHAIC TYPE (Nos. 3. 4. 6, 8-io, 12-15); ADVANCED ARCHAIC TYPE (Nos. i, 2, 5, 11); FREE STYLE (No. 7) Thb Argive Hkrabum Plate XLVIII 22(265) 24(2541 21 (264) 23(256) TERRA COTTA FIGURINES. -ANIMALS -.•' THE TERRA-COTTA RELIEFS THE TERRACOTTA RELIEFS' By CHARLES WALDSTEIN and JOSEPH CLARK HOPPIN The Terra-Cotta Reliefs form a portion of the interesting finds among the ohjects of the lesser arts which the excavations at the Argive Heraeum has yielded. The greater number of them were found during the first season in 1892, in what we termed the " black layer " below the foundation walls of the Second Temple. Though we are not justified in saying that this black layer ran continuously round the foundations of the whole Temple, earth of this consistency, containing objects of the same character, was found at so many points in the filling for the Second Temple that a certain degree of continuity may be assumed. Still, at times, the black earth was found in greater thickness and the objects in greater profusion ; so that we were led to use the term " pockets." Other specimens were discovered, chiefiy on the south and southwest slopes, in the third layer of earth which had been massed as filhng for the Second Temple ; so that on the whole, the provenience of all these objects may be designated as the filling below the Second Temple. The chronology of all objects found in this filling would thus be fixed in the one direction as being prior to the year 423 B. c. The other hmit cannot be determined in the same way by the conditions of excavation ; especially as objects mani- festly belonging to widely different periods were found together in this filling. We may, however, anticipate and mention one fact which will necessitate fuller explanation on some future occasion, namely, that none of these terra-cottas were found on the same level as the Mycenaean walls erected on the bed-rock below the filling. Further points, in so far as they can now be determined, will be noted as we examine each individual fragment. The terra-cotta reliefs will be considered in two groups : first, those that ornamented flat objects of terra-cotta — plaques, tiles, innakes, bricks, etc. ; and, secondly, those that evidently formed parts of terra-cotta vessels or vases decorated ui relief. We shall in this chapter confine our study to the former group. These plaques (Plate XLIX.) are ten '-' in number ; they are all, with the exception of Nos. 9 and 10, in a more or less fragmentary condition. 1. Winged figure to right, legs in profile, head and body en face. Upper left corner missing. Length, 0.14 m. ; width, 0.085 m. ; thickness, 0.008 m. Found on the Second Temple Terrace. 2-4. Rectangular reliefs, bordered by incuse circles, the centres of which are raised, inclosing two square fields. In each field, a winged figure running or flying to left. 2. Length, 0.07 m. ; width, 0.085 m. : thickness, 0.026 m. 3. Length, 0.10 m. ; width, 0.085 ni. ; thickness, 0.027 m. 4. Length, 0.117 m.; width, 0.073 m. ; thickness, 0.023 m. All these were found in or back of the South Stoa. 5. Form similar to No. 2, rectangular field. Two men facing each other with uplifted arms. Leno-th, 0.096 m. : width, 0.086 m. : thickness, 0.031 m. Second Temple Terrace. ' This chapter, descriptive of them, is here reproduced = An eleventh fragment, similar in form to No. 2, but as published in the American Journal of Archaeology with only a foot showing in the field, was too much muti- (vol. II. 1898, pp. 173-186), with slight corrections and lated to be published, additions. 48 THE TERRA-COTTA RELIEFS 6. Part of a rectangular relief,*i-epreseiiting a bearded man in a chariot, driving a quadriga. Length, 0.085 m. ; width, 0.087 ni. ; thickness, nearly 0.01 ni. 7. Fragment with similar subject, only one horse, mnch smaller and less well preserved. Length, 0.0G9 m. ; width, 0.43 m. ; thickness, 0.01 m. Both Nos. 6 and 7 from the Second Temple Terrace. 8 a and b. Two fragments from the same relief. Incuse circles as in No. 2, bordering square fields. Also a rosette with eight leaves. (a.) Fore part of Centaur to right, holding a branch. In the field, rosettes formed of four leaves and four diamonds alternating. Length, 0.091 m. ; width, 0.079 m. ; thickness, 0.018 m. (b.) Lower portion of a kneeling figure to left. Length, 0.067 m. ; width, 0.056 m. ; thickness, 0.02 m. Found back of the South Stoa. 9. Small square relief, well preserved. Above, an extension pierced by a hole. In the field, a lion to right. Length, 0.05 m. ; width, 0.045 m. ; thickness, 0.014 m. From east of the Chambers on the Second Temple Terrace. 10. Small rectangular relief. Two women in a chariot to right, driving a quadriga. Upper part pierced by a hole. Length, 0.054 m. ; height, 0.046 m. ; thickness, 0.01 m. Pi-obably found back of the South Stoa. All these plaques, with the exception of No. 8, are made of a fine reddish clay. No. 8, however, is made of a very coarse greenish yellow clay, in which small stones are visible. The question what purpose these plaques served is not easily determined. There are two classes of ancient metal relief work : rejMiisse, or beaten work, and pressed, or stamped work. The stamped work, especially in soft and thin gold, was evidently made from a mould * upon which the thin metal was pressed. The use of so brittle a material as clay, however hard the baking may have made it, and however thick such brick-like plaques as Nos. 2, 3, 4, and 5 may be, would be strange. Still it is not impossible that the examples just mentioned may have served that purpose. All these reliefs are influ- enced to some degree by the style suggested by metal work. It is not impossible that they may have served as " backing " for a thin coating of metal. But this is not likely, because for this the same objection of the softness and brittleness of the material might be adduced. Moreover, there is no trace whatever on their surface that such a covering existed. And, considering the peculiar adhesive effect the oxydization of bronze exerts on extraneous objects (many vases being found at the Heraeum to which bronze objects were stuck fast), it is inconceivable that no traces of such a covering should have been found. A specimen of a mould in terra-cotta was found at the Heraeum, 0.11 m. in length, 0.7 m. in width, and about 0.01 m. in thickness, flat on one side and curved on the other. In the flat side were several holes of varying shapes, sunk to a depth of several millimetres. A cast taken from this mould showed a series of objects, one of which undoubtedly represented a moUusk, the murex, or purple shell, so common on Mycenaean vases. The artist has made the common mistake in representing this shell sinistral instead of dextral ; he has also made the anterior canal too wide. Whether the others represent nuts, fruit, or seeds of various shapes, we cannot decide. The nearest analogies to such a mould are found in Naucratis,' and the so-called " cake-stamps " {Naukratis, I. p. 45, pi. 29), which are clearly of a very late date. Mr. Cecil Smith ' See a good instance of this in the bronze mould in the ^ Cf . also Evans, J. H. S. VII. p. 44, and Introduc- Ashmolean Museum, Oxford, published by H. Stuart tion, p. xliv. Jones, J. H. S. 1896, p. 323. RELIEF NO. 1 49 informs us that there are in the British Museum, thus far unpublished, stamps of a similar shape to ours, but none with the same subject. The presence of the murex might point to a somewhat early date for our mould, but there is no further clue to the date. That all these plaques are ex votos is the only natural explanation of their existence at the Heraeum. Nos. 9 and 10 certainly serve that purpose, as the holes in their upper parts for hanging them up show, and Nos. 1, 6, and 7 can hardly be anything else. The technical method seems to have been the same in all, namely, that the clay was stamped while soft, and afterwards fired. Nos. 8 and 9 present the peculiarity of being treated with a sharp knife after firing, in order to express details and to emphasize lines ; this is apparent from the knife-marks on the surface of the clay, and from the fact that in several places the outlines have been trimmed down, leaving a fainter line at the back of the relief. No. 1. — This relief is especially interesting in that while, as we shall see, it manifests Hellenic elements, it has traces of Oriental influence more strongly marked than the others. In fact, we may say that the terra-cottas in relief manifest the ' Oriental ' charac- ter in far greater proportion and pronouncedness than is the case with the terra-cotta figurines in the round. While we found but few specimens among our many hundred figurines which bore clear traces of Oriental influence, the proportion of terra-cotta reliefs which show this influence is very large, and, as in No. 1, this Oriental character is clear and pronounced. This is a very significant fact, strengthening, as it does, our convic- tion, based upon much testimony from other quarters, that Greek sculpture in the round, the statue of pure art, is a specifically Greek development ; while Oriental influences only find their way into Greece through the channels of decoratim art, especially through ornamental vessels in ceramic art and metal work. The subject represented is, at first sight, quite simple. It is a nude, winged figure. The upper part of the body, as well as the head, is in fidl face, while the lower part from the waist downwards is in profile. This want of unity in composition is the rule with reliefs, as well as figures in the round, of this early period. Such inconsistency in atti- tude, by a curious effect of conventionalism, survived long after the artists had advanced beyond this point of archaic awkwardness. Ceramic art is a case in point, since not till after the Persian wars was this conventionality abandoned, and instances in sculpture are too numerous to mention. The head ' is surrounded by a mass of hair, which falls down to the shoulder on both sides in a heavy, ribbed mass, while over the forehead it Hes in waves. On the left eye is a slight indentation which at first sight seems like an iris, but which on a careful examination proves to be merely accidental, no such hole being apparent in the other eye. The nose is flat and the mouth hard and straight, a slight effort being made to model the lips and chin. The scheme of the hair is strikingly like that of the Melian or the Tenean Apollo. The wings present this peculiarity, that they grow directly from the breast, in front of the shoulders, which they entirely conceal. The arms are comparatively thin, a result, perhaps, of the difficulty the artist found in dealing with several different planes, the arms being drawn behind the wings. Whether the figure is male or female is difficult to decide, but it is more probably the > The head of this figure resembles slightly that of the female Centaur on the Relief Vase in the Louvre. Cf. De Ridder in B. C. H. 1898, pis. iv. and v. 50 THE TERRA-COTTA RELIEFS former. When it is carefully exaininetl, the traces of a very short wavy chiton can be discovered, at a slight distance below the waist. Female figures in archaic art are never represented, as far as we know, hi a chiton of such shortness. Though there are many points of difference, a coni2)arison of this plaque Avith the bronze relief from the Acro- polis' induces us to beheve that our figure is male. The thighs are large in proportion to the body, with careful modeling of the mnscles, especially about the knee. The nates are small in proportion to the thighs, a peculiarity seen also on the Selinus Metope of Heracles and the Cercopes. The legs from the knee downwards are extremely thin, similar to those of the Tenean Apollo. In fact, our relief seems to afford a mixture of the exaggerated muscularity of the Selinus Metope and the slimness of the Tenean Apollo. The attitude of the figure is not necessarily that of one walking, but is due rather to the inability of the artist to represent the legs and feet from the front. What the hands hold is clearly not a wreath, but a conventionalized flower or branch. This forms part of the action itself, since each hand holds an end of the ornament, and it thus becomes a sort of contaminatio of a plant and scroll ornamenting the background, introduced at the same time into the action of the figure itself. This action is, in reality, a reminis- cence of the well-known " Thierbandigerschema," ' of which the Persian Artemis,^ falsely so called, is a good example. The winged figure in this connection was one of the types adopted by the Greeks, and was remodeled to suit their own peculiar needs, since, though the Hellenic character of the Persian Artemis is now established, no one denies that the original type was a foreign importation. The addition of the wings seems to be a fea- ture of the later archaic art,* since they are certainly unknown as attributes of the human figure in Mycenaean or Geometric art, nor do we find them on the "Island Stones." ° The mythological significance of the relief, if such it has, is not clear to us. We can only say that it is a winged figure treated in an ornamental manner. We use the term " ornamental " advisedly, with a more literal signification than is generally given ; for one of the most peculiar and striking features of this representation is the action of the figure. The winged " genius " is holding in both hands a mere ornament, a decorative design, which certainly is no implement of use, nor has it any special significance. But the use made of it here is quite distinctive and original. It is as if the decorative feeling of the modeler of the plaque had been so strong that it obtruded itself into his represen- tation of a human figure, becoming an integral part of the attitude and action, at the expense of the logic of representation, as far as the human figure itself is concerned. This introduction of the ornament seems to point to two customs in the decoration or framing of scenes and representations of figures on vases, bronzes, terra-cottas, etc. ; namely, the continuous scroll or other ornament bordering these scenes, as with a frame, and the introduction of ornaments into the field or background, which the Germans have called " Flillornamente." Such framing bands and " FuUornamente " are the composite motives which appear to have led the artist to introduce this new form of ornament into the field of this reUef . We have had occasion to cite the " Persian Artemis " as an analogy to our figure. A direct repetition of the motive of our plaque may be found on a bronze plaque from 1 J. H. S. 1893, p. 259, fig. 26 (Bather). ■£</>. "Apx- 1891, pp. 34 f. Cf. Langbehn, Flugelfiguren, 2 E. Curtius, Gesammelte Abhandlungen, II. pp. 110 ff. etc. (' Wappengebrauch uiid Wappenstil im Alterthum'). ^ We have just heard from Mr. Hogarth that winged ' Studniezka, Kyrene, p. 155. human figures which he considers to he of ' ^lycenaean ' * Milchhofer, Anfange der Kunst, p. 86 ; Tsountas, style have recently been discovered by him in Crete. RELIEFS 2-5 51 Dodona,' but of a manifestly later date. Here, however, the ornament has heen treated in a precisely sunilar fashion, the ends of the scroll l)ein<>- held in each hand. In spite of the lateness of the Dodona plaque, a quasi-Oriental influence may easily be detected. A similar position of hands may be noticed on the Euphorbus plate." An almost exact duplicate of our figure may be found on the gold hormus from Camirus.^ Though the centaur, with whom our figure invites comparison, has no wings, the most striking similarity of style is noticeable when we compare the two. The treat- ment of the hair, body, nates, and legs is identical. The wings are supplied by the Arte- mis on the same jewel, and, moreover, present the same peculiarity we have commented on before, namely, that they grow directly from the breast, concealing the upper part of the arm and shoulder. Our figure, in fact, might almost be composed by a combination of the centaur and the Artemis on the hormus. This jewel, which is certainly Rhodian, exhibits more than any other work known to us so marked a similarity of style and technique with our plaque as to suggest some very close connection between them. More than this we do not venture to say, since it has not yet been determined whether Rhodians were influenced by Argives, or vice versa. It may well be the case that the Rhodian types are derived from Argos, since Camirus, according to legend, was a colony of Argos, founded by the Heraclid Tlepolemus,* and was counted as one of the towns of the Doric Hexapolis. The presence, also, of the Argive alphabet in Rhodes is well known. (Cf. the Argive lambda on the Euphorbus plate.) As a last comparison, we might mention a relief somewhat similar in style, but probably earlier, found at Aegina.' This relief exhibits the greatest similarity in the treatment of the hair. That it is Pelo- ponnesian and not Aeginetan seems fairly evident. In summing up, we may say that, while our plaque exhibits Hellenic features, espe- cially in the modeling of the figure, the spirit of the composition and the introduction of the wings are distinctly of Oriental origin. Moreover, we find absolutely no Mycenaean or Geometric elements, but those which are characteristic of the early Corinthian vases. We are forced, however, to assign our relief to a slightly later date than those vases which exhibit this " Thierbiindigerschema," since the conventional and decorative treat- ment of the ornament of our plaque, admittedly without a meaning, is certainly later than this schema, not earlier. Therefore Ave may assign it approximately to the begin- ning of the seventh century b. c. Even in the best period of Greek art, such a deco- rative solecism may be met with, as, for instance, in the beautiful red-figured vase of astragalus shape, signed tvpi(TKo<; eVoteo-e, in the Papa Giulio Museum at Rome. Nos. 2-5. — These are all of similar technique, and evidently contemporaneous. Frag- mentary as they are, we have still enough to show that their dimensions were from 0.10 to 0.12 m. long and 0.7 to 0.9 m. wide. Of aU our rehefs, these show the metal influ- ence in a most marked manner, the incuse circles being probably an imitation of the nail- heads used to fasten bronze sheathing to wood, while the division into fields, as well as the technique, finds its parallel in the series of bronze reliefs from Olympia," Dodona,' the Acropolis,* and the temple of the Ptoan Apollo.^ The subject of the reliefs Nos. 2-4 ' Carapanos, Dodona, pi. xviii. fig. 3. ' Carapaiios, Dodona, pis. xvi.-xviii. 2 Salzmann, Necropole de Cameiros, pi. liii. ; cf. also « /. H. S. 1892-93, p. 249 (Bather) ; Wolters, Atken. pl_ xxvi. Mitt. 1895, p. 473. 8 ii,i,i_ pi 1. s 5 c. H. 1892, p. 348, pis. x., xi. (HoUeaux). * Diod. IV. 58 • V. 57 ■ Find. 01. vii. A survey of such reliefs is given by De Ridder in his 6 Stais, 'E.^. -Apx- 1895,' p. 263, pi. xii. monograph, " De Ectypis quibusdam aeneis, quae falso 8 Furtwiingler, Olijmpia, IV. (' Die Bronzen ') pis. voeantur Aegino-Corinthiaca." XXXVII., XXXVlll. 52 THE TERRA-COTTA RELIEFS is the same : two winged figures moving iaj)idly to the left in the usual " knielaiif- schema." ' What the objects are they hold in their hands cannot be determined with certainty ; that in the right hand is paralleled by a similar object in the hands of the figures on the terra-cotta reliefs from Sicily " of later style. Kekule, however, refrains from defining these objects. That in the left hand resembles an axe. To identify these figures as gorgons seems impossible, and we must be content merely to term them winged daemons. They are similar in style to the reliefs from Olympia,' which contain figures called by Furtwiingler " Daimons," retracting the view he had previously expressed in Roscher's Lerlcon* It is impossible to tell whether the figures are male or female, though the latter seems more probable considering their similarity to the figures on the relief previously cited, which are certainly female. It may be here noticed that these figures bear a strong resemblance to the " Nike of Archermus," but with the same differ- ence that we shall jilainly see when we compare them with gorgons. Such monsters, wliile generally represented in the " knielauf-schema," ' have always the body in profile and head en face, and in the case of our figures both head and body are represented in profile. Moreover, our figures are not holding the usual bird or animal. Only one relief in terra-cotta of similar technique, though of different subject, is known to us ; " it repre- sents the Persian Artemis in profile, holding a bird in each hand. The incuse circles are precisely similar to our reliefs, and a rosette there is similar to those on No. 8. Though little connection may be assumed between the Persian Artemis and our daemons, we see that the same Oriental influence was at work at the time of their manufacture. No. 5. — What the subject of No. 5 represents is extremely doubtful, and several inter- pretations are open to us. It might be a boxing match, if such be the correct identifica- tion of the two Olympia reliefs (Olympia, IV. pis. 39, 703, 704 a ; cf. also Furtwiingler, Bronzefunde, p. 91), or the similar group on the Geometric vase of Copenhagen.'' A certain similarity may be detected if we compare our relief with a group on the well- known Tripod vase from Tanagra in Berlin.* Save that no traces of a wreath can be discovered, the scene on our relief further resembles one of the Ptoan reliefs already cited {B. O. H. 1892, pi. xi. 3 ; No. 45 in De Bidder, op. cit). Again, were it not for the lack of the tripod, a connection might be assumed with the scene on the mould in the Ashmolean Museum we have already referred to. Perhaps we might recognize in it the a/cpoxei/3to-/Aos or the preliminaries of the wrestling match. Instances of this have been identified in several monuments.'' Or perhaps the action of our relief may be only a salutation. More than this we cannot say. In style there is a closer analogy between this relief and bronze relief work than is the case with Nos. 2, 3, and 4. The treatment of the figures is precisely identical with that of the above-mentioned Ptoan relief, both in the hair and in the build of the bodies. The profiles, too, are similar. Thus the question of dating our plaques becomes much ' Ciirtius, ' Die Knieenden Figuren der Altgriecliischeu Society, 1882, p. 2, where, in connection with Arist. Eth. Kunst,' Winckelmanm Programm, 1870 (Gesam. Abhand. Nic. III. 1. 17, p. 1111 (Bekk.), the following illustrations II. pp. 116 ff.). are mentioned : A vase in the possession of the late Camille 2 Kekule, Terracotten von Sicilien, pi. Iv. Lecuyer at Paris, a relief in Clarac, Mmee de Sc. II. pi. » Olympia, IV. pis. 39, 699 a, p. 102, No. 1. 184, No. 55 ; Krause, Gymnast, und Agorxistik, II. pi. x. * Ibid. I. p. 1709, s. V. Gorgonen. 29. Cf. also Vase Catal. Brit. Mus. III. E. 78, as well as = Milchhofer, op. cit. p. 86, fig. 56. Gerhard, Ant. Bildw. pi. 89, also Baumeister, Denhnaler, • Lenormaut, Arch. Zeit. 1866, p. 258, pi. A. I. p. 502, fig. 544. I have recognized another represen- ' Furtwangler, Arch. Zeit. 1885, p. 137, pi. viii. tation of snch a scene in a vase in the Mnseuui of Fine « Loeschcke, Arch. Zeit. 1881, pi. iii. Arts, Boston, which 1 hope soon to publish. ' Waldstein, in Proceedings of the Cambridge Philological RELIEFS G-8 53 simplified. For, though we cannot say with absohite certainty whioli of the two is eavHer, the fact that tliey are contemporaneous (at least they would both belong to tlie same decade) is too evident to be doubted. These bronze reliefs belong approximately to tlie beginning of the seventh century, which gives us a similar date for our rehefs. No. 6. — This design here seems rather more advanced in style. The subject of a char- ioteer in his chariot was not only common through Egyptian and Assyrian art, but was a favorite theme throughout the whole Mycenaean period, as is shown by the grave-stelae found by Schliemann at Mycenae. We owe the introduction of the quadriga to the Dipylon period, from which certain features of our relief are evidently derived, namely, the connection of the pole to the wagon by a rope or staff extending from the dashboard.' The dvTvi has the usual curved form seen on Egyptian wagons, and the wheels are of the common type found on most of the Dipylon chariots. A counterpart of this group may be found on the Frangois vase,^ the figure of Zeus in his chariot at the marriage proces- sion of Peleus and Thetis. In fact, we are in a position to date this relief between the Dipylon period and the Francois vase. Its Hellenic origin need not be questioned ; there is certainly no trace of Ionic influence. No. 7. — The technique of No. 7 is vastly inferior. It has also suffered far more from abrasion. This makes it difficult to decide whether one or two persons are represented in the chariot, more probably only one. Doubt also exists as to whether the charioteer is holdmg the reins or is in the act of shooting an arrow, and from the curious posi- tion of the body with regard to the chariot, a certain similarity may be deduced with the marble relief^ from the Acropolis of the figure mounting a chariot. At all events, the chariot shows the same influence as No. 6. Its chief peculiarity lies in the fact that but one horse is represented, which fact must be assigned to the incompetence of the artist, since never, in Greek art, does a chariot of this form, drawn by one horse, occur. Nos. 8 a and b. — That these fragments fit together is fairly evident ; but unfortunately the sides of the fracture, owing to the soft texture of the clay, have been considerably worn away, and thus the breaks, while following the same lines, do not coincide exactly. Still, the foot in the upper right-hand corner of a is the continuation of the leg of the figure in b. Aside from its peculiar technique, to which we have already called atten- tion, its chief interest lies in the seeming irregularity. Portions of three fields are pre- served, two of which are separated by the same incuse circles met with in Nos. 2, 3, 4, and 5. The rosette of a falls beneath the field of b, being used instead of an incuse circle, as we found in the case of the Lenormant plaque from Mycenae. It is impossible, however, to tell the general form of the relief, how many fields it contained, or what subject is represented. As in the case of Nos. 2 to 5, the connection between this and the metal reliefs is obvious. The centaur, the chief figure of the relief, is represented with a horse's fore legs plainly visible, though their lower part is missing. This type of a centaur with the fore legs of a horse is later than that with human fore legs.* When exactly the later type was introduced cannot be determined, there being no distinct dividing line between the two types, which often appear side by side.'' On archaic gems," however, only the later type occurs. Evidently the two styles continued together for a considerable period. So far 1 Helbig, Das homer. Epos, p. 141, note 3, fig. 83. Cf. ' For example, on the frieze from Assos, Clarke, In- Furtwangler, Arch. Zeit. 1884, p. 108, pi. viii. 4. vestigations, etc. pis. xv., xx. See also on a Cyrenean Vase, 2 Wiener VorlegehUUter, 1888, pis. ii.-iv. Arch. Zeit. 1881, pi. xi. » Collignon, Hist, de la Sculpture Grecque, I. fig. 194. « Rossbach, in Arch. Zeit. 1883, pi. xvi. fig. 16. * Roscher, in Reseller's Lexicon, II. p. 1076. 54 THE TEURA-COTTA RELIEFS as can be judged from other nionumettts which ilhistrate the later type, we are justified iu regarding our rehef as one of the earUest examples of that type. To restore the kneeling figure is impossible. It suggests faintly the " knielauf," but that can hardly be the action. The position of the legs resembles to a certain degree that of the figure on the metal relief from Olympia.* As no trace of a bow or arrow can be found on our figure, its reconstruction as an archer is impossible. Apart from the carefulness of detail, the di-ess of the figure is a most interesting feature. The figure wears a broad belt, and above it the fines of the chiton are indicated by incisions in the clay. No trace of any lower garment can be discovered. Whether this, as well as the fact that the sex of the figure is not indicated, points to any intention on the part of the artist to denote a close-fitting undei--garment is doubtful. Carelessness is again the probable explanation. This garb is paralleled by the figures on the Vaphio cups," which wear the broad belt continued below the waist as a sort of breech-cloth, though the upper part of the body is left bare. The portion remaining of the third field is so small that we cannot determine what the figure was. It might be the rear jjortion of a bird, and the whole a figure of the Persian Artemis type. While assigning an Hellenic origin to No. 8, we must nevertheless class it among the works of the so-called Oriental Greek style, of which the Argive-Corinthian is a part. The incuse circles we have already discussed ; the rosette is a favorite form of decoration all through the " Corinthian " period, besides occurring on bronze reliefs.^ The orna- ment above the centaur is characteristic, and may be found on a Melian vase.* Here we have the use of the ornament in the field as an instance of the horror vaciii so charac- teristic throughout the " Corinthian " period, a feature which does not obtain in the succeeding periods. We may, therefore, be justified in assigning this relief to the end of the seventh century, perhaps a little later. Earlier than the spread of the Oriental influence it cannot be. Nos. 9 and 10. — These fragments properly belong together, since both are complete and illustrate admirably the pinax form. This in itself Avould show their connection with the Corinthian pinakes ® in Berlin, since the pinax does not seem to have been used prior to this period. The lion on No. 9 is chiefly remarkable for the enormous size of his head in proportion to his body. From the absence of any ornaments in the field, as well as from this peculiarity of head, which strongly resembles tliat of the Hons on " Early Attic " vases," we should feel incHned to assign this plaque or pinax to a later stage of the Corinthian period, perhaps about the early part of the sixth century. No. 10, on the other hand, shows traces of Dipylon characteristics. In spite of the abrasion from which the rehef has sufi'ered, it is evident that two women are represented, but in a more advanced style than is characteristic of the Dipylon period ; the horses, however, show the feeling of Dipylon art. Nothing of the chariot is plain, except that the wheels are probably four-spoked, though even this cannot be decided. No. 10 is to be assigned to a later date than the Dipylon period, but probably earlier than No. 6. > Olympia, IV. pi. 40. * Conze, Meliscke Vasen, pi. ii. * Tsountas, in 'E,p. *Apx- 1889. ' Furtwangler, Vasensammlung, I. p. 47. » J. H. S. 1893, p. 240, fig. 18 (Batlier) ; Olympia, « Biihlau, Jahrhuch d. Inst. 1887, pi. iv. IV. pi. xxxvii. 712. The Argive Heraeum Plate XLIX :•■*':■:*'- THE ARGIVE HERAEUM.— TERRA COTTA RELIEFS THE VASES AND VASE FRAGMENTS THE A ASES AND VASE FRAGMENTS By JOSEPH CLARK HOPPIN PREFACE The publication of the Vases and Vase Fragments from the Heraeum has proved a far more arduous task than I expected when the work was intrusted to me eight years ago. As the time for committing it to the press approaches, I am keenly conscious of its many shortcomings, for, in the six years which have elapsed since the completion of the work and its final revision before going to press, the quantity of new material found in Greece, which might serve to throw new hght upon many of the problems the vases from the Heraeum afforded, has increased enormously, and the greater part of it, being as yet unpublished, has been inaccessible to me. I could also wish that it had been possible for me to make the final revision of the work in Athens, with the actual material at my command, since I realize very clearly that many of the points which have presented themselves to me during the six years since I left Athens must remain all too supei-- ficially treated, owing to incomplete notes and a memory sometimes treacherous. Such cases, I hope, will not prove of vital detriment to the value of the work. Two facts must be mentioned to secure a fair understanding of the work. First, owing to the smallness of the means at my command, I am unable to publish the mate- rial as I had originally hoped, and consequently have omitted much which, though not of supreme importance, would have been of distinct benefit had it been possible to retain it. Secondly, I see very clearly that the enormous mass of material should, to secure the best results, have occupied the attention of several workers for at least twice the length of time I have been able to devote to it. Thus the work as it now appears is a small selection of the total material, and cannot in the strictest sense be called either final or complete. I have endeavored as far as possible to present, or at least to mention, all the types and classes that were found; but important omissions must inevitably occur when over two hundred thousand fragments are to be dealt with. Classes like the Geometric or Argive should each be treated in as much space as the whole of this volume, if their elaboration were to be considered final. But the desire to make known to the world as soon as possible the results of one of the most important modern excavations has induced us all to hasten the completion of our several tasks, and leave the various finer points to be more carefully investigated by our successors. It is a great pleasure, on the completion of my task, to look back on the warm friends it has brought me, and to thank them for the many and various kindnesses received at their hands. First and foremost, I wish to thank most heartily my friend and chief, Professor Charles Waldstein, for the ready help and encouragement he has always given me, for his keen and intelligent interest in my work, and for the various suggestions and theories he has outlined for my benefit. The departure from Athens at the end of the first year's work of Dr. Theodore Woolsey Heermance, who had shared with me the dis- agreeable task of cleaning and sorting the fragments, deprived me of most valuable assist- 58 PREFACE ance during the more interesting stage o£ studying and classifying the material ; and I feel sure that, had it been possible for him to have shared the responsibility of the whole work with me, the result would have been far more satisfactory. I am indebted to hun for the Appendix on the inscriptions that have been found on some of the fragments. Mr. Cecil Smith of the British Museum will always have my warmest gratitude for the many kindnesses and useful suggestions so ungrudgingly given me during my work iu London. Further, I wish to express my thanks to Messrs. Kavvadias, Stais, and Tsountas of the National Museimi at Athens, for their courtesy and kindness in extending to me, during my three years in Athens, the utmost privileges of the Museum ; to Professor Wolters and Dr. Zahn of the German Archaeological Institute at Athens, for putting the material from the Acropolis and Aegina so openly at my disposal ; and to Mr. David Hogarth, sometime Director of the British School at Athens, for an opportunity of seeing the pottery found iu his excavation at Melos. I have also received many valuable suggestions from Professor Ernest Gardner of University College, London, Mr. Paul Perdrizet of the French School at Athens, and Mr. Carr Bosanquet, Director of the British School at Athens. I must not forget to thank Mr. Gilheron and Herr Rohrer for their careful and pains- taking work in preparing the drawings and photographs. I am especially grateful to Janni Papadakis, vase-mender in the National Museum at Athens, who for three years helped and spared me in every way. Last of all, I wish to express to my dear friend. Professor Richard Norton, Director of the American School at Rome, my grateful thanks for reading the manuscript of the work during its preliminary stage (a labor of love), and for the valuable suggestions he has given me. NOTE. In fairness to this section of the Heraeum publication the following facts should be stated. The manuscript was delivered to the committee on April 1, 1899. Two years later, in the hope of a speedy publication, it was revised as thoroughly as circumstances permitted. At the present date (November, 1903) the larger part has been in type for a year and a half, and any further revision is impossible for many obvious reasons without a complete rearrangement of the material. That it should not have been possible to consider the vases and fragments from the Heraeum in connection with the new phases of the Mycenaean era revealed by the recent excavations in Crete, no one regrets more than the writer. ABBREVIATIONS EMPLOYED. PUBLICATIONS. AOien. Milt. Mittheilungen des Kaiserlich Deutschen Ar- chaeologiscben Instituts, Athenische Abtheilung. Riim. Mitt. Mittheilungen des Kaiserlich Deutschen Ar- chaeologischen Instituts, Roniische Abtlieilung. Jahrb. Jahrbuch des Kaiserlich Deutschen Archaeologi- schen Instituts. B. C. H. Bulletin de Correspondance Hell^nique. /. H. S. Journal of Hellenic Studies. A. J. A. American Journal of Archaeology. Re«. Arch. Revue Archdologique. Gaz. Arch. Gazette Arch^ologique. Ann. dell' Inst. Annali dell' Institute di Corrispondenza Archeologica. Bull, dell' Inst. Bulletino dell' Instituto di Corrispondenza Archeologica. Mon. dell' Inst. Mouumenti Inediti dell' Instituto di Cor- rispondenza Archeologica. Not. d. Scavi. Notizie degli Scavi (Academia di Lincei). Mon. Ant. Monunienti Antichi. Arch. Zeit. Archaeologische Zeitung. Arch. Anz. Archaeologischer Anzeiger (in Jahrbuch). 'E<))r)/ii. 'Apx- "E^ij^f pis 'Apxi'o^oyiK^ . Class Rev. Classical Review. Berl. Phil. Woch. Berliner Philologische Wochenschrift. CATALOGUES. Brit. Mus. Cat. Catalogue of the Vases in the British Museum. Vols. II. and III. by II. B. Walters ; vol. III. by Cecil Smith. Louvre Cat. Pottier, Catalogue des Vases du Lonvre. Berl. Cat. Furtwiingler, Beschreibung der Vasen im An- tiquarium, Berlin. Mun. Cat. Jahn, Beschreibung der Vasensammlung Ko- nig Ludwigs in der Pinakothek ziir Miinchen. Naples Cat. Heydemann, Die Vasen des Museo Nazio- nale zu Neapel. Boston Cat. Robinson, Catalogue of the Vases in the Museum of Fine Arts, Boston. MISCELLANEOUS WORKS. Ray. et Coll. Hist. Rayet et CoUignon, Histoire de la C^ramique Grecqne. Dum. et Chap. Dumont et Chaplain, Les C^ramiques de la Grfece propre. Per. et Chip. Hist. Perrot et Chipiez, Histoire de I'Art dans I'Antiquit^. Myk. Vas. Furtwangler und Loeschcke, Mykenische Va- sen. Myk. Thong. Fnrtwitugler und Loeschcke, Mykenische Thongefiisse aus dem Gebiete des Mittelmeers. Samm. Sab. Furtwiingler, Sammlung Sabouroff. Aus. Vas. Gerhard, Auserlesene Vasenbilder. Baum. Denk. Baumeister, Denkmiiler des Klassischen Alterthums. Roscher's Lex: Roscher, Lexikon der Griechischen und Riiraischen Mythologie. Daremberg-Saglio, Diet. Daremberg-Saglio, Dictionnaire des Antiquit^s Grecques et Romaines. INTllODUCTION Eakly ill the autumn of 1895, Dr. Heermaiice and I began cleaning the fragments of vases. These had been brought up from Argos to Athens in small baskets,' and included fragments varying from a foot square to pieces no larger than a finger-nail. None had been touched save a few baskets of the first and second year's digging, which had under- gone a very superficial cleaning two winters before. Every fragment was thickly coated with a hard crust of dirt, and hydrochloric acid proved the only satisfactory cleanser. Large bowls were used containing a solution, two thirds acid and one third water, into which a basketful of fragments was put, stirred round for a few minutes, and then care- fully rinsed in fresh water to avoid any after-effects of the acid. The fragments were then sorted, each class in separate baskets, laid out, and a search made for the component parts of a vase, a task attended with very unsatisfactory results, since in only a few cases could a vase be reconsti'ucted in anything like entirety. The number of fragments was enormous. No less than two hundred and sixty-five baskets, all more or less full, Avere cleaned. Owing to the removal of a good deal of waste du't and closer packing of the baskets, the total material was compressed into about one hundred and fifty baskets. Just how many fragments were included is difficult to say ; some baskets which contained large fragments held perhaps several hundred, while in others containing small fragments of the Argive and Corinthian classes, the number amounted to two or three thousand. Roughly speaking, the total number of fragments was about two hundred and fifty thousand. The final sorting gave the following propor- tion to the various classes : — Plain 50 Baskets. Glazed 30 " Primitive ........... 2 " Mycenaean ........... 13 " Geometric ........... 25 " Argive ............ 20 " Corinthian ........... 5 " Black- and Red-Figured ......... 2 " Miscellaneous .......... 3 " 150 " By " plain " are meant all fragments which bore no decoration of any kind what- ever.- These included a large number of fragments of Mycenaean and Geometric vases, but the biUk consisted of coarse fragments of wheel-made vases, which might belong to any period. In " glazed " are included all fragments which had been covered with a dark brown or black glaze, generally dull, but without other decoration. Some few 1 The baskets were those usually employed in excava- on the rubbish heaps. If, therefore, it had been possible tions, about 25 cm. in height and 30 cm. in diameter. to preserve all the fragments of such description, which * It must also be remembered that the bulk of the undoubtedly included a large number belonging to the fragments found at the excavation had no decoration of " Primitive" class, their number would have been mate- any kind, and were therefore not preserved, but thrown rially increased. LOCATION OF THE FRAGMENTS 61 of these belonged to vases of the Black or Red-Figured period, but the bulk bore no distinguishing characteristics. In addition to the whole number, five baskets contained vases in an intact or in sHghtly broken condition. The proportion of the various sites was as follows : — South West Corner 50 baskets \ S. Stoa S. E. Side 6 " ( 1894 & S. Side, outside peribolos wall 3 " ) '95 Old Temple 10 " Above Cyclopean wall . 7 " Above Upper Stoa ........ 6 " E. of chambers 20 " S. E. of Second Temple . ...... 5 " W. Building 2 " N. W. Building 5 " First year of excavation 30 " Unknown .......... 6 " 150 " Baskets of the first year's digging contained fragments from the Second Temple Ter- race, but no more definite information is preserved. Fragments of every kind were found in all these places, but certain classes were more common in certain spots than in others ; for insttmce, on the Old Temple Terrace ' and in the corner of the second platform below the Cyclopean wall and east of the Chambers, the Geometric fragments were much more numerous than any other kind, while in the southwest corner (east of the retaining wall of the West Building) Mycenaean predominated. The Argive (so-called Proto- Corinthian) style was found in greatest quantities in the southwest corner and on the Old Temple Terrace. Other classes were divided with comparative evenness all through the excavation. Little advantage is to be gained from the knowledge (in the case of the vases, at least) whence each separate fragment came. Throughout the preliminary stages of the work the fragments from each spot were kept carefully separate, but this proved impossible after the task of piecing them was begun. In several cases fragments from the same vase were found several hundred yards apart. It must be remembered that no apparent ti-ace of layers which could be chronologically divided was observable in any of the different finding places, for the pockets where the fragments abounded in greatest quantity showed that the material had been shoveled in helter-skelter, and it was quite usual to find Mycenaean fragments near the top and Argive near the bottom. In the case of vases of any especial importance, the finding place will always be given in the course of this work. It is extremely difficult to make a precise estimate of the number of vases which these fragments represent. The number of vases intact or only shghtly broken amounted to a little more than a thoustmd. From the fragments about ten vases were reconstructed almost entirely, about fifty partially so ; while those represented by only half a dozen fraoroents apiece amounted to between two and three hundred. If we should make the extremely modest calculation of four or five fragments to a vase, we should have about fifty thousand vases. 1 Professor Waldstein informs me that to the best of which were clearly labeled " Old Temple Terrace," his recollection tliis statement is not correct. My only contained numerous fragments of the Geometric style, authority for the statement is that several of the baskets, 62 THE VASES AND VASE FRAGMENTS In planning this work, I have endeavored, as^far as possible, to model it on Furtwangler and Loesehcke's Mi/kenische Vasen. The chief difficulty experienced among such an enormous mass of fragments was in making a collection for publication which should be representative. In doing so I preferred to err on the side of presenting more fragments than were absolutely necessary, believing that such a plan was better than presenting too little, but I have not hesitated to omit types which are universally known. Roughly speaking, the Heraeum vase fragments cover almost the whole period of Greek ceramic art from its birth to its decline, a period of perhaps over two thousand years. It will be further shown how the bulk of this material is characteristic of the Argolid, a genuine home industry, and that little was imported. The indigenous types represented are: — Prehistoric. Mycenaean. Geometric. Argive. Local (or types which have no prototypes). Foreign (i. e. imported) types are : Corinthian. Early Attic. Attic Black-Figured. Attic Red-Figured. Cyrenean. Megarean ware. Roman. Now, though the list of imported types is larger than the local, the amount of such material is scarcely a tenth of the latter. The reason for the exclusion of foreign pro- ducts, especially Attic, will be shown later. The reader may notice that the discussion and exposition of the various vase-classes given here differs materially from that given by Professor Waldstein in the Introduction to this publication. In order that there may be no room for misunderstanding with regard to this point, I desire to state specifically the reasons for this discrepancy. After two years' work on the fragments, I had independently reached the conclusion that the so-called Proto-Corinthian style included more varieties of vase-classes than had usually been assigned to it, and that it was probably of Argive origin, — a conclusion which confirmed the views expressed by Professor Waldstein four years before. Also that the style was an offshoot of the Mycenaean style, and contemporary with the Geo- metric, having as its chief motives the arrangement of parallel bands so characteristic on Mycenaean vases. After communicating this view to Professor Waldstein I found that he not only had reached the same conclusion, but was disposed to regard the style as one of the links in a chain which went much farther back, as a development of the principle which he calls " Linear," and which he claims existed in an unbroken sequence from the earliest times. He has already developed his theory so carefully that I do not feel the need of repeating Avhat has been better expressed by him in the Introduction. This is not the place for me to discuss this theory, but simply to define my own attitude in regard to it. I agree perfectly with Professor Waldstein in the general application of his " linear " theory, and accept his views as to the presence of linear motives in the Mycenaean vases, SYSTEM OF CLASSIFICATION 63 and as to the Argolic origin of the so-called Proto-Corinthian style. I do not, however, agree with him in his contention that this style can be traced in an inibroken line throughout the Mycenaean period as far back as the earhest primitive vases, nor do I accept the general classification he has already outlined. It is, of course, fitting that in the General Introduction he should have discussed the relation of the Heraeum vase-fragments to the whole excavation and history of the sanctuary. In so far as the work concerns myself, it seems to me more suitable, avoiding general theories as far as possible, to confine myself to a careful classification of the material, omitting those conclusions which, from my unfamiliarity with all the finds of the excavation, would not have as stable foundations as those expressed by Professor Waldstein. The classifi- cation adopted here is the result of careful study of the material for three years ; and though I am far from claiming that the results reached or the theories expressed are the only ones possible, I have not hesitated to base the whole of this division of the publication upon them, since they have forced themselves upon me from the material. In fairness, therefore, to both Professor Waldstein and myself, I would state that, as I cannot accept his entire theory, it has seemed to me best to abide by my own exposition and leave the decision to the judgment of scholars. As to the classification, I may say that I have endeavored, as far as possible, to hold by the old rules and to be as conservative as I could ; at the same time, such a large amount of new material demanded in many cases an entirely new arrangement. But it must not be forgotten that no classification can be regarded as final ; its main object is convenience ; and although several groups may be distinguished, they invariably have, in the case of one group succeeding another, their connecting links, and shde into each other almost imperceptibly. On the Lake of Geneva there is a threefold division, the Petit, Grand, and Haut Lacs, each a sej^arate sheet of water ; but one can scarcely draw a straight line across the lake and say where the Petit Lac begins and the Grand Lac ends. In the case of our vases, we have the separate classes, but so gradually removed from each other that we can only emphasize the fact of their division, without being too specific as to the actual point. I have tried to treat the vases from a relative, not an absolute, point of view. Unless it be recognized that the study of Greek vases is a means to an end, not the end itself, their value is lost. They are not among the grand arts ; in fact, the scanty references to them in Greek literature show that the vase-painter or potter, while pursuing an honor- able trade, could not claim to be an artist in the great sense of the word. It is true that during the earliest periods vase-painting was one of the few forms of artistic expression, but not the highest form. A civiHzation which could produce the fortresses, the Bee- Hive tombs of Mycenae and Orchomenos, the gold cups of Vaphio, the golden ornaments and sword blades of Mycenae, is not to be measured by mere vases of terra-cotta ! Not but that the vases are artistic ; whatever work a Greek did was never anything else. As historical documents they are extremely valuable. They respond to the various influences at work on Hellas from the earliest times, and reflect in a thousand different ways the life and tastes of its people. It is not a lofty ideahsm we find in them, but a realism in all its forms. Owing to the crude state of plastic and pictorial art through- out the periods to which our vases belong, we cannot place them on a level with Attic vases of the sixth and fifth centuries ; but we can see the steady development towards the later, finer forms. For their presence at the Heraeum, where there was only a temple and not a settle- 64 THE VASES AND VASE FRAGMENTS ment/ there is but one explanation : they were dedicated to the use of the goddess." In only a few cases has a dedication been scratched on a vase. It is very probable that such offerings were those of the poorer classes, who could afford nothing more expensive than a vase. As the centuries went by, the temple must have become overcrowded with such a mass of pottery, and the presence of numerous rubbish heaps proves that there were periodical cleanings out of the sanctuary. Whether such vases were ever in actual use is doubtful. From the fact that at the Kabirion in Thebes, potters sat outside the walls of the Temenos and sold their wares, it is probable that such was the fact at the Heraeum, and that from the earliest times the pottery of the Argolid was in great demand, and so cheap and common as to make the introduction of foreign vases rather like bringing " coals to Newcastle." At first sight the vases from the Heraeum are somewhat disappointing ; there is a noticeable lack of strange or beautiful designs, and none of the vases seem to rise above a common level. But the chief value lies, not in the fragments or vases them- selves, but in their relation to the results of other excavations, especially those in Sicily and Aegina. It is Aegina that is shown in the closest relation with Argos, since many of the more unusual types found at the Heraeum may also be found in Aegina and nowhere else. Such a connection forms a valuable contribution to the history of the two states. We know that it was during the reign of Phidon of Argos that the two came together, since Phidon had his mint at Aegina. Phidon's date is variously given — b. c. 770 by Holm,^ 745 by Stais,* while Beloch ° asserts that he could not have lived earlier than the sixth century. Whether he really introduced the coining of money into Greece, as tradi- tion relates, is another question. As most numismatists are agreed that this could not have taken place before b. c. 700," we must either abandon tradition or else side with Beloch. Holm ^ thinks that he may have introduced Oriental weights and measures into Greece, and thus the introduction of coinage, which followed shortly after, was attributed to him. But the main question does not centre about Phidon so much as about the actual date of the coming together of the two states. In the recent excavations at Aegina conducted by Stais ^ fragments of the Mycenaean style were extremely scarce, but large quantities of Geometric and Argive (so-called Proto-Corinthian) ware were found. Moreover, all the Geometric ware found there resembles that from the Heraeum much more closely than that from Attica, and several types occur which can be duphcated only at Argos. It is also most probable that most of the vases of the Geometric and Argive styles found at Aegina were imjiorted, not manufactured there. As Argos, through the Mycenaean as well as the later periods, was one of the chief centres for the manufacture of vases, it becomes almost certain that after the Mycenaean period she exported her wares into Aegina." Whether Phidon or 1 Dr. Waldstein believes that there was here a primi- ' Op. cit. I. p. 214, Note 3. live settlement to which primitive cooking-pots, etc., ' Loc. cit. and Pallat, Athen. Mitt. XXII. (1897), pp. would belong See General Introdnction.vol. i. 265 if. 2 For discussion on this point cf . Cecil Smith, Naukratis, ' As the question of the origin of such styles as the I. p. 47 ; Furtwangler, Introduction to Geuick, Griechische Mycenaean, Geometric, Argive (so-called Proto-Co- Keramik, p. 3. rinthian), and Corinthian, which are found in Aegina, « History of Greece, I. p. 203. Argos, and Attica could not be settled with positive cer- * 'F,<t>7IIA. 'tipX; 1893, p. 236. tainty from the outward appearance of the fragments ' Griechische Geschichte, I. p. 282. themselves, a microsc()i)ieal analysis of the clav by an ' Cf . Head, Historia Numorum, Introduction, p. 38. expert geologist seemed to offer tlie only solution of the MICROSCOPICAL ANALYSIS OF THE FRAGMENTS 66 some other man was ruler of Argos at that time is not essential to tiie argument, but it would seem that Holm and Stais are right in considering him king of Argos and assigning to him a date hi the first half of the eighth century. That this connection lasted until the sixth century, when the Argives assisted the Aeginetans in repelling an Athenian invasion,' the presence of the vase fragments at Aegina would seem to show. So far as can be judged from the extremely small number of fragments of vases of foreign (i. e. outside of the Argolid) make, such as Dipylon, Early Attic, Corinthian, Cyrenean, Attic Black and Red-Figured, etc., it seems probable that almost all the rest were made in the Argolid. Without doubt all fragments belonging to the Mycenaean style were of such origin. This I judge from the fact that all such vases bear the closest relation to those from Mycenae and show none of the features characteristic of other Mycenaean types difficulty. Accordingly I sent to Dr. H. S. Washington a number of slierds of the Primitive, Mycenaean, Geo- metric, Argive (Proto-Corinthian), Corinthian, Black and Red-Figured styles selected from the following sites : the Heraeum, Mycenae, Tiryns, Nauplia, Melos, Attica (Athens, Menidi, Markopoulo, Thorikos), and Eretria. His report is to appear elsewhere, but we may here men- tion the following facts quoted from Dr. Washington's letter, which were given in reply to the following ques- tions : (1) Wliat is the connection between pottery found at Argos and Aegina ? (2) What is the connection be- tween pottery found at Attica, Argos, and other sites ? (1) " The primitive fragment from Aegina is coarse and contains fragments of raica-sehist. As this rock is not found on the island, the presumption is that this especial piece is of foreign (i. e. extra-Aegina) manufac- ture. In the Peloponnesus the nearest localities of these rocks are in the Xiris mountains, and near and soutli of Doliana towards Sparta; they also occur in Attica. Tlie Geometric fragments of Aegina are coarse with quartz and feklspar grains, possibly derived from schist, but no remains of this or of volcanic rock are visible ; it may be native. The two Mycenaean fragments from Aegina are rather coarse, and show fragments of quartz, feldspar, and augite (a mineral of volcanic rocks on the island), so that these are almost certainly of native (Aeginetan) manu- facture. " The Proto-Corinthian and Corinthian fragments from Aegina do not differ in any essential respect from the corresponding specimens of Argos. The Proto-Corin- thian are much finer than the Corinthian, and I sliould say that the Aegina Corinthian are rather coarser than the Corinthian fragments found at Argos. " Of tlie Argive specimens, the Proto-Corintliian are by far the finest, though the Corinthian come close to them. A fragment of a ' local type ' (v. p. 161, Plate LXVII.) is also dense, as are two of the red-figure style. A prim- itive fragment is also quite fine, but shows carelessness in manufacture in areas of coarser material and in the presence of limestone fragments. The fragment of the Geometric plate (v. p. 116, Plate LVIII.) resembles the corresponding one from Aegina, and may possibly contain a little augite (in very small crystal fragments), which would indicate an Aeginetan derivation. All the Geo- metric fragments are coarser than the Proto-Corinthian, but the Mycenaean are rather better, though less dense, than the Proto-Corinthian. " On the wliole, the only pottery which is, one may say, undoubtedly Aeginetan is the Mycenaean from there, the Argive Mycenaean being distinctly different. There is practically no difference between the respective Proto- Corinthian and Corinthian, and either miglit come from any good clay bed, and witli the material well washed, which would tend to eliminate all heavy portions such as augite. (2) " Argive, Attic, etc. The fragments of the Myce- naean style from Mycenae, Tiryns, Daulis, Markopoulo, and Nauplia are all practically identical with that of Argos. Tlie Geometric fragments from Markopoulo and Eretria are coarser and resemble the Geometric from Argos. The only specimen from Athens (Dipylon) is also coarse, and contains much feldspar, an evidence of carelessness in washing. "The pottery from Melos is quite distinctive in con- taining fragments of minerals of volcanic rocks of the island, as well as here and there fragments of tlie rocks themselves, such as obsidian, etc." " Pottery is made of decayed rocks in which most of tlie original and distinctive minerals have disappeared, and those left are still fiirtlier done away by the washing of the clay, so that it would be only in the most early and archaic periods that we should expect to find tell-tale minerals." Although Dr. Washington's report is somewhat incon- clusive, the following conclusions would seem to be jus- tified: (1) That with the exception of a few fragments of the Mycenaean style and the Geometric plate mentioned on p. 116, none of the sherds of the Geometric, Argive (Proto-Coriiithian), or Corinthian styles are indigenous to the island, but were most probably imported there. (2) The Geometric fragments from Attica, except the pure Dipylon style, resemble very strongly those from Argos. But these same conclusions do not seem to militate against the theories advanced on pp. 119 ff. and 103 ff., that the Argive (Proto-Corinthian) style originated in the Argolid, and that the earliest beginnings of the Geometric style proper are to be souglit for in that locality, al- though, as stated on p. 104, no attempt has been made to prove that the Geometric style necessarily originated in the Argolid, and was taken up by the other states, but rather that the external influence which probably created the Geometric style first made its appearance in .4rgos, and penetrated into Attica in a very short time. 1 Herod. V. 87 ff.; ef. also chapter VII., below. 66 THE VASES AND VASE FRAGMENTS such us Tlieran, Cypriote, and those from the^various islands. At the same time, the complete absence of the first class, lustrous finish of the Mycenaean style, would seem to exclude Mycenae itself as the only factory. With such a large output of Mycenaean vases there must have existed a dozen vase factories scattered through the plain. Though it is a point that admits of much discussion, I am inclined to follow Furtwiingler and Loeschcke's view that the Dorian invasion proved the downfall of the Mycenaean civiUzation, and that the Geometric style arose in consequence of this invasion ' and can- not be traced to the lonians, Carians,- or Egyptians.'^ This point will be further elabo- rated in the third chapter, but we may anticipate a little. It is only recently that the connection between the Mycenaean and Geometric styles has been satisfactorily estab- lished. Wide* has analyzed the survival of Mycenaean elements in the Geometric style and shown how much more the Island types of the Geometric followed the Mycenaean than those of the mainland, but his assumption that, owing to the lack of these Myce- naean features in the Geometric style of the mainland, the centre of the Mycenaean fabrics lay in the islands is not necessarily true, since these same features can be detected in the Geometric vases from the Heraeum. The recent excavations of Fhnders Petrie in Egypt, Cecil Smith and Hogarth in Melos, and Evans in Crete have thrown new light on the beginnings of the Mycenaean style, since a large number of sherds have been found there which, though akin to the Myce- naean, are nevertheless earlier, and have been called " Aegean." These sherds would seem to show a greater tendency towards a naturalistic or pictorial form of ornamentation than towards a linear. As yet the connection between them and the Mycenaean sherds of Greece proper has not been thoroughly established, but it may be assumed that such a connection exists. The Mycenaean style in Greece proper would seem to show the following development. As Professor Waldstein has already pointed out, the earliest Mycenaean fragments show the embodiment of a Linear principle which would naturally be expected in vases which succeed the so-called "Primitive " style. In a very short time, owing no doubt to the maritime character of the Mycenaean civilization, and more probably to the influence of the Aegean pottery, the tendency to employ a species of pictorial or naturalistic orna- mentation, especially of marine life, became widespread. Later again the spirit of conventionality crept in, and in the last days of the Mycenaean style we find a wealth of ornament treated in a purely conventional and rather flamboyant manner, but without the introduction of any fresh ideas ; in fact, the general tendency seems a return to the early Linear feeling. After the Dorian invasion we see this same linear principle marching off in two different styles, the Geometric and the Argive or Proto-Corinthian, which flourished side by side in the Argolid for a considerable ])eriod. During the seventh century the increasing commercial and trading facilities brought about the introduction of foreign, especially eastern influences, into Greece, with the result that the Geometric style died out, while the Argive absorbed the new influences, from which the Corinthian style probably had its origin (see p. 123). Attica in the meanwhile, after experimenting largely with the Geometric style, had worked her way into the Early Attic style, and after absorbing the Oriental influence from Corinth through the so-called Corinthian-Attic, evolved the Black-Figured style which, during the best part of the sixth century, was the most prominent form of Greek ceramic art. ' Myk. Vas. p. xii. * Athen. Mitt. XXII. (1897), pp. 233 ff. and XXI. 2 Rayet and CoUignon, Hist. p. 32. (1896), pp. 385 ff. a Kroker, Jahrb. I. (1886), pp. 95 ff. NATURAL DEVELOPMENT OF VASE-CLASSES 67 A glance at the accompanying figure shows the way in which this development took place, and how the elementary principle, starting from the Mycenaean style, and working its way through two separate channels, united again in the Attic Black-Figured style. Geometric Mycenaean Argive I. Dipylon Early Attic Argive II. Argive V., Coriuthian So-called Coriuthian- Attic Corinthian witli Attic elements Black-Figured Style. It cannot, of course, be said that the progress traced above is absolute, nor can it be worked out exactly with mathematical precision. It is the result of a careful study of the progress of ceramic art as illustrated by the two great centres of Argos and Athens, with especial reference to the vases from the Heraeum. To trace the development of ceramic art in any given place is, after all, largely a matter of theory, and though I do not wish to maintain that the theories of development here expressed are the only ones possible, it is upon them that the whole of this work is based. As the term " Proto-Corinthian " is now so thoroughly unsatisfactory, I f ally agree with Professor Waldstein that it should be abandoned, and that, as the Argohd offers greater claims for being the home of this style (as I shall endeavor to show later), it is best to call the style Argive. But, as stated before, the use of the term in this part of the publica- tion is rather more restricted than that employed by Professor Waldstein, since it is here used entirely with reference to those vases which are post-Mycenaean, and does not include all the varieties which Professor Waldstein would include under the title of " Arsfive Linear." CHAPTER I. PRIMITIVE VASES. The pottery representing the " primitive " period, i. e. all pottery (prior to the Myce- naean civilization) which shows the employment of the usual incised linear decoration, so common on the various sherds in the lower settlements of Hissarlik, is rather scarce at the Heraeum. For this fact two explanations are possible : (1) As we have pointed out before, a large number of fragments, without decoration of any kind, wheel or hand made, which might well belong to vessels of the earliest periods, were not preserved, since there was no evidence of any kind to show what their age was. (2) Sites in which such ware has been found in greatest quantities, such as Hissarlik, Tiryns, Thera, etc., were all settlements where the pottery was in actual use by the inhabitants. The Heraeum was a sanctuary and not a settlement ; and as the cult of Hera was probably not developed until the Mycenaean period, the custom of dedicating pottery would not have been general, and hence we should not expect to find many primitive sherds on the site. If, however, there had been here an early settlement, vases would have been in constant use, and would therefore not have been found in large quantities, as where in a sanctuary only there were constant deposits of votive offerings in vases. Of the plain pottery already mentioned in the Introduction, a great number of frag- ments seem to have belonged to large pithoi, sunilar to those at Hissarlik. Such large vases, of coarse unpainted clay, are characteristic of all periods, and may just as well fall in the later as in the earlier Greek times. Many of them, in fact, are not dissimilar to common pots used by the people of Greece to-day. The bulk of this " primitive " ware consists of fragments of yellow or reddish clay with incised ornamentation, only a few vases being preserved entire. Most of the frag- ments are hand-made ; but as wheel-made ones are found, no general law can be stated for the use of the wheel. It is probable that some of the wheel-made prehistoric frag- ments are older than hand-made Mycenaean vases. Beside the fragments already men- tioned, a large number of very coarse wheel-made fragments were found, decorated only with a series of incised parallel lines very close together, evidently made by some pointed instrument held against the clay as it revolved on the wheel. To date them is impos- sible ; the nearest analogies may be found in Phoenician ware in the British Museum and the Louvre.' Not more than two baskets of the total number contained sherds with incised decora- tion, and it was extremely doubtful in the case of many of the fragments whether they could with propriety be classed among the primitive vases. Some, by their similarity to the vases from Hissarlik, the tumulus at Bos-iijiik in Asia Minor (Koerte, Athen. Mitt. xiv. (1899), pp. 1 ft'., pis. i.-iii.), and the Cyclades (Tsountas, 'Ecprjfi. 'Apx- 1898, pp. 137 ft"., pis. ix., X.) seemed to belong to the last of the third or the beginning of the second mil- lennium B. c. Probably none of them are earUer than the second settlement at Hissarlik, and the majority contemporaneous with the sherds of the third, fourth, and fifth settle- ' Puttier, Vases Ant. du Louvre, pi. iv. fig. 2. PRIMITIVE VASES 69 ments. Some, again, though essentially primitive in their technique, show Mycenaean and Geometric elements without seeming to helong to either style. No attempt can be made to give anything but an approximate date to the fragments of the primitive period from the Heraeum. As the primitive vases from the Cyclades furnish the nearest parallels to many of our fragments, we may date them as Tsountas does the island tombs, b. c. 2500-2000 ; those which show a more advanced style of decoration may be placed in the first part of the second millennium B. c. ; but it is doubtful if any of our fragments to which the term " primitive " can be applied are later than the beginnings of the Mycenaean civilization. FRAGMENTS. PLATE L. 1. Bright red clay, fairly fine. Wave pattern between bands. Cf. Flios, figs. 298, 317. Koerte, loc. cit. pi. iii. 19. 2. Coarse dark yellow clay with small black stones, hand-made. Probably part of amphora. Handle missing. Decoration of small circles, evidently done with some instrument like a punch. To left of the handle, breast-like protuberance. 3. Reddish clay. Hand-made. Same decorations, large circles. Cf. Ilios, fig. 162. 4. Top of oinochoe, of fine dark reddish clay. Same decorations with very small circles, which extend over the lip. Fragments of this type were common at the Heraeum. They possess, however, no affinity with the sieve vases from Troy, as the circles do not go clear through. Two variations in the type occur; (1) where the circles have a central part left (as in 2) and (2) where the circle is a complete hole. (3) Similar decoration occurs on ivory objects from Troy {Ilios, p. 566), as well as from the Heraeum, and on fragments from graves of the Hallstatt period in Vienna. Nos. 5-8 are all of a yellow or reddish clay, hand-made and ornamented with scratched lines. Similar fragments were quite numerous at the Heraeum, and probably belong to the earliest period. The plastic baud on 5, with its perforated ornament, is a common feature in prehistoric ware.i Cf. Sohliemann, Jlios, p. 491, fig. 454. 9. Low bowl ; height 0.27 m. ; diameter 0.17 m. Reddish clay. Incised dots on outer edge of base. Rim has an incised herring-bone pattern. Two holes through edge of base. 10. From shoulder of large vase, coai-se reddish yellow clay. Ornamentation probably ran round the shoulder like a frieze. 11. Handle of vase, coarse reddish yellow clay, hand-made. Plastic ornament resembling the knuckles of a hand. 12. Same. Coarse, dark reddish clay, hand-made ; probably burnt. In plastic form, with four divisions, two large and two small. Several other similar handles, belonging to large anaphoras were found, which vases, as far as can be told from the fragments, possessed no decoration whatsoever. 13. Handle, reddish yellow clay, ornamented with a series of V-shaped incisions. Cf. Ilios, fig. 1000. 14. Reddish clay. Wave patterns incised in similar fashion to 1. In left-hand corner of the fragment a projection like a button. 152 a and b. Bright red clay, hand-made. Base of tall jar. Around bases plastic mould- 1 This would seem to be one of the earliest examples xxii. (1898), p. 441, note 1 ; B. C. H. xix. (1895), p. 179, of a snake, wliicli, though common in Mycenaean and later note 6. times, is probably not of Mycenaean origin ; cf. B. C.H. ^ A fragment very similar to No. 15 has been found ^ ■> -^ l,y YV^ijg at Aphidna (Aiken. Mitt. 1896, pp. 394, 407, pi. 70 THE VASES AND VASE FRAGMENTS ing with scratches. Above, narrow band of triangularjndentations and two bands of ray pattern, incised. The same triangidar indentations occur on a fragment from the Cyclades 'E0j;/a. 'Apx- 1898, pi. ix. 10 and 23. The following fragments are all wheel made, and belong certainly to a later time, probably just prior to the beginning of the Mycenaean period. 16. Greenish yellow clay. From base of a vase with tall foot. Double herring-bone pattern and lines, incised. 17. Greenish-yellow clay, from vase similar to 16. Rays and bands incised. 18. Reddish clay, inside entirely covered with a dark brown glaze. Rays and scratches incised. As the technique of this fragment is distinctly Mycenaean and rather advanced, it seems most probable that it should fall in that period. 19. Light yellow clay, probably from an aryballos. Herring-bone pattern and lozenge contain- ing a swastika, incised. 20. Bright yellow clay, with black core. From neck of a vessel. Three raised bands, cross- hatched, inclosing two bands, upper with rosettes and herring-bones, lower with herring-bones alone. 21. Reddish yellow clay. From neck of aryballos. Plastic tongue pattern on shoulder. The technique of 19-21 is decidedly more advanced than that of the other fragments^ At the same time it is open to question whether, with the possible excej^tion of 21, they can be regarded as later than the Mycenaean period. VASES. 1. (Fig. 1.) Three-legged bowl, two of the feet restored. Hand-made ; height, 0.092 m. ; diam- eter of opening, 0.08 m. Black clay with brick-red slip, burnt dark brown. Scratched ornamentation on rim, neck, and belly. The handle is divided into two twisted divisions, also ornamented with scratches. On the front, two breast-like protuberances. This arrangement of scratches is similar to that of 'E<f>r]iJ.. 'Apx- 1898, pi. X. No. 1, but not so regular, and is undoubtedly earlier. The shape is similar to lUos, Figs. 59 and 1130. (Cf. also Koerte, loc. cit. pi. ii. 9.) Such vessels have usually been classed among cooking-pots. ^P «' 2. (Fig. 2.) Vase with long neck, without handle ; height, 1^ _, Ik 0.07 m. Black clay, polished and burnt. The ^^> shape of our vase is similar, tliough a trifle ^^^ j,,jg J more advanced, to a vase from the same place. Cf. Koerte, loc. cit. pi. ii. 8. The polishing of the clay is a common feature among primitive vases. It may be observed on a large number of vases and sherds from Hissarlik, and also on the vases from Bos-ojiik. Cf. Koerte, loc. cit. p. 25. ^y . ^ 3- (Fig- 3-) Oinochoe ? Neck and handle missing : height, 0.025 m. Black clay, burnt, with incised border on shoulder. Shape of vase cannot be determined, but it rather recalls the latter oinochoe type. Similar decoration may be found on a Trojan vase of the fourth settlement. Cf. also, "Ec^j/ja. "Apx- 1899, pi. ix. 24. XV. 3), and assigned by liini to his early indigenous Geo- ments in tlie (ieonietric style in Attica are to be found in metric class. This would seem significant for dating our the Argolid. This point will be discussed more thoroughly fragment, and also for showing that the pre-Dorian ele- in Chapter III. Fig. 2. Fig. 3. THE MYCENAEAN STYLE 71 Further, about twenty very small hand-made vases were found, which had no decora- tion, but were uniformly of a dark gray or reddish clay, showing traces of burning. They included two-handled vessels, similar to 'Ecft-q/j.. '\px- 1898, pi. ix. 2, and Koerte, loc. cit. pi. ii. 7 ; several three-handled jugs of similar shape to those on p. 100, and a vase without handles, similar to Koerte, loc. cit. pi. ii. 9. A series of vases, which, though not prehistoric in character, seem more nearly related to the earlier than the later vases, is represented by 4-6. They are all of a very dark clay, hand-made and burnt black in most cases, some with a decoration of incised lines running from neck to foot, others without decoration, and chiefly remarkable for a very brilliant polish, giving nearly the effect of dull jet. These vases are very similar to several from Rhodes in the British Museum, and it may be questioned whether our vases are not as are those, a product of the eighth century. At the same time, owing to their scratched ornamentation, they are more fitly included in this chapter, and if of a later date are probably the last links of the chain begun in prehistoric times. 4. (Fig. 4.) Aryballos ; height, 0.045 m. Black clay, burnt, with decoration of parallel lines. Much of the outer surface worn off. Several others were found at the He- raeum. An aryballos precisely similar is in the British Museum (A. case 6, No. 658, from Rhodes). 5. (Fig. 5.) Aryballos ; height, 0.034 m. Fig. 4. Fig. 5. Fig. 6. Black clay, with very brilliant polish. Below handle parallel scratches and a figure shaped like a bow or a crescent. 6. (Fig. 6.) Oinochoe ; height, 0.045 ni. Black clay, polished and burnt, without ornamentation. Several similar jugs were found. CHAPTER II. THE MYCENAEAJ^ STYLE. The development of the Mycenaean style at the Heraeum finds its parallel in the Mycenaean fragments from Mycenae and Tiryns. But unfortunately the Heraeiun frag- ments throw no light on the question whether the Mycenaean style is of Achaean (as is usually supposed), Pelasgian (cf.J. //. *S'. XVI. [1896], pp. 77 ff.), or foreign origin. That prototypes of the Mycenaean style exist in the Aegean pottery is now, I think, gener- ally admitted, and it may well be the case that the beginnings of the civilization are to be looked for in the Aegean islands, especially Crete. The connection, however, between the Mycenaean ware at the Heraeum and the Aegean pottery is not easy to establish ; the earliest elements of the style in the former site are, as Professor Waldstein has pointed out, distinctly linear in character ; and that, as far as I can judge from the island pottery I have seen (e. g. Melos), does not seem to be the case in Aegean pottery. If anything, the characteristics of the Aegean pottery are more naturaUstic than linear. Perhaps" the most probable explanation is that the Mycenaean style in the Argolid developed directly from the primitive linear tyjies, and that the influence of the islands brought about the change to naturahstic forms of ornamentation. The two great classes of the Mycenaean style, vases with dull decoration {Mattmalerei) and with luStrous decoration {FirnlssmaJerei), are both represented at the Heraeum, the former in very small quantities. Although it is usually supposed that " dull" vases are fz THE VASES AND VASE FRAGMENTS older than " lustrous," certainty on this point is jmpossible. The character of the dull style is linear and as such more in keeping with the prehistoric motives ; also as the invention of a lustrous glaze is a distinct innovation in ceramic art, it is probably an improvement on the " dull " technique. But that the manufacture of dull vases continued almost as long as that of lustrous vases, is proved by the fact that both dull and lustrous fragments lay side by side in the same levels. VASES WITH DULL DECORATION. Four small vases (to be described later) and about fifty fragments of this style were found, none belonging to very large vases. To establish a classification with a material so scanty and unsatisfactory is impossible. Furtwiingler and Loeschcke divide this type into two classes: (a) vases made of red clay (" Rotthonig ") and (b) vases of pale clay (" Blassthonig "). Such a classification in the case of the Heraeum fragments proved un- possible, since the clay ran through the various shades of red, yellow, gray, and green, and no vital difference in the decorative forms of fragments of red or pale clay could be distinguished, and the polish, according to Furtwiingler and Loeschcke, a characteristic feature of class n, was noticeable only on vases of j^ale clay. It is true that the majority of our dull fragments show a decoration decidedly linear in feeling, and also that some show a pictorial or naturalistic decoration very similar to vases of the lustrous style. That the first are the earlier of the two seems probable. At the same time the difference is not so great as to warrant our assigning the fragments to different classes, since many fragments bearing linear motives may well have belonged to vases which also showed pictorial ornamentation and vice versa. The classification of fragments alone is far more unreliable than that based upon entire vases. Hence, in the case of " dull " fragments, no attempt has been made to separate them into two classes, though they have been arranged with a view to the character of their ornamentation, be it linear or pictorial. Only the most important fragments are shown in Plate LI. There were many which bore no decoration, though clearly belonging to the same vases as some of the fragments here reproduced ; many again bore only a part of a band or stripe running around the belly. A certain difference can be detected in that in some of the fragments the clay is covered with a iine slip and in some not. The majority of the fragments show this feature, which seems to have been a characteristic of the style from its beginning, since even those vases with the simplest linear decoration show it. As it runs through all the shades of clay, it offers no ground for classification. The polishing of the surface seems to have been the general custom. The clay runs through all varieties and colors, from a very coarse variety with black stones still apparent on the surface to extremely fine clay, cleaned and polished. The majority of the vases and fragments were made on the wheel, though a few are hand-made. FRAGMENTS. PLATE LI. 1. From bowl with large opening. Form, Myk. Vas. XLiv. 48: height, 0.165 m. ; width, 0.125 m. Coarse pale greenish clay, polished on outside and inside. On inner part of rim series of three short parallel dashes. Decoration in black, faded. Cf. MyL Thong, iv. 17. 2. From one-handled-cup. Form, Myli. Vas. XLIV. 98: width, 0.118 m. ; height, 0.068 m. THE MYCENAEAN STYLE: "DULL" DECORATION 78 Coarse brownish yellow clay with whitish yellow slip on exterior. Dashes on rim similar to 1. In i-im hole pierced for suspension. Black decoration. 3. Lip and handle of bowl similar in shape to 1: width, 0.112 m. ; height, 0.095 m. 4. Neck and shoulder of jug. Form, Myk. Vas. XLiv. 20 : height, 0.105 m. ; width, 0.125 m. Coarse reddish yellow clay ; yellowish slip with decoration in violet brown. Cf. Myk. Tliony. IV. 13. 5. Reddish yellow clay, fairly fine in texture. Pale slip on exterior with violet black decora- tion. Cf. 'E(^rj/j,. 'Apx- 1899, plate viii. 104 (vase from Syra). 6. Neck and shoulder of jug similar to 4. Fine reddish clay with greenish yellow slip. 7. Fairly fine pale reddish clay, dull polished surface. Three narrow black bands inclosing a dark red zigzag, and a broader band of dark red ; traces of another zigzag also in dark red. This fragment is almost identical with Myh. Thong, v. 20. This use of two different colors, one (in this case the red) added after the first firing, is an extremely common feature in the Heraeum fragments, and will be analyzed more thoroughly in a later chapter. No. 7 was the only fragment of the dull finish which showed its employment. A few fragments of a very dark red clay with a dark red slip on exterior and interior and black decoration were found. They are in all respects similar to the vase from the Bee-Hive tomb near the Heraeum. Cf. Mijk. Thong, xii. 52 ; Athen. Mitt. III. p. 271. 8. Neck, shoulder, and base of handle, probably from an amphora. Fine greenish clay, pol- ished, with light brown decoration. 9. From large bowl, similar in form to 31yk. Vas. XLiv. 75, but with single handles, about 25 cm. in height and 24 cm. in diameter. Coarse reddish clay, which has received a very high polish and presents a peculiar soapy feeling to the touch. Zigzag band below rim in dark red, very much faded. (Fig. 7.) A few other vases and fragments from the Heraeum showed this same peculiar surface, which otherwise is ^^^ ^ unknown to me. It is probably due to some accident either in the composition of the clay or the peculiarity of the soil in which it is found. 10. Coarse brownish clay with white slip outside. Violet brown decoration. All these fragments show a decided " linear " tendency ; the rest (10-16) are more nat- uralistic in character. The similarity of decoration in 14-16 to vases of the lustrous finish would warrant our assigning them to a later period. They evidently do not belong to the earUest essays in dull decoration. 11. Shoulder and handle of small three-handled amphora (form Myk. Vas. XLiv. 35), with brown black decoration. , , x,- -iao ir 12 Rim and spout (broken off) of a bowl, similar in shape to Myk. Vas. XLiv. 103. Very coarse brick red clay, with a pale white slip on exterior and interior. Decoration in black, laid on heavily and badly worn. 13. Shoulder of small jug. Similar technique to 11 and 12. 14. Spiral in black. . ^ , ,. , i j 15. rL of bowl. Fine brick red clay, with bright yellow slip, highly polished ; decoration in '' This^'frrgment, from the excellence of its technique, as well as the form of the spiral, is clearly synchronous with vases of the best period of the lustrous style. _ • •, ,„ 16. From bowl similar to 15. Greenish clay, with brown decoration. Technique similar to vases of the later (not the best) lustrous style. i THE VASES AND VASE FRAGMENTS VASES. *■ 17. (Fig. 8.) Height, 0.06 m. : diameter, 0.07 iii. Bowl with high foot, being a cross between a kylix and a slivphos, one handle missing. Bands and a continuous spiral. Another vase of exactly the same shape and dimensions, but in- tact, was found in the same tomb. The surface is completely cov- ered by a hard incrustation of lime, but as far as can be seen the decoration is the same as 17. 18. (Fig. 9.) Height, 0.028 ra. ; diameter, 0.066 m. Slight variation of Myk. Vets. XLiv. 87. Bands on rim, belly, and han- dle, and row of dotted semicircles. 19. (Fig. 10.) Height, 0.036 m. ; diameter, 0.056 m. Three-legged bowl, handle restored. Continuous spiral on sides, and on bottom be- tween the legs a design like an S reversed. Clay shows traces of burning. These four vases come from the grave found April 13, 1894, back of the peribo- los wall above the South Stoa. The technique is the same in all, a fine yellow clay, highly polished, without a slip, the decoration in violet black, laid on heavily. All are hand- made. For photograph of tomb see vol. I. p. 41, fig. 13. VASES WITH LUSTROUS DECORATION. The introduction of lustrous paint in the manufacture of Mycenaean vases was revolu- tionary in its effects. Henceforth all vases exhibit this feature. But as we have said, on its introduction the dull finish was not at once abandoned, but continued side by side with the lustrous technique for a considerable period ; just how long it is imjjossible to say, but certainly until after the best period of the lustrous style. Until lately Furtwangler and Loeschcke's division of the lustrous style has been univer- sally followed. This classification has, however, proved inadequate for our vases, and for those from the Acropolis and Thoricus. A second classification,' proposed by Wolters, is more satisfactory, and has proved a better standard for arranging ours. But in spite of its many advantages, it is somewhat unwieldy, so that I have ventured to adopt a middle course, in order to reconcile both Furtwangler and Loeschcke's and Wolters' arrangements. The three are here presented side by side. Wolters. I. The same. Only a few frag- ments of this style were found on the Acropolis. HOPPIN. I. The same. This class is not found at the Heraeum at all. Furtwangler and Loeschckg. I. Wheel-made vases of a very coarse clay, entirely covered with a fairly dull black slip, on which the ornamentation is thinly laid in white and dark red. This class was chiefly found in graves iv. and v. at My- cenae, at Tiryns, and at Thera. ^ This classification has never been published, so far as I know, and has been drawn from the author's lectures. THE MYCENAEAN STYLE: CLASSIFICATIONS 75 II. Vases of coarse clay, covered with a thin slip of finer clay, now white (in the case of vases from the Bee-Hive tomb near the Heraeuni),nowyellowish brown, (v. Myh. TJiong. vii. 42 ; Myk. Vus. p. 21, fig. 7.) The de- coration is painted on this slip in dark brown, with the occa- sional addition of white, as if an echo of the technique of the first class. III. Fine cleaned clay, with a pol- ished surface of warm yellow color. The colors in the deco- ration run through all shades from yellow to dark brown. This, through action of fii-e, becomes bright red in many cases, which, especially in vases of the finest technique, is evidently inten- tional. Details are occasion- ally added in white. II. No radical change. The main point of difference lies in the close connection between II. and the following class, which varies from it only in the qual- ity of the work. The style throughout is distinctly a natu- ralistic or pictorial one, there being no thought of convention- ality. The vase from Thorikos is an extremely good example of the class as a whole. Cf. 'ZipniJL. "Apx- 1895, pi. xi. No. 1. III. 1. Fine clay, with purely pic- torial ornamentation, precisely similar to II., save that the exe- cution is more delicate, which is natural, seeing that the vases are uniformly smaller. The pictorial motives have become conventionalized. In proportion, however, the tech- nique has also advanced and reaches its highest point in this division, and may easily be de- tected by the extreme fineness of clay and extraordinary bril- liancy of the glaze. This class is not the most common of the Mycenaean styles. II. 1 and 2. In the first division of II. have been included all vases wliich in the other two classifi- cations are counted as II. In the second division are in- cluded a number of vases be- longing to Furtwiingler and Loeschcke, III., and Wolters, III. 1. III. 1. Seeing that vases of this class differ from tlie succeeding one only in superiority of technique, while the principle of ornamen- tation remains the same, there seems to be little gain in sepa- rating them by the boundary line of a class. 2. In this division are included those vases which form the bulk of III. of Furtwiingler and Loeschcke. They are the most common at the Ileraeum. The clay has become inferior and the lustre duller, while the orna- mentation has become conven- tionalized in the extreme, and presents a certain mechanical air. It is only in its technical features that it differs from class III. 1. 7G THE VASES AND VASE FRAGMENTS IV. The surface of the clay is greenish or a somewhat dull yellow, occasionally reddish yel- low, but far duller than the pre- ceding class. The decoration is black or yellowish brown, occa- sionally burnt red, but never at- tains to the brilliancy of vases of class III. In the case of vases with a wide mouth, the interior is always glazed. IV. * 1. In this division are placed those vases included by Furt- waii":ler and Loeschcke in III., and bv me in III. 2. 2. rv. This class remains undis- turbed as in Furtwiingier and Loeschcke. Only a few frag- ments were found at the He- raeum. In this division fall all vases in IV. of Furtwiingier and Loeschcke. Vases of this type show a more florid and perhaps decadent style of decoration. The ornamentation abounds in a wealth of detail not seen in the other classes. It has been found in fairly large quantities at Athens, on the Acropolis, and at Aegina and Thoricus. It may be noticed that in Professor Wolters's classification the separation of Mycenaean kistrous vases into two great periods is hinted at, if not expressed. We may commence by drawing a sharp line between the pictorial and the conventional, since the progression from natnralism to conventionalism is a greater one than from one plane of technical skill to a higher. We may therefore keep Furtwiingier and Loeschcke's four classes and separate them into two groups, I. and II., III. and IV. It will thus be seen that dividing III. and IV. into two classes each, as Wolters does, and connecting III. 1 with II., and III. 2 with IV. is somewhat awkward. Also the line between naturalism and conventionalism is not properly emphasized, since they occur in the same class, which is clearly impossible if we are to follow a classification based on ornamental development. By separating II. into two divisions we adhere to the close connection emphasized by Wolters while confining the principle of naturalism to one class. With III. we enter upon the second or conventional group, and we divide this into two classes. III. 1 and III. 2, which correspond to Wolters's III. 2 and IV. 1. Their connection is too strong (the difference between them being purely technical) to warrant their being separated as they are by Wolters, while at the same time we are more in accord with Furtwiingier and Loeschcke, except for the transferring of a small part of their class III. to II. There seems to be no valid reason for separating IV. into two groups. Cer- tainly the difference between IV. 1 and IV. 2 (Wolters) is as great, both from a technical and ornamental standpoint, as between III. and IV. (Furtwiingier and Loeschcke). The following table will show briefly the results thus obtained : — FURTWANGLEB AND LOESCHCKE. WOLTERS. I. n. ni. IV. I. IT. IIL 1 III. 2 IV. 1 IV. 2 HoppiN. Naturalistic or picto- rial style. Conventional stjde. THE MYCENAEAN STYLE: EVIDENCE FOR DATE 77 What the chronological ditfereiices are between these varions classes it is difficult to say. According to Furtwangler and Loeschcke, I. is the oldest; but it is extremely doubt- ful whether any difference in time exists between I. and II. As I. is found in but few other places outside of Mycenae, it is fair to suppose that it was a style more or less local. Judging from the entire lack of this class at the Heraeum, and tlie fact that the style of ornamentation of II. 1, which is the oldest class of lustrous vases at the He- raeum, is practically identical with that of I., there seems good reason for sui)posing that the two are synchronous. Moreover, the difference between II. 1 and the dull vases is so extremely slight that it can be detected only by a carefully trained eye, and even then cases occur where the decision is doubtful. This would show that the lustrous technique at the beginning did not differ materially from the dull, and is another point in favor of assigning II. 1 to the beginning of the lustrous style. II. 2 differs from II. 1 mainly in the technical advance, but this advance is sufficiently apparent to enable us to see in II. 2 the successor of II. 1. The step between II. 2 and III. 1 is even greater, since the style of ornamentation has radically changed, and all motives which II. drew from the marine life that played so great a part in the mari- time civilization of the Mycenaean epoch have become conventionalized and are used more with a view to their decorative effect than as an attempt to reproduce nature. Such a change could have come only during the acme of the Mycenaean epoch. With IV. we see the point to which the bad taste of a decadent art had come. The wealth of ornamentation, elaborated from a given motive, with the introduction of foreign motives, illustrated by " Heraldic " designs and those taken from Oriental embroideries, may be accounted for by this rampant spirit of conventionalism combined with the increased commercial activity of the age. A few words may here be said as to the latest results in dating the whole Mycenaean period. Through the numerous excavations recently conducted on Greek soil, and the corresponding increase of Greek pottery brought to light, the chronology has been worked backwards to the fourteenth century. But the excavations of Flinders Petrie * in Egypt, and the finding, in the towns of Illahun, Tel-el- Amarna, Kahun, and Gurob, of rubbish heaps containing large masses of " Aegaean," i. e. Mycenaean pottery, may be said to have absolutely established the date of the Mycenaean civilization, since the objects of Egyptian workmanship lying in these rubbish heaps, along with the pottery, can be dated not later than this eighteenth dynasty, circa b, c. 1450. Pseud-amphoras of Class III. were found in tombs of a date not later than b. c. 1300. The placing of the acme of the Mycenaean period during the fifteenth century thus accords with the hypo- thesis advanced by Furtwangler and Loeschcke on the basis of Egyptian wall paintings.'^ As Class IV. was not found by Petrie, we may assume that it is later than b. c. 1300 ; but there is no reason for assuming that the manufacture of Class III. ceased after that date. We must also allow at least a hundred years for the development of the Mycenaean style from its beginning, so that to place the manufacture of dull vases as early as the sixteenth or seventeenth centuries is permissible. We thus obtain a period of about five 1 For a more detailed account of Petrie's results, v. hm, and Gurob, p. 10, pis. xvji.-xxvi. ; Kahun, Gurob, and Tel-el-Amarna, pp. 16, 17, pis. xxvi.-xxx. ; Illahun, Ka- Hawara, pi. xxviii. 2 Myk. Vas. p. xiii. 78 THE VASES AND VASE FRAGMENTS centuries from the rise of the Mycenaean style to its downfall at the time of the Dorian invasion, or whatever the upheaval was which led to its decline. That all the vases of the Mycenaean style found at the Heraeum were manufactured in the Argolic plain, and not imported from some other centre, seems unquestionable. Whether the Argolid was the chief centre of the Mycenaean civilization or not cannot be absolutely affirmed, though the evidence seems to point to this supposition. At any rate, the amount of vases found at Mycenae, Tiryns, Naupha, and in the various tombs throughout the plain, is so great that we can safely assert them to be of home manufacture. That Class I. is found at Mycenae and not at the Heraeum may perhaps be accounted for by a difference of feiste in the two places. The Heraeum, which was one of the largest sanctuaries in Greece, must have kept several potters' factories busy to supply the faithful with the requisite vases for ex votos, and that cerfciin styles should have been popular there and others not so is not extraordinary. Several facts may be noticed in regard to our fragments which are significant. The singular uniformity of all the fragments of the Mycenaean style, both in clay and technique ; the complete absence of Class I., and, lastly, the equally complete absence of any foreign variations (e. g. Theran, Melian, Cypriote, etc.) of the regular Mycenaean types. These facts woidd seem to indicate : (1) That the potteries which produced the ware found at Mycenae were not the same as those which produced our fragments ; had such been the case it is scarcely credible that no traces of Class I. should have been found. (2) That all Mycenaean pottery found at the Heraeum was the product of one or more particular centres of activity, situated near the sanc- tuary. (3) That this manufactory reserved its wares exclusively for home consump- tion, and neither exjjorted its product or imported similar wares. This last assumption is based on the fact that those particular varieties which are indigenous to some foreign spots are hardly represented at the Heraeum, and the few exceptions to this rule are probably accidental. On the other hand, no Mycenaean vase found outside of the Argolid can be proved to have been manufactured near the Heraeum. CLASS II., DIVISION 1. Of this class only a few dozen fragments were found, and no whole vases. Only a few fragments clearly belonged to the same vases, and the reconstruction of any vase proved impossible. The characteristics of these fragments are similar to those of a jug in Athens {Myk. F«s. p. 49, fig. 29) and a three-handled vase from Thoricus {'E(f>r}iJ,. 'Apx- 1895, pi. XI. 1), though the decoration is extremely simple and monoto- nous, with no attempt at any elaborate design. Only a few of the fragments are here reproduced. The technical features are : very coarse quahty of clay, Avith frequent small stones, varying through the different shades of red and brown to green ; a thin wash of white, yellow, or red ; ornamen- tation in violet, brown, or red. 1. (Plate LI. 17.) From belly of vase, form uncertain. Dark red clay, with a darker central core, and liglit red wash, dark red decoration. Two fishes advancing towards a monster of some sort whose head and forelegs alone are visible. To identify the cliaracter of the fishes or the monster is impossible. Cf. Myk. Vas. x. 63, XXXIX. 401a ; Schliemann, Mycenae, Fig. 317 ; Imhoof-Blumer and Keller, Tier und PflanzenhUder, vi. 47, 48, vii. 3. 2. (Plate LI. 18 a-d.) Four fragments from same vase, form uncertain, clay dark red, THE MYCENAEAN STYLE: CLASS IL, DIVISION 2 79 white wash, with faded violet brown decoration. The design cannot be restored, but that the spiral formed part of it is certain. 3. (Plate LI., 19 a and b.) Two fragments of same vase. Thick brown clay, yellow wash, brown decoration. Flower pattern. 4. (Plate LI., 20.) Greenish clay and wash, violet black decoration, with a greenish tinge. Flower pattern. 5. (Plate LI., 21.) Brown clay, with yellow wash. 6. (Fig. 11.) Fragment of large amphora: height, 0.108 m. ; width, 0.12 m. ; very coarse brick red clay, pale reddish wash, dark red decoration. (Flower pattern.) The height of the amphora cannot be determined, but the opening mea- FiG. 11. sured certainly 13 cm. in diameter. Cf. Myk. Vas. XXL 156. CLASS II., DIVISION 2. The fragments of this particular class were far more numerous than those of the pre- ceding, and are uniformly of a finer character, though the pictorial style of ornamentation remains the same. They represent throughout smaller vases. The clay is of a better quality, carefully cleaned, running through all the shades of red, brown, yellow, and pale green, the decoration generally red or brown. The technical method is the same in all. Over the natural clay a slip is laid, of very fine red or yellow clay, on which the design is painted and the surface polished. In some cases the slip is on the inte- rior as well. Occasionally the decoration is laid on so thickly as to stand out from the surface of the clay, and be easily friable. Cases occur where a whitish wash has been applied, as in Class II., 1. Fragments which resemble this class very strongly have been found in the Bee-Hive tomb near the Heraeum, already referred to. In only a few cases were fragments clearly part of the same vase and no entire vases were found. The style of ornamentation is uniformly naturalistic or pictorial, the motives being taken almost exclusively from flowers, while marine subjects are hardly used at all. In but a few instances can the form of the vase be determined. PLATE LIT. 1 a and b. Two fragments from the same vase, form uncertain. Eeddish clay, yellow slip, red brown decoration. Flower wreath, a is perhaps the best example of Class II. 2 at the Heraeum. Fragments very similar have been found in Crete ; v. Haussoulier, Rev. Arch. XL. (1880), p. 359, pi. xxiii., and more recently by Evans at Cnossus. Cf. Arch. Anz. 1900, p. 149, fio-. 6. Cf. also Myk. Vas. p. 23, fig. 12, and the fragments from the Bee-Hive tomb near the Heraeum: Myh. Thong, xii. 64; cf. also Myh. Vas. xxvri. 213, 217; A. J. A. VI. (1890), pi. xxii. 2. Reddish yellow clay, brilliant red brown slip, red brown decoration. Two other small frao-ments from the same vase were found. This fragment is' an exact duplicate of one from the Bee-Hive tomb ; v. Myh. Thong, xii. 57. 3. Similar to 1 but slightly coarser clay, slip, and flaky decoration. Probably part of shoulder of a three-handled vase (form, Myk. Vas. XLiv. 32). A similar smaller fragment, but from a different vase, was also found. 4. Yellow clay, flaky black decoration. Cf. Myk. Thong, xii. 60. 5. Yellow clay, brilliant black decoration. 6. Double flower. Cf. Myk. Thong, xii. 73 ; B. C. H. X. (1886), pi. iii. 5. 7. Rim of small bowl. Flower ; similar technique to 2. 8. Black border to stalks, changing to red in the centre. 80 THE VASES AND VASE FRAGMENTS 9. Whitish wash on exterior, with i-ed brown glaze on interior.- 10. l*art of a plate, with branches on the reverse similar to 12. Clay has a peculiar ribbed surface. 11. Decoration ranging from brown to red. Presents same ribbed surface as 10. 12 a and b. Two fragments from a vase of clay and technique similar to 8. Cf. Myk. Vas. p. 58, fig. 34. 13. Rim of a large bowl. Clay rather coarse. 14. Grayish clay, evidently burnt. A few other fragments of similar technique and decoration were found. Cf. Myk. Thong, iii. 10 ; Myk. Vas. xxi. 153, xxvi. 195. 15. Flower pattern in dark red. Polish on exterior and interior. 16. The same. 17. From shoulder of three-handled vase. Arrangement of petals similar to 1. Cf. Myk. Vas. XXXVI. 376. 18. Dark core in the clay. Decoration in red. 19. From a vase of form 3Iyk. Vas. xliv. 59, probably. Whitish yellow slip. Cf. Myk. Vas. XIII. 89. 20. Combination of branches (watergrass, perhaps), similar to 12, and snakes or eels. 21. Rosette. 22. Form uncertain. This fragment, as well as several others, shows a peculiar technique, the addition of a white streak to the decoration after firing, but before glazing. For similar technique, cf. Myk. Vas. xxvi. 203. 23. Black and red decoration. Uncertain what the complete design was. 24. From a plate. Reddish clay, with yellow slip on obverse, red on reverse. Flower pattern, the same on both sides in dark red. Cf. Myk. Thong. Xil. 68. 25. Part of plate, with pattern similar to 3, in flaky dark brown. 26. Greenish clay. Cf. Myk. Thong, ill. 9. 27 a and b. Two fragments from the same vase. Reddish clay and decoration. Alternate buds and flowers. In spite of a certain conventionality of treatment, and indifferent technique, the resemblance of these fragments to those from Therasia (^Myk. Vas. xii. 78) justifies their position in this class. A considerable number of fragments representing three-handled vases similar to Myk. Vas. vii. 45, ix. 52, were found. This seems to have been one of the favorite shapes of this class. As a rule, ornamentation is introduced into the space above the scallops. Fig. 12 a and b. Fragments of large bowl which measured 45 cm. in diameter. Form uncer- tain. Whether it had a foot is doubtful, so that the height cannot be estimated. Coarse reddish clay (thickness 0.012 m.), with a yellow slip, more reddish in tone on the inside. Decoration in light and dark red. On the outside (b) a row of tall leaves, separated by stalks ; on the inside, on rim a variety of " tongue pattern," with a milled border. In field two fishes and an eel. Fig. 12. This plate is perhaps the most interesting specimen of Class II., 2. The form is unknown to me in any other vase. It varies from the regidar plate type in that the subject is different on the two sides. The pattern on the rim is clearly a very early instance of the " tongue pattern " so common THE MYCENAEAN STYLE: CLASS IIL, DIVISION 1 81 in later vase-painting. Cf. M;ik. Vas. xxxiv. 344. The leaf pattern on the outside is only a symmetrical arrangement of the design on Myk. Vus. xxxii. 314. The species of the fishes (perhaps carp) remains douhtful, but the naturalness of the drawing is an advance of No. 1 on Plate LI. CLASS III., DIVISION I. We now cross the great bridge wliich divides the Mycenaean style and find that the old naturalism has given place to conventionalism. But this is not the only dif- ference between Classes II. and III. ; technique itself has made great progi*ess. The clay used is generally red or yellow in color, of a fine clear quality, carefully cleaned from impurities. The quality of the decoration is more lustrous and glossy, while in the best specimens the glaze takes on an intense brilliancy. The style of ornamentation changes, and certain motives used occasionally in the previous classes now become the favorites, such as spirals, etc. The old plant and marine motives with a few exceptions (e. g. the murex) have disappeared, and those which are now prominent are paralleled in other forms of Mycenaean art in stone and metal. Though the conventionalism of Class III. follows the naturalism of Class II., and in this sense is a later style, we are not to assume that the manufacture of Class II. ceased. In fact there is positive proof that the two flourished side by side, since in the Bee-Hive tomb which we excavated (cf. p. 91 ff) were found vases of both classes. Such a radical change as that from naturalism to conventionalism is not the work of a day ; it must have extended over a long period. But such a fact does not argue against a classification based on ornamentation, where a difference of time is not necessarily demanded. Nor should the relation of III. 1 to III. 2 be misunderstood. We cannot regard them as two absolutely distinct and unrelated styles. I do not for a moment intend to imply that for a certain period of years Mycenaean potters continued to manufacture vases of a certain " hall-marked " excellence and then suddenly began to go down hill. What we call III. 1 is undoubtedly the best work of a certain period of Myce- naean ceramic art ; III. 2 is the general average of such a period. As the best work of a given period is only a small proportion of the whole, the small amount of III. 1 com- pared to III. 2 bears this out. III. 1 bears the same relation to III. 2 that the work of Euphronios, Hieron, Duris, and Brygos does to the mass of red-figured vases produced in Athens prior to the Persian wars ; it is the best ceramic art of the time. The fragments reproduced on Plate LII. are not all that were found, but the number was not large, and those represented illusti-ate the group. The technique of all the fragments is substantially the same, the clay red or bright yellow (more often the former), the decoration varying from red to brown, with black occasionally used and the glaze of surpassing brilliancy and fineness. The favorite ornaments seem to be the spiral and the murex. The majority of the fragments are rims of two-handled bowls or kylixes similar in form to Myk. Vas. xliv. 76, 83. 28. Rim of bowl. Decoration shades from brown to red. Part of a spiral row, spirals uncon- nected. Cf. Myh. Vas. vin. 46. 29. Rim of one-handled bowl. Clay extremely delicate. Outer band of spiral brown, inner bands red. Cf. Mijk. Vas. xxix. 258. Both form and decoration seem to be the same as on the vase in Cassel. Arch. Anz. 1899, p. 57, fig. 1. 82 THE VASES AND VASE FKA(iMENTS 30. Rim of bowl. Spiral chain. » The majority of the fragments of this class bore the decoration of spirals, in chain or single. Most of the fragments were extremely small. 31 a-C. Tln-ee fiaginent.s from rim of a bowl. Reddish clay, with bright yellow slip and red decoration. For the toothed edge of c, cf. Myk. Vas. xxiii. 171. The murex or piu-i)le fish ornament is extremely common. It is curions that we cannot trace its origin to an earlier class, since it does not occur in Classes I. and II. Even in Class III. it is impossible to trace any preliminary steps unless we recognize one of the early essays in the cup from lalysos ; * that, however, seems rather too conventionalized for a prototype. The arrangement of the pattern varies, the most common being a row of double murexes as in Myk. Vas. xxxi. 297, or else they are laid diagonally on their sides. 32. Rim of bowl. Series of murexes placed diagonally to the line of the rim. Cf. Myk. Vas. vili. 47. Decoration bright red and technique extremely fine. 33. The same. 34. Clay a trifle coarser than is usual in this class and appears burnt. Glaze extremely brilliaut. Undoubtedly part of a large vase, as the murex is far larger than is usual. Various Designs. 35. Rim of bowl. Series of parallel zigzags, decreasing in size towards the base. Cf. Myk Vas. XXX. 273. 36. Greenish clay. The design cannot be reconstructed. The nearest similarity lies in two fragments of Class IV. (^Myk. Vas. xxxvii. 378, 379), and it is possible that we may have here as there the combination of a spiral and a ray. 37 a and b. Two fragments from same vase. Net ornament. Cf. Myk. Vas. xx. 146. 38. The same. 39. Rim of bowl. Undoubtedly a suggestion of a guilloche. Cf. Myk. Vas. xxxiv. 337 ; xix. 134. 40. Rim of bowl. Probably a design similar to Myk. Vas. xxx. 280. 41. Rim of bowl. Cf. 3fyk. Vas. xxxi. 288. 42. Fine red clay, with whitish slip, red brown decoration. Similar pattern to 37, save that a cross is placed at each corner of the mesh. 43. Similar clay and slip. Half circle at corner of each mesh. CLASS III., DIVISION 2. Class III. 2 forms the bulk of all Mycenaean pottery, no matter where found. Though the largest class of all, it is, as a whole, uninteresting save in regard to its bearing on contemporary art. There is little exercise of skill on the part of the potters, who seem to have been contented to turn out large numbers of vases varying little in form or decoration. The class differs principally from its predecessor in that the quality as a whole is decidedly inferior. The clay is coarser and not so carefully cleaned, varying in shade, red and yellow predominating. The decoration is of many colors, the different shades of red and brown being the favorites. The paint also is of a very inferior quality, and on many vases has almost entirely faded. The glaze is generally dull, and never reaches that lustrousness so characteristic of the preceding class. In the arrangement of the fragments, the order adopted by Furtwiingler and ' Oaz. Arch. 1879, pi. xxvii., also in Myk. Vas. ix. 56. THE MYCENAEAN STYLE: CLASS IIL, DIVISION 2 88 Loeschcke has been followed, since the general succession of our fragments differs in no great degree from that of other Mycenaean fragments. Only a few of the Heraeum fragments are here presented, but they illustrate all the principal types and the gen- eral development of the ornamentation. The condition of the fragments was extremely bad. Out of the whole class we succeeded in reconstructing only three vases, one of which coming from a small tomb had almost all the fragments preserved. Another half dozen had about half their fragments remaining, and twenty or more were represented by perhaps a dozen frag- ments apiece. More than a hundred were represented by from two to five fragments apiece. What the total number of vases represented by the fragments was, is impos- sible to say, as no calculation can come near the truth ; the number certainly ran into the thousands. As a rule where several fragments from the same vase were forth- coming, only one or two are here given, unless the scheme of ornamentation was not plain. As so many cases occurred where fragments of identically the same clay and technique clearly belonged to different vases, the relation of several fragments to one vase has been ignored, unless their common origin was clearly shown by a joining or otherwise. The field in which ouv fragments lie is practically covered by plates xvii. -xxxv. of the Mykenische Vasen. The fragments corresponding to plates xxv. and xxvi. we have already assigned to Class IL 2. It has proved impossible in all cases to follow Furtwangler and Loeschcke's arrangement exactly, and such variations as are introduced are justified by the demand of the particular frag- ments under discussion. I have endeavored to discuss the natural sequence of the ornamentation, i. e. to treat the conventionalizing of naturalistic motives first and the linear ones last, since we find the Mycenaean style as a whole passing from linear to naturalistic motives and through the conventionahzing of the latter, rececUng to linear themes again, as illustrated by Class IV. Fig. 13. Amphora, from 3Ij/k. Vas. xliv. 45 ; height, 0.33 m. ; diameter of opening, 0.105 m. Reddish clay, with yellow slip and dull black decoration. Prom a grave back of the peribolos wall above the South Stoa. Small part of vase re- stored in plaster. This was the only vase which allowed a fairly complete reconstruction. The ornamentation is shnilar to that of the amphora in Myk. Vnx. XXVI. 223, save that the central design and the buds of the flower are treated m a slightly different fashion. PLATE LIU. Flower Buds. 1. Rim and handle of bowl. The bud has here been cut in half. Another smaller fragment of this vase was found. 2. Flower buds in series. Cf. Myh. Vas. xxvi. 218, 220. 3. Flower bud and diamond. Cf. Myk. Vas. xxviii. 24(3. --^«^^w<s& Fig. 13. 84 THE VASES AND VASE FRAGMENTS Circles, Diamonds, etc. 4. Circle ; dotted row outside. Cf . Myk. Vas. xxviii. 229. 5. Same ; dotted row in.side. 6. Handle of bowl. Same as 4, with addition of central dot to the circle. Cf. Myk. Vas. XXVIII. 237. 7. Rosette in simplest form. 8. Lozenge with central cross ; similar to Myk. Ws. xxvill. 240, save for omission of outside lines. 9. Diamond chain. Cf. Myk. Vas. ill. 18. 10. Diamond, diagonals inside. Cf. Myk. Vas. xxix. 256. 11. Diamond circle in centre. Spiral Forms and their Development. 12. Rim of bowl. Another bowl of precisely similar form and decoration is in the Louvre, Room A, left hand window case (no number given). Spiral in its most elementary form. 13. Spirals in chain, simplest form. Cf. Myk. Vas. xxviii. 242. Fig. 14. Jug with spout and one handle. Height, 0.135 m. Pale yellow clay, with greenish slip, faded black decoration. Row of spirals unconnected. For form, cf. Myk. Vas. XI. 66. This jug was labeled as coming from the same tomb as the four small vases of the " dull " finish mentioned on p. 74. Professor Waldstein informs me that the label is wrong, as a photograph of the tomb shows only the four vases already referred to. It is probable that this jug comes from the same tomb as Fig. 13. 14. The same. Introduction of ornament (semicircle) in field. Similar to Myk. Vas. xxviiii. 245, save that here the ornament is unconnected with the spiral. 15. The same ; greenish clay, with faded decoration. 16. The same ; ornament resembling an E reversed. ^"^- ^*- Cf. Myk. Vas. xxxii. 312. 17. Two semicircles in field. Cf. Myk. Vas. xxxiii. 319. 18. Double row of spirals, alternating. 19. Similar arrangement of spirals to that of our amphora (Fig. 13). Cf. Myk. Vas. xxix. 256. 20. Cf. Myk. Vas. xxix. 253. Several other fragments of this bowl were found. Its chief peculiarity lies in the greenish yellow slip of the exterior. The decoration is considerably faded. Though no fragments fitted each other, it is plain that the scheme of decoration was a row of spirals, the loops curving upwards and unconnected. 21. Rim of bowl. Reddish clay and red brown decoration. 22. Yellowish clay, pale slip. Stripe on inside. Cf. Myk. Vas. xxix. 255. 23. Shoulder of three-handled vase. Double row of spirals. 24. From a cup of precisely similar form and decoration to one from Nauplia. Cf. Myk. Vas. XXI. 150, p. 146 ; also xxxii. 302. 25. Rim of bowl ; dull red glaze on interior. Undoubtedly same shape and design as Myk. Vas. XV. 93. 26. Clay of a bright brick red color. Whether the floral or marine influence is most prominent here is hard to say, as the movement strongly suggests the arms of a polyp. Cf. Myk. Vas. XVII. 110. 27. Handle with spirals. Introduction of swastikas. THE MYCENAEAN STYLE: FLOWER PATTERNS 86 Flower Patterns. The flower pattern in Mycenaean art is one of the earliest motives ; we have ah-eady found it in Class IL 2 (cf. Plate LII. 6 and 7), and similar though not identical plants may be found on the sword-blades from Mycenae.' A comparison of our fragments with those in Myk. Vas. xxx. shows many points of similarity, but the progression towards conventionalization mentioned by Furtwiingler and Loeschcke (p. 60) can hardly be seen here, since in none of our fragments is the drawing of the pistils, seed-bags, and leaves at all apparent, and all our fragments have arrived at the conventional stage. The pseud-amphora is the commonest form, though many fragments of bowls like Myk, Vas. xxx. 276, were found. 28. Shoulder of pseud-amphora. Two clays together, pale red inside, dark red outside, with a bright yellow slip. Cf. Myh. Vas. xviii. 124 ; xxx. 266. 29. The same. 30. From belly of a bowl. Cf. Myh. Vas. viii. 43 ; xxi. 157. 31. Rim of bowl. Several other fragments found. Cf. Myk. Vas. xvill. 122. 32. Similar design. 33. Similar design. Top of flower ends in a loop. Several other fragments from the same vase were found. 34. The zigzag corona of the flower now stands by itself. This is one of the commonest motives, and may be found on dozens of our fragments. Cf. Myh. Vas. ix. 51. 35. Series of zigzags used as border lines. 36 a and b. Two fragments from the same vase ; another smaller fragment was also found. Brick red clay, probably burnt. The zigzag pattern is used almost exclusively, and suggests a transition style to the geometric. The half circles as ornaments in field are used in similar fasliion to the " Running Dog " pattern. Cf . Flinders Petrie, Tel-el- Amama, xxvii. 41-43 ; Myh. Vas. XV. 96. 37. Flower branch. Cf. Myh. Vas. xxxi. 286. PLATE LIV. Flower Branches. 1. Simple form. Dotted row above and below corona. 2. Flower branch similar to Plate IV. No. 32, corona of zigzags. In field toothed wheel. 3. Similar treatment to No. 1. 4. Interior dotted row omitted. 5. Similar to No. 1. Clay burnt to a dark gray. 6. Lower part in form of two spirals curved like volutes. 7. Waving branch to each side, probably as an imitation of a palm-tree. Cf. Myh. Vas. xxxL 293. 8. Top arranged in triple form, half circles unconnected. 9. Part of large bowl. Triple arrangement of corona, in elaborate form, combined with the murex. Cf. Myh. Vas. xxxl 298, p. 61. Whether the murex is paired or single cannot be told with certainty, but probably the former case is true. This combination is a common feature on vases of this class. 10. Cf. Myh. Vas. xl 72. Similar, but not as elaborate. 11. Half a dozen other fragments of almost identical clay and technique were found. Those fragments, however, had different motives, and it seems impossible that so many different themes should be combined on one vase. Hence no reconstruction was attempted. The arrangement of the motive here is clearly connected with the flower branch just discussed. 12. For the scroll ornament cf. Schliemann, Mykenae, p. 291, fig. 369, Jahrb. II. (1887), p. 55. 1 Athen. Mitt. VII. (1882), p. 245, pi. viii. ; cf. also the wall paintings from Thei-asia. Myk. Vas. xn. 73, 74. 86 THE VASES AND VASE FRAGMENTS Murex, or Purple Fish Patterns. * The murex pattern is generally arranged in series of single ones, or pairs, the tops touching the border of the rim of the vase. Furtwiingler and Loesehcke's statement,^ " das Ornament wird besonders haiifig an Bechern und Niipfen verwendet, doeh nur an solchen deren Innenseite ungefirnisst ist," is not correct, since on several of the Heraeum fragments, with glaze on the ulterior, was this ornament found. 13. Side of a large bowl, form 3fi/k. Vas. xxxi. 297, about 30 cm. in diameter. Height cannot be determined, but it was probably the same as the diameter. The clay is of a warm yellow, but with a tinge of red and yellow slip. Decoration changes from black to brown and dark i-ed. Murices in single series. Several other fragments of same vase were found. 14 a and b. Two fragments of a kylix, yellow clay. Murexes in single series, rosette under handle. Several other fragments of same vase were found. 15. Rim of bowl. Simple murex combined with another diagonally. Cf. Myk. Vas. vi. 30. 16. Murex laid diagonally. Dark red glaze on interior. 17. Murex arranged diagonally. Cf. Myk. Vas. viii. 47. Dotted circle inside head of murex. 18. Murex laid horizontally. Brilliant dark red glaze on interior. 19. Rim of large bowl. Dotted circle outside head of murex. The zigzag pattern at the side is curious and suggests the imitation inscriptions found on black-figured work. Cf. Myk. Vas. XV. 96, XX. 145, XXI. 154. 20. Murices in pairs, points curving outwards. Dotted circle between the pairs. This varia- tion where the points curve outwards, and the pair combined in single form, is extremely rare. Cf. Schliemann, Tiryns, pi. xxii. b, Myk. Vas. xxxi. 299. 21. Alternate murex and pistil. 22. Reddish clay, brown decoration. This ornament is pi'obably a development of the nmrex pattern. Linear and Textile Patterns. The conventionalization of the pictorial ornaments now leads to a steady employment of linear and textile patterns. Whether the latter are really drawn from those used in embroideries is impossible to say." 23. From a bowl of precisely the same form and decoration as Myk. Vas. xxxii. 306, save that here the lozenges of the upper row are double instead of triple in form. 24. Alternating larger and smaller arches. Cf. Myk. Vas. xxxii. 308. 25. The arches are bounded above and below by circles. 26. Small arches in series. 27. From a large bowl. Suggestion of a guilloche. Cf. 3Iyk. Vas. ill. 21. 28. The ornaments in field here consist of circles whose diameter is broken. Cf. Myk. Vas. XXXII. 313. Perhaps a representation of a mussel-shell open. Cf. B. C. H. 11. pi. xv. 10. 29. Similar, with the addition of a smaller circle inside the larger one. 30. Ornament resembling a reversed E surrounded by semicircle. Cf. Plate LV. 16 ; Myk. Vas. XXXII. 304, 307, 312. Perhaps the half of a mussel-shell ; cf. Myk. Vas. p. 61, No. 304. The principle in the following fragments seems to be the division of the vase into a series of metope-like squares by vertical lines, the vacant spaces being filled by natural ornaments conventionalized or else textile motives. 31. Simplest form ot design. Series of semicircles back to back separated by a series of lines, not quite vertical. 32. Similar decoration. Outer vertical lines bounded by series of loops, which is one of the commonest motives at this time. 1 Myk. Vas. p. 61, No. 297. a Cf. Myk. Vas. p. 62. THE MYCENAEAN STYLE: LINEAR AND TEXTILE PATTERNS 87 33. Semicircles almost contiguous. Between the vertical lines, series of parallel zigzajrs. Cf Myk. Vas. xxxiii. 317. f & fa 34. Double semicircles in upper right-hand corner with central dots. Cf. No. 29. 35. Vertical lines interlaced by diagonal lines. 36. Double semicircles, large and small. Herring-bone pattern between the vertical lines. Cf. Myh. Vas. xxxm. 318. 37. Three double semicircles bounded by mussel ornament in series. 38. Interior semicircle in loop form. Cf. Myk. Vas. xxxm. 324. 39. Dull brown glaze on interior. Square divided into four triangles by diagonals, with arcs. Cf. Myk. Vas. xxxiir. 316. 8 .r s , 40. The same ; double semicircles in each triangle. 41. Diamond in square, divided by double semicircles into three sections. Another smaller fragment from the same vase was found. Cf. Myk. Vas. i. 5, xv. 99, xxxiv. 345. PLATE LV. 1. Reddish clay with darker core, plastic moiddings on upper and lower edges. Form uncer- tain, but probably that of a three-handled pyxis with rounded base, as in Myk. Vas. XLIV. 34. Ornament a variation of the " Running Dog " pattern (see below) growing out of the loops already mentioned. Perhaps the ornament may have been intended for a guilloche. Cf. Myk. Vas. xxxiv. 347. 2. Similar clay, but from rim of a bowl. ^ ^^^^^^^H^Ti? ^Sr~^V\ X^^IP!^ i Perpendicular loop pattern, the ends car- ried out to a point (as in No. 1) with rosette in field. Fig. 15. Bowl. Form, Myk. Vas. xxxm. 328 : height, 0.104 m. ; diame- ter, 0.145 m. Light red clay, with warm "^fe ^ yellow slip, circular bands on interior. Sim- p j- ilar decoration on both sides, three perpen- dicular series of lines, the outer inclosing a herring-bone pattern, the central one zigzags with a dotted semicircle on each side. Cf. Myk. Vas. xxxm. 327, 330. Numerous other fragments of precisely this scheme of ornamentation were found. 3. Rim of bowl, reddish yellow clay, continuous zigzag between the perpendiculars and dotted rosettes. Good technique. 4. The ornament is probably a variation of the palm-tree (cf. Myk. Vas. xxxi. 296). I have been unable to find another instance exactly similar. Lozenge chain introduced between the perpendiculars. Another small fragment of the vase was also found. 5. Exactly the same clay and technique as Plate LIV. 11, but it is doubtful whether it belongs to the same vase. The herring-bone, as on Plate LIII. 34, is here introduced. 6. Herring-bone between the verticals ; one of the commonest Mycenaean motives. Cf. Myk. Vas. XXXIV. 336. 7. Lozenge chain between the verticals as on 4. 8. Continuous zigzag between the verticals as on 3. 9. The zigzags are here separated. Cf. Myk. Vas. xxxiv. 340. A common motive. 10. The same, broader lines, verticals in pairs. 11. The same, verticals in pairs, closer together. 12. The lines between the verticals are here treated as on 2. 13. Interlacing diagonal lines between the verticals. This form of decoration is especially characteristic of the shoulder decoration on pseud-amphoras, or three-handled vases. Cf. Mijk. Vas. XXII. 159. 88 THE VASP:S AND VASE FRAGMENTS 14. " Running Dog" and scale patterns. Cf. Mi/k: la.s. xix. 136: xx. 146. ^, 15. Red glaze on interior. Dotted scale pattern. Cf. Myk: Vas. x. 62 ; Schliemann, Mykenae, pi. xiii. 63 ; Dumont and Chaplain, Cer. pi. iii. 19. 16. Scale pattern with dotted circles. Similar in character to Myl: Vas. xi. 70. 17. The same, double lines to each scale. Cf. Schliemann, llryns, fig. 36. la Scales in branch form. Cf. 3IyL Vas. vi. 32 ; xi. 70. 19. Rim of bowl, another fragment from the same vase also found. Parallels inclosing cross lines meeting at an obtuse angle. 20. Lozenge chain laid horizontally. Cf. No. 7. 21. The same, chain bounded by an outer line on each side. 22. Similar to 21, the interior pattern a " Running Dog." Another fragment of the same vase also found. It is very probable that in 21 and 22 a guilloche was represented. Cf. Myk. Vas. xxxiv. 328 ; Schliemann, Tiryns, pi. xxvi. c. 23. " Running Dog " pattern. Cf . Myk. Van. xviii. 131. 24. Probably the same pattern. The ends, however, do not overlap. Cf. Myk. Vas. xv. 96. 25. " Running Dog " pattern, simplest form. 26. The same, double pattern arranged vertically. 27. Probably a development of 24. Cf . Myk. Vas. xv. 96 ; xxi. 154. It is well to consider here the line decoration, or linear principle, which runs all through Mycenaean decoration from the earliest times. The examples here cited (Plate LV.) ought strictly to fall in several of the classes already considered ; but as this linear prin- ciple has a pecuhar bearing on later periods, especially the Argive style, it is best to treat these fragments for a moment in one special class. Only those fragments have been selected which afford an example of fine line decoration ; fragments of large vases with broad lines as their only decoration were extremely common. Though some line decoration occurs on almost every vase of the Mycenaean epoch, it is only during this period, when Class III. flourishes, that we find the steady employ- ment of the fine as well as the heavy fine, used chiefly on vases of the pseud- amphora type. The shoulders of such vases are generally filled with various motives, but from the shoulder to the base the vase is ornamented with a series of parallel lines, heavy and fine. Such fragments were extremely common at the Heraeum. Their rela- tion to vases of the Argive style will be more fully discussed in chapter iv. 28. Lower part of funnel-shaped vase, of the so-called " lalysos " type. Cf . Myk. Vas. vii. 42 ; XI. 71. The opening runs the whole length of the vase. Fragments of such vases were fairly common at the Heraeum, and as far as could be seen the only form of decoration on them was linear. 29. Belly of pseud-amphora. 30. Base of pseud-amphora. 31. From shoulder of pseud-amphora. On shoulder traces of a flower pattern similar to Plate LIH. 28. The central lines are bordered by a loop pattern, in series. 32. The same. The spii-als are treated in a more naturalistic fashion, which, combined with the excellent technique, would seem to denote its standing between Classes II. 2 and III. 1. There still remain several kinds of vases to be mentioned, which, to judge by technique and style, are certainly not older than Class III. 2, and most probably belong to it. Vases of a reddish yellow clay, carefully polished, but without decoration. Frag- ments of these vases were numerous and filled several baskets. Some specimens were found in the Bee-Hive tomb (see p. 91). Only a few vases could be restored from the fragments, and represent mostly kyhxes and amphoras. THE MYCENAEAN STYLE: CLASS IV 89 Another series, which seems to belong to a separate class, was formed by fragments exactly similar to the one-handled cups from the Bee-Hive tomb. (Of. Fig. 32.) The clay is generally fine, red or yellow in color, with a brilliant glaze, the rims and feet being ornamented with bands. The body of the vase is covered with a decoration hard to describe ; it seems as if the color had been sprayed on the surface of the vase. Whether this is the same technique exhibited by the vases from Ahki {Myk. Vas. xvni. 119, 125), I cannot say ; it is not common, and thus far the only similar speci- mens I have been able to note are from the Acropolis at Athens and Aegina. Just what position they occupy in the Mycenaean style is doubtful, but the several points of technique, clay, glaze, and color would seem to assign them to Class III. in the first division rather than the second. Of the class of small, hand-made vases, mentioned in Myk. Vas. p. 63, the Heraeum furnished one intact (Fig. 16) and numerous fragments of others. This is a one-handled jug (height, 0.065 m.) of hght reddish clay, with yellow slip, the decoration varying be- tween black and red. Cf. Myk. Vas. xv. 101, 102, XVI. 108 ; Schliemann, Mycenae, p. 66, No. 27. Another vase belonging to Class III. is Fig. 17 ; height, 0.06 m. ; diameter, 0.051 m. Two legs and handle restored. Greenish clay, with pj^ ^g j-,g. 17. black bands on body and legs, dots on rim. The form of this vase is identical with Myk. Vas. xliv. 192, save that the legs do not turn up at the bottom. Cf. Jahrh. I. (1886), p. 134, fig. 3006. CLASS IV. The number of fragments belonging to Class IV. is very small. In only a few cases were more than three fragments of the same vase found, and even then the design could not always be reconstructed. The peculiar fact is that the fragments of this class which we possess do not show the wealth of ornamentation we should have expected. A very slight study of vases belonging to this class from other localities shows that it is thor- oughly decadent, with a wealth of ornamentation linear rather than pictorial in charac- ter. The Heraeum fragments shoAV this linear feature, and even .though the wealth of ornamentation be lacking, the similarity between them and other vases of the class, in decoration, clay, technique, etc., is strong enough to warrant our placmg them in Class IV. without question. The clay of all our fragments of this class is much coarser than in the previous class, and is generally greenish in tone. The color has lost the lustre which was prominent before, while the glaze is generally dull and never rises to the level of that of Class III. in point of brilliancy. Though the forms of the vases cannot always be determined, it is safe to say that bowls were the most common. It may also be stated that fragments of this class which belong to vases with large openings do not always have the interior glazed {Myk. Vas. p. 63), or, to put it differently, the absence of glaze on the backs of krge fragments of bowls is no proof that they are not connected with Class IV. In fact, among the Heraeum fragments of this class, glaze on the back was the exception rather than the rule. . As the number of fragments is so few, no attempt has been made to arrange them in no THE VASES AND VASK FRAGMENTS chronological order, nor, indeed, conld any siitisfactory classification of ornamentation be fonnd, since in an over-elaborate style like Class IV. it is a difficult tasE, when deaUng with a few scattered fragments, to evolve any definite scheme. In the main the order is based upon plates xxxv. and xxxvi. of the Mykenische Vasen. 33. Pattern similar to Plate LIV. 38. Alternate diamonds in the interspaces of the arches. Cf. 3Ii/k. Vas. xxxv. 351. Two other fragments of the same vase were found. 34. From side of bowl. Greenish clay. Cf. Myk. Vas. xxxv. 857. Fig. 18. Half of bowl. Form, Mi/k. Vas. xxxiii. 318 ; height of fragment, 0.069 m. ; clay light red, with warm yellow slip, red decoration fairly brilliant. Though half of the design is miss- ing, there can be no doubt that it was somewhat similar to Mi/k. Vas. xxxi. 296, with a lozenge of interlaced lines on either side. *'"• ^^' Throughout Class IV. the " Heraldic " scheme or " Wappenstil " plays a prominent part and is illustrated by several of our fragments. 35. Similar " Heraldic " scheme. Band on interior just below the rim. Cf. Myk. Vas. XLI. 424. 36. Similar arrangement. Continuation of design to a circle of interlacing lines. 37. Decoration much faded. Evidently a development of Myk. Vas. XXIX. 256 ; cf. Mi/k. Vas. xxxv. 357. 38. Possibly a development of Mi/k. Vas. xxxi. 296. Though dots are frequently employed as the groundwork of a design, their use in place of the line is very curious. A somewhat analogous use occurs in fragments from Tiryns. (Cf. below No. 46.) The " Heraldic " scheme suggests Myk. Vas. XXXV. 356. 39. Shoulder of pseud-amphora. Ellipse inclosing series of zigzags. 40. Perhaps similar arrangement to Myk. Vas. xxxvi. 376. 41. From a bowl. Band below rim on inside. Similar " Heraldic " scheme to Pig. 18. 42. From a large vessel. 43. Reddish clay, fine technique. Elliptical lines inclosing sei-ies of zigzags curving to a point. 44 a-d. Four fragments from same bowl. Greenish clay, with black decoration. The arrange- ment of the design is extremely difficult to restore. That a bird was intended seems fairly evident if a (the head) be compared with Myk. Vas. xxxvi. 364, and C (the wing) with xxxviii. 383. Perhaps like Schliemann, Tiryns, pi. x. 45. BIRDS, ANIMALS, AND HUMAN FIGURES. Though the use of figures, animal or human, is one of the later innovations of the Mycenaean style and a characteristic feature of Class IV., all vases which show this use do not necessarily belong to that class. Such figures seem to have been introduced during the latter period of Class III. Only half a dozen of such fragments were found at the Heraeum, but no satisfactory explanation of the lack of such fragments is forth- coming. All except the last (No. 47) belong to Class III. 45 a and b. Two fragments from the same vase. Brick red clay, with dark red decoration. Two birds, somewhat resembling a wild duck, are represented. Both fragments show an ex- tremely advanced style of technique and belong to Class III. 46. liear part of base of the well-known " Tiryns " type. The legs and body of the horse are filled in by dots as in No. 38. An identical use may be found in a fragment from Tiryns (Schlie- mann, Tiryns, pi. xxi. a). Fig. 19. Fragment of a large bowl of 34 cm. in diameter. Height of fragment, THE MYCENAEAN STYLE: THE BEE-HIVE TOMBS dl 0.125 m. ; length, 0.22 m. ; thickness of clay, O.OOG m. Reddish clay, with lighter core and pale reddish slip, dull red decoration. Upper part of two fishes represented. Class III. The arrangement of fishes in a symmetrical series with a flower branch between them is a natural characteristic of the third class. The use of crosses instead of dots as decoration for the body is quite akin to that on plate xi. 48 of Schlie- mann's Mycenae. A similar arrangement may be noted on a lebes from Cyprus in the British Museum. (^Excavations in Fig. 19, Cypncs, p. 35, fig. 66, No. 1038.) Cf. also Petrie, Kahun, Guroh, and Ilawara, pi. xxviii. 9. 47 Base of a bowl. Reddish yellow clay, with black decoration. Lower part of human figure with a tree or branch. This is the only Mycenaean fragment from the Heraeum on which the human figure occurred. Its treatment is somewhat curious, and savors more of those figures on Geo- metric vases. However, it seems doubtful whether any Geometric influence can be here recognized. Fig. 20 (Form, Myk. Vas. xliv. 63 ; drawing on Plate LXIV. 1.) One-handled jug ; height, 0.054 m. Found at the southeast of the Second Temple. Yellow clay and slip with decoration varying from black to red. On base rays, and figure zone on the body ; another figure zone on the shoulder, and stripes on the neck and rim. Vertical zigzag on the handle between two perpendicular lines. Intact. This vase is of special interest, since it shows the transi- tion between the Mycenaean and Argive styles. Clay and technique are unmistakably Mycenaean, but the general scheme of decoration savors more of the Argive style, since we have an animal zone and a very primitive example of a ray pattern around the base. On the shoulder are represented three animals, but so rudely drawn that it is impossible to identify them. They have large open mouths, three projections above, which may represent ears and horns, and a tail almost like an extra leg. division. Three animals precisely similar are represented on the main zone, inverted. This inversion gives an extremely curious effect, as the animals thus resemble boats with their sailors, the ray pattern serving as waves ; but it is plahi that such an effect is acci- dental. This inversion of the figures is paralleled by two skyphoi from the Heraeum (v. p. 151), and does not, as far as I know, occur on any Mycenaean vase. This vase has been assigned to the Mycenaean class purely on technical grounds, for it shows so perfectly the characteristics of both the Mycenaean and Argive styles that it is difficult to decide to which class it really belongs. It must be admitted that no animal on any Mycenaean vase resembles these figures in any way ; but, for that matter, no similar animal can be found on any Argive vase. Fig. 20. Each leg ends in a triple THE BEE-HIVE TOMBS NEAR THE HERAEUM. Two tombs were discovered on the 20th and 23d of April, 1894. Both have been described in the previous volume of this publication, and consequently there is no need of devoting any time to the consideration of their architectural features. The contents of these tombs was as follows : — 92 THE VASES AND VASE FRAGMENTS TOMB I. 49 vases, 32 entirely or nearly whole, and 17 in fragments. TOMB II. Fi-agments of vases. 4 steatite whorls. 1 red agate bead. 1 bronze ring. 3 terra-cotta figurines. 1 terra-cotta chair. 1 island stone. 4 steatite whorls. I ivory needle. Glass beads. The few vase fragments from the second tomb were mostly undecorated, and pei*- mitted no reconstruction. A few fragments of bones were also found ; but in the main the contents of the tomb are not worth a detailed consideration. In the first tomb the results were more gratifying. Three layers were found, the upper containing bones and four vases (mostly cups), at a depth of 2.79 m. ; the second, which contained a skull, some calcined bones, and the majority of the vases and frag- ments, at a depth of 3.28 m. Ten cm. below the second came the last layer, which, besides a few fragments of bones and a cup (No. 18), contained a small separate grave near the entrance of the dromos (length, 1.26 m, ; width, 0.4:5 m. ; depth, 0.66 m.). Through some accident during the transportation to Athens, several of the labels were lost, and it is now impossible, in the case of some of the vases or fragments, to decide from which layer they came. The classes represented were II. 2 and III. 2. No specimens which showed the advanced technique of III. 1 were found, nor were there any rough pieces which could belong to II. 1. At the same time there appears to have been no distinction in the matter of layers, since vases of both classes were scattered about in each layer. Certain it is, however, that all these vases are products of the same period. The shapes represented were as follows : — II jugs with handle. 9 cups with handle. 2 jugs with handle and spout. 4 kylixes. 9 three-handled vases. 1 pseud-amphora. 6 amphoras. 4 vases, represented by fragments, but not 3 cups. permitting a complete restoration. Jugs with Handle. 1. Height, 0.207 m. Form, Mijl: Vas. xliv. 63. Coarse reddish clay, without any decoration. Base of handle pierced by a hole. Intact ; first layer. Two more jugs (height, 0.235 m. and 0.22 m.) were found in the second layer, of precisely simi- lar form and technique. Both had a small hole at the base of the handle. 2. Height, 0.155 m. Similar shape to 1, but with handle a lit- tle more flaring. Reddish clay, with a highly polished yellow slip, without any decoration. Half of another vase of same form and technique was found. 3. (Fig. 21.) Heigiit, 0.103 m. Same form as 1. Red clay, with herring-bone pattern in a series of four on the shoulder in red brown. Fig. 21. Fig. 22. THE MYCENAEAN STYLE: THE BEE-fHVE TOMB 93 One-handled vase with Another jug precisely similar, but with spirals on the shoulder and plastic boss under the handle was also found. 4. (Fig. 29.) Height, 0.078 m. Handle missing. Form similar to 1, but rather more rounded and squat in appearance. Reddish clay, with black decoration laid on very thickly and cracked in places. Bands and continuous spiral chain on the shoulder, with dots in the field. Another similar jug with leaf pattern on shoulder was found. 5. (Fig. 23 a and 6.) Height, 0.074 m. ; diameter of cover, 0.075 m. cover. Form- similar to Myk. Vas. XLiv. 115, except that here the foot is lacking. Coarse red clay, with dark core pierced by two rows of holes, with a single row in the cover. The use of this peculiar ves- sel has never been satisfactorily ex- plained ; similar examples have been found at Troy in the fourth city (Schliemann, llios, p. 557, fig. 11, 96), and at lalysos. {Mt/k. Vas. ii. 15.) According to Furtwiingler and Loeschcke these ves- sels were used to burn incense or some sweet-smelling powder for sanitary reasons, and this expla- nation seems very plausible, especially as we find the cover also pierced with holes. Fig. 23. 6. (Fig. 24.) Height, 0.17 m. Amphoras. Form, Myk. Vas. XLIV. 44, except for a slight variation in the foot. Coarse brick red clay without decoration. The exterior r-^fJI^^XX has been entirely covered with a white wash, almost entirely ^ \ worn away. I The principal feature of this amphora lies in the white wash. ' ] This, as is shown by many of our fragments, was a very common feature, and half a dozen baskets are filled with fragments showing it. Generally it seems to be employed on vases of coarse clay, large pithoi or small wheel-made saucers. While this technique is employed very often on Mycenaean terra-cotta figurines and later vases (e. g. the white lekythoi), it is ex- tremely rare on archaic vases (two in Berlin, 1309, 1629), and I know of no other instance in the Mycenaean style. From its frequency in our fragments, it may perhaps be considered as a local fashion. Another small amphora (height, 0.09 m.) similar in shape, of yellowish clay with a reddish tinge, but without the white wash, was found in the second layer. 7. (Fig. 25.) Height, 0.105 m. Light red clay, with yellow slip. Faded black glaze over all, inside and out^ side (except base). This vase differs from the ordinary Mycenaean am- phora in that the handles have no connection with the mouth of the vase. The absence of all decoration and the presence of the glaze proves that a large part of the glazed fragments which the Heraeum furnished must be placed in the Mycenaean period. a Height, 0.07 m. ; form, Myk. Vas. XLiv. 73. Yellow clay, with red decoration very much faded. Second layer. This vase was identical with Myk. Vas. x. 64, xv. 99, except that around the shoulder ran a series of Fig. 24. rnaments similar to Plate LV. 24. Fig. 25. 94 THE VASES AND VASE FRAGMENTS Three-handled Vases. j^ Three varieties were found : (a) form, jMijk. Vas. xliv. 25, tall with wide opening ; (b) form, Mt/k. Fas. xliv. 32 ; (c) form, 3/>/k. Vas. xliv. 33. 9. (Fig. 26.) Height, 0.213 m. Reddish clay, with bhick decoration turning to dark red. One handle missing. Three pairs of spirals on shoulder, sejiarated by the handles and bands, heavy and fine. Second layer. The form of this vase does not quite correspond to Myk. Vas. XLIV. 25, as the mouth is a little wider and the foot less pronounced; it lies between forms 25 and 27. When found it was completely full of ashes, which are probably human. The upper part of a similar vase, but much smaller, was also found. Shoulder decoration, interlaced lines. b. 10. (Fig. 27.) Height, 0.098 m. ; diameter, 0.172 m. ; diameter of opening, 0.098 m. On shoulder flower pat- base, the upper joined Fig. 26. Fig. tern similar to Plate LIT. 6, with dotted rosettes in field. Bands on with a wave pattern and series of radiating lines on base. The decoration varies from black to red brown. 11. (Fig. 28.) Height, 0.05 m. ; diameter, 0.185 m. ; diam- eter of opening, 0.085 m. On shoulder three fishes. Usual decoration of bands in dark brown. First layer. Four more vases of this type were found varying in height from 38 to 84 cm. The decoration was the same in all, a wave pattern on the shoulder similar to Myk. Vas. viii. 45. Fig. 28. C. 12. Height, 0.06 m. ; diameter, 0.109 in. ; diameter of opening, 0.07 m. Dark and light red decoration, identical with Mi/k. Vas. xxii. 159, save that ^Hflfif-. the stripes on the body run horizontally, not vertically. gr - i^-Ja*'*- '^ Interlaced lines on shoulder. %. ^.■-^•■J&^: .:-.,, ^^^ 13. (Fig. 29.) Height, 0.065 m. ; diameter, 0.10 m. ; diameter of opening, 0.07 m. Brilliant black decoration. On shoulder double line loop pattern, bands on body in red and on base spirals in light brown. The form of 13 is a cross between Myk. Vas. xliv. 32 and 33, being higher in the body than 32, but lower Fig. 29. than 33. Teapot-shaped Jugs. 14. Height, 0.13 m. Form, Myk. Vas. xliv. 68. Bands on body and handle and series of spirals on shoulder precisely similar to Myk. Vas. ix. 54. 15. Height, 0.12 m. Form, practically the same as Myk. Vas. xliv. 15, except that the foot is flat. Entire vase covered with a dark red glaze except under handle and on the bottom. THE MYCENAEAN STYLE: THE BEE-HIVE TOMB 95 Pseud-amphora. 16. Height, 0.115 m. Form, Myk. Vas. XLiv. 50. Ordinary decoration of heavy and fine bands similar to Plate LV. 29, in red and brown. First layer. Kylixes. Four of these were found, of red or yellow clay, without any decoration. Two were of the same form as Myk: Vas. XLiv. 83 ; one, the same as Myk. Vas. xuv. 82 ; while the fourth probably corresponded to form 85, but as one side was completely broken away, it cannot be told whether the vase possessed one or two handles. They were all from 10 to 12 cm. in height and from 11 to 15 cm. in diameter. One (form 82) was elliptical in shape, having a diameter across the handles of 13 cm. ; at right angles to the handles, 14.| cm. Cups. Two varieties of cups were noted (a) without handle ; (b) with handle. Fig. 30. 17. (Fig. 30.) Height, 0.027 m. ; diameter, 0.07 m. Brilliant red stripes around body, with another stripe on the rim inside. Two others of similar shape were found, one a little more rounded at the base, measuring 43 and 44 mm. in height, 9 and 11 cm. in diameter, with no decoration whatever. Two varieties of b were noted : (1) Those with a handle curving upwards, above the rim, and (2) those with a handle curving downwards, below the rim. 18. Height, 0.06 m. ; with handle, 0.14 m. ; diameter, 0.153 m. Form, Myh. Vas. xliv. 102. Entire vase, ex- cept base and under the handle, is covered with a brilliant red glaze. 19. (Fig. 31.) Height, 0.06 m. ; diameter, 0.104 m. Red clay, red glaze on interior, outside plain. Three other cups of similar shape but with the rim less emphasized, were found, two undecorated, one ornamented with stripes. It is probable that 18 and 19 are direct imitations of metal work.^ Fig. 31. 20. (Fig. 32.) Height, 0.045 m. ; diameter, 0.113 m. Form, Myh. Vas. xliv. 87. Red clay, with light red stippling. This cup is a good instance of the technique men- tioned on p. 89. Three more cups were found, all of the same shape, one showing the same technique, one with a row of ornaments on the side similar to those on No. 8, and the third plain. The terra-cotta figurines and the chair are described in the chapter on Terra-Cotta Figurines (p. 42). 1 Cf. Perrot and Chipiez, Hl^t. VI. p. 965, fig. 533. Fig. 32. 96 THE VASES AND VASE FRAGMENTS MISCKLLANEOUS VASES. Of the vases found intact, or partially so, mentioned in the Introduction, by far the greater part consisted of small, hastily made vases, with or without decoration, the exact date of which is extremely doubtful. The quality of the clay is distinctly inferior to that of the Mycenaean vases, and in many respects corresponds to that used in vases of the Geometric style, so that the majority of them may well belong to that period. A comparison of the vases without decoration with the decorated vases is the only means for esttiblishing a date for the former, and in the case of the latter the following data are available: (1) The most primitive examples (the saucers) show a use of " didl " color, which, together with their rather hasty technique, may perhaps assign them to the earlier periods of the Mycenaean style. (2) One small three-handled jug, which, from its pohshed surface, belongs more properly to the class discussed on p. 99, had a linear band of '"dull" color. (3) Practically all the jugs, amphorae, and bowls were made of a coarse clay, mox*e resembling that used in the Geometric vases, and if they are to be included in the Mycenaean style at all, would seem to belong to the very latest period. These, I shall endeavor to show later, constitute the real link between the Mycenaean and Argive styles. At the same time, I do not regard their sequence from primitive times as existing without a break, and for that reason I do not consider them as examples of the "Argive Linear" style which Professor Waldstein would consider them to be. It cannot, however, be assumed that their influence on any particular style was marked ; they are simply the examples of the low level to which the potter's trade could sink, and from their large number it is reasonable to suppose they were the cheapest vases which the devotee at the shrine could buy. Being rough and small, they were not so easily broken, and were thus in far better preservation than larger vases of better technique. To assign an exact date to them is impossible, nor can they be arranged in any exact chronological order, except where the development of the shape can be readily seen. Three general types may be distinguished : — Saucers. Bowls (with or without handles). Jugs (with one, two, or three handles). Sance7's. Three distinct varieties may be noted : — a. Plain, hand or wheel made. b. With central boss. c. With handles. a. 1. The most primitive examples seem to be a series of small hand-made saucers, of grayish or reddish clay, pierced by a hole near the rim, and varying from 5 to 7 cm. in diameter, and 2 to 3 cm. in height. Several hundred were found in- tact or in fragments. The ornamentation is of the simplest character, consisting of two lines crossing each other at right angles. Whether they belong to the dull or lustrous Mycenaean vases is difficult to decide, as the decoration is extremely poor, Fig. .32 a. ^^^ ^hey are more prf)bably examples of the " dull " technique. Diameter, 0.05 m. Variety is obtained by increasing the number of cross-lines, THE MYCENAEAN STYLE: SAUCERS 97 wliith vary from two to eight. As a rule the ornamentation is confined to the interior, but cases occur with the same design on the exterior as well. The earliest of the series is a saucer of red clay 5 cm. in diameter, with ribbed edges and two incised cross-lines on the interior. This probably falls in the primitive period. Similar saucers were found in large numbers at Hissarlik, Mycenae, and Tiryns. In some respects they resemble saucers from Bos-ojiik (Koerte, Athen. Mitt. 1899, pi. iii, 7) and Syra ('E<^t,/x. 'Apx- 1899, pi. ix. 25). 2. As a development of these we have a series of wheel-made saucers, of reddish clay (e. g. Fig. 32 b), from 4 to 7 cm. in diameter, of which about fifty were found. The inner edge of the rim is either sharply defined or rounded, and none have any decoration. The clay is of a slightly coarser variety than that used in the preceding examples. 1. The earliest examples are hand-made, of a reddish clay, without decoration, having a small boss in the middle ; they resemble the earliest ex- amples of class a, save for their lack of decoration and the addition of the boss. 2. The wheel-made type (Fig. 32 c) corresponds in character of clay (wheel-made) and lack of decoration to a, 2. The principal variations lie in the height of the boss, which in some cases rises above the edge of the saucer. As in a, 2, the edges of the saucer are either rounded or sharp. A few examples were found with a series of parallel dashes of pauit (dull) on the rim. Fig. 32 b. Diameter, O.CKi'i m. Height, 0.013 m. Fig. 3l' c. Diaraeter, O.Otil m. Height, O.OKi ni. Height of boss, O.Oai m. Two varieties may be noted with one handle, and one with two handles. The simplest form of the one-handled saucer is hand-made, the edges rolled over so as to form a curve down to the bottom of the vase, and the handle, of the simplest form, parallel with the sides. (Fig. 32 d.) This type was extremely common. A varia- tion is introduced occasionally, where a plas- tic dab of clay is placed on each side of the handle along the rim, similar to those on the shoulders of our terra-cotta figurines (Fig. 32 e) ; this saucer was entirely covered with a brown glaze, worn away in places. A few specimens were noted -ndth one handle, and three plastic dabs of clay, arranged so as to form a symmetrical scheme on the four sides of the vessel. While the majority of these saucers are un decorated, some few are covered with a didl dark glaze, badly worn. All the saucers with two handles show these plastic projections on each side ot the Fig. 32 d. Diameter, 0.04 m. Height, 0.012 m. Fig. 32 e. Diameter, 0.044 m. Height, 0.015 m. 98 THE VASES AND VASE FRAGMENTS haiulle. Tlie earliest example is haiul-macle, of precisely similar clay and technique to those first mentioned under a, with four cross-lines on the interior. The Afteel-made types show a decided advance in execution. They are almost flat (a trifle over a centi- metre in height, and from six to seven centimetres in diameter), the edges well defined, and they are made of a dark red clay without decoration. About a dozen specimens were found. As the quality of the clay resembles more closely the Geometric clay, and as they are exactly paralleled in shape by several of our bronze saucers, it is certain that they must fall in a rather later period, perhaps about the eighth century. Bowls. The earliest examples of bowls seem to be contemporary with the most primitive siiucers mentioned above ; they are hand-made, of very rude technique, the sides raised a trifle. In some cases a small dab of clay is added to the rim to represent a handle. None have any decoration. The total number of bowls found was between one and two hundred ; of these a few are hand-made, but the greater part are wheel-made. Slight variations in the forms may be detected, but hardly enough to warrant a regular classification. The hand-made bowls are either plain or decorated, rounded at the base or else provided with a flat base. None have handles. The decoration invariably consists of one, two, or three narrow bands encircling the bowl just below the shoidder. A dozen examples were found where the vase had been covered with a white paint (as in No. 6 of the Bee-Hive tomb : Fig. 24). On one example a straight or a wavy band had been jjainted in red ; this was added after the final firing as in the case of terra-cottas, and is easily friable. None of these bowls exceeded 3 cm. in height or 5 cm. in diameter. Something over a hundred wheel-made bowls were found, the majority of a coarse reddish clay, similar to that of the wheel-made saucers. Only a few examples bore any decoration. The form is generally the same with the following variations. The rim is either flat, slightly curved, or sharply emphasized as in the saucers ; the profile sometimes flares sharply outward and downward so that the diameter is greater at the base than at the opening. As a rule the profile curves inward towards the base, which is flat. The usual form of decoration consists of one or two broad or fine wavy bands on the shoulder. The following is the best example : — Fig. 33. Height, 0.024 m. ; diameter, 0.047 m. Dark clay, seemingly burnt. On shoulder dark brown wavy band with yellow dots. Similar dotted band on rim. This is probably one of the latest of the series, as the clay is more like that of the Geometric vases in texture and the technique more advanced. It is probably one of the earliest instances of the Geometric snake. The use of yellow dots seems to be a peculiarity of the Argolid and will be discussed later. A few bowls were found, the rims ornamented by three plastic heads of animals, probably calves. Fig. 34. Height, 0.026 m. ; diameter, 0.043 m. Red clay. Further another class may be mentioned of a technique slightly more advanced, where the bowl is provided with a foot and two large handles rising above the rim, similar in form to Iluk. Vas. xliv. 16, Fig. 34. but without any decoration. THE MYCENAEAN STYLE: JUGS 99 Jwjs. These numbered over five hundred and formed the majority of all the Heraeuiu vases intact or partially so. The greater part of them are wheel-made. As usual two classes of these were observed, those with decoration and those without. The