EXPLORATIONS IN THE PISCO VALLEY MAX UHLE'S IREORTS TO PHOEBE APPERSON HEARST AUGUST 1 TO JANUARY 1902 EDITED AND ANNOTATED BY JEAN-PIERRE PROTZEN AND DAVID HARRIS NO. 63 CONTRIBUTIONS OF THE UNIVERSITY OF CALIFORNIA ARCHAEOLOGICAL RESEARCH FACILITY BERKELEY ISBN 1-882744-17-9 EXPLORATIONS IN THE PISCO VALLEY MAX UHLE'S REPORTS TO PHOEBE APPERSON HEARST AUGUST 1901 TO JANUARY 1902 Edited and Annotated by JEAN-PIERRE PROTZEN AND DAVID HARRIS Number 63 Contributions of the University of California Archaeological Research Facility Berkeley COVER PHOTOS Front: Tambo Colorado, North palace from SE Back: Huaytara, S fagade of church, niches 4 and 5 (counting from east) Both were taken by Max Uhle in 1901. Library of Congress Control No. 2005938874 ISBN 1 -882744- 17-9 ? 2005 by the Regents of the University of California Archaeological Research Facility University of California at Berkeley Printed in the United States of America. All rights reserved. Preface .........................v Introduction ........................ vii Letter of August 23, 1901 ......... ................ 1 Letter of September 11 & 24, 1901 .........................3 Letter of September 30, 1901 ........................ 55 Letter of October 9, 1901 ........................ 56 Letter of January 1902 ........................ 60 Illustrations ........................ 89 Acknowledgements ........................ 90 Table of plates ........................ 91 Plates ........................ 94 Plans ........................ 123 PREFACE My interest in Max Uhle's work in the Pisco Valley in southern Peru is concomitant with my interest in the Inca ruins of Tambo Colorado in this same valley. After many years of studying Inca architecture in the highlands around Cuzco, and at Ollantaytambo in particular, I thought of investigating how the Incas, in their conquest of the coastal regions of Peru, adapted the distinct architecture of cut stone they had developed in their homeland to the local materials and construction practices. Tambo Colorado, undoubtedly the best- preserved Inca architecture on the coast, seemed a most appropriate object for such a study. I was aware that Uhle had mapped the ruins of Tambo Colorado and photographed them sometime in 1901. Therefore, before starting my own work there, I consulted his plans and his notebooks at the Ibero-Amerikanisches Institut Preussischer Kulturbesitz in Berlin.' Subsequently, I also read Uhle's reports from the field and inspected such photographs as are preserved at the Phoebe A. Hearst Museum of Anthropology at the University of California at Berkeley. Because the site has suffered noticeable deterioration and damage in the intervening century, Uhle's plans, photographs and notes are of particular importance to anyone investigating Tambo Colorado. Humidity, windblown sand, vandals, the flow of tourists, and a recent road construction have taken and continue to take their toll on this fragile site. Even more important than the historic value of Uhle's documentation is the superior quality of his work. His plans are of an astonishing accuracy and his observations and reflections most thoughtful and astute. In the course of studying the Uhle materials on the Pisco Valley, it occurred to me that these materials would neatly complement Uhle's work in the adjoining valley of Chincha and should, therefore, be made public. For this purpose, I transcribed the five letters Uhle wrote to his sponsor, Phoebe Apperson Hearst, in which he reported on his work in the Pisco Valley. Subsequently, Dr. David Harris and I edited these letters. The fact that Harris was unfamiliar, both with the sites discussed by Uhle and Andean prehistory was, I believe, most fruitful for our collaboration. If he did not understand what Uhle tried to communicate, nor would the reader who might be equally unfamiliar 1 These pians, originally published in Max Uhie (1856-1 944), Plane archaologischer Staaten im Andengebiet, AVA-Materialien 56, W. W. Wurster, ed. Verlag Phillip von Zabemn, Mainz am Rhein, are reprinted at the end of this volume. with the sites. Whereas from my own knowledge of the sites I might have been tempted to reinterpret Uhle, Harris kept us honest and as close to Uhle's text as comprehension would warrant. I express my gratitude to both the Ibero-Amerikanisches Institut Preussischer Kulturbesitz (IAI-PK) in Berlin and the Phoebe A. Hearst Museum of Anthropology (PAHMA) at Berkeley for allowing me to peruse their collections, and for having facilitated my access to their respective archives. Furthermore, I acknowledge the IAI-PK for granting me permission to reproduce the plans accompanying this publication, without which the text would be, in places, hard to follow. I also acknowledge the PAHMA for granting permission to reproduce most of the photographs Uhle mentions in his letters. I would like to thank personally Dr. Peter Masson, Dr. Gregor Wolff and Norbert Knossola for helping me navigate the archives in Berlin, and Dr. Douglas Sharon, Joan Knudson, and Therese Babineau who have done the same here in Berkeley. And last, but not least I express my appreciation of the Archaeological Research Facility for undertaking this publication and thank its editor Tanya Smith, for her support and professional advice. JPP Berkeley, March 2005 vi INTRODUCTION UHLE'S INTENTIONS Max Uhle's investigations in the Pisco Valley are related to his earlier work in the adjoining valleys of Ica to the south and Chincha to the north. Uhle had a grand plan that he described as follows in his introduction to the manuscript "Das Thal von Chincha."i It was the author's plan-beginning with the exploration of the valley of Chincha-to make a broad cross-section through the regions of southern Peru cultivated in ancient times. In this, the valleys of Pisco and Ica, which have never been studied before, were to be touched first. . . . Next, progressing towards Cuzco, the interesting regions of Lucana and the department of Ayacucho, the former domiciles of the Chancas made famous through ancient history, were to be touched. And further, one could have hoped to find on the way a key to the historic culture of the Incas and the style of Cuzco through the study of its neighboring cultures. The rather extensively conceived program could only be realized in its first part regarding the valleys of Chincha, Pisco and Ica. After one year of work on this [the first part] from September 1900 to October 1901, he [the author] was reached by a recall to California to give a preliminary account of his accomplishments to date.ii Uhle meant to report the results of his work in three publications: "Das Chincha Thal," "Das Thal von Ica," and "Die Inca Ruinen des Thales von Pisco."iii None of the three ever made it into print. Uhle's work in Chincha was later published by A. L. Kroeber under the title "Explorations at Chincha,"iv based on a report Uhle wrote to Mrs. Hearst on August 30th 1901. However, this publication is incomplete, for as Kroeber wrote in his Editor's Note: "[a] section of some fifteen manuscript pages on the Inca Palace at La Centinela has been greatly condensed, ... because its details are unintelligible without the map prepared by Dr. Uhle." Uhle had taken the plan with him to Peru and later to Germany. It is to be hoped that the above-mentioned manuscript "Das Thal von Chincha" will be published some day in its entirety. Regarding Uhle's work in the Pisco Valley there exists a manuscript entitled "Inka-Ruinen des Thales von Pisco"v housed in the Ibero-Amerikanisches Institut Preussischer Kulturbesitz in Berlin that treats, among other things, the lower Pisco Valley and the ruins of Tambo Colorado. Another manuscript "Huaitara," also in Berlin, is dedicated to the upper Pisco Valley,vi the church at Huaytarai, and the ruins of Inkawasi. Underlying these manuscripts are five letters Uhle wrote from the field to his sponsor at the time, Phoebe Apperson Hearst, which are preserved at the Phoebe A. Hearst Museum of vii Anthropology of the University of California at Berkeley. It is these letters that are the subject of the present publication. To avoid any possible confusion with the German manuscripts we chose the title "Explorations in the Pisco Valley"vii in analogy to Kroeber's "Explorations at Chincha" for this publication. Judging from Uhle's original plan, it seems that he intended to report on his work on the ruins in the Pisco Valley in only one publication, viz. "Die Inca-Ruinen des Thales von Pisco." Whether this publication was to include Huaytara or whether Uhle always conceived of "Huaitara" as a separate publication is not clear. The fact is that Huaytara was on his mind already in January of 1901 while in Ica, but that he thought it unadvisable to visit the site at this time because of the rains in the mountains.viii Later in the year, on July 27, he wrote to Phoebe A. Hearst, again from Ica "I would prefer to climb up ... to Huaitara, which seems to be a very curious place by some ancient ruins .... Huaitara is also famous by curious fables of ancient migration ..." He definitively added the site of Huaytara' to his investigations after he grasped its importance while working at Tambo Colorado. Watching the heavy traffic of people and animals passing daily through Tambo Colorado along the ancient Inca road, a fact "I must confess ... before unknown to me," led Uhle to the conclusion "I see now that Huaitara is a natural mark for nearly all immigration to the southern Highlands of Peru in the direction of Ayacucho ..." Furthermore, Uhle observed "as some monuments of architecture of stone which exist at Huaitara are of some renown here all about, ... a visit made to them will appear to be of some inportance.'x Uhle added the site of Inkawasi to his investigations after he heard of it at Huaytara'. TIMELINE OF UHLE'S WORK IN PISCO VALLEY Uhle carried out his explorations of the valley of Pisco from August 22 to October 23 of 1901. In this time, he covered the entire valley, from the mouth of the River Pisco-then called Chunchanga-to Inkawasi, an Inca ruin at an elevation of ca. 3800 meters and some 1 15 kilometers from the sea. The timetable of Uhle's activities and movements is given below:x 22 August travels from Ica to the hacienda Pallasca in the Pisco Valley on whose lands stand the ruins of Tambo Colorado, 23 August to 1 1 September spends his time investigating and mapping Tambo Colorado and its surroundings, viii 12 September goes to Pisco, 13 September travels by train to Ica, 14 September returns to Pisco and goes on to Lurin Chincha, 14 to 21 September explores huacas in the southern Chincha Valley and returns to Pallasca, 22 September to 6 October continues work at Tambo Colorado, 7 to 8 October travels to Huaytara, 9 to 19 October investigates the Inca ruins of Huaytara 20 to ? October investigates the Inca ruins of Inkawasi and visits the Puna beyond, 23 to 26 October travels from Huaytara, via Tombillos to Ica, 29 to 31 October returns from Ica to Lima, 7 November embarks on the SS. "Chile" in Lima's port of Callao, 3 December arrives in San Francisco. SUMMARY OF UHLE'S LETTERS FROM THE FIELD Uhle wrote the first four of the five letters to Phoebe A. Hearst from the field; the last one he composed shortly after his return to San Francisco. The original hand-written letters and notes are found in the ORIGINAL CATALOGUES, M. UHLE, Volume IV, pp. 3l193,xi and Volume V, pp. 1-17. Letter of August 23, 1901, written from Pallasca, 4 pages. In this letter Uhle announces his arrival at the Hacienda Pallasca in the Pisco Valley from Ica and relates his first impressions of the ruins of Tambo Colorado, which he found to be "pure Incasic," "wonderfully preserved," and embedded in a rich and very pleasant environment. Letter of September 1 1, cont'd on Sept 24, 1901, written from Pallasca, 101 pages. At the core of this letter are Uhle's description of, and his observations on Tambo Colorado and its immediate lX surroundings. The last few pages are dedicated to the description of the agricultural fields, ruins and burials opposite Tambo Colorado on the other side of the river, to the account of the old Inca road down the valley from Tambo Colorado and the sites along it, such as Lima la Vieja, and to Uhle's visit to the southern part of the Chincha Valley. There he visited the Huaca Centinela near Lurin Chincha,xii another, larger huaca about 1 km from the former,xiii and the Huaca Santa Rosa. On his return to Pallasca, Uhle also came by the ruins of Chongos along the Pisco River. Letter of September 30, 1901, written from Pallasca, 2 pages. Letter of transmittal of the above letter. Letter of October 9, 1901, written from Huaytara, 7 pages. Here Uhle describes the upper Pisco Valley and the ancient Inca road from Tambo Colorado to Huaytara'. The church of Huaytara' built over the remains of an Inca building attracts his attention for the high quality of the Inca stonework and because these ruins were hitherto "scientifically unknown," that is' they have not "been presented at any time in any scientifical work." It is at Huaytara' that Uhle learned about the ruins of Inkawasi (Inga-huasi) and other ruins, which he promised to investigate. Based on some informant's notice, he also expressed the "hope that perhaps another of the still unknown ancient peruvian languages may be discovered by me at a distance of about 22 leagues from here."' Letter of January 1902, written from San Francisco, 61 pages. The letter starts with an elaboration on the previous letter by discussing the Inca road and its importance in Inca times, the environment of the upper valley, and petroglyphs Uhle had noticed on his way to Huaytara. Subsequently, Uhle discusses the layout of Huaytara' and its immediate environment, the ayllu structure -or social organization -of the place and the customs of its inhabitants before moving on to the description of, and reflections on the church of Huaytara' built over the remains of an Inca building and its immediate surroundings. The last fifteen or so pages are dedicated to a report on an ancient settlement on top of mountain Runiayoc near Huaytarai and the description of the ruins of Inkawasi. x NOTES ON UHLE'S APPROACH In one of the lettersxv Uhle emphasized that he will not be bound to the exploration of a single site alone -here Tambo Colorado, but that he will want to look at the context of the whole historical development of human habitation in the larger region of that site here the entire Pisco Valley. He argued that "[o]ne period deserves so much attention from our part as the other, and we never will learn anything about historical development in Peru,- of interest for all the history of civilised mankind too, -if we are not willing to comprise all periods in our study." Uhle's approach to archaeology was indeed a very broad one. Reading his reports it becomes obvious that he was not only preoccupied with ancient architecture, burials and artefacts, but that geomorphology, works of infrastructure- terraces, irrigation systems, and road networks-toponymy, linguistics, and ethnography were equally important factors in his considerations. NOTES ON THE GERMAN MANUSCRIPTS "INCA RUINEN DES THALES VON PISCO" AND "HUAITARA " The first part of the manuscript "Inca-Ruinen des Thales von Pisco" (pp. 1-122)Xv does not concern the Pisco Valley at all but deals with Uhle's explorations in Paracas, Ocucaje, and the valley of Ica. It is only in the secona half of the manuscript (pp. 123-199) that Uhle turns to the Pisco Valley and the ruins of Tambo Colorado in particular. Here the manuscript follows very closely the contents of the August 23 and September 9 letters. In its descriptions of, and observations on, Tambo Colorado the German text appears, by and large, as a literal translation of the corresponding English letters. The German "Huaitara" manuscript is also an almost verbatim translation of Uhle's January 1902 letter. The only part deleted from the letter is the introduction and at the end the conclusion Uhle drew from his observations of the Inca road with its "tambos" that "something has been also done for improving our understanding of the governmental machine of the interesting Incasic empire." Also not pesent- at least in the transcript of this manuscript- are the sketches Uble interspersed in his letter to Phoebe Hearst. It appears that Uhle wrote the German manuscripts in Berkeley during the years 1902 and 1903 while sorting out his collections.xvi The fact that Uhle wrote these manuscripts shortly after his return from Peru may explain why the German text so closely follows the letters to Phoebe Hearst. He simply used his notes, as well as the letters he wrote to Phoebe A. Hearst, as the basis for the German manuscripts. It was the understanding that Uhle's American born wife Charlotte, -would later translate the manuscripts. "[S]he completed the translation xi of only two of the seven reports, and the manuscripts were never sent back to California ..." xvii NOTES ON EDITING UHLE Max Uhle was a native speaker of German, but the letters presented here were written in English to his American sponsor, Phoebe Apperson Hearst. One wonders what she must have made of them. The letters, especially the two long letters, are filled with difficult passages and, even with the insight of Protzen, who has visited these sites, it was no easy task to edit these manuscripts into a readable and comprehensible form. The task was not precisely one of translation; it was more one of attempting to understand and reframe Uhle's intentions. No editor can be certain of the intentions of an author, but we know Uhle's basic intention was to report on Peruvian architecture, history, and culture, and, in the letters presented here, he is largely describing specific sites, providing the editor with an opportunity to check Uhle's often-confusing words against the actual site or documentation of the actual site. Ideally, Uhle's words would have been presented without editorial intervention -but Uhle's original is heavy going most of the time and virtually incomprehensible at times. Therefore, we attempted to clean up his language as much as possible by clarifying and simplifying constructions, and correcting grammatical errors and the errors in idiom that are common in authors writing in a second language. Many problems are obvious and easily repaired. For example, Uhle never pluralized "meter" and always wrote "circonstance" instead of "circumstance." Adding an "s" to the end of "meter" when appropriate and correcting the spelling of "circumstance" only remove unnecessary distractions. Some unidiomatic phrases were altered for clarity. For example, Uhle used the word "confining" where an English speaker would typically use "bordering" and "adjacent." Though the roots of the word do provide the correct meaning, in English "confining" has the connotation of containing-thus hindering understanding when Uhle speaks of one geographical feature "confining" another, as he does, for example, on the fifth page of the Sept. 11 letter: "Lower down the valley is open, and the left side confining with the wide desert, which also separates Pisco from Ica." "Confining with" is clearly not idiomatic English-and thus we edit it to say "... the left side borders the wide desert, ..." Or for example, Uhle would sometimes say "on the other side," for example on page 30 of the Sept. 1 1 letter, when clearly he meant "on xii the other hand," which we edit his text to say. Saying "on the other side" instead of "on the other hand" can be particularly confusing in a discussion of an architectural/archaeological site as "on the other side" commonly refers to physical location. In some cases, Uhle is attempting to describe architectural details without using architectural terms. In many of these cases we have changed Uhle's language for ease of comprehension. In cases where comprehension was not troublesome, Uhle's language has been left as is. Here is a brief table of such terms used by Uhle and their corresponding current architectural terms: Uhle Common Receding frame recessed frame Inside of niche (or doorway) reveal Roof (of a doorway) lintel Chamfer chase or wall chase or groove Embossed (stone) pillowed (stone) Direction (of a wall) alignment (of a wall) One particularly common change that we made was in Uhle's use of the word "room." Uhle, as a native speaker of German, used "room" much in the way one uses the word "Raum" in German. Sometimes when Uhle speaks of a room, he means a space that would be called a "room" in English, but often he is speaking of other spaces- passageways, courtyards, and other spaces that are not "rooms" to an English speaker. In these cases we have changed his use of the word "room" to more closely accommodate the space he is describing, thus sometimes changing the word to "space" or to "passage" or "court" as appropriate. Since most of these spaces are indicated on the plans, the reader will be able to see what kind of space is being described. In yet other cases, sentence structure and usage combine to make a difficult and ambiguous statement. In these cases we have done our best to refine the sentence to express the interpretation that makes most sense. As, for example, in the following passage from the beginning of the Sept. 11 letter, in which Uhle wrote: "I knew and heard casually at Pisco, that Mr. Bandelier in about 1893 has been for about 4 to 5 days here starting from Pisco, and the owner of the hacienda has told me afterwards while being at Ica that Mr. Bandelier had made a large map of the ruins in Lima after his measurements made at Pallasca (and that he himself had seen it?) but as nothing has been published since by Mr. Bandelier, I did not know, of which kind of responsability that work of Mr. Bandelier might be." xiii It has awkward phrases like "Mr. Bandelier in about 1893 has been for about 4 to 5 days here starting from Pisco," which we attempt to untangle by placing them in more common constructions: e.g., "in about 1893, Mr. Bandelier was here for about 4 to 5 days, visiting from Pisco." Due to awkwardness with the construction, confusions result: the syntax suggests that the map made by Bandelier was "of the ruins in Lima" -but, of course, it is the ruins at Tambo Colorado that interest Uhle here, and though the map was made in Lima, the ruins it depicts are those that interest Uhle: Tambo Colorado. After editing, the sentence becomes: "I knew and heard casually at Pisco, that, in about 1893, Mr. Bandelier was here for about 4 to 5 days, visiting from Pisco, and the owner of the hacienda has told me that Mr. Bandelier had a large map of the ruins made in Lima after the measurements he took at Pallasca (and that he, the owner, had seen it himself), but as nothing has been published since by Mr. Bandelier, I did not know of what degree of accuracy that work of Mr. Bandelier might be." But though we tried to clarify his intention, we attempted to leave his work as little altered as possible. Thus, though we might wish to edit the preceding example for structural simplicity -for example we could pare the phrase "I did not know of what degree of accuracy that work of Mr. Bandelier might be" down to "I do not know how accurate it is" -we did not, feeling that, though it may retain some awkwardness, it is not confusing. Nor, when he speaks of "night-harbors" on the Inca roads, as he does on page 11 of the January 1902 letter, do we alter it to "way station"' or "rest stop"-the meaning seems abundantly clear. Such attempts to intervene were the extent of our efforts to alter the text we present. In some places, we believe that Uhle's report is incorrect or somehow flawed with respect to actual conditions at the site in question. These are noted, but we do not correct Uhle if his meaning seems clear and we believe that he is wrong. This is an attempt to present his work. Rather than correcting his text, we write an explanatory note concerning his conclusion and the evidence as we know it. Thus, for example, in his assumptions about Mochic on page 15 of the January 1902 letter, we believe he is incorrect and note this, but leave the incorrect implication in the text. Similarly, when Uhle is puzzled, e.g., by the Inca building at Huaytara, in the January 1902 letter, because he had never seen a similar building, we do not attempt to correct his reasoning based on xiv more current knowledge of these buildings, known as Cuyus Mango. We do, however, note where Uhle's conjectures and uncertainties are or are not supported by later research. Finally, in some cases, Uhle's letter lacked detail or clarity and so our intervention in the text is either minimal or lacking detail. Here, we attempt to supplement the text with footnotes explaining our best interpretation. Uhle's original, handwritten field notes, field catalogs, letters and other notes have been assembled in the ten volumes of the ORIGINAL CATALOGUES, M. UHLE deposited in the Phoebe A.Hearst Museum of Anthropology. The letters published here are found in volumes 4 and 5. To conserve the originals and to facilitate our work each page of each letter has been digitally photographed. In the text of the letters the original page number- in brackets-is followed by the corresponding digital photograph number. Throughout the text of his letters Uhle makes references to photographic plates he made to illustrate his work. Negatives of most of these plates are preserved and catalogued in Catalog 15, also at the Phoebe A. Hearst Museum of Anthropology. There are 103 plates mentioned throughout the letters here presented, 60 of which are included in this publication. Among the mentioned photographs are several that represent the same subject but from different angles. We have selected the photographs that seemed to be the most relevant to the text. A list of all photographs at the Pnoebe A. Hearst Museum of Anthropology and their correspondence to Uhle's plate numbers is included in this publication. i Max Uhle, "Das Thai von Chincha," ms. (no. 26), held in the Ibero-Amerikanisches Institut Preussischer Kulturbesitz (IAIPK) in Berlin, transcribed by Gerdt Kutscher, also in the IAIPK, (no. 27). The numbers refer to the catalog no. of the inventory of the Uhle Bequest at the IAIPK. ii Translation by JPP. The original text is as follows: "Der Plan des Schreibers bestand die Erforschung des Thales von Chincha zum Ausgang nehmend einen grossen Durchschnitt durch die in alter Zeit cultivirten Religionen [Regionen] des sudlichen Peru auf Cuzco zu zu machen. Bei diesem waren die Thaler von Pisco und Ica die noch nie fruher studirt worden waren, vorerst zu beruhren . .. Bei dem weiteren Vorstoss gegen Cuzco hin, waren die interessanten Gebiete von Lucana und dem Departement von Ayacucho, einst die Wohnplatze der durch die alte Geschichte beruhmten Chancas zu beruhren gewesen. Und weiter hatte man hoffen durfen, auf geographischem Wege durch das Studium der Nachbar-culturen einen Schlussel zur historischen Kultur der Inca und des Stiles von Cuzco zu finden. xv Das ziemlich ausgedehnt gedachte Programm konnte nur in seinem ersten die Thaler von Chincha, Pisco und Ica betreffenden Theile gelost werden. Nach einjahriger Beschaftigung mit diesem vom September 1900 bis Oktober 1901 erreichte ihn der Zuruckruf nach Californien um vorlaufige Rechenschaft von dem bisher erreichten zu erstatten." Ibid. p.l. Ibid. p. 1. Max Uhle, "Explorations at Chincha," A. L. Kroeber ed., in University of California Publications in American Archaeology and Ethnology, Vol. 21, Nos. 1 and 2, pp. 1-94, plates 1-28, 28 figs. in text, University of California Press, Berkeley, 1924. v Ms. (no. 34), transcribed by Gerdt Kutscher (no. 35). vi See ftn. below. Vii When subsuming both the ruins of Tambo Colorado and of Huaytara under the title "Exploration in the Pisco Valley" we also follow Uhle, who wrote: "The upper part of the Valley of Pisco up to Huaitara .... (Letter of October 9gh, 1901, ms. p. 1). Strictly speaking, this is not accurate, for the Huaytara River flows into the Pisco River near Pampano. In other words, Huaytara is not located in the Pisco Valley proper, but in a tributary to it. viii Letter to Phoebe A. Hearst dated Ica, January 7h, 1901. ix The quotes are from the September 11, 1901 letter, ms. pp. 20, 21. x The timeline given here is extracted from both the letters presented here and Uhle's Notebook no. 57 at the IAI-PK in Berlin. It differs slightly from the timeline presented in "Max Uhle (1856-1944) Plane archaeologischer Statten im Andengebiet," W. W. Wurster ed., AVA Materialien 56, Verlag Philipp von Zabern, Mainz am Rhein, 1999, pp. 80, 81. x Only the right side pages in the catalogues are numbered. Numbers are consecutive. Left side pages are without numbers. Catalogue pages (left or right) may contain more than one actual document (e.g., several pages of a letter). xii Not to be confused with the Huaca La Centinela near Tambo de Mora,'also in the Chincha Valley. xiii Possibly the huaca that today is known as Huaca Medina. xi' Letter of September 11, ms. pp. 8-9 xv Pages 1 19-21 of this manuscript are missing. xvi Rowe, J. H., Max Uhle 1856-1944. A Memoir of the Father of Peruvian Archaeology in University of California Publications in American Archaeology and Ethnology, Vol. 46, No. 1, 1954. xvi Ibid. xvi [p. 1, (PICT0378)] ' Pallasca in the Valley of Pisco August 23d 1901 Dear Mrs. Hearst I have the honor to advise you that I arrived yesterday in the valley of Pisco in the hacienda Pallasca, where lie the ruins of Tambo Colorado, of great fame in this and the neighboring valleys. My baggage, which has not arrived here yet, contains an unfinished letter for you, but, having a rare occasion today to send down to Pisco (about 9 leagues or 25-27 miles from Pallasca), I take the liberty to give you this notice upon the only kind of paper which I can get here. This part of the valley of Pisco is about 9 leagues (25-27 miles) distant from Ica. The road follows the railroad line about 2 leagues, up to the upper end of the valley and there crosses a wide sandy plain, very little trafficked to the north west due to the difficulty of marching on it. At the end of this plain (the "pampa of Chunchanga," the same which extends to the west about 40 miles in the direction of Pisco), in the direction of my march there lies a high mountain pass, on which one has to climb about 300 meters from the level of the plain, which itself lies about 70 [p.2, (PICT0379)] meters above the level of Ica, so that on the pass, one rises to a height of about 800 meters above the sea. On the other side, one descends in the direction of the ruins of Tambo Colorado along a ravine about 4 miles long. The buildings of the Hacienda Pallasca are about 2 miles farther down the valley, but still lie about 40-50 meters higher than Ica, so that one might estimate that the ruins of Tambo Colorado are about 100 meters above Ica. From all these observations it may be concluded that the traffic between the two valleys of Pisco and Ica, though not very long (about 25 miles) is not free from difficulties, but there is no doubt, as it seems to me, that the spot where Tambo Colorado has been built, has been especially chosen with respect to the circumstances that from there, the traffic to Ica by the mentioned pass was easiest, considering that the mountain pass is exactly in front of the ruins, though not visible from there. I would also like to mention that I collected a few painted pottery fragments about 6 miles from Guadelupe, in the middle of the sandy plain. These belong to one of the oldest periods, though not representing the same style for which I especially came to the valley of Ica. But the ages of the two styles seem [p.3, (PICTO38O)] to me to be very close. The painting is interesting; it has something in common with the so-called epigones of v"Each page of UhIe's original letters has been digitally photographed. For the reader's reference, the page number of the original letter followed by the number of the digitized photograph of that page appears in brackets throughout the text of this volume. 2 Letter of August 23, 1901 Tiahuanaco. It also is well preserved. Considering the matter now, I cannot but believe that the original vessel may have been buried (perhaps in an ancient cemetery) there in the middle of the pampa, and the fragments preserved, until discovered by the wind or other circumstance. Perhaps I shall gather more information at that spot if the opportunity arises. But the ruins of Tambo Colorado are from a quite different time. They are pure Incasic. Tambo Colorado means: "painted harbor." Indeed, various paintings still exist there. All the walls seem to have been painted formerly too, the walls red, the interior of the niches yellow. It is curious how much misunderstanding can exist in a country like this about the age and the paintings. It is incomprehensible that people could tell me that the color of the walls might have been the natural color of the clay used for forming the brick. It is incomprehensible that nobody clearly acknowledges that the ruins are Incasic. But in no place of Incasic ruins can the matter be more easily decided. The ruins are wonderfully preserved, better than any Incasic ruins I know of in the coastlands, and they represent, apparently, something like an Incasic palace, perhaps a palace of [p.4, (PICT0381)] the Inca himself, because we find here excellent long galleries with niches, wall painting, and much other artistic decoration of the walls, a splendid court, and more. Being here and taking photographs, it seems to me necessary to measure the ruins, which do not exceed about, as I calculate, 200 meters long and wide. The country is wonderful, the climate agreeable, plenty of water waters the valley (which is only about 100 meters wide). Mr. Bandelier measured the ruins too, but I have not seer, anything of his work, though people here tell me here that he did so. I calculate that I will stay here for about two weeks. Plenty of fine andenes are in front of the palaces on the other side of the valley, and other ruins too. Other ruins are in the valley below. I shall visit this part rapidly and after that go up to Huaitara, about 40-50 (?) miles from here Believe me, Dear Mrs Hearst, Yours Very Sincerely Max Uhle. My unfinished letter notifies you of my discovery of petrified wood, worked by human hand before its petrification, at the time of the oldest style of the valley of Ica, for the study of which I was staying there. I believe this petrified wood proves an age of at least 1.5 - 2 thousands of years for that period, perhaps thereby proving an higher age than I was able to ascertain directly in any other way myself. [P. 1, (PICT0382)] Pallasca, September 1 I', 1901 Dear Mrs. Hearst, I have the honor to relate to you my work done at Pallasca Hacienda, about 9 leagues (about 40-45 kilometers) above the port of Pisco in the valley of Pisco. As I told you in a brief letter on August 23rd, I left Ica on the 22nd rejoining my animals, which had left Ica earlier, and, crossing the desert of Chunchanga, I arrived the same day at Pallasca in the valley of Pisco. The river bears the name of Chunchanga, which as I said before, reminds one of the name of the Chincha tribe, after which the neighboring valley (of Chincha) has been denominated. As mentioned in my last letter, the ruins which belong to the hacienda of Pallasca surprised me extremely by the extraordinary preservation of their masonry and wall painting! Till I arrived here, I had no clear idea to which period the ruins of this part of the valley might belong. I knew and heard casually at Pisco, that, in about 1893, Mr. Bandelier was here for about 4 to 5 days, visiting from Pisco, and the owner of the hacienda [p.2, (PICT0383)] has told me that Mr. Bandelier had a large map of the ruins made in Lima after the measurements he took at Pallasca (and that he, the owner, had seen it himself), but as nothing has been published since by Mr. Bandelier, I did not know, of what degree of accuracy that work of Mr. Bandelier might be.' Perhaps he never might publish any of this, and further I thought it was necessary for me to see the ruins while studying the valley of Chincha and Ica, as indigenous Peruvian literature has done much abuse to the name of the "ruins of Umay" (which is the village about 1 league down from the ruins in question, and no other ruins of any uncommon appearance exist in the district to which the ruins of the Hacienda Pallasca also belong2). In articles written in Lima and published as separate papers, the importance of the stone of Ancash, a mold of which I sent to the University,3 has been presented in the manner that Peruvians are wont to treat science. In that paper, the author argues, among other things, that the marks of the stone belong to the same system of old Peruvian hieroglyphs as those painted on the walls of the old ruins of Umay in the [p.3, (PICT0384)] valley of Pisco. As the stone of Ancash is very old-it may have an age of at least 1.25 thousand years-my 1 Adoph Bandelier did indeed make a map of Tambo Colorado. That map is kept in the American Museum of Natural History in New York. This map gives a general idea of Tambo Colorado's layout but is far from being accurate (editor's note). 2 The name of this village today is Humay (editor's note). 3 This mold cannot currently be located (editor's note). 4 Explorations in the Pisco Valley interest was naturally awaked to see those ancient painted hieroglyphs of Umay. Instead, at Tambo Colorado I found evidence of the most recent prehistoric period of Peru, and instead of hieroglyphs, nothing but wall paintings which, though interesting, have no more hieroglyphic interest than white washed walls, because the system of wall painting does not follow any principle other than coloring walls and niches in uniform colors, with stripes on the walls (at the top of the walls), and in coloring the niches and the different parts of the interior of the niches in a distinctive and particular way. This is the system of science in Peru. Everybody adorns himself with plumes of different colors, without having any reason for wearing them in the scientific matter which he so casually treats at that moment. It is sufficient that something exists for someone to give it momentary appreciation, in which he needs only to show that he knows anything more than others, no matter if he himself has any more knowledge about the matter than word of its existence. [p.4, (PICT0385] It is hardly necessary to say, Dear Mrs Hearst, that I did not find anything of any similarity in time, appearance, technique, etc., with the stone of Ancash. Notwithstanding, the ruins were of very great interest to me, for they taught me much about Incasic civilization, and therefore, I made this small part of the valley the object of a small, methodical inquisition into ancient Peruvian history. This part of the valley of Pisco has many charms. It is rich in water. Though we are now living here in winter, the valley seems ever-green with its fields of alfalfa, growing corn, and evergreen trees. Though all haciendas are small in extent, the owners may make a good deal of money by rich harvests of aguardiente, production of oxen and cows, horses, etc. Some cotton is also produced here. The condition of the valley of Pisco is quite different from that of Ica. This part of the valley of Pisco is narrow; the valley of Ica is wide. The valley of Ica is an open plain. Here, we are between mountains, upon the uneven soil found in mountain districts. The valley of Ica is dry, wanting water; the valley of Pisco abounds in water. For this reason the haciendas of this valley are flourish-[p.5, (PICT0386)] ing, while those in the valley of Ica are nearly starving. About 8 leagues above the port of Pisco, the river of Chunchanga begins to be bordered by mountains on both sides. Lower down the valley is open, and the left side borders the wide desert, which also separates Pisco from Ica. There is no doubt that great terrestrial catastrophes have passed through this valley. As in the valley of Ica, there are clear marks of its having belonged formerly to the open sea, forming a bay, and, in the valley of Pisco, one sees clearly that its terrestrial history was shaped by immense torrents coming down from the mountains. To about 60 meters of height, layers of boulders still cover the slopes of the hills which border the valley, forming river terraces.4 Undoubtedly the river once formed an immense channel for 4~ The valley is characterized by both terraces made by natural forces and terraces made for agriculture (andenes). Uhle refers to both as simply "'terraces." We have, in some cases, added the word "river" to indicate the terraces made by natural forces (editor's note). Letter of September 11 and 24, 1901 5 conducting the water of immense rainfalls in the mountains in long-past periods. These terraces have also been the subject of agricultural work done by prehistoric man in Peru. That part of the valley of greatest interest has an extent of about 1 league, or 2-3 miles, from east to west. The valley has there a width of about 500-700 meters, and is cultivated on both sides of the river, [p.6,(PICT0387)] which every year tries to destroy the human work done on both sides. Prominent rocks separate smaller sections of the valley. Such a one is that section of about 1300 meters of length from east to west, in which those ancient palaces were built, the ruins of which now bear the name of Tambo Colorado, for the rare, well-preserved, ancient wall painting all over them. These ruins are Incasic, nothing but Incasic. Whoever has gotten an impression of what is Incasic architecture, will not vacillate a moment in declaring that these ruins are pure Incasic, and one of the most interesting, and, in character, most decided, monuments which the Incas have left to us. Incas settled there and built what was probably one of the most admired palaces of their empire, which we are happy enough to see well-preserved up to our times. Incas may have been attracted by the natural attractiveness of the valley, with its green fields, high, dry mountain ridges, plenty of good water, gurgling river, and temperate climate. Though the heat is rising now, it is less immoderate than at Ica, and, in the evenings, one always breathes the cooler air rising up the valley from the direction of the sea. This is not possible at Ica, because the valley of Ica turns near that town to the south instead of going on to the west, and it is separated by the high desert from the western coast and its cool airs. [p.7, (PICT0388)] I enclose with my report three maps [pp. 132-34 in this volume]: 1. one map of the whole section of the valley in which the ruins of Tambo Colorado are lying, scale 1:4000 2. one map of the main part of the ruins, scale 1:500 3. one map of the principal, and, at the same time, the best preserved of the three palaces of Incasic architecture existing there, scale 1:200. Besides this I made numerous plates, which I have not yet put in order. But I trust that they will give quite a clear idea of Tambo Colorado, not withstanding that color, so important here, can only insufficiently be expressed by photograms. So far as I know, at the time of his visit, Mr. Bandelier did not take photographs nor order any to be made for him. Only lately I heard that he had a photographer here for one day, who left immediately after that for Lima. I venture to express, that at least this small circumstance would give some advantage to my visit over his, and to my report of the ruins over his, if he were to publish them. It seems to me, that of a ruin so well preserved as this, which reveals to the naked eye thousands of things about what has been Incasic architecture, Incasic taste, Incasic refinement, photograms cannot be overlooked, [p. 8, (PICT0389)] and it seems to me good to have at least saved photograms, should the ruins afterwards decay through negligence or abuse of the indigenous people, or the historical events so commonly passing over these countries. 6 Explorations in the Pisco Valley But I did not bind myself to the study of those ruins alone. As I saw several other remains of the pre-Spanish evolution here, too, I tried to get a more or less clear impression of the historical traces which civilized mankind left in this valley before the invasion of the Spaniards, and I am glad to say that I found out that in this small section of the country alone, prehistorical interest is in no way limited to the well preserved pre-Spanish buildings of Tambo Colorado. It would be a very superficial manner of study to wander around and observe only the localities of ruins which are prominent in the view of the uneducated person. At least it is my idea that we have to penetrate now, as with X-rays, all the system of historical development of these countries, because the period which most calls our attention, the last, the Incasic period, has in no way been the most prominent period in pre- Spanish times. One period deserves as much [p. 9, (PICT0390)] attention as the other, and we never will learn anything about historical development in Peru - of the history of civilized mankind - if we are not willing to comprise all periods in our study. For myself, the Incasic period, considered alone, is a poor thing, considering it in a comparative way with the entire pre- Spanish history of this soil, which probably has been peopled for at least 2000 years by civilized mankind before the arrivals of the Spaniards, while Incasic history within this space embraces only 150-200 years (the primordial development of Incasic empire in the mountains comprising about another 200 years). History of the small section of the valley around Tambo Colorado embraces at least 1500 years. At the beginning of that time, people were crowded in large but irregular settlements, built up of broken stones. I found at least three, of about 150 x 100 meters in extent, between the hacienda of Pallasca and Tambo Colorado (about 2500 meters from one to the other), two on the right, one on the left side of the river. The houses or huts made of stone were small and narrow, crowded together in a small space, no means of communication between them being [p. 10, (PICT0391)] recognizable now, though they must have existed. These settlements occupy dry places at the side of the river and among the cultivated fields. One of the three which I found had been laid out upon one of the boulder terraces (formed by the river in an older geological period, about 60 meters above the river), the second upon the dry bed of a wide ravine, which was formed in older periods between the hills bordering the valley, the third on a lofty plateau projecting from the hill, and surrounded by green fields on two or three sides. The walls of the original houses of this period are entirely destroyed. Only more accurate or closer observation realizes that the many stones and boulders lying around formed an ancient settlement. All is now so destroyed that only the traces remain. Walls have fallen, forming now only heaps of stones in series. In other walls, the stones are still one above the other, in the original condition, but the clay originally binding the stones together has disappeared nearly entirely. Regular walls enclosed a part of such settlements, as far as no hills were protecting them, for instance, at the back. Letter of September 11 and 24, 1901 7 I inquired closely as to the age of these most ancient settlements. Nobody seems to have made any excavation there at any time, but it seems also that it is difficult to [p. 1 1, (PICT0392)] discover ancient objects by such excavations. During my various trials, I found fragments of painted pottery, which in time corresponded to the style found by me in the valley of Ica, and, as petrified wood forms a mark of its age, one may form an idea of how old are such ruins. One very small piece of pottery was extremely fine, quite so fine as the best of that period found in the valley of Ica (as at Santiago). In one excavation, I discovered those half globular adobe 4I-Iii which were the material of wall constructions and are so absolutely indicative of that style of Ica, previously mentioned. At the same spot, I found, among the other pieces of adobe, pieces of rectangular adobe, painted on two sides in colors, of which the Incas, to my knowledge, made no use. Several of these pieces showed regular painting, as of geometrical ornaments, on two bordering sides. This proves to me, that at that far-gone period, people already understood artificial wall painting quite as well as the Incas, who lived about 1000 years afterwards. That older people may also have possessed painted houses, built of adobe, and adorned with niches, as those mentioned pieces of painted adobe are showing two colored faces. In any case, it was an already stupendous- [p. 12, (PIC70393)] ly developed people, even in house construction, and decoration of the houses. Another style period represented in this part of the valley is that which I denominate the proper Chincha style, according to what I found near Tambo de Mora. This style exists partly in the smaller houses, joined to the principal Incasic palaces. These smaller houses are distinguished from the Incasic by the more careless construction and layout of the walls which form them. I am in no doubt that in these houses of inferior quality were living Chincha caziques, subjected to the Inca by conquest. I preserved only a very small piece of the pottery of this period, collected in one of the mentioned palaces (south of the main Incasic palaces), as proof. But about 1/2 to 1 mile above Tambo Colorado, in the neighboring hacienda, and at Guaya grande,5 about 1.5 leagues higher up the valley, there are cemeteries where pottery of this period may be excavated, perhaps in great quantity. But I do not like to lose time by repeating in this valley, studies made before in other valleys. This is the reason I do not intend to expand my collection of that period by studying those palaces. The third period is the Incasic, of the Incasic conquerors from whom originated those interesting palaces specially mapped in map 2. 5 Today called Huaya Grande (editor's note). 8 Explorations in the Pisco Valley [p. 13, (PICT0394)] Pallasca, September 24t 1901 Having come to the end of my supply of fresh photographical plates, I decided to go down to Pisco on September 12?, to bring several dozen more plates from Ica by train, for observing the archaeological particularities of the lower part of the valley of Chunchanga and also for studying the southern part of the Chincha valley, bordering with that of Pisco at the north, as want of animals had previously made it impossible for me to see the large huaca of Lurin Chincha and to study more closely the large huaca of Santa Rosa, two leagues from the latter. The distances are these: 9-10 leagues Pallasca to Pisco 5 leagues Pisco to Lurin Chincha 2 " Lurin Chincha to Santa Rosa I went up to Ica by train on the 13th, returned on the 14th, and arrived on the same day at Lurin Chincha. I returned from there on the afternoon of the 16", and arrived at Pallasca on the 21st, where I am now occupied in developing, at night, a large lot of plates taken of various antiquities, and preparing the end of my definitive report on the ruins of Tambo Colorado. I return now to the continuation of my report. [p.14, (PICT0395)] Two of those ancient settlements mentioned before bear the interesting names: "Pantaico," which is undoubtedly a very ancient name, and the other: the "heights of the sorceresses" ("Altos de las brujas"), which seems to indicate, that in recent times there were sorceresses existing in the valley, as one of the small villages near Ica, Cachiche, is still known today as a village where sorceresses or witches are living. These sorceresses living in Cachiche may dedicate themselves especially to curing sicknesses, but it is not improbable that they are still occupied in some old witchery too. According later reflection and growing experience, I am not so sure, at the present time, that those mentioned ancient settlements were really settlements and not only graveyards, as all the place surrounded by walls is full of ancient graves, though nothing is found in the graves but some bones here and there. Contrary to my former affirmation, I am now of the opinion that a great deal of the ancient graves have been opened in later times, though I cannot tell for what purpose, as I cannot discover any contents of value in any of the ancient graves, and it is commonly known in all this part of the valley too, that nobody has yet discovered anything in these ancient "stony places." Notwithstanding it must be so, that in older Spanish times many ex- [p. 15, (PICT0396)] cavations were tried, a proof of which is this, that I found in one of the old, reopened graves, a single playing card of older Spanish origin, of a type which is not in use anymore. Besides this, I am not sure that the people of the Chincha period did not also open ancient burials, if only for making newer burials in them. I found one square (parallelepipedal) burial Letter of September 11 and 24, 1901 9 chamber, with the faces of the walls smoothed in the interior, which was not the custom of the civilization of about 1500 years ago. The grave had been filled by the half-globular pieces of adobe, which are undoubtedly that old, but they may have been turned out by opening the grave and then used again for closing a burial of more recent origin. Wooden beams were lying above the half-globular pieces of adobe, while in the older graves, they are lying below, as support for the adobes. And besides this, the grave was not at the bottom of the parallelepipedal burial chamber, but in a sort of cavity below one of the walls which form the burial chamber, a type of grave which corresponds especially to that kind of graves which were made in the later 1117,., Chincha period, so that I have to conclude that this grave was probably made at the time of the later Chinchas (about 1000-1300 a.d.), not at the same time of the older [p. 16, (PICTr0397)] civilization of this spot at about 1500 years ago, notwithstanding that adobes and the characteristic fragments of pottery of the older period were found mixed in the grave. All this shows how difficult it is sometimes to discover the real age of ancient remains. The mentioned grave was that in which those mentioned small painted fragments of painted adobe walls have been found. I am, therefore, equally of the opinion, that these painted adobe fragments may be also of later date. This might seem to be corroborated by the fact that some small remains of a building constructed of adobes of later times exist near that place. It is a curious circumstance that the later Chinchas generally followed the custom of dedicating ruins of older periods to a different use in their own civilization. I mention again the fact that the huaca of Alvarado near Tambo de Mora is of much older origin (originally having been a palace or temple which was afterwards completely destroyed) than the graves made upon it The same can be said of the huaca of Santa Rosa in the valley of Chincha about 1.5 leagues south of Tambo de Mora, and of the huaca half (or three quarters) destroyed by the river Pisco, near Pisco, of which I made various panoramas for illustrating [p. 17, (PICT00398)] its inner construction. This huaca is about of the same age as the huaca of Alvarado near Tambo de Mora. It is built up of the same kind of globular adobes as the other. During my two days' excavation, I found at least one characteristic fragment of pottery which, according to its painting, belongs to the same period as the "new found style of Ica," which corresponds to the age supposed by me for the huaca, according to the material used in its construction. But the graves on it (and there are a large number there) all represent a much later type, that of the so-called later Chincha period. 10 Explorations in the Pisco Valley I return now to the description of Tambo Colorado. The ruins occupy a small plain which commands the cultivated fields on the right side of the river, about 220 meters from its bed. The plain is irregular in shape (see map 1), about 550 meters long and 180 meters wide, elevated about 6 meters above the cultivated fields and about 9 meter above the river and confined by the hills rising at its back. Apparently this plain is of natural origin. It represents, so to speak, the second terrace (counting from above) formed by the river bringing down enormous masses of earth and stones, and sinking after that between them to its new level. In this section of the valley, the plain at this level is only on the right side of the river. But on the left side of the river, extensive [p. 18, (PICT0399)] grounds, cultivated in ancient times, correspond to the same level as the cultivated fields on the right side of the river. The plain on the right side of the river does not seem to have been cultivated in times before the construction of the Incasic buildings, though this would not have been impossible, given the widespread use of azequias6 by the ancient Indians. The southern front of the principal palace, whose northern face leans against the hills, is more or less directed in the direction of 243?20' of the needle [of the compass], so that this palace may be said to face more or less South-South-East. The sun shines very little on the front of the palace at this season of the year, crossing from the front to the back of the palace at about 7:30 in the morning. The gorge in which the "road" used by travelers from Ica descends from the summit of the mountain, lies a few minutes lower down on the other side of the river (see map 1). But the ruins of Tambo Colorado are visible from some higher parts of that gorge as well. A side road leads also down to the bank in front of the ruins, and by chance, I descended myself to that part, coming from Ica, as the road used more generally was not entirely clear. There is no doubt that the selection of this spot for construction of the ruins has something [p. 19, (PICT0400)] to do with the general line of communications with the valley of Ica. Because, from the mountain pass (which lies about 900 meters above the sea, while Ica is at 400 and the ruins of Tambo Colorado at about 450 meters above the sea), one descends on the other side to an ancient road, which crosses the "pampa of Chunchanga" in the direction from the pass to the so called "Cerro Prieto," which stands like a guard at the western entrance of the valley of Ica. Undoubtedly this road was used by the Incas in their visits to the valleys of Ica and Nazca when coming from the valley of the river Pisco, and from their abodes at the present "Tambo Colorado." In this way the ruins lie, so to speak, at the spot where two different lines of communications are meeting: the line of communications with the valley of Ica, and the general road leading through the valley of Chunchanga, up to the mountains (Huaitara,7 Castro Virrey, Huancavelica, Ayacucho). This latter road leads up through the valley along the right bank of the river, and the same road leads also through the ruins of Tambo Colorado, which lie on the road, bordering it on both sides. The road crosses the ruins. Openings for 6 "Azequia" is a Peruvian Spanish word for irrigation canal (editor's note). 7' Uhie's Huaitara is today spelled Huaytarai (editor's note). Letter of September 11 and 24, 1901 11 its entrance and exit have been left at the eastern and western side of the trapezoidal plaza that lies between the ruins and [p.20, (PICT0401)] is entirely closed on all other parts of its perimeter. As in ancient times, travelers are still passing in front of the principal palace on their voyage from the mountains to the coast or on their return. This road is much trafficked. As best I could tell, this road is more used than any other for the traffic to the mountains, as the ascent through this valley is the smoothest. The importance of Ica consists in its character of being the upper end of a railroad 74 kilometers long leading to the interior. But before its existence and besides it now, still the most important road for traffic (not excepting the way by Ica) is leading through this valley. Every day I see here many troops of mules and llamas, and many single travelers descending down to the coast by this road. I calculate that every day about 50 to 100 people pass in each direction with 200 to 400 animals. I must confess, that this fact, before unknown to me, makes me a little suspicious of the position occupied by Huaitara in the ancient fables. Because I see now that Huaitara is a natural mark for nearly all immigration to the southern Highland of Peru in the direction of Ayacucho, it is not necessary to attribute to it a particular kind of importance because of the existence of ancient fables which mention that the oldest immigration [p.21, (PICT0402)] from Central America to Peru went to Huaitara. But as some monuments of stone architecture at Huaitara are of some renown here about, at any rate a visit to them appears to be of some importance. The whole complex group of ruins at Tambo Colorado occupies about 400 meters in length, 205 meters in width. But these are not all ruins of equal importance. The principal buildings are those bearing pure Incasic character. They occupy the middle part, while those of inferior character form only something like annexes east and west of them. The principal ruins border a quadrilateral plaza of irregular shape.8 The sides of the plaza are (taking into account that one building is projecting into it): 192 meters on the northern face, 137 meters on the southern, 102.50 on the western, and 62 on the eastern. The irregularity may have originated from the difficulty of building palaces on both the northern and the southern side of the triangular plain. For that reason the northern face of the buildings at the south of the plaza was turned more to the west. In the other case, the ground at the back would not have been sufficient for the construction of the proposed buildings. Though the ground has been broken off at the river-side by the river during occasional flooding in the valley, after the construction of the buildings, [p.22, (PICT0403)] which, therefore, are now mostly incomplete at that side, there is no doubt, that the ground may have always 8 In speaking of the wide central plaza between the three palaces-the plaza across which the main road used to run -Uhle often refers to as "court," or as "wide central court." We have replaced Uhle's "court" with "plaza" for clarity, so that the central plaza is distinct from the courtyards in the palace interiors. Note that above [p. 19, bottom of original letter] Uhle called it "the trapezoidal plaza" and that in his plan of Tambo Colorado he used the word "Plaza" to designate that very same feature (editor's note). 12 Explorations in the Pisco Valley had such a circumference as to make it necessary to reduce the width of the plaza at the east. The most prominent building occupies about the middle of the northern side of the large plaza; in length, it is about 57 meters. Two similar buildings, side by side, faced nearly the entire length of the plaza's southern side. A wide terrace fronts the plaza at its western side, with the original plaza wall at its back, and only the square building of the eastern side seems to have been of some inferior importance. Though it contains many niches which are characteristic of Incasic architecture, the disposition of the apartments is of a simple character and the material (quarry-stones bound together by clay) differs from that of the buildings of higher rank. It seems, therefore, that this building may have served guardians or other people in the general service of the buildings comprised by the plaza. It seems that the thick wall which borders the plaza on the west, and probably went on along the southern border of the complex of buildings before the destruction of this part of the complex by the river, served only to support the higher walls within, as the exterior ground was much lower at these sides. It had, therefore, probably nothing [p.23, (PICT0404)] to do with the defense of the buildings. But the general meeting of all principal buildings around the plaza, closed by walls on all sides, with the exception of the openings left for the mountain road passing through it, shows notwithstanding that the owners of those buildings wished to remain among themselves by exclusion of other people, however prominent the others may have been. This seems to stand in relation with the fact that the Incas were the owners of the country as its conquerors and imposed themselves on the indigenous people as a superior race, with the declared intention, at least as far as the persons of the supreme power were considered, to remain separate from the others. I shall now give a detailed description of the buildings from which two things are hoped: first a more detailed knowledge of Incasic architecture in all its different branches of constructive and decorative work, as far as Incasic architecture in the materials of the coast, clay and adobes, is concerned, second a more detailed explication of the purpose and use for which these special buildings were constructed. The ancient buildings are extremely well preserved, at least as far as the palace at the northern side of the plaza is concerned. My numerous photograms will give ample proof of this fact. [p.24, (PICT0405)] I go as far as to ascertain that this particular ruin enables us to get a very complete notion of the manner in which the Incas were accustomed to lodge, and of the general ancient appearance of their buildings, so rare now to observe or even to imagine, given the condition in which ruins are mostly found. If these ancient buildings were roofed, they would give us a nearly complete impression of ancient Incasic buildings as they were in their original condition, and according to my opinion, it would be a wise measure of any government which appreciates the marks of development which human civilization has made on its soil, to roof them, for protecting such an Letter of September 11 and 24, 1901 13 important monument of national history. But we are very far from such measures, and it is to be considered at the same time, that - perhaps because one is here so far from such measures - we are profiting of all liberty for making archaeological studies for our own, in which very little is really done here for the advance of science. I think, it might be easy to complete the impression which those buildings must have given originally by drawings which add roofs to the walls preserved. Destruction of the northern palace is restricted nearly to a few walls that have been broken for new entrances and passages in this building, which originally possessed very [p.25, (PICT0406)] few of them from the outside as well as in its interior. Besides this, the floors of several depressed rooms, which I con-sider to be ancient bathrooms, have undergone a nearly complete destruction, owing to the circumstance that these rooms were depressed and floored by large stones, which seem to signify to treasure-hunters that treasures must be hidden under these large stones. Treasures below a bathing-tub! It is unnecessary to say that well pre- served bathrooms, however interesting it might be to observe them in their original condition, do not exist in any of the three ancient Incasic palaces of these ruins, though originally there may have been at least five of them. Several of the ancient doors of minor importance are well preserved, with lintels of stone or of soft, light wood, perhaps "Pakai." All more important entrances (with receding frames9 at the sides and at the top) are lacking now their former lintels. Wall painting is marvelously preserved, owing to the wonderful, nearly rainless climate of this valley. The buildings bordering the large plaza at the south are much less preserved than the northern, due to the fact that the most recent former tenant used them as stables for cattle, which enormously damaged the walls. These latter are no better now than those in other parts of the country. It is now difficult [p.26, (PICT0407)] to verify the exact position of the doors or the kind of wall ornaments (such as niches etc.) used. The ancient wall painting has been nearly entirely lost, and it is only with difficulty and eager study that one can recognize certain classes of distinguishing ornaments used here and there among these latter ruins. The southern palaces are similar in prominence to the northern, in the most general kind of arrangement and adornment. Notwithstanding, the former were not at all reaching in importance the latter, as can be seen by the more simple disposition of its courts and rooms, and the apparently less defined manner of decoration. We have, therefore, reasons for being satisfied that at least the most prominent of the three Incasic palaces originally existing there, the northern, escaped destruction so well. The wall bordering the plaza at the west is much ruined in its upper part, and it is difficult to reconstruct all its original details. The material used in the principal palaces is only adobe bricks, quarry- stones and wood. Bricks of adobe compose nearly all the wall work. They are well made, very hard and heavy, and of different sizes, though always 9 What Uhle describes as the receding frame of a doorway, is better explained as a recessed frame, or, in today's usage, as a "double-jambed" doorway (editor's note). 14 Explorations in the Pisco Valley large. Innumerable are the sizes of adobes used. No rule exists in this matter. But to give an idea of the sizes used, I measured a few bricks, the sizes of which are given by me here: [p.27, (PICT0408)] 38x26x13 44x30x1O 53x24x13 58 x 14'0 62x28x1 1 80x26x12 It is to be understood that such extreme sizes as the last mentioned are absolutely most uncommon and that their application was by no means for certain particular purposes, as they are found in composing walls like all other sizes. The sizes given above justify the general observation made sometimes on ancient ruins, that the bricks used in Incasic constructions were generally the largest of all used in pre-Spanish Peru. Quarry-stones compose mostly or everywhere the foundations of the walls. The lower part of the walls is built up of the same material, to different heights in different rooms, according to the necessities or particular purposes of the rooms. Walls consisting of quarry-stones up to the height of one meter from the ground, while the upper wall work is made of bricks, are not uncommon. Besides this, large quarry stones were used, in addition to wooden beams, as in other Incasic ruins, for roofing doors, niches and windows with lintels. Quarry stones were also used for flooring and for giving the lower part of the side walls to bath-rooms, as well in the construction of the water channels in the subterranean cana- [p.28, (PICT0409)] lisations. Some quarry-stones used in the construction of the bathrooms are of extraordinary size. In particular, there is one in the bathrooms of the southwestern palace (see plate 1188), which measured originally (after this modern mutilation) about 1.5 meters." It has been moved in search of treasures below it. Another large stone is that which stands upright in the eastern side wall, under the inner opening of a water channel let through the wall in one of the baths of the northern palaces (see pl. 1187). The quarry-stones in the construction of walls are laid in mortar of common clay. All walls of stone and brick also had a cover of clay (below the painting), which in some parts (of higher importance) was apparently better made than in others (of lower importance), as in much of the latter it has been destroyed by the passage of time. This is, for instance, the case at the back of the eastern prolongation of the front wall of the northern palace 10 In this case, Uhie is presenting the dimensions of a block with only two dimensions accessible to measure (editors's note). 1 Uhle's prose is sufficiently tangled that we cannot, from what he has written, ascertain whether he meant the original dimension of the stone, or the dimension after the mutilation. Evidence from Tambo Colorado suggests that it is the dimension after mutilation, such that the original dimension may have been even greater than 1.5 meters (editor's note). Letter of September 1 1 and 24, 1901 15 and at the western facade of the western wall of the northern palace. Boulders from the river were sometimes used in construction outside the principal palaces. I do not remember to have observed any wall of the principal palaces constructed of tapia. It seems that tapia-work did not correspond to the taste or the practice of the Incasic architects. Outside of the principal palaces there are a great deal of houses made of tapia, both east and west of the palaces, and in other [p.29, (PICT0410)] parts of this section of the valley and other neighboring sections of the valley too. But in all these cases, it may be observed that they may not be of Incasic origin, though the houses may have been built under Incasic influence, because constructed after the Incasic conquest. The latter is in some of the cases proved by the existence of square niches, the lowest type of those types of niches used by the Incas. But the niches are few in number, so corresponding less to the principle of the adornment of more or less all walls by long series of numerous niches. Besides this, walls are mostly less well made, worse in cover, worse in elevation, worse in alignment, so that it seems generally to be easy to decide if a house has been constructed by Incasic architects or only under Incasic influence by other people of the valley civilizing themselves under such influence. That this is the case is clearly proved by the parallelism with the development of indigenous manufacture after the Incasic conquest. Because we observe there the same continual mixture with and reception of Incasic stylistic elements in objects made by the indigenes. There are, nevertheless, some curious remains in the narrower district of the principal palaces, which are consisting of pure tapia. These are visible in plates I 1 82 and I 1 83, behind the northern front wall of the large inner plaza, east of [p.30, (PICT041 1)] the principal northern palace. They consist of several irregular terraces arising in about seven steps, one above the other, according to the gradual elevation of the natural slope of the hill. They are much destroyed, but consist of pure adobe, and have walls of tapia. They are separated from the eastern wall of the principal palace by a street-like area, which, however, is much wider at its upper end. Beside this is a very high isolated tapia wall, 4-5 meters high, with an entrance left in it is to be observed (see pl. 1 183). According to the openings visible in the southern front to this wall, a roof, perhaps in the shape of a roofed gallery, may have originally been in front of it. Those terraces and this wall do not fit well with the system of the other ruined palaces, as the orientation of their walls diverges from the general direction of the walls of the palaces. On the other hand, they are standing in some connection with those of the palace; the sign of this is the four doors provided in the long eastern prolongation of the solemn southern front-wall of the northern Incasic palace. It was undoubtedly the intention, in the construction of this prolonged wall, not to allow any disturbance of the refined impression given by the principal plaza by structures of inferior quality standing openly at one side of it, but at the same time to give full access to the plaza from those structures [p.31, (PICTO4 12)] of inferior quality, as from back-rooms of less importance. The 16 Explorations in the Pisco Valley simplicity of the four doors left in front of those structures stands in accord with such an explanation of the facts. Though higher than any other doors in any other part of the ruins (as far as such are preserved)-2.20 meters high, clear above a threshold of stone, which is about 0.40 meters high -these doors are not wider than the smallest and narrowest doors existing in the palaces (in the middle 0.56 meters, at the base 0.65), while at the top they are narrower than any of the latter (0.43 m.). Besides this, though being in the wall facing the plaza, these doors have no adornment in the form of receding frames. According to this, I may express my opinion, that even the shape of the doors left expresses that they only served as entrances to back-rooms, and I might easily admire the manner in which this fact was expressed by the shape of the doors without damaging, on the other hand, the solemn impression of the whole northern front of wall of the principal plaza. For all that which I discussed above, I might express the idea that those back rooms were occupied by indigenous people of the valley or coast, standing in some nearer relation to the Incasic household, but without participating in the principal honors of it. The wall construction made of tapia indicated a coastal origin of the people staying there. An indication in the same direction is given by the contents of a child's burial discovered by me there. This burial contained a black [p.32, (PICT0413)] Incasic plate and a black bottle- like vessel with some small indications of corn-ears. The latter kind of vessel is not especially Incasic. But both vessels are such as were used on the coast after its conquest by the Incas, apparently more to the north of the valley of Chincha, than south of it. Both small vessels, which have no especial importance besides that, have been incorporated in my collection.'2 The principal palace forms more or less the middle part of the northern side of the central plaza. This circumstance alone would be sufficient to indicate its superior importance over the other two palaces. Its decorated front is prolonged, as much to the east as to the west, so as to form a unified front along all the northern side of the plaza, about 192 meters long, while the palace itself occupies only about 57 meters of the front-line. So, to a certain extent, all constructions of the northern side of the plaza appear as having been brought under the domination of its prominent importance (see plate 1171-1175 with the supplementary plates 1182-1183). While the southern palaces are lying between the plaza and the cultivated plain of the valley, the principal palace faces all the valley, leaning with its back against the hill- terrace, which rises about 40 meters above the plaza. The palace itself rises in steps, gradually, from the front backwards. This secures a splendid pro- spect from various [p.33, (PICT0414)] of the interior rooms, behind the front rooms. Apparently the construction was made in a way that the foot of the hill has been dug a little lower than the elevation of the floors wanted by the architects. Gradually the floor rises from the front to the back, in correspondence with, but not according to, the original, natural elevation of the ground. For instance, the floor of room 65 (see section 4) is much lower than what corresponds to the natural declivity of the slope of the hill. In this 12 We have not been able to locate these two vessels (editor's note). Leffer of September 1 1 and 24, 1901 17 way this effect has been produced, that the back wall of the palace behind gallery 66 is more or less exactly on the same level as the adjacent part of the hill. This manner of leaning house constructions against natural elevations of the ground, by inserting them, to a certain extent, in the foot of those elevations, is not unknown from other Incasic architectural works. For the same manner of construction is to be observed in the convent built for the use of the Sun virgins at Mamacona (Pachacamac, valley of Lurin), and in a similar way the corresponding convent has been constructed upon the island of Coati (Lake Titicaca). The general disposition of the rooms is, to some extent, the same in all three Incasic palaces. It seems that the Incas had developed a certain ruling type for the construction of houses. [p.34, (PICT0415)] We observe something like that in the general uniform character of the three palaces. This character is quite different from that which is to be observed in the houses east and west of the central plaza, which were probably dwelling places of the indigenous caziques (chiefs) of the valley. The disposition of the rooms of the latter houses does not seem to observe any rule in them. This circumstance would be sufficient for proving how much more developed Incasic architecture was than that of the coast. The three Incasic palaces are uniform in the following respects: 1. A solemn gateway at the middle part of the front with a short kind of gallery behind it, leading to a court 2. A wide, square court, which shows some similarity with the Roman atrium and the court of Spanish houses, by having 3. Rooms on all four sides of it. 4. Only a few narrow, long gallery-like rooms at the front of the palace, between the front and the court, more numerous rooms at the sides of the court, the most extensive rooms at the back. 5. A roofed, and somewhat elevated gallery at the front of the palaces. Undoubtedly the southwestern palace was of a higher importance than the southeastern, as it reveals a few more similarities to the northern palace: first, a second, more intimate [p.35, (PICT0416)] court at the back of the first wide court of the palace;13 secondly, a tower-like building at a corner of the front of the palace; third, a similar division of the rooms at the one, less important side of the courtyard (with back-rooms communicating with the court by small passages between other rooms). The appreciation of the value and significance of the special rooms of the northern palace is to be preserved for later a part of my report. I wish to discuss now only the following: This northern palace has three courts succeeding each other, in decreasing size, from the front to the back. The principal rooms are apparently those 13 Uhie conceived of each palace as a single building, thus when referring here to the 'first court of the house' he is referring to the first court of the northern and southwestern palaces (editor's note). 18 Explorations in the Pisco Valley lying at the back and at the left (western) side. Nearly all rooms were roofed. Even the front of the palace formed a roofed gallery, about 7 meters wide and about 0.30 meters above the level of the wide middle plaza. This gallery runs from about 10 meters west of the western end of the palace, where it was closed by a side wall, over all the northern front of the wide middle plaza, nearly to the end. The wooden bases of the wooden pillars (up to 0.1 [to] 0.15 m thick), which formerly supported the roof, are still existing in the ground and visible on the surface. I indicated the places where they can still be observed in my map 1:200 of the principal palace. They are found, successively, at these distances from the end point, starting [p.36, (PICT0417)] from the western first one: distance distances between from end point 3.10 3.10 6.20 6.40=2x3.20 12.60 20.00=6x3.30? 32.60 3.50 36.10 10.50-3x3.50 46.60 7.00=2x3.50 53.60 3.30 56.90 2.60 59.50 14=4x3.5014 11.40 70.90 4.00 74.90 It may be said, therefore, that the wooden pillars which supported the roof were originally standing, in general, at distances of 3 meters, more or less: 3.10 - 3.20 meters on the west of the principal gateway of the front, and about 3.50 meters on the east of the gateway. The wooden pillars stood about 14 The 14 that Uhie uses here results from adding the 11.40 to the 2.6, and intended to show how, despite the apparent variation, the basic module is retained (editor's note). Letter of September 11 and 24, 1901 19 0.30-0.40 meters inside the ledge, 30 cm high, built of quarry-stones. Remnants of original wooden pillars may be occasionally observed close to the front wall, too. The palace forms a square which is not quite regular, as the western side and the northern measure about 59 meters, the southern (front) and the eastern side about 57 meters. The main entrance is laid exactly to the middle of the southern front, but the first court, which is nearly square, as its length measures about 29 meters, is entered about 12 meters from its western side, as the court is lying in the eastern part, rather than exactly in the middle, of the palace.'5 Forming a square (rhomboid) with sides 57 - 59 meters long, [p.37, (PICT0418)] it is clear, that the angles of the walls of the building are not right, that is: none of the principal angles is right: the angle of the southwestern corner has only 89 degrees, the northwestern about 93. It is true, that in this the palace corresponds little to the rule dominating Incasic buildings, that the angles are right. However, it never could be ascertained that this rule was always followed strictly. Besides this, the angles given to meeting walls in this building are quite different from those given to walls in buildings of other than Incasic origin." For, it cannot be denied, that even so, a certain regularity is ruling among the angles at which the walls meet. One has the impression, standing inside, that all the building follows a rectangular plan and only the use of a compass proves the contrary, and the sense for regularity followed in the construction of the building can even be observed by looking at the map of the building (see map 3). The inner part of the building is divided by several main walls into several (about 7 different) blocks (see map), which are each subdivided in many different rooms of very different size. I observe the following distribution of rooms among the blocks:'7 1)2,8 2) 4, 5, 6 (with 7) 3) 10-14 4) 43-52 5) 53-65 6) 19-30 7) 31-41 [p.38, (PICT0419)] This block system seems also to indicate that the work is planned by a thoughtful architect, who proceeded, after much deliberation, in the direction that each room of a certain purpose might get the right location within a larger whole of similar rooms of similar purpose, in order that each block 15 As can be readily seen from Uhie's plan, the central axis of the court lies to the east of the main doorway, which lies on the central axis of the palace as a whole (editor's note). 16 Compared to Peruvian ruins of non-Incasic cultures, 93 degrees is very close to being a right angle (editor's note). 17 The following numbers refer to the room numbers Uhle assigned to the rooms in his plan of the northern palace (editor's note). 20 Explorations in the Pisco Valley comprising several rooms might correspond to a certain category of rooms which were to be represented here. Without yet beginning a special appreciation of single rooms, I may, however, express here as curious in every way, how different are the facilities for entering the various rooms from the southern main court. Rooms 2-8 have a direct entrance from the court. Rooms 43-52 for instance, can be entered only from the inner court 42. The same is the case with rooms 53-65, even for the gallery 66. But it is still stranger that for reaching the entrance to the western rooms 19-30 one had to go back to the middle court 9, and even rooms 31-41, lying at the front at the palace, can be only reached in the same way, from the middle court 9; to enter them, it being necessary to pass all the block of the rooms 19-30. In a work apparently so well disposed, it is clear that such a disposition of entrances answers to superior reasons, so it is to be hoped that, as the whole building has been [p.39, (PICT0420)] well laid down in the disposition of all the existing details, the accurate observation of these details will perhaps teach us their original uses in this palace. Before entering into the manifold manner of decoration preserved in this unique ruin, it may be proper to treat of roofs and doors, as far as anything can be said about them. If one looks at the section given by me as No 4 of the palace, one will observe that generally each two walls belonging to the same room are of the same height, while other adjacent ones are of different heights. From this, as from the observation of the photograms, it can be assumed that the Incas did not follow our system to roof one house by one roof all over,'8 but that the different rooms had separate roofs, more or less each like separate pavilions. In this way the inner coherence of the house may seem to have been the loosest possible. It is known that, for instance, in some provinces of the Netherlands, the different parts of one house are of different heights and roofed each separately by roofs of small extension. But the roofing is at least coherent there in itself, it is, therefore, always a building in itself on all sides. The manner of roofing used by the Incas may perhaps more closely correspond to the system used by the Chinese, if I have a correct impression of their architecture of houses. With this pavilion system, the coherence [p.40, (PICT0421)] of the building is only produced by the walls enclosing blocks from the outside, and separating them from others. According to section 4, it is to be supposed that all parts of the house (excepting only the courts) were roofed. For instance, of those rooms whose walls are represented there in section, only rooms 32, 27, 22, 60, 54, and 65 seem to have possessed regular roofs. Passages, like 63, 59, and 18, may have been entirely unroofed and sheltered perhaps only by the projecting roofs of the adjacent rooms. Rooms like 3 1 (a small terrace is visible in this room) or 29 may have been unroofed; at least it is not clear what kind of roof may have been there. On the other hand, it is to be supposed that many rooms, which 18 As mentioned before, Uhie considered the entire compound of the northern palace (as well as of the southern ones) as a single building, or house, instead of understanding each room as a single building, and the whole as an agglomeration of such buildings (editor's note). Letter of September 11 and 24, 1901 21 might easily seem to have been unroofed, were sheltered at least partly by roofs. The northern side of the widest court,'9 in front of the southern main entrance, possesses a terrace about 6 meters wide and 0.70 meters high. It runs the length of the court from east to west (see pl. 1 164). This terrace may have been roofed. It is still much surer that the smaller court 9 behind it must have been roofed, in part or for its entire extent. A chamfer,20 which can be recognized at one of the northern walls of that court (see pl. 1 164), is the proof of this fact. Besides this, the side wall of this court, rises symmetrically in steps in the west and in the east from south to north, which may indicate that this symmetrical [p.41, (PICT0422)] growth of the wall on both sides answered to some special purpose, as for instance a common roof joining both side walls and sheltering the inner side of the court, -or only its northern part, or its southern part, too. One can imagine how pleasant may have been the stay in this court, with its low-pinnacled southern front wall, which left a nice prospect over the lower court 1 in front of it, if it was roofed gallery-like. The same may be supposed of the inner court 42. Other chamfers,2' which indicate that rooms had been roofed, have been preserved in room 33, on its southern wall, and in room 3, at the eastern side. In other rooms one distinguishes that one high wall forms a ledge at the height of other adjacent walls of the same room: always an indication that the ledge served for supporting the beams of the roof of the room. This is the case, for instance, in rooms 3 (west), 7 (East), 11, 12 (west), 61,62, 64 (west), and 65 (all sides). The doors are of two classes: those adorned by a receding frame and other, simple ones. The former are, it is to be regretted, all open at the top.22 Many of those of the latter kind are complete. The photograms taken allow us to recognize several of them. All doors, and framed and unframed gates answer to the known Incasic door-type, by being trapezoidal in their elevation. Three of the doors measured by me are represented in their [p.42, (PICT0423)] basic form in figures 6, 7, and 8. Number 8 is a typical door of those of the eastern extension of the northern front wall of the wide plaza of the ruins. The gateways, which possess a receding frame for decoration, generally show it only on the outside. An exception is that wide gateway (it is 3.3 meters wide at the middle part) which intersects the mountain road for closing the wide plaza at the eastern end. This possessed a receding frame as decoration on both the inner and the outer side. Gates with receding frames always signify a greater solemnity of the entrance for which they are made. It is, therefore, worthwhile to state the places where they are found in the northern building: 1 9 Uhie used the term plaza here. To not confuse the courts within the palaces with the plaza without, we changed it to court (editor's note). 20 What Uhie means by chamfer is actually a horizontal chase or groove in the wall that accommodated the thickness of the roof abutting to that wall (editor's note). 21 What Uhle means by chamfer is actually a horizontal chase or groove in the wall that accommodated the thickness of the roof abutting to that wall (editor's note). 22 I.e. the doorways are missing their lintels (editor's note). 22 Explorations in the Pisco Valley Principal gateway at the front of the palace between the courts 1 and 9 between the court 9 and the passage serving for interior communications at the entrance to the small passage which communicates between court 9 and room 10 between court 42 and room 53 and between court 42 and gallery 53. Therefore, the principal entrances leading from the main courts to the inner rooms are decorated by receding frames around the doors. It is remarkable that the entrances to rooms 2-4, 6-8, do not take part of this particularity.23 [p.43, (PICT0424)] Respecting the southern palaces, I was able to identify only one entrance of the same superior kind in the interior of them, that is that which served for the communication between the two large inner courts of the southwestern palace. Besides this, the main entrances of these palaces show the same manner of decoration. It seems to me remarkable that, while the northern palace possesses five gates of this kind, the southwestern palace possesses only two of them, and the southeastern only one. Undoubtedly this fact illustrates the difference of rank of the inhabitants of the various palaces, more than any other detail. If we assume that the decorated frame of the gates indicated, to those outside, that the inhabitants were of the rank of Inca, then the inhabitants of the northern palace are marked in this way even in the most concealed rooms of the interior, those of the southwestern palace were so marked as far as the second inner court, while those of the southeastern palace were only marked in this way at the main entrance of the palace, to indicate their rank to the general population in the market and plaza.24 I come now to various manners of adornment and decoration used in the construction of this building. If one is not convinced by the general inner disposition of the rooms, that this was a building made with the intention to allow many kinds of convenience and to show luxury, one should be [p.44, (PICT0425)] by looking at the various kinds of adornment. The following details have contributed to this effect: innumerable niches of various kinds in all rooms, combined with the use of windows, which, of course was more practical than ornamental. ornamental crests of masonry on various walls, in two kinds. In several of the cases, in which they have been applied, they may have at the same time served as windows, forming openings below the roof. Figural ornaments in relief on walls 23 Uhle may not have noticed that in Inca architecture the double-jambed doorways almost always lead not into a room but into an open space or court from which other rooms are accessible (editor's note). 24 There exist the remains of yet another double-jambed doorway beyond the second court in the southwestern palace at the west side of the court, which may not have been visible to Uhle at the time (editor's note). Letter of September 11 and 24, 1901 23 Galleries forming a kind of belvederes, provided with artificial open-work balustrades Setting, in a few cases, a second floor above the lower one for providing better lookout or access to galleries which served as belvederes. Wall-painting in different colors, and Painting the interior of niches, and partly of windows too The special map, scale 1:200, of the northern palace lends ample proof of the extensive use of niches for the adornment of rooms in this building. Several hundreds of them may be found in it. Very few large rooms lack niches entirely, as do rooms 5, 7, 14, 10. In addition, niches are lacking in some small rooms, like 48, 49, 51, 26, 64, 20, 28. [p.45, (PICT0426)] All other rooms are opulently outfitted with niches, which often form long series of up to 1 1 niches on each wall, if the room is large. These niches are applied mostly at a convenient height above the floor. I observe three kinds' of niches in this palace: 1. square ones: about 40-60 cm long and 30- 50 cm deep 2. higher than wide, in different sizes 0.30-0.40 meters wide, 0.40-0.70 meters high. _Y :i Nc 3. Niches with two recesses the one in the other, the larger outer one frames the smaller inner. These are mostly 0.60- 0.70 m high and receding about 0.40 m in depth. While the former two kinds are always in convenient height above the ground, so as to enable one to make also practical use of them, these of the latter kind are nearly purely ornamental, and often applied at such a height as to secure best this ornamental effect. For this reason, those of the southern front of the palace are found with their base 1.95 meters above the floor. Those in front of room 60 are about 2.45 meters above the floor and so forth. These are also found in some rooms which seem to have pos- [p.46, (PICT0427)] sessed a superior ceremonial importance, such as on the three 25 [Uhie added in margin]: It is worthi mentioning that high niches, rising immediately from the floor, with receding frames around them as ornament, as they are found in the convents at Mamacona and of the island of Goati (Lakce of Titicaca) as well as in the palace-like Incasic building near la Centinela (of Tambo de Mora, valley of Chincha) (also a convent??.) have not been applied in these palaces. Their application served, perhaps in all cases, still higher ceremonial, or perhaps religious purposes. 24 Explorations in the Pisco Valley wall-sides of the courts 9 and 42, on the western wall of room 16, which presents itself to the visitor who passes from court 9 through its western door, and in front of and on the northern and southern walls of room 54. In the latter cases these niches are found, similar to the others, being, in parts, only 0.85-1.30 meters above the floor.26 There is no doubt that these niches of the mentioned third type have a representative character. For this reason, we meet with them in rooms of superior ceremonial character, as in courts 9 and 42. They are absent in court 1 which lacked this higher importance as is proved by the absence of framed ceremonial doors in front of rooms 2-4, 6-8 too. Owing to their higher decorative and representative character they were applied as wall decoration along the southern front of the palace, and the same kind of niches, applied in the fronts of rooms 60, 65, 47, 50, in each case double, overtopped the roofs existing before them (as can easily be observed from section 4, in the case of the niches in front of room 60 and 65), which shows that they were decorative and representative for the whole building from the outside, not only to the inhabitants of the palace themselves. Having proved this prominent cere- [p.47, (PICT0428)] monial character, we are allowed to lay especial stress upon the fact, that besides the courts 9 and 42, which though intimate, have some higher public ceremonial character, the only rooms quite at the interior of the palaces which possess this kind of niche are rooms 16 and 54. As room 54 was still more hidden to the outsider, than [room] 16, I venture to declare with all certainty that 54 was the principal room of residence of the authoritative person or persons who inhabited this building. Ornamental crests made of masonry are met with in the following parts of the northern palace: Upon the wall fronting the rooms 47, 50, and 52 between the room 45 and the eastern palace wall upon all four walls of rooms 47 and 50 and upon the back wall of room 45. These ornamental crests represent two types, one in the shape of pinnacles, falling off at one side in three steps in this way O.3 / directed to the left, which may also be found in a vertically inverted variation. [p.48, (PICTO429)] 26 [added in the margin]: Between the first and second court of the southwestern palaces there are found niches of the third kind, the inner part of which is open window-like through the wall, as the niches, for instance, in the Incasic palaces of la Centinela (Tambo de Mora). Letter of September 1 1 and 24, 1901 25 The other: that is: the top of the wall cut out in this way, that both sides rise in 4 steps each.27 The first type is represented: upon the wall between court 1 and 9 (and east of them) and upon the eastern wall of room 50 and " " northern " " " 47 The first type, inverted, is found: upon the wall fronting rooms 47, 50 and 52 " the northern and western wall of room 50 and " " eastern wall of room 47 The second type (in each case only one figure of this) upon the southern wall of room 50 "9 ". southern and western wall of room 47 "4 " northern wall of room 47 The same figure, twice: upon the northern wall of room 45. For the representation of these wall-ornaments please compare photograms 1 160, 1 164, 1169, 1177. I may mention at the same time, that the front walls of the southern palaces were decorated with the same kind of pinnacles of the first class.28 I measured some which were left, and found that they had these proportions: A A W ~~~~~~~~~~~~~~~~~~~~. v1 _ [p.49, (PICI0430)] Now it is clear that upon the walls of rooms 45, 47, and 50, these pinnacles, as they seemed to have formed openings below the roof, may have served as windows for providing light to the interior at the same time.' But the ornament is of no practical use originally. Upon the wall at the back of court 27 The step motif described here by Uhie has in fact only three steps. The opening most likely had a lintel on top and thus formed a stepped window (editor's note). 28 Te"pinnacles" that Uhle saw on the front wall of the southern palaces can today no longer be observed (editor's note). 29 In room 50 it is by no means clear that the "'pinnacles" were below the roof. The walls form a ledge all around the room just below the "pinnacles," which very well could have served to support a flat roof structure (editor's note). 26 Explorations in the Pisco Valley 1, and upon the front wall of the southern palaces its application was purely decorative. The same is to be said of those cases in which the pinnacle appears as ornament rising above the roofs of rooms lying before it, and standing in front of other walls, as, for example, behind rooms 4 and 5, and standing in front of rooms 10, 12, 13 and 14. It might be possible that the shape of these pinnacles had no other significance than to serve as an ornament. We have seen that Incasic architectural ornamentation seems to have followed a certain system, a certain code. Many architectural details were expressive for signifying the rank of the inhabitant of certain prominent palaces. But this observation by itself, though leading to the idea of a similar expressive character of those ornaments, would not alone be sufficient to let this signifying role be asserted with a high probability. It happens that I sent several scepter-like symbols to the University from Ica. All were collected from graves [p.50, (PICT043 1)] of caziques. All showed a distinctive manner of decoration or figures of totem-animals at the top, or plating30 and gilding. You will remember, dear Mrs. Hearst, that one of these scepters, one of the most principal ones, bearing some carvings at the top, had no other ornaments besides a series of pinnacles, falling off on one side in (I believe) three steps, so: These pinnacles were silver-plated and gilt alternatively. Undoubtedly these very pinnacles had a signification, and undoubtedly they were marks of a high rank, as of caziques, as on a great deal of the other prominent scepter-like symbols the same ornament often is repeated, while on common objects it never occurs. I conclude from this that this pinnacle- ornament was symbolic of a high rank wherever it was applied. It follows from that, that the pinnacles in those architectural works had the same value, the same signification, and that its having been applied on any of the palaces was sufficient to show every visitor: "There lives a high cazique, respect him!" [p. 51, (PICT0432)] That is, to my mind, the signification of this ornament. This alone sufficiently declared that the residents of the three palaces were prominent caziques. If the pinnacle is expressive, the other ornament, in the shape of two stairs, one meeting the other in the centre, undoubtedly will have likewise been symbolic. I remember that I met with this, or a similar ornament, often in combination with the figure of frogs, as for instance in the "Sun-pillars" 30 By plating or plated Uhle meant silver-plating or -plated (editor's note). Letter of September 1 1 and 24, 1901 27 near Hatuncolla (N.W. of the Lake Titicaca, see the work of George Squier). This ornament is very common in Peruvian especially in Incasic archaeology: I consider it as being the complete representation of the cross-figure, which had religious signification. Its existing in combination with frogs upon the mentioned "Sun-pillars," and in the same manner of combination elsewhere, is sufficient to demonstrate this religious value. Now, the figure is the lower half of the figure shown above. It must have been unavoidable, that if the whole figure had widely known religious signification, the presence of the half of that figure must have reminded people of the same signification. The latter (half) figure easily cannot be imagined without one being reminding of the complete figure. If this is so, it is clear, [p.52, (PICT0433)] that the figure _ ~~~g in architecture must have also served a symbolic purpose. Of course, the half figure took the place of the complete figure by adaptation to architectural difficulties. If one were to attribute to the previously mentioned pinnacle- ornament the power to signify some rank in the political hierarchy, the half-cross symbol would express, according to my mind, something which has to do with religion or religious hierarchy. Though the symbols mentioned above signify a high rank of the resident, or of several residents, of the palace, I suppose that the most valuable symbol, that which marked still a higher rank, is found in another detail of the ruins. Number 32+41 is a tower-like building at the southwestern corner of the northern palace. The southern front-wall is incomplete at the top. According to many parallelisms it is to be supposed, that it showed originally more or less the same finish as the northern corresponding wall of the same tower. The latter seems to have been roofed at the height of 5.50 meters above the level of the terrace which passes in front of the palace. The wall is there about 0.70 meters thick. Above the outer face, at about its middle, this wall shows a shield-like ornament made of adobes,3" about 2.40 meters long, 0.70 meters high, about 0.30 meters thick. It is not quite complete at [p.53, (PICT0434)] its western end. It has two fronts, a northern and a southern one. Both show a curious ornament in bas relief, each six times repeated. The ornament has more or less this shape: 31 This ornament is made of tapia, or rammed earth (editor's note). 28 Explorations in the Pisco Valley * w _- tt - J _I - Plates No. 1158 and 1159 show it clearer. This ornament seems to possess some similarity with a plumed helmet. The location on the wall where this figure is represented, in order to make it visible from afar like the sign-board of a merchant, and the manner in which it has been applied, by setting it on a special small wall, which served no other purpose but to bear these signs, upon the architectural walls of the tower, excludes, to my mind, the idea that this board made of bricks followed any other purpose than to signify, to the people looking at the palace from the outside, the character of the building with respect to the rank and position of its residents. Among the three symbols used, that in shape of pinnacles, that in shape of the half of a cross, and this, this last seems to have been the finest both in technique and in the particular way of its application. I am in no doubt that the figure really represented a helmet with two plumes. This is a symbol quite appropriate for expressing [p.54, (PICT0435)] a rank of high degree. Undoubtedly such a rank, like that of an imperial general, or the first person of an empire, deriving his authority from the divinity itself, may have been expressed by this symbol when applied this way. Without the intention to go too far, I wish not to pass over the fact in silence that the Inca bore, as mark of distinction, two hawk-feathers, and he has been represented infinite times as bearing them, of course without the right knowledge of the manner in which he put them on (as up to the present the Inca never has been represented well with the distinctive llautu,32 because all such pictures and drawings have been only made by people who had never seen a llautu themselves). Two other kinds of particularities of this palace are tower-like constructions with two floors each and open galleries enclosed by balustrades. There are two tower-like buildings in this palace, one (No. 32 and 42) at the southwestern corner, the other (No. 65) near the back of the palace. Both are different in construction and with respect to the purposes they served. But it may be considered that both were constructions made for greater convenience, probably both with a certain interest in giving a view over the environs of the palace. The first one, No. 32+41, had two floors, the first [p.55, (PICT0436)] is that floor labeled No. 32 in the map (scale 1:200), as well in section No.4, the other is labeled No. 4 1. The lower room was entered by a door from the side of room 31 (see also plates 1 158,1 159). The other was attained indirectly by passing through rooms 33-37 and 38 and through a door at the eastern end of 32 The llautu is a headband worn by the Inca (editor's note). Letter of September 11 and 24, 1901 29 room 41. Room 32 had been roofed at the level of the floor of room 41 (see section 4). The lower room has been decorated with niches, and light was brought in through six windows. The upper room, if it had been roofed at the level of the common top of the walls would only have had a height of about 1.40 meters or less. But the southeastern corner shows a square pillar-like elevation 0.40 meters above the common top of the wall. I might suggest that originally all four corners had such pillar-like elevations of masonry work, and it was upon these that a roof rested. A height of about 1.70 meters between the floor and the roof of room 41 would have been quite sufficient to stand upright, and the difference of about 0.40 meters between the height of the pillars supporting the roof and the common level of the walls would have secured enough room to profit from the excellent prospect which was to be had over the ancient settlement from this height of about 5.80 meters above the level of the large middle plaza in front of the palace. [p.56, (PICT0437)] The other tower-like building is represented by No 65 on the map. It had only one entrance, from the South, on the lower floor. All four walls show a step-like landing, which must have supported a roof 3.10 meters above the lower floor. Above the landing the three walls of the Eastern, southern and western sides are crowned by open-work balustrades, which prove that the roof served as a bellevue-like open gallery intended for access. A door-like interval exists in the wall which separates gallery 66 from the interior of room 65, 3.20 meters below the level of that gallery. Besides this, it is to be observed that gallery 66 had no other entrance than from room 65. For the small one, which exists now at the back of court 42, bears all marks of recent origin in its masonry work, which is confirmed by the present tenant of the hacienda Pallasca, who has lived about 20 years in this hacienda, and affirms that he opened this entrance for erecting the cross upon the gallery, which adorns it still. For entering gallery 66 it was, therefore, necessary in ancient times, to enter room 65 and to ascend from it, perhaps by a kind of ladder, to the roof of the same room, which was on the level of gallery 66 and itself formed an open gallery. This conclusion is unavoidable. But no certain marks of what this ladder was, or in which special part of room 65 the ladder was placed, are visible now. [p.57, (PICT0438)] We find three open galleries, adorned by open-work balustrades about 1 meter in height in front of them, in the northern palace. Several of my photograms show them. They are each in some way different from the other. No. 66 is an open gallery, 57 meters long, in the east 2.10, and in the west only 1.30 meters in width. It crosses the entire width of the palace, at the back, with the 2.60 meter-high back wall of the palace behind it, which separates the palace from the slope of the hill, which there abuts the back wall, about 2.10 meters above the floor of the gallery (0.50 meters below the top of the back wall).33 The gallery rose above all the walls and roofs in front 33 cf. UhIe's sections 1 and 2, which are also labeled figures 4 and 5 (editor's note) 30 Explorations in the Pisco Valley of it and nearby, being itself 10 meters above the level of the large irregular central plaza of the ruins. The view from there, therefore, commands all the building, its environs, the cultivated plain, and the river, as well as the other bank of it. It must have been an ideal place for an authoritative person like the Inca for enjoying the view of that beautiful valley, which like others, stood under his command. When there, nothing could trouble his noble enjoyment, the gallery being removed from all the trouble of the house and of the environs, which it overlooks, by having only one small access from one of the most innermost rooms of the palace, and nothing at the side diverting his attention by the narrowness of the gallery and its protection by [p.58, (PICT0439)] a higher wall at the back. The second open gallery is that which was formed by the roof of room 65. It was enclosed, as I noted before, by an open-work balustrade on three sides, while at the fourth it was bordered by a low, niched wall with a door- like entrance from gallery 66. It served, therefore, as an extension of gallery 66, and provided, thereby, a good view to three sides. The third open gallery is found contiguous to bathroom 57, indicated as No. 58 on the map. The bathroom is about 1- 1.50 meters below the level of the gallery. The gallery is in front of it, elevated above the general level of the ruins about 5 meters, but rising only about 1.10 meters above the small passage, No. 18, in front of it. This gallery is only 4.50 meters long, and 2.50 meters wide, having passage 18 in front of it, the bath-room, about 1-1.5 meters deeper immediately at the back of it, and the walls of other rooms (17 and 60) at the sides. It can be scen in photograms 1161 and 1 163. An open-work balustrade shuts it up at the front. This open gallery was apparently destined for the convenience of the more intimate private life and was undoubtedly of use after the resident of the palace took his bath, for exposing himself to the fresh air or to the sun- shine according to his convenience. Besides this, I find it to have been in nice taste, that the resident of the palace did not [p.59, (PICT0442)] take a bath in a narrow, damp room, entirely closed by walls as we do, but in the open air and in front of the pleasant view provided by the country. Undoubtedly the construction of such open galleries provided by artificially worked balustrades implies the existence of a high development of civilized taste, which we heretofore had no right to ascribe to the indian antecessors of the Spaniards upon this soil. And it seems to me that the conquering Spaniards treated them too badly, in considering them only as brute heathens. It may be, perhaps, that not all inhabitants of the vast empire possessed such artful galleries or, perhaps, had not even the right to provide their houses in this way - which would not be wondered at, as we know by the work of Garcilaso, that the Inca determined the length of the hair as worn by the inhabitants of his empire, according the rank held by each and that, for instance, the right of use of golden or silver vessels was connected by the Inca as a privilege, and Garcilaso tells of the existence of many other such regulations more, -in every case we observe that Indian taste was developed in the same direction, as in our highest European civilizations, and it is really a pity, that the world's history was so unjust, as to let perish a Letter of September 11 and 24, 1901 31 useful member of advancing humanity under the pitiless hands of the perhaps more brutish Catholic [p.60, (PICT0441)] Spaniards. Another sample of an open gallery provided with an artful open-work balustrade existed at the southwestern corner of the southwestern of the three Incasic palaces (see map 2, scale 1:500). That corner of the palace is occupied by a terrace-like elevation, the details of which are now nearly entirely lost. At the southern front of it, above the two walls, which border the palace step-like on this and the western side, I observed the traces of the former existence of a complete open-work balustrade and a fragment of this perforated balustrade-work was, besides this, found by me as thrown away in a neighboring court of the same building. This gallery, with its balustrade, proved the same refinement of taste as the other galleries, mentioned before, in the first palace. The terrace elevates itself about 8 meters above the cultivated fields of the valley, which are nearly immediately at the foot of it. Only a refined taste could teach these Indians to construct there a terrace with an open gallery for enjoying the wonderful prospect, which one must have had from there in olden times. It is the same spot from which my cyclorama of the valley, plates 1200 sqq, has been taken, as this spot seemed to me by far the best suited for this purpose, long before I had become aware of the existence of an ancient belvedere at the same spot. [p.61, (PICI0443)] All open-work balustrades follow two patterns, of which one is represented by those of galleries 65, 66 and of the belvedere of the South- western palace, the other by that ot the gallery 58. The patterns are these: Both patterns correspond in some way to the pattems used in the textile art of the Kechua civilization. Both may have been produced by leaning each pair of bricks of adobe against each other.? Before proceeding to treat of the decoration made by painting, I may note here the use of windows in this palace, having missed the occasion to discuss them above. My map 1:200 will show the extent of use made of windows in this building. There are about 70 in the northern palace. Some large rooms contain many windows, up to about 10 of them. Nearly all are about 2 meters above the floor. In shape and size most of them correspond to the 1 type of niches mentioned above (pA.4), some of them (like those of the tower-like building 32+41) are similar to the 2d type of niches mentioned there. They served only practical uses, and nothing decorative can be found in them, though in various rooms they are found alternating in a higher level with 34 At close inspection, the open-work proved to be of tapia (rammed earth) and not made with bricks (editor's note). 32 Explorations in the Pisco Valley niches (of the first or second order) in a lower level. But their appearance little offended the eye in rooms of solemn character, as on [p.62, (PICT0444)] the eastern side of court 9, where the walls are pierced for windows, which give light to the rooms 11 and 12. The openings of the windows are horizontal.35 In the case of room 32 (the tower-like building), I observed that the inner openings of the window are widened below in comparison with the outer opening. 36Whether all rooms from which light was sent through to contiguous rooms were unroofed, is a question which I am not able to decide satisfactorily, as evidence of the former existence of roofs seems to appear in rooms from which others received light.37 Painting was applied as decoration in two ways: as painting of the inner [parts] of the niches and as painting of the walls. No patterns of painting are found in all the building.38 On the other hand, it should be said that there is, perhaps, no room which had no painting or at least white washing. Nearly all rooms, with few exceptions, show various colors of painting, and most rooms show a high degree of this sort of decoration. The colors used were white, yellow, and red, in general, undoubtedly all of mineral origin. Sometimes the yellow has an orangey cast or, in other cases, an olive-colored one. Some kind of lead-coloring was met with in one case, which I will discuss later.39 Samples of the principal coloring substances were taken [p.63, (PICT0445)] by me from the walls and preserved for future chemical examination. I observed that the uppermost covering color had not been the only one applied. Many covers of colors, sometimes separated by thin iayers of uncolored clay, had been applied in most of the cases (the same can be observed on the walls of the Sun temple of Pachacamac). Very often a different, older color shone through the later, covering color. When I am discussing in the following which colors were found in the different apartments of the building, I usually mention only that color which forms the most recent painting, and which corresponds to the decorative condition in 35 It is not clear what Uhle meant by 'horizontal' (editor's note). 36 By "widened below" Uhle could have meant that the window sills on the inside are lower than on the outside, or that the window sills on the inside are wider than on the outside, i.e. that the windows are more trapezoidal in shape on the inside than on then outside. However, neither of these interpretations are borne out at the site; thus what Uhle meant to say remains unclear. The sill of these windows slopes downward to the interior. Thus, on the inside the edge of the window sill is lower than the edge of the sill on the building's fa;ade (editor's note). 37 Uhle conceived of each palace as a single building rather than as a group of buildings, and thus imagined a roof that covered more of the palace compound than ever actually was roofed. As best as can be ascertained from studying the site and from knowledge of general Inca practice, all rooms in the palace receive light directly from an exterior, unroofed space. It is uncertain what evidence Uhle took to suggest that some of these spaces were roofed and whether he would have interpreted similarly if he was not thinking of the palace as one building (editor's note). 38 What Uhle probably wanted to say is that there are several patterns of paintings and that no single pattern was applied throughout the building (editor's note). 39 Later in the text, Uhle refers to this lead-coloring as "?blue" or "bluish-lead" (editor's note). Letter of September 11 and 24, 1901 33 which the palace was left at the time of the destruction of the empire. I do not need to explain, that, of course, all coloring is of old, Incasic origin, as it also answers only to Incasic taste, not to a modem one. The painting of the walls, as well as of the interior of the niches is, of course, different in all the rooms. Windows are mostly white on the reveal, as in rooms 3, 4, 12, 13, 22, 25, 27, 29, 62, and 65. The only case in which they were red on the inner side was in room 23, of which I shall say more below. In room 8, the windows were painted yellow, beneath which older, red paint can be seen. The southern window in room 13, was the only window [p.64, (PICT0446)] of the room which was red. Niches of type 1, on the inner side were uncolored, in room 8 red, in rooms 12, 13, 17, 23, 38, 44, 45, 60, 62, 63 white, in rooms 47, 50, 65 They were red with the exception of one window in the eastern wall, which was whitewashed, in room 11. They were red in the southern and northern ends, and white in the middle of room 3. There were niches of this type in two and three colors: Bottom (base)40 white, sides red: room 29, northern wall, and room 37, eastern wall. The niches of the northern wall in room 37 were yellow, and those of the western wall, red, and those in the southern wall in room 29 were yellow! Bottom yellow, sides red: rooms 55 and 62. That is, in room 55 only one of the niches on the western wall (red, covered there by white).4' The other niche of the western wall is entirely white, while all other niches of the eastern and northern side of this room are yellow at the base, and white on the sides, and on the inner upper side! Niches of the second type were red in rooms 32 and 41 white in rooms 4 and 5, red in room 6 with the exception of those of the three walls south of the southern entrance which were all white. It is not impossible that these originated from a partition of rooms existing there. 40 What Uhie means by "bottom" or "base" of the niche is what he later refers to as the "ground." It is the back face of the niche (editor's note). 41 Uhle means here that an older layer of red paint is covered by white paint. As white is a color of lesser importance, Uhle's interest in the older coat of paint, in this case, is understandable (editor's note). 34 Explorations in the Pisco Valley The greatest variety presents itself in the inner [p.65, (PICT0448)] painting of the decorative niches of the third order, which, by their manner of inner partition, favored such an inner variety. In the description of these niches, I call the innermost part of them their ground, the back of the receding frame, their frame, and the sides of the frame, contiguous to the outer face of the wall: their sides. The colors have been applied differently to the various surfaces of these niches. 42 1. Niches of this type applied twice in a pure decorative way to high walls prominent over roofs before them. Ground: yellow, Frame: red Sides: yellow Niches of passage 46 in front of rooms 47, and 50 and in front of room 65. Ground: red, Frame: yellow Sides: yellow Niches in front of room 60 (passage 18) Entirely yellow Two niches in eastern wall of room 44, in front of 47. 2. Niches of private rooms Ground: red, Frame: yellow, Sides: white Niches of room 16. The niches of northern and southern wall of stateroom 54 form an exception of all other of this order, as their coloring is in four parts: Background of the receding inner niche: yellow Sides " " f i " red Background of the framing outer niche: white. Former red painting is shining through white here. Sides t it i " red One of the niches of this room forms a further exception, insofar as the two colors of the inner receding niches appear as having been exchanged on the last western niche of the southern wall of the room, so that this now presents [p.66, (PICT0449)] this succession of colors: red, yellow, white, red. 42 To account for cases where the ground is different from its sides, and there are several more such cases, one needs a further distinction in the description of the double-framed niches Uhie describes here: the outer reveal (Uhle's sides), the outer face (Uhle's frame), the inner reveal, and the inner face (Uhle's ground) (editor's note). Letter of September 11 and 24, 1901 35 3. Niches of ceremonial courts These niches are very particular in respect to their painting. in court 42, the western niches present the colors: Ground: white, Frame: red, Sides: red the northern and eastern niches: niches:43 Ground: red, Frame: yellow, Sides: white In court 9 the first and the fourth of the western niches: Ground: white, Frame: red, Sides: red the second and the third of the western niches, both contiguous to the gate conducting to interior of apartments, and all niches of the inner northern wall:44 Ground: yellow, Frame: red, Sides: red the first and the second of the'outer northern wall: Ground: yellow, Frame: red, Sides: yellow the first and the second of the outer northern wall: Ground: white, Frame: red, Sides: yellow 4. The niches of the southern front wall of the northern palace These have the ground and the frame bluish lead colored, and the sides red. 5. The niches of the northern front of the southwestern palace These show the ground red, the frame yellow, and the sides red. The doors are many times white-washed. The upper part of the inner sides is red, the lower part white, for instance in those doors, which lead from room 54 to rooms 62 and 55. In the doors between court 42 and passage 43, and between court 42 and room 53, the upper part of the inner sides had been yellow, covered later on by red. I will now discuss the coloring of the pinnacles of the wall [p.67, (PICT0450)] and the open-work balustrades. The pinnacles, which fall off in steps on one side, seem to have been painted red in the interior of rooms 47 and 50. Those above rooms 11, 13, 14, and in front of gallery 46 present, it seems, the only example of an attempt made to apply the decorative colors in a sort of pattern. Though the original color here, cannot be very well distinguished now, the wall with the pinnacle must have been painted more or less in this way:45 43 UhIe adds in the margin: "These may have been red all over, before their last painting, according to the colors shining through yellow and white." 44 Uhle adds in the margin: "These may have been entirely yellow at some older time, according to a similar observation." 45 Uhle shows red squares painted onto the "pinnacles." Today there is no trace left of this decoration (editor's note). 36 Explorations in the Pisco Valley UL 44-~~ The open-work balustrade of room 58 shows traces of its originally having been painted red and yellow. That of gallery 66 had been decorated in stripes in this way: ,~~~~~~~7 zZn; -, -- / s-;,- ,/ 'i,/{ ,! The inner faces of the open work in the wall correspond in their coloring to the painting-color of the horizontally contiguous faces [p.68, (PICT0454)] There are various reasons for which I gave all the details of the coloring of niches and doors. First, it is interesting to take note of the curious variety met with in this respect, to give the right impression of which, the complete relation of all details seemed to be necessary. Secondly, it might be important to know which of the colors were given to the niches of each room of the building, as it might have been the case that the various colors had been applied in a significant way in relation to the different quarters of heaven, of which they were expressive. In this respect I found no proof for an application of the colors in such a significative way. Third, it seemed to me to be preferable to take notice of all details of the painting used, in order to appreciate easier the different degrees of distinction of the various rooms, as expressed by the superior or inferior kind of decoration in painting. Fourth, it was possible that single colors might have expressed a particular function for a room, the adornment of which they served, and therefore a general survey given of all manners of decoration used in this building seemed to be important. Letter of September 11 and 24, 1901 37 I exposed in another work (the poor "Pachacamac," not yet printed) that various colors were significative in the following way: white for the worship of the Sun ?blue46 perhaps for the infinite origin of the world red for the Inca, [p.69, (PICT0455)] red and blue also for the Inca (the color of his llautu, for instance) red and yellow for the "crown prince" as Garcilaso expresses the rank of that person with the right of succession who wore the "royal" llautu in the colors red and yellow, also for other immediate relatives of the Inca. brown for the caziques, chiefs etc. of the provinces, etc. Now, I must call attention to the fact47 that the only case in which bluish-lead coloring has been used for the decoration of the niches, in the combination red- bluish-lead - red, is the decorative niches of the southern front of the northern palace. It is, therefore, as red and blue were the distinctive colors of the Inca, that the painting of those frontal niches of the foremost of the three Incasic palaces was to be significative for indicating to the outside people: "This is the powerful Inca's habitation! " In that case the painting of the frontal niches of the southwestern palace, second in importance, in the colors red-yellow red might have in a similar way indicated, that that was a palace of near relatives of the Inca.48 If all this is so, as I suggest, one might also suggest that in every case of niches of the third ordcr, the painting of that part which I denoted as frame was more determinative for indicating the rank of an [p. 70, (PICT0456)] occupant, or for indicating the degree of importance of a certain room alone, than the colors of the other parts of the same niches. Of course, it would be an error to expect that the niches of all rooms of the interior inhabited by the Inca should have been painted red and blue. It would have disgusted him to be surrounded on all sides by the significative marks of his own dignity. The indication to the outside people of the painting of the frontal niches of the palace; "There lives the Inca! " may have been deemed as sufficient for this purpose. This may have been the reason that the bluish-lead colored painting is only found in the frontal niches of the palace, and in no other part of the palace. But the case was perhaps different with yellow. Yellow, red and white are found as painting in most of the principal rooms of the building, and, as we shall see later on, red and yellow were the 46 The question mark preceding the word blue is taken from Uhle's manuscript. Uhle was uncertain whether to call this color "blue." At other times he speaks of this color as "lead coloring" (editor's note). 47 In the margin Uhle has written: "M. The color seems to be the effect of a conflagration before the front." Uhle's intention is unclear, both the meaning of the "M." and the suggestion that the color is from a fire, when the rest of the text seems to regard this color as a specific symbolic color of paint (editor's note). 48 Today, the bluish-lead color cannot be observed in the niches of the southern front of the northern palace, but it can be found in several niches on the northern front wall of the southwestern palace. Could it be that Uhle was mistaken here (editor's note)? 38 Explorations in the Pisco Valley painting-colors of most of those parts of the walls of the interior which overtopped other walls and were visible from the outside. In private rooms inhabited by the foremost Inca there was perhaps no other Inca privileged to enter than one of those who had the right to put on the colors red and yellow in their distinctive kind of headdress. I wish to say, that I consider all the rooms west of courts 1, 9, and 42 (with the exception of the rooms 3 and 2, which had different entrances from court 1) as the private habitation of the Inca, who inhabited the building. For there has been much more use made of the red and yellow than in the rooms nearer to the entrance from the street. In particular, [p.71, (PICT0457)] room 54 awakes in me the impression it was the special state-room among the private inner apartments of the Inca Room 54 is distinguished by niches which are colored in four partitions, the only niches painted in this way in all the building. Room 54 is the innermost of all rooms outfitted with niches of the third order, and for this reason it must have possessed an especial importance. The Inca appeared here as in his private room, not as in a ceremonial state-room and as surrounded by walls decorated with niches of the third order on all sides.49 He appeared there as though conveniently stepping forth from a still more private room at the back and had such niches only at his right and left side, which impressed even his visitor of his authoritative position. The southern and northern inner sides are visible in photograms 1161 and 1186. I likewise consider remarkable the painting of the inner sides of the side doors leading from room 54 to rooms 55 and 62. And further: rooms 55, 62, 22 and 25 are the only rooms of this building in which even niches of the first order (see above) show painting in two colors each. Room 55 is the only one in which there occurs any niches painted in the three colors.50 This proves that much more attention has bee paid to the decoration of niches of the rooms west of courts 1, 9 and 42, than to the rooms east of those courts (including rooms 2 and 3). We have already seen that the rooms [p.72, (PICT0458)] of the blocks comprising 53-66, 19-30, and 31-41 were accessible by doors that were less accessible than all the other rooms of the wide building. It is only on this side that we find access to galleries 65 and 66, which undoubtedly were reserved for the use of the principal resident of this building, and only by passing through the rooms on this side of the building could the rooms of the tower-like building 32+41 (which are of a similar character to galleries 65 and 66) be entered. The only bathrooms of the building, 57 and 24, are found on this side. We may form now our more or less detailed opinion with respect to the determination of the special rooms of this building. According to my mind, court 1 served for the first general reception of visitors on the building; rooms 2-8 around it, which lack special decoration 49 There are niches only on the northern and southern wall of this room, as is clear from Uhie's subsequent writing (editor's note). 50 What Uhie was referring to is that the three niches in the east wall of this room have yellow outer faces and red reveals with the top reveal or lintel in white. This is indeed a unique and interesting detail (editor's note). Letter of September 11 and 24, 1901 39 on doors and niches, may have served for the general house-keeping of the palace. Guards, posted upon the terrace which faces the court from the north, to the sides of the entrance of court 9, may have received visitors there. Court 9 was of higher formality. People of the more intimate service of the Inca, and who were also probably Incas of lower rank, may have occupied rooms 10-14, which, though lacking special decoration on the niches of the interior, were, however, distinguished by a framed door at the western entrance of the passage that led to these rooms. I do not know if room 16 and court 42 (which may have only been entered [p.73, (PICT0459)] by means of the western gate of court 9 and by passing through the small gallery 15) were particular reception rooms for the Inca. But it seems to me to be certain that the series of apartments particularly inhabited by the principal resident of the palace began with room 53. I do not know with certainty the purpose for which spaces 43-52 may have been determined. But I take the peculiar shape of the decorations of rooms 47 and 50 as a hint that this series of rooms may have had something to do with religious purposes.5' Room 54 was, in my opinion, the private state-room of the resident of the palace, who was probably the Inca himself. The persons of his most intimate service were, perhaps, waiting on him in room 53. One observes that the only entrance to apartment 54 leads from court 9 (western gate) through passage 15,52 court 42 and chamber 53, and one further observes that, as No 53 had no other exit to the interior of the building than to the apartment 54, it may not have possessed any other value than that of an antechamber for No 54. But the Inca could easily pass from No 54 to ai-y other room of this block, comprising spaces 53-66. He might have passed through No 63 to No 64, to tower 65, with the belvedere on it, and to the open gallery 66, which is closed from all other sides of the building: He might have entered room 62 or 55 which served his own personal conveniences; he might have passed through room 55 to the bathroom 57 with its open gallery 58. [p.74, (PICT0460)] He could likewise enter into the rooms by passing through the southern door of room 54 and passage 59. But the Inca could not leave the rooms of this block in the direction to the south, nor could anybody have entered his apartments from that direction. The block was entirely closed off from passage 18. I consider room 60 as having been the bed-chamber of the Inca, on account of the step-like elevations (see section 4) existing in that room. These might well have served him as bedstead and pillows. From there he might have easily passed to the contiguous bathroom 57, at his convenience. 51 The decoration to which Uhie refers here is the "half-cross" ornament, which he identifies as "type 2" (editor's note). 52 Uhie's plan (scale 1 :200) does indeed show an opening between room 18 and 19. However, at close inspection, this opening proves to be a breach in the wall made at some undetermined time, but it was not a doorway, there are no signs of a sill or frame on the inside of 19. Thus, the only entrance into the group of rooms 19-29 was through the anteroom, 29, and room 27 (editor's note). 40 Explorations in the Pisco Valley It does not seem to me easy to explain the particular uses of the rooms of the blocks 19-30 and 31-41, but it is sure that nearly the same proofs are to be found in their details as in those of spaces 53-66 for their also having formerly been occupied by a resident of the highest class. The access to these rooms was nearly as difficult as to spaces 53-66. It may seem that the passage to them is a little shorter than that leading to spaces 53-66, but there is not much difference in this respect. To some extent it was perhaps more difficult to enter the rooms contained in block 19-29 than to enter the rooms of the Inca, 53-66, because one has to pass through passages 15 and 17 to get to the gallery-like passage 18. No direct entrance existed from there to the main spaces 22-29. Instead, one had to pass through the small anteroom, 19, (with No. 20 on its right side) [p.75, (PICT0461)] for getting access to the first of the rooms, No 23. The difficulty for reaching rooms 25 and 26 from the main entrance of the palace, or for reaching to room 37 from the same part-to reach these rooms one had to take a round-about way, 5-10 times as long as the direct route from entrance would have been -proves sufficiently that these rooms south of passage 18 were highly guarded against sacrilegious contact by outsiders. It is to be observed that spaces 31-41 were not directly accessible from the adjacent spaces 19-29, and that they were easier reached from spaces 15-18 by passing the long corridor 30, which served only the purpose of giving direct access to spaces 31-41, without the necessity of entering any of the rooms 19-29. I conclude from this that rooms 31-41 formed a group of apartments, in the access of which the inhabitant was only partly interested, and in some way perhaps less so than the inhab- itant of some other group of apartments in the building. The latter would in this case have been the Inca, the inhabitant of spaces 53-66, though he had to take a long, roundabout way to get from room 54 to room 31. Among the rooms in block 19-29, there are two rooms, 23 and 25, which may be con- sidered as resting rooms or bed-chambers because of the step-like elevations existing in them. The room between them is the second bathroom of the building. It received, as we shall see later on, its pro- [p. 76, (PICT0462)] vision of water through the wall which separates it from room 3. The resident of these rooms, therefore, was able to avoid contact with outsiders even in his bathing necessities. In comparison with rooms 53-66, which contain that splendid stateroom 54, and give direct access to belvederes 65 and 66, rooms 19-29 fall short in size and exterior importance. The size of these rooms corresponds more or less to those of the convent of Sun-Maidens at Pachacamac, or to those which I considered as women's apartments in the Incasic palace-like building near Tambo de Mora. Two wide openings, larger than doors, that serve as the eastern entrances of rooms 22 and 27, are similar to those previously observed by me in the convent of the Sun Maidens on the Island Coati (in the Lake of Titicaca), though I do not know what purpose they may have served. In the work of George Squier, these halls of Coati figure as chapels of the divinities of the Sun and the Moon. As in the palace-like building near Tambo de Mora, we find here two kinds of rooms, each provided with separate bathrooms and a bed-chamber (or bed-chambers), the one more easily accessible than the other and Letter of September 11 and 24, 1901 41 provided with larger rooms and state-rooms, the other still more removed than the former from contact with outsiders, and consisting of smaller, less imposing rooms. Undoubtedly both groups of rooms had separate inhabitants, according to the separate and parallel conveniences offered in them. As in the case of the previously mentioned building of the Chincha valley, [p.77, (PICT0463)] I would not oppose here the interpretation that the former group of rooms served as apartments for the male, the latter for the female inhabitants of the building, and I will defend this opinion unless faced with proof of a better explanation of the existing facts. With respect to the painting of the wall, I have the following to observe: Three colors are used: white, red, yellow. In cases when the painting is not uniform, e.g. when the painting is unlike the simple white washing in many rooms of the interior, or the red on the southern front-wall of the palace,53 it is applied in horizontal stripes, setting one color above the other, nearly without exception in this fashion: there is a wider lower part that is uniform, and the remaining colors are applied as if they were wide border stripes (up to about 1 meter in width) above. Only in one case is the painting set off in steps, appearing like the stripes of different flags, that is on the eastern outside of the eastern front wall of the palace (the western outside of the western front wall is remarkable, unpainted).54 Both side walls of the palace fall off in steps according to the descent of the hill slope against which the palace has been built. The painting of the walls follows the steps of the walls; each section of wall delimited by a step is painted in horizontal stripes, giving an appearance like a series of flags (see plates I 1 84-1185). Undoubtedly, the painting of the faces of the walls was in some way also expressive of the [p.78, (PICT0464)] value or importance of the rooms for which the painting served as decoration. White (as white-washing) may be said to have been the most inferior kind of painting, as it is found in many smaller rooms and the larger rooms dedicated to ordinary housekeeping. Red, especially where used alone, may be considered as that color which expressed the highest importance or rank. The southern wall of the northern palace, which faces the wide central plaza, is painted entirely red,55 in the way that the whole exterior of the Sun-Temple at Pachacamac was painted only red. This may have expressed that this habitation was that of the Inca. 53 The all red painting of the southern wall noted by Uhle represents the last coat of paint applied and it may not have been applied to the entire wall. Today one can observe that originally the wall was painted red above to about 40-45 cm below the sill of the double framed niches, and white below (editor's note). 54 Uhle may not have noticed it, but the western enclosure wall was also painted. It is difficult today to determine the pattern of the painting as only specks of paint, red, yellow and white, can be found here and there. Remaining paint near the outside north-western corner of the palace, on the outside of gallery 66, suggests that the western outside wall may have been painted in the same manner as the eastern outside wall. It is to be noted that the paint is generally more eroded on west facing walls than on east facing ones probably due to prevailing winds (editor's note). 55 See the previous footnote. 42 Explorations in the Pisco Valley No room in the interior shows the same uniform coloring. The rooms of the interior which are not uniform white are all painted in a variety of colors, in white and red, or white, red and yellow. There is only one part of the interior of the building, in the painting of which no white has been used (at least in its present condition), that is the western part of the solemn court 9, comprising the western wall and the contiguous inner wall of the northern side of the court which show only red and yellow in their painting.56 An especial ceremonial solemnity of this part of the building, in which was the entrance to the inner apartments of the Inca, seems to me to be demonstrated in this way ad oculos. I proceed now to describing the painting of various rooms in detail. Rooms for ordinary housekeeping, like 3-5 and 8, are only white- washed. In other rooms of this [p.79, (PICT0465)] kind, the color with which they may have been painted, if they had any, cannot be recognized now. White-washing is also the painting of rooms 13-14 (that is: of the back rooms, east of court 9), while the front rooms on this side, 10-12, show a red border stripe above the wide white lower stripe. Narrow passages or unimportant small rooms, like 43, 48, 49, 15, 17, 61, 64, and 63 (on three sides), and the passages before 5 and 14 are mostly white-washed. But one has to be curious that also the interior of the dormitory, 60, and of the tower, 65, as well as the western wall of 53, are only white-washed. Rooms 47 and 50 are also only white on their inner side, if one makes no account of the ornamental parts at the top of the walls, which seem to have been red.57 Room 52 is also only white at the inner side.58 Even white with a red border stripe above does not seem to have marked any higher importance of the room in which it is found, thus it is met with (besides the cases cited above), in the back rooms 51, 26, 28, in the passages between 43 and 44, 46 (in front of room 52), 63 (southern wall), and 30 (northern wall). But the northern side of room 44 and the southern side of room 53 (and perhaps to be considered there as a passage) are also found painted in this way. Red with a white border stripe above, which has something distinctive, as it is the only case (besides the front wall of the palace and the western parts of court 959) in which red figures as painting at the foot of a wall, is only found around the outer sides (east, west, north) of the curious rooms 47 and 50 (that [p. 80, (PICT0466)] is, the wall between gallery 46 and rooms 47 and 56 Uhle's observation holds also for the eastern wall of that same court (editor's note). 57 The walls of rooms 47 and 50 are topped with crenels, or as Uhle called them "pinnacles," that are painted red only in the reveals, but are all white on the interior and exterior faces (editor's note). 58 Room 52 is indeed white at the bottom, but it has a red band all around at the top of the wall (editor's note). 59 This passage is confusing. It is not clear what Uhle means here by the "front wall of the palace," which by his own account should be all red. As to the court 9, again by his own account, it has no white. Perhaps he meant to say the western wall of court 42, which indeed has a white band on top (editor's note). Letter of September 11 and 24, 1901 43 50, and eastern side of 44).' Besides these, we find the following variations in the pattern of wall painting: I white, red, white in three stripes 2 yellow, red, white in three stripes 3 red, yellow, white the former two colors as wide border stripes above the white color 4 yellow, red, white, " " " " " 5 red, yellow, red, white the former three colors so used as border stripes 6 red, yellow, red, yellow, red, in five equal stripes. - white, red, white is only found in room 16, on the contiguous western wall of court 42, and the partly corresponding eastern and northern wall of room 53. The rest of the walls of court 42 have been painted in three stripes: yellow, red, white (it has to be remarked, that several older yellow and red paintings shine through the lower white stripes). I do not doubt, that this manner of painting must have seemed to be appropriate to the certain solemnity which this court undoubtedly possessed. It is curious that in some rooms different walls show different painting. For example, the western wall of room 45 and the northern and eastern walls of room 62 are painted in the colors, succeeding from above to below, red yellow, and white; the northern wall of 45 and the western and eastern walls of 62 are painted in the colors yellow, red, and white. The latter are also the colors of the western and eastern walls of the state room 54, probably for their correspondence with the neighboring rooms 55 and 62), while the northern [p.81, (PICT0467)] and southern walls of this stateroom, distinguished by niches of the third order, are painted in the colors red, yellow, and white. Red, yellow, red, and white may be seen to have been the most prominent and most typical pattern of colors in the painting of the walls of this palace, after the uniform red color of the southern front wall and the yellow and red painting of the western part of court 9. The outer eastern side of the eastern side wall of the palace shows the same colors, distinguishable from afar to the traveler who approaches the palace from the east (see plates 1184, 1185). And it seems to me that many high walls which show these colors too, in the interior, do so more for their visibility from the area surrounding the palaces than for the importance inherent to the rooms to which they properly belong. Many walls show painting above the level of the roof level of adjacent rooms, for example, north of 11, 13, 14, 1, 4, .5, west of 3, room 18, 26(north), and 28 (north). Several small rooms and passages show the colors red, yellow, and white, with more or less clearness, for instance spaces 19, 20, 34, 35, and 40, a part of the eastern walls of passage 60 UhIe's description of the painted walls seems to be erroneous. It is clear from his text that he could not be referring to the north walls of rooms 47 and 50, but rather the southern outside walls of 47 and 50. The northern walls are painted white. Evidence from the site shows that the eastern wall of space 44 (the exterior of 47), and the northern walls of gallery 46 (the exterior south walls of 47 and 50) were painted in the manner he has described, except that there is evidence of an additional yellow stripe above the red base on the south outside wall of 47 (editor's note). 44 Explorations in the Pisco Valley 30, and the eastern walls of 3.6' As examples of more prominent rooms in which the coloring is also mostly more distinct, the following should be mentioned: the eastern side room 55, of the stateroom 54, the bathing apartments 56-58, rooms 25 and 27, the north and western walls of 31, and room 38. Red, yellow, red, and white are the pattern of colors in rooms 22, 23, 29, 37, and the passage 21, and on the western wall of room 44. None of the rooms 22, 23, 25, 27, or 29, nor the passage 21, nor the smaller spaces 19 and 20 shows coloring inferior to either red, yellow, red, white or red, yellow, white. [p.82, (PICT0468)] But in prominence of painting, all these rooms are surpassed by court 9 (see plate 1164), with its colored stripes: red, yellow, red, yellow, red, at its western and inner north-western wall,62 and with the colors: red, yellow, red (also without any white) at the eastern and outer north-eastern walls. This was the most ceremonial room. It was probably there that the Inca, stepping forth from his private apartments, went in to meet with the prominent chiefs of his empire, when he received them. It remains to me to discuss the bathrooms, which have been an interesting feature of the last pre-Spanish periods of Peruvian civilization. Incasic baths are not uncommon in the mountains, where prominent Incasic remains exist, and they are known to the indigenous people, together with the original use made of them. Several of those so-called "baths" once were undoubtedly baths. This is, for instance, the case with an Incasic bath which exists in a hacienda about 1/2 league southeast of Copacabana. At least two other Incasic baths are known upon the island of Titicaca. Another exists near Tarma, another at Cajamarca, if I remember the literature, another is said to exist near Huaitara. And, upon the hill of Marca Huamachuco, there exist details which prove undoubtedly the former existence of a beautiful bathing place. The last mentioned of these, [p. 83, (PICT0469)] beyond doubt, did not belong to the Incasic civilisation, from which fact we are to presume that the institution of baths was common to various, perhaps all, periods of Peruvian civilization. I remember that near Tiahuanaco there exists also a place which may be considered as a bath of that long-gone period of Peruvian civilization. Bathing places especially made of single stones hollowed out, or of masonry-work, will have been a feature belonging only to chiefs. The other people may have taken baths in the rivers or azequias, or they may have taken no baths at all. Bathrooms are less known on the coast than in the mountains, even though the climate of the coast disposes much, the climate of the mountains very little to bathing. This may possibly stand in relation with the peculiarity of structures made of stone in the mountains. But however this may be, the only baths known to me on the coast, are those two 61 UhIe probably meant the western wall of room 3 above the roof line, for all walls of room 3 below the roof line are white (editor's note). 62 [Uhle added at bottom of page]: On these two walls the niches (of the third order) are also placed lower than on the other two rooms, which proves a more general distinction of the two parts of the court in use and decoration (see section 5). Letter of September 1 1 and 24, 1901 45 which I first observed, together with their system of canalizations, in the Incasic buildings near "La Centinela" (Tambo de Mora). The ruins of Tambo Colorado present about 6 more samples of ancient bathrooms, two of which are represented by rooms 57 and 24 of the northern palace, two others by those which are marked as d and e in the southwestern palace, on the map 1:500 of the ruins; one is that marked as f in the south-eastern palace (now still more destroyed than the others) and one exists in one of the eastern houses, outside of the principal ruins, near "g" on the map 1:500, apparently situated there in a building [p.84, (PICT0470)] which was not inhabited by an Inca of the mountains, but by an indigenous chief of the coast, who may have been influenced by Incasic civilization.63 These bathrooms (two have been photographed by me: that of room 24 of the northern palace in plate 1187, and "d" of the southwestern palace in plate 1188) are of no great extent. The bath 57 is, for instance, 3 meters long and 2 meters wide; "d" of the southwestern palace is about 2.5 meters long and wide. The depth amounts generally to 1 to 1.25 meters. The shape is square, the floor and some of the lower part of the side walls consists of stone; as I mentioned already, some very large slabs have been used in the construction of these baths, which gave treasure hunters the idea that treasures might be hidden below them, the consequence of which was that none of the baths of the three palaces is now undamaged. The higher parts of the walls consist of masonry work of adobe. Curiously enough, in various of these bathrooms, which are generally girt by low breast-walls, as in the case of 24, in the bathroom in the southwestern palace, and in bathroom 57 of the Inca apartments, there are a few small niches in the walls at the lower floor of the bath. These baths are further distinguished by their canalizations. In bath 57, now much destroyed, it is still possible to recognize a water channel which emptied the water from the bathing room in the direction of passage 18. There the canalization ends, [p.85, (PICT047 1)] only because the floor of passage 18 is so much destroyed that nothing may be discovered if not by serious digging.'M Bathroom 24 shows another kind of channel (see plate 1187), one which conducted the water into the bath through the wall which separates room 24 from the service room 3. Outside, on the corresponding place on the wall, there exist some remnants of a low masonry-work of quarry stones, which might originally have represented a small reservoir. On the inner side, the water fell from the channel over a slab, which still stands upright, into the bath. No channel to drain the water from the bath can be discovered now, owing to the serious destruction of its floor. But both kinds of channels can still be observed as existing in the bath "d" of the southwestern palace. A 63 One may seriously question whether the feature "g" was indeed a "bath." It is constructed entirely of adobe, and it lacks all evidence of a water entry channel, as well as of a drain. Its characteristics are those of a tomb more than of a bath (editor's note). 64 Indeed, later clearings of passage 18 revealed a channel continuing from the outlet below 57 to the west and through the western outside wall of the palace (editor's note). 46 Explorations in the Pisco Valley channel which crosses beneath the room contiguous at the east, passes the wall and poured its content from above the upper level of the bath into it. At the base of the floor, one observes another channel, by which the bath was evacuated after its use (see pl. 1 188). The western wall of the southwestern palace and of the large central plaza, which is about 3 meters high, shows at the outside three openings of channels which formerly poured out their contents to this side of the building. They are marked on the map 1:500 as "a", "b", and c "a" and "b"" end about 0.50 meters below the landing of the outer supporting wall, [p.86, (PICT0472)] and about 1.5 meters below the general floor level of the interior of this palace. It is difficult to say to which of the baths each of the channels belongs, but it seems that the contents of bath "c" were poured out near "a", the contents of bath "d" near "b". The subterranean path of the channel which ends near "c"' is enigmatical. It is near "c", at a distance of eleven meters from the terrace-like building,65 which interrupts the wide, low terrace bordering the central plaza at its western side. No clear marks of the former existence of a bath in this building are visible. The opening of this channel is situated 1.5 meters below the landing of the outer supporting wall. It is, therefore, not to be considered yet as impossible that this channel belonged to the system of canalization of the northern palace, and that the contents of bathrooms 57 and 24 were poured out at such a distance from the palace. Careful examination of the outside of the northern palace did not show any openings where the contents of the bathrooms might have poured forth. The channels are all of equal construction, similar to those I described in the Incasic buildings near Tambo de Mora. They are higher than wide, and constructed of small slabs of stone set together, more or less in this way: [p.87, (PICT0473)] The large central plaza between the three palaces may have served for commerce, meetings, displays of parts of the Inca army and festivals and dances executed partly in the presence of the Inca. The wide platform at the west of the plaza, with some small ramps leading up to it, with an enclosing, niched wall at its back, and a square, terrace-like building about I meter above the level of the ledge, more or less in the middle of it, may have served for the authoritative persons which were present at these occasions as 65 Uhie is here referring to what today is identified as an "usnu" -a ceremonial platform of the Incas. Letter of September 11 and 24, 1901 47 spectators.66 Both sides, north and south, of the plaza will have been roofed, gallery-like; at least the platform in front of the northern palace has been so, and a similar terrace-like platform is also to be found at the fronts of the two southern palaces. The country around the palaces must have presented a much nicer aspect in Incasic times than it does now, because the foot of the hills of the other side of the river are still bordered by andenes,67 arising terrace-like in steps, which now are lying dry, but were hemmed in ancient times by pleasant green stripes produced by their being irrigated and cultivated. As on the right side of the river, the left bank elevates itself in terraces, 2, 8 and about 20-40 meters above the present bed of the river. All three are effects of the work of the river in different geological periods. The lowest is fertile soil, the second terrace nearly the same, the third, now broken by ravines in separate terrace- like hills which border the rocky hills, at their feet, is composed of boulders [p.88, (PICT0474)] of varying size -- some of which are immense-mixed with fertile "loess"-like clay, which all have been brought down by the river at the time of former geological catastrophes. It is more or less the same kind of soil which fills so many other coastal valleys of Peru, especially that of Lima, down the coast. The first and lowest terraces are now being used for cultivation; the extensive plains of the second terrace were cultivated as fields in ancient times, but are lying dry now. The slopes of the third set of hill-terraces are bordered by steps of terraces, which are works of pre-Spanish times, with ditches of parts of one or two ancient azequias running along them. The upper surfaces, resembling a table-land, has been also, for the great part, transformed into fields, where their elevation allowed their being irrigated by azequias conducted from higher parts of the river. The latter has much descent in this part of the valley, which allows irrigation of fields up to the height of 20 meters above its present level, about 2 miles below the spot at which water has been diverted from it in ditches. All the fields of ancient cultivation upon the second and third river-terraces are lying dry now, though projects of individual people exist for returning them to their original use. These ancient cultivated lands on the left side of the river may comprise about 50-60 hectares, and besides this, andenes extend about 1-2 kilometers downriver. I took several photograms which show these extensive effects of pre-Spanish laboriousness and industry: plates 1178-1180, 1194-1195, 1196- 1198, 1189 and 1190. Besides this, [p.89, (PICT0475)] the panoramic view of this section of the valley, taken by me from the southwestern corner of 2nd (southwestern) Incasic palace, shows the fields on plates 1201-1205, in nearly their whole extension, from a distance. The work of the ancients consisted in the following occupations: 66 Again, Uhie is referring to the "usnu." 67 "Anden(es)" is a Spanish word for platform, which in Peru is commonly used to refer to artificially constructed agricultural terraces (editor's note). 48 Explorations in the Pisco Valley The removing of boulders from the grounds which were to serve for cultivation. In this way the fertile clay remained in the proposed fields alone. The boulders were heaped along the edges of the fields or served for bordering the fields descending in low steps, one below the other, or for the construction of the walls of the andenes which bordered the slopes of the terrace-like hills. It is enchanting to see the effects of the immense patience and of the great care in even the most minute details of the work undertaken by the ancient Indians. Parts of the fields often possess no larger extension than that of modern garden-beds, if the continually-changing level of the ground did not allow them any wider extension, and even in this case much art and industry was necessary to secure these small portions to profitable agricultural work. Gutters were provided where the water was to fall down from terrace to terrace, for irrigating one after the other. Stairs were prepared so that people might reach the higher terraces from below. The latter were made partly from the stones cleared from the fields. Ancient gutters can be recognized in my photogram 1 190. Sections of the andenes were partly separated by walls, as if having belonged to different proprietors, as can be [p.90, (PICT0476)] seen in German vineyards leaning against hills. The azequias conduct the water from higher parts of the valley where it was taken from the river. They are turning now around the hills, the slopes of which they formerly irrigated, following exactly all curves of the country, passing ravines, crossing the surface of the hills plain above, if these were not to high for them, and branching in manifold ways. There is another extensive ground of ancient field work and andenes on the right side of the river upon the ground of the hacienda San Ignacio, about 1.5-2 leagues below Tambo Colorado. There I also observed an ancient dike, which served to conduct an azequia over a small ravine about 30 meters wide and 2 meters deep, because no other way existed to bring the irrigation to the ground on the other side of the small ravine; other similar particularities are to be observed here and there. The andenes on the left bank of the river rise up to about 25 meters above the river on the west end. I observed up to 7 terraces of andenes, one above the other, which formerly had been irrigated by one, and in parts by two, azequias at different elevations. Several ancient buildings, partly indicated on my map of this section of the valley, have been constructed amidst the andenes, which mostly may have belonged to the Incasic period. The andenes cannot be all work of the Incasic period alone. Many ancient walls from the oldest period observed by me in this valley, belong to ancient p.91, (PICT0477)] settlements upon the upper plateau of most of those terrace-like hills, surrounded by andenes on three sides, so as to make it probable that construction of andenes took place in very remote times of Peruvian civilization. One of the ancient buildings of the Incasic period on one of the slopes of the hills provided with andenes has been placed on across the course of an ancient azequia (the higher one of two), so as to cut it in two pieces. Its construction would have paralyzed the use of the azequia further on, if it was still in use then. This proves that the azequia was of an older period than the Letter of September 11 and 24, 1901 49 building, and out of service already. All such observations prove that undoubtedly not only the Incas extended so much ancient agricultural work in this valley, but many generations and probably also nations before them. I count about 9 different buildings of ancient times on the other side of the river, amidst the ancient fields or behind them. Some of these are constructed of boulders mixed with clay as mortar, others of tapia. None of them deserves any close attention. The walls of one of these appear in the foreground of plate 1181. Besides these, there exist some ancient burial places upon the plain of the most western hill-like terraces. One of them has the shape of a cemetery. It has been, for the most part, reproduced in the views of plates 1192-1193. It has partly the appearance of a building consisting of various square rooms, sunk in the ground, with a some-what higher wall at the south at the back of it. To the east, these rooms are adjoined by some [p.92, (PICT0478)] depressions of more irregular shape, and mostly rounded at the corners, more or less in the shape of a honeycomb. Other graves lie separately, possessing the shape of barrels. They have inner walls built up of small boulders. Their diameters vary widely, from 0.60 to about 3.20 meters, their depth from 0.50 - 2.00 meters. On the inner side nearly all, like the square rooms of the building-like construction, show several (two to three) stones projecting like steps from the walls, which served to descend into them on the occasion, perhaps, of making an interment in them. All these tombs belong to the Incasic period, though it seems to me, according to fragments of pottery found on the site, that indigenous people took part in the use of this cemetery. In the panoramic view (plates 1194-1195) one can recognize the remains of several burial buildings of the Incasic period at the front of the principal hill. According to my mind, because I have observed some of somewhat similar type upon the island of Titicaca, and because they seem to represent some higher type of burials, these would deserve to be considered as pure Incasic. As appears to be the case with the burials mentioned before, all these have been excavated and destroyed in former times. Human bones are mixed with the disturbed soil within them. There exist about a dozen such ancient burial houses on that spot, mostly forming a row. With the exception of one building, which is round, they are all square or rectangular. The inner rooms measure between 2.50 and 3.50 meters in length and width, according to the [p.93, (PICT0479)] seven which were measured by me: 1. 3.40x3.40 meters 2. 2.80x2.60 3. 3.x2.55 4. 2.90x2.90 5. 3.40x3.50 6. 5.50x3.55 7. 4.90x3.50 The round one is 3.50 meters in its inner diameter. 50 Explorations in the Pisco Valley The thickness of the walls of surrounding these inner rooms is mostly of about 1 - 1.5 meters. The exterior of the houses rose very little above the surface of the surrounding ground, while the depth of the interior amounts, in general, to about 1 meter. The roof was supported by wooden beams, remains of which still exist in some cases. The entrances of these houses were formed by small doors about 0.45 - 0.55 meters wide and 0.65 - 0.95 meters high, nearly indifferently directed to north, east or south. As the entrance is below the level of the ground around the house, the ground has been excavated in front of the doors in the shape of an inclined plane. A cross-section of one of this half subterranean burial-buildings would have more of less this appearance: 4~~~~~~~~~~~~~~~~~~~~~4. ~~~I I -. ... [p.94, (PICTO480)] While the lower part of the valley, up to about 5 or 6 leagues inland, is occupied by huacas made of tapia of the common type of the coast and of the valley of Ica (see plates 1051 and 1052), such ancient remains are not found higher up. There the shape of ancient settlements was different. About 10 minutes west of the very unimportant village of Umay, a village of about 200 inhabitants belonging to the district that occupies a large part of the middle of the valley, and which may have as many as 5000 total inhabitants, there are, near the road, some tapia walls found in an area of about 100 meters in length and width. Though these ruins are considered as ancient (they would in this case be the only ruins in the immediate neighborhood of Umay, Tambo Colorado is lying about 1 league above it), I could not confirm this fact on account of the many shard of large, earthen vessels now in use for holding aguardiente laying among the ruins, and the regular street pattern typical of modern settlements. As these ruins, if judged by their appearance, deserve very little attention, I leave this fact in doubt, without occupying myself in detail with them. About a league and a half below Umay, there are the remains of another ancient settlement known in the valley as Lima Ia Vieja. They are bordering the road which runs through the valley on its southern side, and lie Letter of September 11 and 24, 1901 51 upon a dry plain situated between the northern hills and the cultivated plain of the valley. They may be extended over about I kilometer in length, and about 300 meters in width. A large, square, middle plaza occupies its center, separated from [p.95, (PICT048 1)] the road only by a small gallery-like room. Elevated terraces bordered several of the sides of the plaza. An altar- like elevation in the middle of the place is of modern origin, having formerly borne a Christian cross. Lima la Vieja is much destroyed. It forms now only heaps of debris and extremely little would result of a more detailed study of them. I took photographs 1245-1247 of it, at the time of the sunset. In the northern part of the ruins, the remains of a somewhat higher building rise above the others. Entering there, I found it contained square niches of the type introduced by the Incas in the valley, and which had been formerly painted red. Lima la Vieja is, therefore, to be considered as a settlement which was still inhabited at the time of the Incas, or which, perhaps, began in that time according to the regularity of the large regular market-place in the middle, which seems to show also traces of Incasic influence in its elevated side-terraces, and in the circumstance that only long narrow rooms bordered it on one side. The ancient road through the valley followed, as I said before, the right bank of the river. It can be traced down about to the hacienda Manrique (4 leagues from the coast), where it leaves the dry plain to run between the fields. Between San Ignacio (about 1- 1.5 leagues above Lima la Vieja) and Manrique, the road crosses a dry open plain, in which nearly no stones are to be found. Notwithstanding, the line of the ancient road can still be traced in many parts, for small obstacles of stones have been cleaned out of the way [p.96, (PICT0482)] and now border, in more or less clear lines, the sides of the road. The present road there is the same as the ancient, as it is the direct one. Higher up, where the ground is rocky the ancient road is still made clear in many parts by boulders of all sizes, having been disposed along the sides of the road so as to line it. In several parts, this lining has been done of masonry work. The width of the ancient road is about 5 meters, for instance, within the ancient settlement of Pantaico (about 1500 years old) crossed by it: 5.40 meters. One might be disposed to believe that the road originated there from the time of the original foundation of the original settlement. But this is not so. The road was cut through the ancient settlement in later times, which is shown by some ancient walls remaining here and there in the ground, because of their not having been extirpated by the later constructors of the road. For this reason, I have no doubt that the road may be of Incasic origin. [p.97, (PICT0483)] My trip to the southern part of the Chincha valley conducted from Pisco, through the large hacienda of sugar, Caucato, after that several kilometers along hills of pure gypsum, which separate the valley of Chincha from that of Pisco, at their western side, and are separated from the beach of the ocean by a strip of land about 400 meters width, belonging also the hacienda Cauacto and planted with sugar-cane. 52 Explorations in the Pisco Valley From this it is clear that the separation of the two valleys of Chincha and Pisco bears a subordinate character. The plain of the two valleys is uninterrupted and the vegetation alike, though the cultivation of the two valleys suffers an interruption of several kilometers in the southern part of the northern valley. Lurin Chincha is an hacienda of some importance, abundant in water, at least for all necessities. The ground near the sea is swampy for a part and the access, therefore, difficult to some of the old huacas made of tapia, which for a part seem to have been built intentionally upon ground somewhat unsafe. The modern buildings of Lurin Chincha are surrounded on almost all sides by ancient huacas. I counted about 15-20 there. A few minutes south of Lurin Chincha, there is a large huaca known as the "Centinela" (of Lurin Chincha, not to be confused with that of Tambo de Mora). It faces the sea, from which it may be at a distance of about 220-300 meters and is constructed of tapia, and of some interest due to the main building having been placed upon a larger terrace of about 10 meters in height, secured by a supportive wall several meters thick at the outside. The main building, probably having served a chief as dwelling place, is now completely destroyed [p.98, (PICT0484)] and deserves no further attention now. Photograms 1218-1219 present it from the South West. About one kilometer from there, there is another, much larger huaca, belonging properly (as does the former) to the Hacienda Hoja Redonda (an hacienda which lies inland, but with grounds extending down to the sea). I enclose here a sketch made of this large Huaca on the spot. It measures about 260 meters in length in the direction to the sea (ca. 780 east of north), mostly 80 meters wide, but about 180 meters wide for part of its length at the east end. According to the sketch, its shape is more or less that of a hatchet, though this may be an effect of chance. It measures about 20 meters in height at its highest part. It descends in several terraces from the highest part, which is between 30 and 50 meters from its western end, to the east, three of which embrace large court-like depressions in the middle. The lowest part (about 8 meters high) is the eastern end. A projection to the north is still lower than the southern part (about only 4 meters high). The western end falls off about 5 meters from the highest part of the huaca, forming a terrace over against it. Undoubtedly the huaca is much older than the common huacas made of tapia. Its extension is much weather-beaten, and no clear details can be distinguished. The material consists of irregular chunks of clay, by which the huaca seems to have been built up. According to this, the age of this huaca seems to lie between those huacas built up of half-globular [p.99, (PICT0485)] adobes and those of pure tapia work. Upon its surface, I could find no fragments of pottery giving satisfactory indication of its age. It also shares a feature found on some huacas of very great age in the same valley, in bearing on its surface many chamber-like, square holes made for interments at the time of the last period of Chincha civilization, probably, therefore, at a later time than that in which the huaca originated. Several or Letter of September 1 1 and 24, 1901 53 all of these chamber-like graves have been excavated without anything of value or interest having been found in them besides human bones. Human bones are also met with strewn on parts of the surface where such chamber- like graves are missing. I took several photograms of the huaca, which bear the numbers 1210- 1219,68 1220-1222. The huaca of Santa Rosa, about 1.5 leagues from Tambo de Mora and about 1.5-2 leagues from Lurin Chincha, was visited by me on foot from Tambo de Mora on one of the Sundays of last October. No time was left then to measure it and to get more than superficial knowledge of it. At my recent visit from Lurin Chincha, I stated its length as being 332 meters, its width, for instance its eastern part, as being 168 meters. The huaca is, therefore, the largest in all the Chincha valley. Its height may be 25 meters. It must have been an ancient temple at the time of construction of the huaca of Alvarado, of which I treated in my report about the ruins of the Chincha valley. One distinguishes several summits, [p. 100, (PICT0486)] on the highest of which a Christian cross has been erected. The northern foot has been washed away by the river, which in one of the last ten years unexpectedly changed its course for a short time at the season when it is swollen with run-off. The southern foot, of a width of about 70 meters, is lower than the rest of the huaca; it is nearly level and rises about 5-10 meters above the level of the valley. Large courts like ancient basins,69 one of which now contains water, can be seen at the middle and at the western end of the southern side. The eastern end of the huaca opens into a wide court about 98 meters wide and 1 14 long, bordered by dams about 10 meters wide which must have once been thick walls. There are also recognizable traces of an outer court to east of it, about 150 meters long. The whole disposition reminds one of the great huaca of the Rimac Divinity with its large eastern forecourt, in the valley of Lima, though the latter huaca seems to have been of much later construction. A thick dam similar to those which border the forecourts exists also south of the huaca, measuring - from east to west - about 200 meters. I mentioned the many chamber-like burial places made at the time of the later Chincha civilization upon this huaca of much older origin in my report on the ancient remains of the valley of Chincha. Photograms No 1224-1230 have been taken of this huaca from south, north, east, and from above. [p.101, (PICT0487)] On my road from Pisco to Pallasca I found south of it the so-called ruins of Chongos, about 1.5 leagues from the former place. They appear like an ancient fortress, looked at from below from the valley, over which they are elevated about 30 meters (see my photogram 1241). But they represent, in reality, only one palace of the last period of the Chinchas, 74 paces long and 68 This should probably read 1210-1217, since 1218 and 1219 are of the huaca "la Centinela" of Lurin Chincha. See p. 98 (editor's note). 69 Uhle is apparently referring to old courtyards which, surrounded as they are by walls or the remains of walls, have once flooded and show evidence of this flooding (editor's note). 54 Explorations in the Pisco Valley 23 paces wide, girt on three sides by a wall 120 paces long at the western and eastern sides, and 36 paces long on the southern side, where an entrance exists, while the northern side, 44 paces long, has been left open. For this reason, it would not be possible to say that the outer wall was made for defense. This palace crowns one of three small summits bordering the dry plain in front of the cultivated plain. Some ruins beside it indicate the former existence of a few more ancient houses nearby. Photograms 1241-1244 come from the close observation of these ruins. Ancient names found in the valley: Pisco (Kechua = "bird") Chongos Pallasca (seems to be also of Kechua origin) Guaiya Detrayo Huaincano Pampano No ancient name exists now that belongs to the ruins of "Tambo Colorado" (that is: painted harbor on the road) studied by me. [p. 1, (PICT0488)] Pallasca September 30th, 1901 Dear Mrs. Hearst I have the honor to send you my report about my work done in the last weeks in the middle and lower valley of the Chunchanga River. The tenant of the hacienda Pallasca, going unexpectedly down to Pisco, presents a rare occasion to send you my report, missing which, I would perhaps have needed to send another reliable person or to go down myself, losing in either case a few days for my work. For this reason, you will kindly excuse me that I revised my manuscript only up to about page 54, hoping that the end will not contain too many errors. I do not know if it will be possible to me to send my maps at the same time, but it seems to me of greater importance to give you a detailed notice of the work I've done by sending you my manuscript immediately, than to hope for a convenient later opportunity. I enclose at the same time an earlier letter from Pisco, and the list of the photograms I've taken so far. My excavations, notwithstanding great effort, gave good scientific results, but yielded few artifacts. But after various experiments, which aided the general scientific results of my studies, I hope to have still some good results tomorrow or after to- [p.2, (PICT0489)] morrow before leaving Pallasca for Huaitara, which will be perhaps Friday. Hoping, Dear Mrs Hearst, that you are well and might not be too disappointed for my swerving around in the South, while you expected me perhaps long since in the North, I remain, Dear Mrs. Hearst, Yours very sincerely Max Uhle [p.1, (PICT050 1)] Huaitara, October gth 19g0 Dear Mrs. Hearst, I have not yet had the honor to receive notice from you whether you received my last letters referring to the most recent finds of gold objects made by me in June. But perhaps these letters from you are en route to South America. I have the honor to advise you that I arrived yesterday at Huaitara after some difficulties I had in the last days at Pallasca, being compelled to send down to Pisco on my own animals those few boxes containing skulls and other trophies of excavations that I packed at Pallasca, as no other conveyance was available for that purpose in that part of the valley. Currently, I have six sheets of maps and drawings, results of my work on Tambo Colorado, which I shall send to you with the next mail, not with the present letter, as this letter is taken down to Ica by a friend from here, and rolled maps probably would not arrive safe at Ica, if sent with him. The upper part of the valley of Pisco up to Huaitara bears sufficient proofs of its complete taking under Incasic influence after its conquest by the Incas. I suppose that this was dictated partly by political reasons, [p.2, (PICT0502)] as the valley of Pisco was one of the most important on the Peruvian coast for communication to the interior, and of particular importance for the communication between the coast and Ayacucho, one of the most important northern centers of the Incasic power in the mountains. The Incas have constructed a very good road along the valley up to Huaitara. In many of its parts, the work done by the Incas can still be recognized as such by wall lines bordering the road (which is between 5.60 and 5.96 meters wide) - substructions in the shape of walls made at one side of the road above the inclined slopes or steep precipices. I am convinced that the road made by the Incas formed the base of the road passing now from Pisco to Huaitara, that the Spaniards changed its course only where the river destroyed the original road or where the rocks or slopes came down by themselves, and further that the road made originally by the Incas was in a better condition than that which exists now. On the other hand, I venture to express my opinion that many things which have been said about the ancient roads of the Incas may be exaggerated. For, it is undoubtedly not true, at least for the ancient road from Pisco to Huaitara, which probably was one of the most important of the empire, that the ancient road ran straight and at its full width, both over the plains and through mountains. [p.3, (PICTO5O3)] The ancient road possessed its full width where the surface allowed it. Above slight inclined slopes (of earth) they occasionally fortified the road, filling up the ground at one side, so that the road might preserve its full width. But it is undoubtedly not true that, where the road was to be cut in the Letter of October 9, 1901 57 sides of precipice, it possessed a greater width than what is absolutely necessary, about 1- 1 1/2 meter. Nor is the road level in such places. But it was sufficient for passengers on foot, for llamas, and undoubtedly for the able people who carried the Inca seated on a litter. Perhaps an ancient Indian may have felt less horror in passing in such a manner above steep precipices than many modem passengers do. I may say that probably half of the road was much narrower than 5-6 meters, probably not wider than 1.5 meters. Precipices consisting of loose soil were probably not opened wider than about 1 1/2 meters to give space for the road, though it would have been easy to open some wider. Walls were constructed to border the road, it seems to me, only in those places where travelers might wander too easily over neighboring fields and so on. In other parts of the plains only rows of stones, set in lines on both sides of the road, served to border it. But in such case the direction of the road was always straight. Some admirable ancient walls were constructed by the Incas above rocks, as substructions on one side of the road, vertically above [p.4, (PICT0504)] the river between Huancano and Pampano, about 30-40 kilometers above Tambo Colorado. All the upper part of the valley of Pisco is filled by andenes on the lower slopes of the hills and ancient fields, --preserved where they have not been destroyed by the river during the last centuries. Huancano and Pampano seem to be names half of Kechua origin. Huanca = big stone; Pampa = plain (indeed, Pampano forms a plain in the midst of the mountains). "No" seems to be a formative of indigenous origin (that is, an older original language of the valley) siglrifying "spot of." Huaullanca, Callahuasi (probably Qalahuasi = house of stones, or house among stones) are also Kechua names. The same is the case with Pallasca, the hacienda to which Tambo Colorado belongs. I saw ancient remnants in an ancient cemetery recently excavated near Huancano. The remnants belonged to the period just following the Incasic conquest of the valley, showing the indigenous kind of civilization (of the coast) corresponding to that period. "Muchic" or "Muchica," about 3-5 kilometers below Huaitara, is curiously reminiscent of the name of the ancient "Mochicas," who, according to a recent rediscovery of the late Professor Brinton of Phildelphia, occupied the valley of Chincha and spoke the same language as the Chimu of Trujillo (compare also the name of the village Moche, near Trujillo, and the name of the principal azequia, [p.5, (PICTOS05)] "Mochica," near Trujillo, which originally conducted water to the big town of the ancient Chimus, northwest of Trujillo). Huaitara contains some particularly fine remains of the Incasic period of this part of Peru. It has an Incasic bath of stone, which I shall have cleaned, and some ancient stone walls, which now form a large part of the Christian church of the village. These walls consist of quarry-stones (apparently of syenite and porphyre) marvelously worked, so as to fit absolutely together in 58 Explorations in the Pisco Valley a polygonal way,' as they express themselves here in such a manner, that "no knife may enter between contiguous stones." These walls contain about 10 high niches2 marvelously worked and wonderfully preserved. It will be my work, to measure, to study, and to photograph these wonderful remains, which are scientifically unknown, none having been presented at any time in any scientific work. A few leagues from here, there is another important Incasic ruin, which bears the name of "Ingahuasi" (house of the Inca), which I shall study equally. I am not yet sure if this latter ruin consists of bricks of adobe or of stone. It seems that about 12 leagues from here there is a spot where ancient mummies may be excavated. Other ruins are said to exist, upon a mountain about 600 meters above Huaitara, facing it. The road to that part seems to be of the [p.6, (PICT0506)] worst of this part of the country, running in an infinte "cuesta"3 upwards along the steep sides of the hill. The position of those said ruins reminds me much of that of the ruins of Marca Huamachuco, including its relation to the plain which contains a modem village (there Huamachuco, here Huaitara). I am here at an elevation of about 2900 meters above the sea. Huaitara, a poor village of about 300 inhabitants, set beside wide and nice fields of alfalfa, is surrounded by mountain ridges of 500- 1000 meters near round on all sides of modern. The isolated summit, which bears the mentioned rare remains of finest Incasic masonry, possesses a wonderful prospect, reaching from the mountain ridges (along the valley of the river) down to the region of Huancan (a distance through the air of perhaps 8 leagues, or 35-40 kilometers). It would have been easy to receive fire-signs at night from there along the valley of the river. But the river which waters Huaitara is only a secondary left-side tributary of the principal river of Pisco, which has its confluence with its principal left-side tributary near Pampano, about 5 leagues below Huaitara (in a straight line). I venture to note here that I am in some hope that perhaps another of the still unknown ancient [p.7, (PICT0507)] Peruvian languages may be discovered at a distance of about 22 leagues from here. My information is still very vague; I shall try to get a few individuals brought from there to this place. In this case, I would have another important proof against the erroneous conception of the late Professor Brinton about the original linguistic condition of ancient Peru. I have long since prepared a paper on this topic. I must still repeat that my information is still very vague, and for long time I had nearly despaired of discovering any more of the original Indian languages in the mountains of Peru. It will, therefore, be possible, that my recent suspicion of another ancient language still living is wrong. Nous verrons! But you see, Dear Mrs. Hearst, that I do not forget to make all 1 It is not clear what Uhie meant by "polygonal." This term is used today to denote masonry of irregularly shaped stones. The walls of the church at Huaitarai are of regularly coursed masonry, with courses of more or less even height (editor's note). 2 There are in fact 11 such niches, 8 on the church's south fa,cade, and 3 on its west side (editor's note). 3 In Spanish, a "cuesta" is a slope or a path running up/down a slope (editor's note). Letter of October 9, 1901 59 possible efforts in this scientific direction too. At Huaitara, Kechua is still spoken generally. At Pampano, Kechua is also partially used still. At La Quinga (6 leagues below Pampano), Kechua is at least still known to some individuals. I shall profit of the occasion to get familiar here with the interesting ancient idiom. I remain, Dear Mrs. Hearst, Yours very sincerely Max Uhle Please excuse the haste in which the above lines were written. [p. 1, (PICT0508)] San Francisco January 1902 Dear Mrs. Hearst, I have the honor to give you my report of the last part of my work done in Peru, comprising the time from about October 9th to the end of that month, last year. I dedicated this time to a trip to the mountains above Pisco and its valley, especially to the region of Huaitara indicated in the maps as being at an elevation of about 2900 meters. I left Pallasca (near Tambo Colorado) on October 9th1 and arrived at Huaitara the next day after two-days' trip of nearly continual climbing up. I advised you of my presence at Huaitara by a letter from there,2 in which I laid also down some observations made along the road. I left Huaitara on October 23d, arrived at Ica on the 25". The few days between that day and the morning of 29t were fortunately sufficient to break up my entire apparatus of traveling, sell animals, etc. Leaving Ica on the morning of the 29th, I arrived in Lima on the last of the month, and the few days which I had up to the then rather unexpected arrival of [p.2, (PICT0509)] the steamer "Chile" of the English Pacific Steam Navigation Company, had fortunately been likewise sufficient for finishing so many other arrangements necessary before the beginning of a four weeks' sea voyage. On November 7th, I left Callao (the port of Lima) by the mentioned steamer, and as you know, arrived at San Francisco on December third. In my description of the road from Tambo Colorado to Huaitara (given in my letter written from the latter place),3 I did not mention that on many parts of the road there are also ancient petroglyphs. The first I observed on passing along the road were about 4-5 miles below Huancano (which is about in the middle of the journey between Tambo Colorado and Huaitara) on a big rock which constricts the road, at the left side of the river bed. These petroglyphs had been applied to an immense even surface, vertical by nature, of the mentioned block, but consisted only of a few flourish-like drawings, the antiquity of which might be proven, perhaps, with difficulty. More than 5 miles above Huancano, there is a long narrow ravine through which the river of the valley has broken in earlier geological periods. The declivitous rock at the left side of the road (with the right bank of the river to our right and 1 According to Uhie's Notebook N 57, p.142 (in the archives of the Ibero-Amerikanisches Institut in Berlin), Uhie left Pallasca on October 7th. The letter to Phoebe A. Hearst from Huaytara, dated October 9th1, states that he "arrived yesterday," i.e., on October 8"'h (editor's note). 2 Letter of October 9"', 1901 (editor's note). 3 Letter of October 9"', 1901 (editor's note). Letter of January 2, 1902 61 below us) consists of slate. Many large and small boulders are strewn over the slopes where these [p.3, (PICT0510)] are not too declivitous. Slate is a very inviting material for people who are fond of writing or drawing on rock. It is easy to carve writing in the stone, and the written signs, which come out in clear contrast from the stone due to their light color, stand after that for centuries and thousands of years. In this part of the road there are found many petroglyphs, but distributed here and there over many different boulders, different in size, along the road. Undoubtedly, the splendor of nature is overwhelming here in that narrow ravine, where the river broke through. But I would not argue -as it has been done sometimes in the literature -that it is the imposing character of nature that animated (the dull feelings of) the Indians to carve signs and drawings upon the stones. It is simply that the ravine presents good opportunities for writing and making signs or drawing on the native slate, and the Indians passing along this road took advantage by laying down their intuitions. [p.4, (PICT0511)] It can be easily seen that a number of ancient Indians, and perhaps Indians of all times, made signs there on the rock, by the fact that this is exactly the ancient road -the only road in ancient times -from Pisco to the region of Ayacucho and Castrovirreina, and is still a well-traveled road between the same places. Undoubtedly modem travelers on horse-back would not have much time to make signs on the rocks while being on the way. But considering the Indian manner of traveling on foot alone or with llamas, which do not advance more than about 6-10 miles a day, one can easily understand that Indians had plenty of time for making these signs near the road. This becomes especially clear from the consideration that traveling Indians are not accustomed to pass the night in the villages and towns, but remain on a sheltered or unsheltered spot wherever they feel inclined to stop their voyage for the day, along the road, sleeping there exposed to any weather, and maintaining themselves on the small, cold provisions, which are generally carried by them. The petroglyphs on the mentioned part of the road are of a different nature. I would not say that they are in the majority important by themselves, though by looking them carefully through, it would be probably possible to find a few of some greater [p.5, (PICT0512)] interest. They consist partly of llamas, partly of human faces or human figures roughly drawn, insignificant flourishes, etc. The Christian cross is not missing either, which is one of the proofs that the custom of making drawings on the rocks has had more modem followers. But undoubtedly there is also a good stock of petroglyphs which are of pre-Spanish origin. A petroglyph like this: 1! 62 Explorations in the Pisco Valley points undoubtedly to a pre-Spanish period for its origin, because of the feather-dress worn, and the antique weapons held by the figure. I also copied only a few figures of llamas: 'CS and these curious signs (perhaps textile ornaments) and something like I further did not mention in my letter written from Huaitara, some well constructed substructions of the road made of stone by the Incas at a certain spot between Huancano and Pampano. At that spot the road is especially narrow and dangerous, as one has to descend down over rocks along the road while having abysms of about 40 meters, falling [p.6, (PICT0513)] directly down to the river rushing below. At the other (the left) side are steep rocks hanging over the travelers. This was, without doubt, one of the most difficult parts in the construction of the road which leads through the valley up to the mountains, for cutting the road in the living rock could not be avoided. Besides this, substructions of fine masonry of stone were made at the right side of the narrow passage, where the road is hanging over the cliffs. I was told at Huaitara, (I did not know that while passing the spot) that this masonry work is considered as having been done by the Incas. And this appears now quite credible to me, on account of the fact that the Spaniards have not done nearly any work for improving road communications in Peru and lived there only in a style of pure waste of the good works inherited from the Incas. If those substructions, as I believe, are of Incasic origin, they are worth mentioning as prominent signs of the carefulness with which the Incas provided their country with lines of communications, which were undoubtedly splendid for their time. Approaching Huaitara, one passes some mineral deposits. Interest is now awakened for exploiting them by mines in search of gold, copper and silver, and [p.7, (PICTO5 14)] many people hope good results of it. Letter of January 2, 1902 63 The reasons why I laid stress on a visit of Huaitara, have been given by me in my former reports and letters. Huaitara is a small mountain village about 2800 meters (maps say 2900) (Aneroid 547.S)4 above the sea, with about 200-400 inhabitants. Its situation is in the "Cabeza del valle" (as they says in Peru), that is to say at the head- end of one of those mountain valleys which send down streams from the highlands of the Andes to the Ocean. Though, one being in one of these valleys, as for instance at Huaitara, and coming up from below, one does not become aware that he is near to the extensive highland which gives a little rest to the traveler in climbing. Declivitous and gigantic mountains rise on all sides, and the eye finds only with difficulty any horizontal line to rest upon. All is slope, declivity, and precipice. So I prepared the general map of Huaitara and its environs by means of trigonometric calculations, using the individual peaks around me as landmarks. My trip to Ingahuasi5, which lies about 1000 meters higher than Huaitara in the direction of Ayacucho, showed me the highlands of the Andes a little better from above, but I understood fully and became entirely impressed of the situation of Huaitara in a deep valley [p.8, (PICT0515)] cut into the border of the vast highland, only after I had left Huaitara, when I saw its valley, and the neigboring parts of the vast highland aside and behind it, from a mountain ridge, in which I had to pass to the height of over 3800 meters above the sea, on my road back to Ica. It seems that all "cabezas de los valles" in Peru show more or less the same character. I remember of La Paz in Bolivia, which also lies in a deep mountain gorge, scarcely possessing softer slopes along its steep sides, where it is possible to climb out of it. The climate of the "Cabezas de los valles" is generally temperate owing to their deep depression of between 500 and 1000 meters below the neighboring highlands. Huaitara itself does not lie in the immediate neighborhood of the river. The open and sloping plateau between the steep slopes of the surrounding mountains, on which Huaitara is situated, extends about 1/2 mile from east to west and about 1.25 (or counting differently, 1.667) miles from north to south. One and a quarter mile north of Huaitara there is a joining of two small rivers, the one comes from east,6 the other from southeast.7 The former holds more water. After joining, the river8 meets another river, which holds still more water, about 12 miles below, near Pampano and together join with the river Pisco, which falls into the sea near Pisco. In the valley of the small 4 Aneroid barometer: a barometer in which the action of atmospheric pressure in bending a metallic surface is made to move a pointer (Webster's Seventh New Collegiate Dictionary; editor's note). 5 Uhle's "Ingahuasi" is today's "Inkawasi," not to be confused with Inkawasi near Lunahuana in the Caniete Valley (editor's note). 6 According to Uhle's map on p. 127 in this volume, this is the Arma River. Map previously published in Max UhIe (1856-1 944), Pidine archaologischer Statten im Andengebiet, AVA- Materialien 56, W. W. Wurster, ed. Verlag Phillip von Zabern, Mainz am Rhein, 1999, fig. 71, p. 167 (editor's note). 7' The name of this river today is Sanquiniyoc River (editor's note). 8 Today this river is called the Huaytara River (editor's note). 64 Explorations in the Pisco Valley upper river, which comes from the south- [p.9, (PICT0516)] east, is situated Huaitara. But the river runs down along the mountains about 500 meters east of Huaitara, and the direct supply of water for the village is given by one of the small brooks, several of which fall down the high mountain from the east to west, and the channels of which are cut deep in the country, forming many impediments for crossing traffic. From the smallness of the supply of water, it is to be supposed that Huaitara never can have had a great many inhabitants, whether in ancient or in recent times. Even the fields are suffering by the small amount of water which is at disposition for irrigation. Though many houses of Indians are distributed over the valley, it must be noted that the number of people that might live in the valley and maintain themselves, especially if they had to live crowded together as in Huaitara, is not very large. The two small rivers, which join below Huaitara, belong to two separate valleys that join where the rivers join. The rivers join about 400 meters below Huaitara. For the whole it is to be said, that the ascent of the last part of the road before Huaitara is especially rapid. One ascends about 400 meters from the northwestern foot of the valley (the upper end of the county Muchi, about 1 2/3 mile from Huaitara). [p.10, (PICT0517)] The valley of the small southeastern river is filled by debris which resulted from immense destruction of the surface of the earth and of the mountains in long gone geological periods. In its long extension of about 5 miles, it does not form anything else than an immense inclined plain slowly sloping down from about 3400 to 2300 meters. This slope presented many difficulties to an easy administration of agriculture. For that reason, one becomes aware of andenes (terraces which support the artificial horizontal fields) everywhere. The whole valley is filled with andenes, which form the most striking element in the view of the country. The steep sides of the mountains are, of course, barren, and it is only the valley floor which has been subject to agriculture. Upon the upper part of the sloping plateau of this valley is Huaitara, with inclined streets, situated at the foot of the mountains behind it. The fields at this altitude produce mostly alfalfa, wheat and potatoes, but of course, the alfalfa is of inferior quality compared to that of the lower parts of the valley as much due to the greater coolness of the climate, as the scarce supply of water. Huaitara is lying on the road from Pisco to Ayacucho, which passes through the valley of Pisco. Besides this, the road for Castrovirreina, the principal [p. 1 1, (PICT05 18)] town of the province of the same name in the Department of Huancavelica, starts from there. Ancient travelers from Ayacucho to the coast passed Huaitara, as they passed Tambo Colorado and all the Pisco valley. At the present time, this road has lost much of its importance due to the existence of the railroad connecting Pisco with Ica. But still some travelers make use of it, especially those who go down with animals from Ayacucho directly to Pisco and the coast, among which are also many Indians with mules or donkeys. The other road for Ayacucho, of Letter of January 2, 1902 65 greater importance at the present time, ascends from Ica, in the valley of the river of Ica, to Tambillo, which is a night-harbor in the upper part of that valley, and passes from there to the highland directly. But Tambillo is also connected with Huaitara by a road which passes from the one valley to the other by crossing a mountain chain about 3850 meters high, and a few travelers continue to take this way passing from Ica to Ayacucho, by way of Tambillo and Huaitara. The Peruvian mail makes use of the same line, as, starting from Ica, it passes by Tambillo to Huaitara in 2 days, and reaches Ayacucho from there in about 5 or 6 days. All this proves, that for Peruvian conditions, Huaitara is not so unimportant a place. On my own route, I ascended to Huaitara from the valley of Pisco, taking the Pisco-valley- [p.12, (PICT0519)] road which runs to Huaitara, and descended from there to Ica by taking the other road which passes through Tambillo and the valley of Ica. The latter road is recent and has no significance for the study of ancient lines of communications. But, though in some parts good, especially where nature helped, it is sometimes rather bad in others, and it is interesting and amusing to see, what, in some parts, was understood as construction of roads by modem Peruvian people. A low, rocky, hill-like ridge is projecting from the steep mountain at the southern side of the valley of Huaitara to the north (see pl. 1293-1294). One of its summits at the north bears the cemetery of Huaitara (pl. 1293, to the right), another, of pure rocky nature, to the south contains a streak of copper. A small depression, between the latter and the foot of the mountain to the south, bears the church of Huaitara, enclosed by a large irregular court at the present time (see pl. 1293). The eastern side falls off in terraces. These terraces, and others leaning against the northern foot of the southern mountain, bear the modem village of Huaitara (pl. 1293-1294). Huaitara is the capital of the province of Castrovirreina (Dep. Huancavelica) with a subprefect, a judge, a sort of jail, and a guard of four soldiers of inferior rank, as police of the entire province. This preference has been given to Huaitara more for the softness of its climate in comparison with the other settlements [p.13, (PICT0520)J of the province, which are all in much colder climates, than for the importance of the town itself. The inhabitants belong to the mixed race, still inclining much to the Indian habitus. Indians of still purer Indian provenience are mostly living around the town, scattered through the valley in not very small number. People of Huaitara may be said to maintain themselves by agriculture, though, as in many cases in Peru, with many of them it is utterly unintelligible what they live upon in reality. Many people do not do anything. Drinking fills a good time of the life of many of them, and it is not uncommon there to see a good part of the population on a weekday morning arranging cock-fights in the street. Huaitara belongs to the area in which Kechua is still much spoken, in general, much more than Spanish, and, for instance, many women do not understand any Spanish words, though the Kechua is mixed with Spanish words. Kechua can be found on the river down to Pampano. It is worth 66 Explorations in the Pisco Valley noting here, that the dialect spoken at Huaitara is the same as that which is spoken at Ayacucho and Cuzco, that is: the dialect "of Cuzco." The other important dialect of the Kechua language, that of "Chinchasuyu," reaches no farther to the south than Huancayo in the north. Herdsmen, living [p. 14, (PICT0521)] far from the general lines of communication, in the environs of Ayacucho, for instance, are said to speak still differently, for example with a regular change of r into I in a word. And the same thing happens in a small separate Kechua dialect which is spoken only at Chupamarca and Chavin, about 40 miles north of Huaitara. The Indians of Huaitara are still distributed in "aillus" as at the time of the Incas. There are 4 aillus at Huaiatara -Quito, Wayacundo, Chocorbos, and Yunca- which form together to parties -Hanansaya and Hurinsaya. The first aillu, Quito, comprises one quarter of Huaitara, Huatas (the last and highest part of the valley, on the road to Ayacucho), and one part of Armas (a small village in the northern valley). The second, Wayacundo, the second quarter of Huaitara and the other part of Armas. The third aillu, Chocorbos, comprises the third quarter of Huaitara and is also found on the road of Castrovirreina, near the bank of the river. Yunca, the fourth aillu, comprises the last quarter of Huaitara and Muchik. "Yunca" 9 means "warm" in Kechua, but it is also used as a geographical denomination for warm regions. At the same time we know that the Incas called certain nations Yunca, especially those of the western coast, from the hot climate in which they were living. Muchic is the lowest part of all the environs of Huaitara. [p. 15, (PICT0522)] Curiously, the name of the county, Muchik, reminds one of the ancient Mochicas, that is the nation which lived in the valleys of Chincha, Pisco, and so on, as is known by ancient information, and which spoke about the same language as the people around Trujillo, where various names like Moche and Mochica are found. - In the name of the first aillu, Quito, there is a surprising similarity with the name of the modern capital of Ecuador, and with the ancient Quitus, who, as we are told, were subdued by the Incas. I cannot pass without also calling attention to the fact that Quito is also the aillu of the highest and coolest region belonging to Huaitara, as Quito is a high and very cool place in Ecuador. The aillus have still significance for the distribution of the public works and the annual festivals. The Indians still follow various customs of their ancestors in sowing, harvesting, marriages, and so on. The songs in Kechua, however, executed at these occasions, some of which were pointed out to me, mostly do not seem of very old origin. 9 "Yunca o yuncaquinray: los ilanos o valles, Yunca: los Indios naturales de alli" (Diego Gonzalez Holguin, Vocabulario de la lengua general del Peru, liamada lengua qquichua o del Inca, [1608] Universidad Nacional Mayor de San Marcos, 1989) (editor's note). 10 Whether or not the Moche have actually occupied the Chincha and Pisco valleys is unknown, but Moche-style objects and iconographic themes have surfaced in Ica and Nasca (Christopher Donnan, personal communication) (editor's note). Letter of January 2, 1902 67 They still dance in certain festivals, for example on Christmas they dance "Pallar," Kechua "palla"= Seniora, on Carnival, at the festival of sowing etc., "jauca," some kind of round dance, found nearly everywhere in the mountains of Peru and Bolivia, danced without fancy dresses, and on January 6th, "Negritos." The dance of January 6th is undoubtedly of modern [p. 16, (PICT0523)] origin, as it reminds entirely of the negro kings of the holy legend of that day. None of these dances could be seen by me, as it was not the time for them. With their few dances, Indians of Huaitara fall very short, for instance, of the Aimaras of Northern Bolivia, where I noted and partly observed about 36 different dances, most of which were executed with masks and in fancy-dresses. But the observation and recording of all such ancient Indian customs throughout Peru is an investigation not yet carried out, except on chance occurrences when an observer happened to be present at the right time. I was surprised to see women carrying their babies about in cradles hanging on a strip slung over their shoulders. This use is a reminiscence of ancient times, when the bed and cradle of the babies were the same, and mothers carried their babies in the cradles. I took a picture, though the mother did not seem much inclined to expose the baby to the photographic apparatus, and hid it partly behind her. The church and the churchyard of Huaitara are sited on the only ancient ruins at Huaitara. As I discussed, the church and the churchyard are lying west of Huiatara, about 20 meters above the plaza, in a small depression between a small rock at the north and the foot of a high, and in the upper part very steep, [p. 17, (PICT0524)] mountain to the south (see, map fig. 1). 1I The eastern side steps down in terraces from the churchyard (pl. 1293). At the western side, the ground falls by a steep precipice down to a small separate valley-like cut. I cannot believe that it is an accident, that a person standing in the churchyard has an interesting view to the west, over the feet of the mountains, rising from North and South, thus forming the channel of the valley as far down as to Huancano (about 8 leagues below Huaitara), where the view ends, closed there by two small symmetrical summits at the end of the valley. The churchyard is of irregular shape, supported by high retaining walls below on the east and west, and enclosed by low walls all around. It has two entrances: one at the northern side and one at the southeastern corner (pl. 1293 and 1272). I hope, that my maps and photographs will give a detailed idea of the general situation (figure 1, and especially figures 2 and 3). The ancient ruins found here are of stone and consist of several ancient walls, different in the quality of construction and by decoration. The modern church has been constructed by setting walls of adobe bricks upon a thick, 1 Map fig. 1 is on p. 123 and was also reproduced in Wurster 1999 fig. 71, p. 167 (editor's note). 68 Explorations in the Pisco Valley ancient three-sided stone-wall (pl. 1272 and 1293),12 which relieved the modem builders (of about the 16' century) of the need to build a foundation for the church, and to which is owed the preservation of the ancient walls. [p. 18, (PICT0525)] The front part of the church with its side-towers has been added in modem masonry. If some ancient walls stood there before and have since been extirpated, this cannot be determined, because the eastern ends of both the southern and the northern ancient walls are intact (see for instance pl. 1278). Near the southeastern corner of the churchyard there is another ancient wall, forming, with two others of the same age behind it, a sort of chamber (see fig. 2,13 and pl. 1275 and 1299). The first of the three walls, similar, though inferior to the ancient stone wall of the base of the church, is now only a remnant of a larger one. It suffers from continued destruction, little by little, and it may be that in a few years nothing more will exist of it. The high retaining wall supporting the churchyard at its western side between two bordering rocks, that of the north and that of the south, is also old (pl. 1302). Besides these, there is an ancient fountain stone lying near the south- western corner of the churchyard on an improvised modern bath (see fig. 12). It will have served for conducting water to the interior of the ancient building from a water-channel, which still passes along the southern side of the chulrchyard, and which may have been conducted there for that purpose by the ancients. The fountain stone is also carved with two channels for water flow, as in ancient fountains on Titicaca Island and so on.14 Two round pedestal-like ancient stones are now in use within the church for setting holy flags in them (see fig. 13). These are apparently only an extremely poor remnant of what existed before the arrival of the Spaniards. Undoubtedly many old walls, once forming a complicated [p.19, (PICT0526)] and well-disposed building, have been demolished to give room to a plain churchyard for Christian necessities. The innumerable old, worked building- stones laying around in the houses of the modern village, are partly used in the construction of all these houses, in the repairing and construction of the terraces, in the construction of the doorways in the churchyard, as well as in the towers of the church, give us proof of this fact. In all parts, the fine though not sumptuous hewing of these ancient stones contrasts widely from the rough form of the modern stones used with them and from their rough and ordinary application in the modern walls, which says much in favor of the destroyed ancient civilization. 12 What Uhie meant to say was not a wall with three sides, but rather a building with walls on three sides (editor's note). 13 Fig. 2 is on p. 128. It can also be found in Max Uhie (1856-1944), Plane archaologischer Statten im Andengebiet, AVA-Materialien 56, W. W. Wurster, ed. Verlag Phillip von Zabern, Mainz am Rhein, 1999, fig. 74, p. 169 (editor's note). 14 Examination of the site by JPP in 2003 suggests that the fountain stone remains in its original location as placed by the Incas (editor's note). Letter of January 2, 1902 69 Since any ancient walls within the churchyard have been destroyed, it is now very hard to make detailed assertions about the original disposition of these buildings. It seems to me, however, credible that the part preserved as the base of the modern church was the most prominent part of the ancient buildings, both in proportions and in decoration, so as to be selected for serving in the construction of the church. In this way we may be at least consoled by the observation that the part preserved was probably the best of all that existed, and so may give an idea of the chara- [p.20, (PICT0527)] cter of the best and main parts of the old buildings so badly treated. Sculptured stones also formerly existed. The modem adobe wall of the western gable of the church shows several (4 to 6) square stones symmetrically inserted. These stones bear all the signs of having been mutilated on their front side, as though sculptured heads had formerly projected from them, which are now missing due to their mutilation. Although, it cannot be determined from the marks at what time the mutilation occurred. I searched in vain for the sculptured fragments broken off of these stones, which may exist still in some unknown spots.15 There are some other parts of the constructions which seem to be of pre- Spanish origin, according to the style of the architectural decoration represented by them, which, however, are not so old. I would describe them as "pseudoantiquities." To this class of constructions belong the portal at the south eastern entrance of the churchyard (reproduced in plate 1298), which imitates Incasic architecture in both shape and proportions, in the technique of roofing, and in the niches, which serve as decoration to the western back- wall of the churchyard (see some of them at a distance in view pl. 1275). The mentioned portal narrows [p.21, (PICT0528)] at the top, and possesses a recessed frame as decoration, like so many Incasic niches of that kind. The inner height of the gate is 2.80 meters. About 58 centimeters of thickness of the wall, which is 1.60 meters, represent the depth of the recessed frame from the face of the wall. The gate had been roofed by using wooden boards tied together with strings, which is the character of roofing in many true Incasic niches.16 But more accurate observation of the gate, which in the upper part consists of adobe, in the lower of stones, teaches us that in the construction of the portal many stones of ancient Incasic masonry have been used in a secondary way. These stones are mixed there with other ordinary stones and 15 The modern western gable of the church is built over the original Inca adobe gable, the outline of which is still clearly visible, including two original windows near the top. The square stones to which Uhle refers are stone pegs, symmetrically arranged, along the outer edges of the gable. Such pegs, which are typical of Inca architecture, are believed to have served to anchor the roof structure to the building. The part of the pegs set into the gable wall is square, but from that square base projected a cylindrical stub that now is indeed broken off. When searching for the broken off parts, Uhle may have thought of tenon heads as found at Chavin de Huantar. He may never have seen the pegs on Inca buildings in the mountains before (editor's note). 16 The description Uhle gives of the "roofing" corresponds to the construction of the lintel over the doorway. The Incas did not use boards but sticks or logs (editor's note). If indeed Uhle observed boards, then the lintel may not have been of pre-Spanish origin. 70 Explorations in the Pisco Valley in no part fit together as where used in the ancient buildings. It is, therefore, absolutely necessary to consider this portal as a more recent imitation of ancient Incasic niche and door construction, however curious such a counterfeit may appear. The niches on the back wall of the churchyard are distributed at distances of about 1.25 m from each other. They are not quite complete now. Some of the distances may also appear irregular. But the best proof of their more recent origin is in the irregular mixing of stone and adobe in the construction of some of them, and in the main material (bricks of adobe) being of the same kind as that of which the upper part of the church [p.22, (PICT0529)] has been constructed. The high, eastern retaining wall built of stones from the churchyard (see pl. 1272) seems, in some parts, to be of better construction than in others. There may be distinguished about three to four different kinds of quality of construction in it. A few stones seem to represent the ancient system of fitting stones together. I became, therefore, inclined to suppose that the wall was originally old, but was repaired so often over the course of centuries that little of its original character has been left. The lower part of the wall seems to show better fitting of stones than the upper part, so that the possibility that perhaps there were two terraces originally, the lower of which was, in recent times, built up and filled with rubble to the height of the higher, may not be rejected yet as absolutely improbable (see pl. 1272). The material of the building is stone, a light, reddish porphyry-like of no excessrve hardness, which is from the environs, for instance it can be found some miles below Huaitara.17 Stone is the natural material for construction in the mountains, as is adobe on the coast, both by the quality of resources as by the necessities of the climate. It is by no means surprising in this way, that while we found Incasic monuments constructed of adobe at Tambo Colorado, we shall find them constructed of stone, and in a quite different way, at Huaitara. Not all these latter remains show entirely the same technique of masonry. But the general [p.23, (PICT0530)] character of their technique is the same, and consists in fitting stones, embossed artficially or by nature, so closely together that not the slightest slit can be seen between them.18 Mortar (of clay) is invisible at the surface. For fitting the stones so well together, it was necessary to grind the sides down till both neighboring stones fit against each other. The question of the number of the corners which a stone was to have after that was apparently subordinate. It did not matter whether the stones were rectangular or trapezoidal and set up in horizontal rows upon horizontal rows (pl. 1305, and so on), or if they were polygonal and formed an irregular pattern of stones in the perfect building (see pl. 1302 and 1299 at the left). It was, as I understand the matter, only a question of perfection or taste, if the one or the other shape of stone was chosen. From these results, it can 17 The stone to which Uhie refers is andesite, not porphyry (editor's note). 18 Today's scholars refer to what Uhie called "embossed" stones as stones with a pillowed face (editor's note). Letter of January 2, 1902 71 be seen that it is incorrect to base a system of determining architectural period from observation of the different number of corners of the stones, and upon the different manner in which the stones are set together. This faulty dating method has, in this way, been used to antedate the cyclopic style to the rectangular or trapezoidal style, while both are apparently contemporaneous. It would be only possible to separate them chronologically in this case, if it [p.24, (PICT0532)] were possible to prove that the present remains at Huaitara were created as the works of two succeeding periods, the one of them having superimposed its works on those of the other. But there is no probability for assuming this. All remains appear to have been created and left by the same people, the nation headed by the Incas. The difference in technique depended of the different services to which the walls were destined. Those walls which were to hold together the earth behind them were constructed of rougher blocks, showing still more natural shape, than walls of buildings standing free (see pl. 1299, to the left, and 1302). From this point of view we easily understand the difference in technique and stylistic perfection seen in various walls of these ruins. It is true that terraces in front of the convent of the Island of Coati, and another discovered by me upon the Island Intja in the southern part of Lake Titicaca, both constructed by the Incas, have been built after the system seen in view 1299 and 1275 at the right. But this circumstance would not compel us to assume that the Incas should have built walls of similar function in other parts of the country, and in the same luxurious manner, too. The fitting of the stones in the face of the wall is only one of the peculiar features of the construction of these walls. One would be quite wrong in supposing, that the stones used in the construction of these walls [p.25, (PICT0533)] fitted so well together along all their sides with the stones beside them. The masonry work of the Incas, at least that done at Huaitara (but probably that of other places too) was not of so solid a character. The stones fitted marvelously together, but only on the surface of the wall. The shape given to the stones behind the surface was rather arbitrary.'9 Further, the walls are not to be understood as being massive throughout. The stones of the outsides served only as a kind of veneering, but one of marvelous perfection, which do not let one divine the character of the construction at the inside. The inside of the walls consisted of ordinary stuffing of clay and common stones. Such is the construction of the wall at the southern side of the churchyard, and, in view 1275, this manner of construction can be studied perfectly. From this point of view, walls like those represented in plate 1276 and following, which were are adorned by many niches, appear still more costly, as the niches gave much less room for this easy method of construction. But that the same principle, of stuffing the inside by clay, was also applied there 19 If Uhie's observation is correct, it is also true that in many Inca walls the stones were carefully fitted over the entire depth of the wall. Observations at the church of Huaytara suggest that it too had been built with stones fitted throughout the wall (editor's note). 72 Explorations in the Pisco Valley when it could, for instance in the lower parts of the wall which contain fewer niches, seems to me to be beyond doubt. [p.26, (PICT0534)]20 I would classify the walls under three divisions according to the different techniques: first: with stones artificially embossed at the surface. The outside of the stones, rectangular or trapezoidal, set so as to form very good parallel lines of stones, set above each other. See the lower walls of the church (pl. 1272, 1305). second: with stones artificially embossed. The outside of the stones, rectangular or trapezoidal or more or less polygonal, but, in general, more irregular (pl. 1275 and 1299, both to the right), forming irregular lines in the wall. So the same type of construction is found in the walls mentioned above on the Island of Coati and Intja in Lake Titicaca. At Huaitara the type is represented by the remnant of the high wall at the western end of the churchyard (see pl. 1299 and 1275). The single stones are more or less wedge-shaped, with the narrower end at the inner side, as can be well seen in plate 1275. But it is to be observed, that the stones are well worked all over, as well on the covered as on the uncovered sides, and on the inside as well at the front side of the wall. A wall constructed thus is, therefore, far from being a quite common kind of work. [p.27, (PICT0535) third: walls of more or less cyclopic appearance. The front side of the stones has been little worked over or not at all. The covered sides at the back of the stones are equally rough. The edges of the front side have been probably only superficially worked over for facilitating the good fitting of the stones. To this type, belong the wall in the left part of the view in pl. 1299, and the high wall supporting the churchyard between rocks at its western side (see pl. 1302). In the latter view it is also visible that the largest blocks are the lowest, and that the blocks are diminishing in size towards the top of the wall. All stones have been only hewn; no stones show marks of having been polished (pl. 1272). The principal ruin, now the base of the nave of the church-building, presents itself in the form of a large hall in the shape of a horseshoe, with 20 Pae26 shows some crossing out of phrases apparently done at a later time and in pencil. These corrections are not reflected in this transcript, as with the crossed out phrases the text would become unintelligible (editor's note). Letter of January 2, 1902 73 thick walls on three of its sides, south, west and north, while it is open to the east. The eastern ends, as I mentioned above, appear to be intact. There were no doors of the common Incasic shape, nor any continuation of niches at the ends. The hall is at the northern end of the churchyard, near the northwestern corner (see fig. 2 pl. 1293). Its middle part is at only [p.28, (PICT0536)] eleven meters from the sheer foot of the rock in the north, the top of which is about 20 meters above the churchyard. The northwestern corner of the hall approaches within 2 1/2 meters of the rock at the western end of the churchyard, where the rock drops off in a vertical precipice. A strange peculiarity of the hall, never previously found by me in similar ruins is, besides its being walled only on three sides, that it opens slightly to the east in a rather symmetrical diverging of the northern and southern side walls from the longitudinal axis. The wall at the west, which joins the two, is perpendicular to this axis. The alignments of the two side walls differ by about 3.75?. The southern wall, according to my compass-needle, is aligned at about 86015' to the East, the northern at about 82?30' to east. The opening of the hall stands, therefore, very well to the east, the region of the rising of the Sun. Making account of the circumstance that the compass-needle gives erroneous indications in Peru of about 100 too much to the east, the real geographical direction of the geometrical orientation of the central axis of the hall to the east would be about 74-75?.21 [p.29, (PICT0006)] The hall is nearly exactly 28 meters long, about 12.80 meters wide at the eastern end, and about 11.30 meters wide at the western end; all these measures being taken at the outside of the hall. The walls are, on all three sides, about 1.60 meters thick; a small step on the inside of the northern sidewall increases its thickness at the base (see fig. 3 and fig. 8 and 9). The inner length of the hall is, therefore, about 26.40 meters, the inner width at the west end about 7.90 meters, and at the east about 9.40 meters. The height of the ancient wall, upon which the adobe walls of the church have been constructed, originally amounted to about 3.30 meters, according to measurements taken at the inner side of the northern side-wall. Nothing can be known about whether the ancient hall had been roofed or not.22 We only observe that the top of the three-sided ancient wall forms a horizontal line over all its extent, and it is probable, that this was the height of the wall, unchanged since antiquity (see pl. 1276-1278; 1300-1301). A door opens near the middle of the northern sidewall into the small space between the hall and the neighboring rock (see fig. 2). The construction of the adobe walls of the nave of the church upon this wall is not the only change the ancient remains have undergone. Besides that we do not know if there were walls to the [p.30, (PICTOOO7)] east of it, 21 Uhie's original paragraph was particularly hard to understand. We have revised it considerably (editor's note). 22 As noted above (on p. 20), Uhie did not recognize the ancient adobe gable on the building's west side. The hall was roofed in Inca times with a two-sided, or gabled roof (editor's note). 74 Explorations in the Pisco Valley A door opens near the middle of the northern sidewall into the small space between the hall and the neighboring rock (see fig. 2). The construction of the adobe walls of the nave of the church upon this wall is not the only change the ancient remains have undergone. Besides that we do not know if there were walls to the [p.30, (PICI0007)] east of it, which formed an integrating link of the construction and are now missing; several other changes have been made. A chamber used as a sacristy has been constructed of adobe at the outside against the western end of the southern wall (see fig. 2). One of the niches has been broken through on behalf of this, so as to form a door which opens to the sacristy from the church.23 Niches have been covered by building the adobe wall against the ancient stone wall. Other niches have been altered to serve as shrines the sacristy. At the opposite, northeastern end, one of the walls of the left tower of the church has been set against the end of the wall (see fig. 2). A modern gate for the church has been constructed at the eastern end of the southern wall (pl. 1278). Several decenniums ago, the inner parts of the church were burned. At that time many stones of the walls and niches of the inner side of the church were damaged. The present curate renovated the church on the inside. For that purpose he filled in several of the ancient niches, and for giving a better aspect to the damaged stone walls he whitewashed them all, after having mended some of the damaged stones by means of lime.24 Several altars and wooden pillars of the nave of the [p.31, (PICT10008)] church now hide a great many niches of the interior, and considering that so many other niches have been filled up, and all is covered by whitewash besides this, it undoubtedly needs a special and devoted attention to make out all the ancient niches, and to determine their proper positions and the general system followed in their placement. There are 3 kinds of niches applied as decorations to the walls of this hall (see pl. 1276-1278; 1300-1301, fig. 4 and 5). 1. Niches the size of doors, rising from the ground with one step (if not two) at the base, a recessed frame and a small window perforating the 1.60-meter-thick wall at the back: Ir- --_. : f; 23 Tesacristy Uhle observed has now been removed and the broken niche summarily repaired (editor's note). 24 Tewhitewash and the lime mending have since mostly been removed (editor's note). Letter of January 2, 1902 75 The depth of the outer niche amounts [p.32, (PICT0009)] to 0.42-0.45 meters; the depth of the inner or recessed niche, to 0.58-0.62 meters. The niches show the characteristic Incasic inclination of the sides towards the top. All niches, as well as the northern door, share this feature on that building (see pl. 1301 and fig. 7). In this way the outer width of the high niches increases from 1.55 meters at the top to about 1.80 at the base, that of the window from about 0.32-0.35 to 0.38-0.40 meters. Such door-sized niches are not found on all structures of Incasic origin. For instance, they are missing in the palace-like building of pure Incasic origin at Tambo Colorado. They are found, for instance, in the Incasic convent of Pachacamac, in the square Incasic building at the foot of the Sun Temple-ruin of the Chincha valley (but there only at the back of the terraces, which I ascribe to the religious purposes of such niches),in some Incasic ruins of the valley of Caniete (see Squier, Peru, p. 83, 84, Middendorf, Peru p.139), several Incasic buildings of the island of Lake Titicaca (see, for instance, Squier, p. 343), and the ruins of Colcampata at Cuzco (according to a photograph in my possession). As they always signify a higher degree of solemnity, I expect to meet with them in buildings serving to religious purposes. The Sun-Temple [p.33, (PICT0010)] at Pachacamac also shows a line of high niches on a prominent spot, with the unimportant difference that their style there is combined with the architectural habitudes of the coast people. On all the other buildings compared above, the windows perforating the wall at the background of the niche are missing. They remind one of the small window-like openings met with in the Incasic buildings at the foot of the Sun-Temple-ruin in the Chincha valley. I might add here perhaps another observation, that is, that the niches of Incasic buildings of different provinces of the old empire seem to show some differences according to the provinces. These differences may appear to be based partly upon foreign influences altering, more or less, the original pure Incasic types, as for instance in the large niches of the Incasic Sun-Temple at Pachacamac, or those in Incasic buildings of the islands Titicaca and Coati (Lake Titicaca). I do not see much difference in the details and fundamental proportions between the high Incasic niches of the convent at Pachacamac, those in the buildings of the valley of Caniete, those of others in the Chincha valley and those of Huaitara here. On the other hand, those of the ruin "Colcampata" at Cuzco are stiffer, more serious, and open less at the base. But there is another possible explanation for this fact, that is, that chronological differences caused [p.34, (PICTOO 11)] the apparent differences of the general shape. For instance, it would be possible that all the niches of the Incasic buildings between Huaitara and Pachacamac are so alike to each other in their general appearance for the reason that this whole region came under Incasic domination more or less at the same period, and that in consequence the Incasic buildings in them were erected all in the same historical period of the empire and the same period of the stylistic development of Incasic architecture. If this manner of explaining differences might prove to be right, 76 Explorations in the Pisco Valley we would have found an extremely fertile principle for completing a chronological order of the different Incasic buildings of the Inca empire, which, applied to ancient Incasic ruins, might give us still much light for understanding details of the history of the development of the Incasic empire, and perhaps for understanding better the history of its development. The second type of niche occurring in this hall consists of high niches of triangular cross-section (see fig. 8 and 9). They measure a little less than the former in height, as they are only about 1.92 meters high, but as they rise from a pedestal-like step in the wall, about 1.20 meters from the ground (according to the present level of the wall inside the church - the original level was perhaps 0.50 meters higher, as can be seen from the height of the base of the high [p.35, (PICT0012)] niches of the outside), they rise in every case to such a height that their top was by no means lower than that of the high outside-niches. They were, therefore, certainly less decorative than the high outside-niches with their receding frame etc., but undoubtedly did not lack solemnity. It is very curious that I do not know of any other example of the application of the same kind of niches in any other Incasic building. But the niches share the feature common to Incasic niches, having the width increasing below; similarly, their proportions correspond entirely to those usual in Incasic architecture elsewhere. Their pure Incasic character seems, therefore, to be beyond doubt. The third type of niches consists of smaller ones, like small recesses in the wall above the ground (see fig. 4 and 5, fig. 10- 1 1). They are found in all Incasic buildings, but are here of specially fine cor1;truction. I comprise under them also the windows of this shape. They have been applied in horizontal rows. A fine feature, enjoyable to the eye, is the arrangement of two types of niches in an alternating order. Niches of the first and third type have so been arranged on the outside of the southern and western fronts. This circumstance lets the eye enjoy more in their aspect.25 The niches have been distributed over the wall sides in this way: niches of the first and third kind are alternating at on the southern and western front [p.36, (PICT0013)] of the outside. - I wish to mention here that I found the base of all the high niches covered by earth and over 0.20 meters below the common level of the churchyard. I cleaned the foot of all the high southern niches, but found only one intact (No 10). In all the others, the stones of the inner step of the niches had been taken away. Only one niche of those on the western front was cleaned at the base, and it was found intact. Due to the very exact similarity of all those niches, it appeared to be unnecessary to have the bases of all of them cleaned. Niches of the second and third type, alternating, were distributed over the inner side of the northern wall (see fig. 8). As for the decoration of the inner side of the southern and western walls, the windows passing through the wall from the high niches outside have 25 The details of the ancient door in the northern wall are unknown on behalf of its being entirely barred with stones. [Uhle added this as a marginal note.] Letter of January 2, 1902 77 already been described as one element. Undoubtedly the windows were not made only for decoration, but their effect was decorative at the same time. These windows alternate with niches of the same proportions but of somewhat larger size, producing in this way an agreeable effect (see fig. 10). The only side left without decoration was the northern (back) side of the northern wall. It is very curious to observe that not only did one of the outer sides differ with respect to decoration from the others (it being the back side of the building, thus the reason for the difference is intelligible), but that also [p.37, (PICTOO 14)] the three inner sides, two of which undoubtedly were symmetrical about the longitudinal axis, at least from our point of view, were quite different in decoration. It then strikes me, that the outer southern and western sides of the building undoubtedly were especially intended for show, and the corresponding inner sides were treated only as an unimportant back of the building in their decoration, while the third, north, side is treated as one of the principal faces of the building. All this teaches us: first: that the southern front must have had a court of considerable extent in front of it, so that the front of the building could make the right effect, secondly, that both show sides of the exterior screened from view the main body of the hall behind them, revealing only the prominent inner side of the northern wall. To the room in front of that latter side, the windows of the southern and western side served as outlook. But it is very curious and striking that according to this manner of understanding, which I believe is right, the whole building would seem to have been oriented to two fronts, first according to the eastern opening of the hall, [p.38, (PICTOO15)] secondly, according to the direction of the decoration of the inner face of the northern wall, to the south. I am unable to account for this now. It seems that there may have been another link now missing from the hall. This may have existed in the east. The direction of the inner side of the northern wall, like the exterior facade to the south, can be accounted for only with difficulty without supposing the former existence of an entrance leading to the hall from the south. But for the fact that, the southern wall, so far as it remains today, is intact and no door appears to have existed leading through it in ancient times, the former existence of a continuation of the hall to the east, and a gate east of it, might be suggested. But in this case the entrance was unlike the other doors, as the wall comes to a straight and vertical eastern end. And to suggest things which are against the architectural rule as known to me, cannot be recommended to anybody in anyway. As the outer sides served as an outer front for show, hiding the things which were behind it, in the same way, the decorated southern side of the northern wall served, perhaps, as a second front for show, hiding the parts which were behind it and accessible by using the door now barred in the same wall. But behind this wall there was only a narrow space [p.39, 78 Explorations in the Pisco Valley (PICTOO 16)] between the wall, which lacks decoration by niches, and a barren rock (fig. 2). I am, however, unable to refuse entirely the possibility of such an explanation of the facts found. It might be possible that the principal part in the conception of the buildings was not the interior of the hall, but the top of the barren rock immediately behind it. For instance, the top of the rock might have appeared as a principal site in the execution of religious ceremonies. But for the rest, we lack arguments for discussing this question further. Before proceeding I may mention some details respecting the niches. The precise measurements given in my drawings of the niches, which are nearly, though not entirely complete, will show the very exact observation by the builders of the sizes and all dimensions of the niches, which are all alike from one to the other (see figs. 4-6, 8-1 1). Some more measurements were taken by me, but as they could not be indicated in the drawings, and those given seemed to explain sufficiently the systematical exactness, I dropped them, or at least their reproduction here. Niches are alike in generally widening to the inside.26 My drawings of elevations and sections will show that sufficiently. It is possible that this circumstance has something to do with the wedge-like reduction of the sizes of the stones used, as observed in the inner construction of the wall represented in the view in plate 1275. [p.40, (PICT0017)] Though sizes and dimensions are repeated very exactly from niche to niche, one may not believe that the horizontal lines of stone also follow an entirely mathematical order. Slight variations may be observed by one who studies various niches more closely, but the variations are not so great as to appear by simply looking at the work of the ancients. It is no absolutely easy task to determine with certainty the original kind of service of these buildings. As the buildings were such of prominent character we may think of three kinds of uses generally possible for them: as a palace of the Incas as a convent for women or as a temple. The position of the buildings upon terraces little accessible between rocks answers little to what we are accustomed to see as having been observed in the construction of the palaces of the Incas. The curious shape of the main hall and the thick walls perforated by a few windows would, by no means, be accounted for by supposing that these buildings were constructed either for a palace or convent. I would not be able to account for such curious constructions in any way but by assuming that [p.4l, (PICTOOI 18)] the main hall, now the base of the nave of a church, was 26 WhtUhie is describing here is that the niches are flaring open to the back of the niches such that the back of the niche is wider than its front. What is interesting is that this feature is a hallmark of niches at Tiahuanaco. Presumambly, Tiahuanaco sets a precedent for the Incas (editor's note). Letter of January 2, 1902 79 originally the main part of an ancient temple.27 Garcilaso gives, as one of the characteristic features of the temples dedicated by the Incas to the Sun, that they were constructed upon terraces, and, indeed, in addition to the temple of the Sun of Cuzco, which was described by Garcilaso as forming the ruling type of all Sun-temples erected by the Incas in the provinces, the Sun-temple at Pachacamac is also standing upon terraces. I am unable to find agreement between the main hall and Garcialaso's description of the temple of Cuzco. But neither can I find agreement between the details of the temple of Cuzco and the Sun-temple at Pachcamac, and for the latter, its use as a temple is beyond doubt. Besides this, Garcilaso helps us also by saying that the Sun-temples varied in their shape in some respects according to the province, the original customs and architectural styles of the inhabitants found in each province, probably being of some influence upon the final shape of the temple erected among them. Neither the description of the present remains of the temple of the Sun at Cuzco, as given by Squier (p. 440 and sq), nor would the description of the temple given by Garcilaso be sufficient for showing a likeness of these remains of Huaitara with the inner disposition of the Sun Temple of Cuzco upon them. It still seems possible that [p.42, (PICT0O 19)] a more exact observation of the remains of the Cuzco temple will show some closer original similarity between that of Huaitara and the latter. At the least, it is interesting to find that the church of the convent of Santo Domingo at Cuzco also seems to have been constructed by using a part of a wide hall (see the map in Squier's work, p. 441) at the base, and that nothing would be more natural than to assume that this was really the main hall of the ancient temple where the image of the Sun was standing, as the Catholic priests undoubtedly must have been eager to extirpate the worship of the Sun at this central place by erecting upon exactly the same ground upon which the image of the Sun had stood, a temple to their own supreme deity. It is a very strong proof for the original use of the old buildings of Huaitara as temple by itself, that a Christian temple has been erected upon the ancient walls of the main hall. Building a modem temple on the spot of an ancient temple seemed at all times to be the best way for extirpation of idolatry. So it was in the centuries of the ancient northern European heathendom, and it has also been so in Peru at the time of its discovery. Some ancient walls, which might be denominated rather cyclopic for the size of the stones used in their construction, [p.43, (PICT0020)] but which, however, are none other than that of the western supporting walls of the churchyard really, are still found among the walls which support garden-beds (andenes) east of the church at about 300 meters from it, and at least 60-80 meters lower down. I observed at least two of such walls. They have, of course, their front to the north, according to the declivity of the valley. In this part of the valley may have been the fields which, according to the 27 Tetype of building described here has been depicted by Guaman Poma de Ayala and given the name of "sCuyus Mango" (El primer nueva coronica y buen gobierno [ 1615], J.V. Murrua and R. Adomno, Mexico, 1980: 303) (editor's note). 80 Explorations in the Pisco Valley custom of the Incas, were chosen and dedicated to the use and service of the temple, wherever a Sun temple existed in the vast empire (the so called "Inti- chacras," "fields of the Sun"). No old parish-register exists at Huaitara. I was told that they perished in the fire of the church some decenniums ago. But the curate told me, according to tradition, the legitimacy of which could not be determined by me, that Astuarac was the name of constructor of the ancient buildings. That name is Kechua and may have been that of an Incasic nobleman, but the right of its having been brought in connection with the construction of the ancient buildings is not clear. The environs of Huaitara present various remains of ancient settlements, none of which, however, deserves of attention for its artistic prominence. Nearly all of the remains are found [p.44, (PICT002 1)] upon the very summits of the mountains surrounding Huaitara and between 500 and 800 meters above it. We find there, therefore, the same style of ancient settling as in northern Peru between Trujillo and Huamachuco. It was undoubtedly safer for the people of many ancient periods to live upon the mountains, than to live in the valleys. Ruins of an ancient village exist upon the big mountain ridge to the north of Huaitara. The ascent is extremely steep, by means of an infinite zig-zag cuesta28 which extends over two leagues (according to the Peruvian manner of measurement) that my own animals, brought from the coast valley and not accustomed yet to such steep climbing trips in the thinner air of those mountains, were of no use to me for that and other animals could not be got by me at that time. I could not visit, therefore, those ruins. The mountain Runayoj rises about 500 meters above Huaitara, east of it. One ascends to its summit from the side of the road leading to Castro Virreina. There are several summits, often with steep precipices at their sides. Some of these summits are occupied by ruins of small, ancient, village-like settlements. The views in plates 1284 and 1283 were taken in one of them, which is about 60 meters in length and 40 in width. Fragments of worked stone are found here and there among these ruins, as well as arrowheads chopped of obsidian. The ruins of the ancient habitations form an irregular crowding. They are much destroyed now, [p.45, (PICT0022)] as view 1284 can show, and overgrown by shrubs and cactus. They were never very well built or of any superior workmanship. But they present a peculiar feature, for which I cannot account yet, that is that each of those 2 1/2 -6 meter long and wide rooms, which represent one habitation, have a vault-like room at one of its sides, which served perhaps for storing provisions, as there are no signs of their having been used as kitchens anywhere. In this way the 281In Spanish, "cuesta" can mean both a slope and path running up/down that slope (editor's note). Letter of January 2, 1902 81 meter long and wide rooms, which represent one habitation, have a vault-like room at one of its sides, which served perhaps for storing provisions, as there are no signs of their having been used as kitchens anywhere. In this way the general type of the maps of in such a settlement is more or less like this: ..r I '00 __.. ~ . , - The vaults are low, only about 0.60-0.90 meters high, and 0.70-1.00 wide inside. Three of them are visible in view 1284, opened from the outside. The material of the low house-walls is quarry-stones. The space left between the rooms, generally about 1 meter higher than the floor of these rooms, served as roads for communications within the settlement. A wider place has been provided which, though very small and irregular in shape, seems to have served as a sort of plaza. The vault represented in 1283 is somewhat larger, constructed right above the side of a precipice. It was found open and, as it contained some human bones, it may have been constructed as a burial place. Other ruins are said to exist upon mount Huaitara, about 5 miles north east of Huaitara. [p.46, (PICT0023)] The road to that mount is, of course, much longer. The mountain appears in the background of view 1271 and, from its isolated position and general shape, presents some similarity with the isolated hill Ayangai, east of Motil (near the road from Trujillo to Huamachuco), which also bears (2) also (1) ruins of an ancient settlement Excavations by me in the environs of Huaitara gave unsatisfactory results. There is nothing in the immediate neighborhood of the ancient Sun Temple which might be excavated. Some ancient graves, apparently of the last (Incasic) pre-Spanish period, east of the present cemetery and about 500 meters northeast of the temple, had been opened years ago. At the northwestern, lower, entrance to the valley near Chuquimaran, there is a 82 Explorations in the Pisco Valley ancestors, and the people of mixed race seem to be too lazy to undertake or to be interested in something like that. Not a year has passed since about 100 graves of a cemetery were opened near Huancano by the people of that [p.47, (PICT0024)] village. In that cemetery, nothing is left at the present time but skulls. Spanish thirst for treasure-hunting, therefore, instigates people of the mountains continually to excavate here and there, with the result that after 50 -100 years, perhaps, it will be still more difficult to meet with ancient unopened graves there. So much I am willing to claim, at least: the most ancient civilization found by me in the lower parts of the valley, had also extended to the environs of Huaitara. Small painted fragments of pottery found in the excavations upon the summit of the mountain Runayoj, and from the steep sides of Huaitara, are the proof of this. It is interesting to see that the traces of the people of so remote centuries are found upon the mountains, while no similar proofs were found by me in the valley in the neighborhood of Huaitara. All the latter remains observed represent the Incasic civilization or that of the coast, as developed during the period of Incasic domination on the coast. The name Runayoj, of the mountain, is Kechua and signifies "that which has people." It might seem, therefore, that the mountain was still peopled at the time of the Incasic invasion into the valley. But the remains found by me there indicated they were deserted for centuries before that time. People not only lived upon the summits of the mountains, but they had also their fields there. The terraces of [p.48, (PICT0025)] those ancient fields can be observed from Huaitara and give a very striking aspect as appearing 500- 800 meters above the valley in a climatic region which generally seems to produce only shrubs and grasses. Traces of small aqueducts for irrigation- work can be observed on the higher parts of slopes of those mountains. It is difficult to understand how agricultural work could be maintained at such heights, where water for irrigation must have been extremely scarce, in general. An explanation would perhaps be that fruits were cultivated which ripened during the rainy season (like potatoes, ocas, etc.), as is done in many parts of the puna of Bolivia. All this agricultural work, so extensive in ancient times, has been abandoned, and with the aspect of this abandoned work, the feeling increases even more, that much more could be done by the modern Peruvian people for developing the resources of their country. My visit to Huaitara gave, therefore, various results; one I have just mentioned; another is the discovery and study of the small but still important remains of an Incasic temple of the Sun; the third is a negative one, though also important. Though traces of the oldest civilization observed by me in the valley of Pisco, have been found also in the environs of Huaitara, there is no reason, however, for crediting that ancient story that the first immigrants from the North in Peru [p.49, (PICT0026)] settled near Huaitara and left important old ruins there. A much more likely center of the oldest civilization of the valley of Pisco was among the people living in the valley of Ica. With respect to the first appearance of civilization in Peru, Huaitara claims no separate and especial attention at all, and its is probable that the Letter of January 2, 1902 83 fable of original immigration at Huaitara before settling the highlands of Peru had no other good base than the knowledge naturally common among the ancient Peruvians, that one of the best and most frequented roads from the coast to the highlands of Southern Peru leads through the valley of Pisco and that Huaitara was a natural station upon this road. During my stay at Huaitara, the last Sunday of my presence there was dedicated to an excursion to the ruins of Ingahuasi, the fortress- or palace-like character of which I had heard about before. Ingahuasi is about 15 miles away from Huaitara to the east, on the road to Ayacucho, about 1000 meters higher than Huaitara, as the aneroid [barometer] indicates there 480 mm of pressure. The road follows the valley of Huaitara about 7 miles up, along the mountain which borders the valley from South. It climbs then, after that, faster, up to a region more open but poorer in vegetation, and over undulating ground between elevations, which at this height do not seem to be more [p.50, (PICT0027)] than hills, one arrives, by and by, while climbing up to the region of the Peruvian puna, some 10 miles from Huaitara. There the surface of the hills has only little vegetable earth, the rocks are more barren than in the valley below, the vegetation consists of poor grass and a few low, cactus-like plants. One, or perhaps even two, huts of Indians living on cultivating the poor ground with barley, which perhaps does not ripen, can still be observed. The water of the brook, which is crossed on the journey (the head water of the small river of the valley), is chilly. At Huaitara, the rainy season last year29 began about two months earlier than other years, at about the middle of October. Though it did not rain yet as much at Huaitara at that time as it did later on, when the season advanced, the season seemed much further advanced among those hills of the puna, where my excursion was directed, and indeed, a good deal of my work done that day at Ingahuasi, in measuring, taking views etc., was done in a chilly, and not always very soft rain. It was thundering all the time and the sky was clouded, as if the world were to perish. That is the climatic character of the region of Ingahuasi, with the exception of the winter, when the sky is oftener clear, but the air very cold. Ingahuasi, the ruin which I visited, lies in a small, valley-like depression through which a small [p.5 1, (PICT0028)] stream runs from east to west. This depression may be about 300 meters long and about 200 meters wide. For, at the eastern and western end, the rocks draw closer together, thus forming a very solitary, small mountain valley between them (see pl. 1289-1292). A small plain, sloping smoothly down from the ridges of the surrounding rocks extends on the southern bank of the rivulet. Big blocks of stones are blocking its eastern entrance (pl. 1292). The road from Ayacucho leads through this lonely piece of desert, up along the southern (left) bank of the stream. 29 That is 1901, the year Uhie was at Huaitara (editor's note). 84 Explorations in the Pisco Valley The ruins are lying on the sloping plain on the left side of the water, with their front directed to the north, near to the rocks to the east, as though shelter was sought in this way against the stormy weather coming from the east. The back of the buildings is on a level with the surrounding plain, while their front is supported by terraces above the ground, which is lower there (see plates 1292, 1285). There are three buildings existing there (see map fig. 14). Remains of three smaller houses appear upon a projecting rock northwest of the former buildings (see plate 1290), but, for lack of time, could not be visited.30 The three buildings are constructed mostly of quarry stones in a very common and ordinary way (see pl. 1285-1292). But a few parts, especially the main doors of the buildings show all the perfection of workmanship (pl.1285), which we are accustomed to see in constructions made by the [p.52, (PICT0029)] Incas. On these latter parts, the stones are hewn and nicely fitted together exactly in the same way as we can see it on the walls of the main hall of the ancient temple and one the remnants of another small wall in front of it at Huaitara. Besides this, the application of a very great number of niches in all buildings is a proof of the Incasic character of the work (pl. 1286-89, 92). The sides of the doors are inclined, and the one door on each of the two principal buildings is decorated by a receding frame, in accordance with Incasic manners (pl. 1285). Two of the three buildings31 are lying at the right of the road to Ayacucho, as it passes there, with their doors against it. Both are at a distance of about 40 meters from each other, and not quite in the same orieiltation, according to the needle of the compass, to the north. The third building is lying in front of one of the former, on the left of the road, but having also with its front to north so as to lie with its back to the road. The two former buildings are very similar to each other; the third is different (see map fig. 14). The exterior appearance of the former two is very simple. They are courts with low walls. Entering the court by the door at the middle of the northern front, one observes a few smaller rooms, more or less complete at the sides, and traces of several more walls on the surface of the soil. Those latter walls once formed a number of closed rooms, of which [p.53, (PICT0030)] no trace is left. Long series of many niches (of the third type mentioned above) decorate the inner sides of the two side walls and of the back wall of the court. They are now lying free to the view as the front walls of the rooms to which they belonged have been destroyed. 30 Probably for the same lack of time Uhie seems to have missed some beautiful fountains, carved rocks and adjoining structures of very fine stonework along the creek, as well as the incoming lnca road, immediately to the east of the building marked III on Uhle's map (fig. 14). This map is on p.132 and was originally reproduced in Max Uhie, 1856-1944 Plane archaclogicher Statten im Andengebiet, A VA-Matierialien, W. W. Wurster, editor, Verlag Phillip von Zabern, Mainz am Rheim, 1999, fig. 79, p.173 (editor's note). 31 The buildings Uhle refers to would better be described as compounds as they consist of a cluster of buildings (editor's note). Letter of January 2, 1902 85 The eastern of these two buildings32 was the foremost of the three. A separate terrace with a staircase interrupting it in the middle (pl. 1285), extends along its front (pl. 1285, 88, 89). Two houses with steep gables at their eastern and western ends have been built inside, against the front of the building and beside the principal entrance,33 which is decorated by a receding frame. Both houses are divided in two by a high wall running between the two gables, which also served as support for the roof; the two parts do not communicate with each other. Those at the front were entered by separate doors alongside the principal door. These other two doors were a little inferior in perfection and were also lacking the decoration of the receding frames. The two rooms of the inner sides of the two houses were entered from the inner side of the court. Unfortunately, the front walls of the rooms inside the court are missing now (see pl. 1289). The gables of the houses are 6 meters high (pl.1286). The inside of their rooms is decorated by niches (of the third type, as above), about 0.82 meters high, 0.45 wide and [p. 54, (PICT0031)] 0.50 meters deep. In addition, each room possesses two windows, one in each gable. The front wall of the houses rises 3.60 meters above the terrace in front of the buildings, but on the inside of the rooms it is only 1.20 meters high. The gables rise, therefore, about 4.80 meters above the top of the front walls. As the distance from the front of the houses to the long middle wall inside does not amount to more than about 4.40 meters and the ascent of the sides of the gables is still shorter, the sides of the resulting gables were declivitous, as we see them in plates 1288, 1289, and 1290. A small corridor is left between the two houses, joining the terrace with the inner court by passing only through the principal door of the building. These two houses inside the court of the first building seem to have been the only ones among these ruins roofed in this way. The shelter which they gave to their occupants is the most appropriate in the climate of the mountains. Nobody will roof houses flat in the mountains, while on the coast, where rain is very scarce, flat roofs largely prevail. High gables can also be observed among the ruins of the ancient convent of the island Coati, in Lake Titicaca, among the Incasic ruins of Tarmatambo and various other ruins of Incasic origin. For the good shelter given by the declivitous roofs, even to the walls themselves, the higher part of the latter could be constructed of bricks of adobe (see pl. 1286), while this [p.55, (PICT0032)] material is not observed in other parts of the same ruins. The second building (see map and pl. 1288), though similar in its general disposition to the first, is simpler in its details. It was entered by a framed doorway and steps which undoubtedly led up to the level of the court. The interior once contained about 8 different rooms of different size and shape, at the sides and at the back. -The separating walls have mostly disappeared. As in the first building, the middle part of the building must have formned an open plaza-like court. 32 Compound 1 in Uhie's map (fig. 14) (editor's note). 33 The buildings to which Uhie refers flank the doorway to the compound, one on either side (editor's note). 86 Explorations in the Pisco Valley The third building is partially destroyed (pl. 1287). Its northern front wall (pl. 129 1) is of higher perfection than those of the other two buildings. This higher perfection was perhaps given it for practical reasons, on account of its proximity to the channel of the brook, which sometimes swells a great deal in summer time. This is to be believed, also, as the northern door of the building, though well constructed, lacks the receding frame which is an element of decoration on the two other buildings. The building is smaller than the two other, and shows a very symmetrical disposition of the rooms in front and at the back, and to the left and right. A small corridor between rooms led from the northern entrance, probably up steps, to a small plaza which was enclosed by the rooms on its four sides. Apparently several doors (1.00 meter high by 0.85 meters wide) led [p.56, (PICT0033)] to each room and niches served as decoration and probably also for practical purposes, at the back of the walls. Many walls have disappeared or are incomplete, but the general map of the building can still be traced. On comparing more closely the complete map of one of the buildings with the incomplete maps of the two others, I am struck, notwithstanding, by the general similarity among all three. The disposition of rooms is symmetrical in all three: smaller rooms on the four sides of the court embrace a larger plaza-like room in the middle part. A long gallery-like room seems to have existed in the back of all three buildings without enclosing the rooms at the corners, which were separate.34 I like, therefore, to suggest that the general disposition of the three buildings was originally the same, that the disposition of rooms of the third building was also that of the rooms of the other two, and that the three buildings became unlike each other by the disappearance of various walls, which are apparently now missing, in the two former buildings. But we can go farther. Is not the inner disposition of the third building absolutely identical in its general plan with that which we exposed as having been the general type of the palaces constructed by the Incas at Tambo Colorado? [p.57, (PICT0034)] I am no longer in doubt that the buildings are constructed by the Incas according to the technique of masonry, shape and decoration of the doors, and the typical shapes of the Incasic niches applied in the decoration of the rooms. But the general disposition of the palaces of Tambo Colorado and of the third building agree in the following parts entirely: Middle door at the front Court in the middle part of the building, enclosed on all sides by rooms The general proportion between court and surrounding rooms, as far as the latter do not occupy more space in the buildings than galleries would do around the court. 34Th arrangement of buildings around a courtyard, sometimes surrounded by a wall (as in this case described by Uhle) is today referred to as a "kancha" (editor's note). Letter of January 2, 1902 87 The plan of the three buildings, therefore, answers, according to my mind, to the general map of houses, or - as no other houses are known yet than palaces, and some prominent character is expressed on two of the buildings by the framed decoration of their front doors- to the general character of palaces as built by the Incas. The first of the buildings, however, possesses a striking particularity in the position of the gable-houses, which unavoidably were prominent parts of the buildings, in addition to the front-door. There is another curious coincidence between the palaces built at Tambo Colorado and [p.58, (PICTOO35)] these buildings at Ingahuasi. Here, as there, three buildings of similar character have been constructed. As I pointed out, that the inhabitants of all three buildings of Tambo Colorado were different rank, the same can be suggested for the inhabitants of the three buildings of Ingahuasi because of the different exterior appearance of the buildings. An interesting question arises here, the solution of which must be left in doubt: Is it a coincidence that we meet three Incasic buildings of quite the same general disposition at Ingahuasi as well as at Tambo Colorado? And if not, does the number three indicate something respecting the number of authoritative persons, or respecting administrative divisions, in general in this case? Considering the strong formalism observed by the highest authoritative person of the Incasic empire in all its appearances before the public, I cannot help but suggest that the number three was caused by a certain kind of distribution of administrative powers in which the number three occurred. "Ingahuasi" means in Kechua "House of the Inca." I believe tradition is right in considering these buildings as being for the Inca personally. The various expressive decorations in these buildings point to assuming this. [p.59, (PICT0036)] This suggestion is corroborated by an additional consideration. Who in the world would be inclined to build a palace, sumptuous in some way, in a lonely ray-struck desert, far from people and from all better resources of living, if it were for living there continually? It is, therefore, to be assumed, that the buildings were built for transitory service. They are lying on the road to Ayacucho and it is, therefore, probable, that they served as night-harbors on the road. A certain inequality in the technical perfection of the various parts of the buildings points in the same direction: to assume that they were constructed to serve only for transitory purposes. But who would have constructed palaces for night-harbors on the road in Peru, if not the Inca? The Inca traveled along this road. The well-constructed road leading through the valley of Pisco allows us to infer this. The Inca resided sometimes at Tambo Colorado. His way from Cuzco by Ayacucho to Tambo Colorado and the coast was along this road, and it is quite credible that the construction of these buildings arose from his wish to have a good shelter at the end of a journey during the tiresome traveling along the Peruvian puna. Another curious observation is to be made here. Thle map of Peru made by Raimondi shows [p.60, (PICTO037)] a geographical name "Ingahuasi" 88 Explorations in the Pisco Valley about 63 kilometers directly east of our "Ingahuasi" on the road to Ayacucho. That other "Ingahuasi" is the name of an hacienda of cattle, etc. It was denied to me that ruins are found in this latter "Ingahuasi." It is, however, unquestionable that such had existed there if they do not exist there now. Measuring now the distances on the map, it curiously appears that the distance between Tambo Colorado from our Ingahuasi above Huaitara, in a direct line on the map amounts to also nearly exactly 65 kilometers. Our Ingahuasi above Huaitara, therefore, lies exactly in the middle between Tambo Colorado, the most important palace of the Inca on the coast, and the present hacienda of cattle called Ingahuasi, on the road to Ayacucho. It seems, therefore, necessary to infer from that that the distances from Tambo Colorado to the first "Ingahuasi" above Huaitara and from there to the second Ingahuasi, the hacienda of cattle, each represented one journey in the Inca's traveling through his empire.35 If I am right in making this conclusion from all the circumstances surrounding our Ingahuasi, and coming into view during its study, it seems to me something has been also done for improving our [p.61, (PICT0038)] understanding of the governmental machine of the interesting Incasic empire. 35 Uhie noted in the margin: "measuring from the Incasic palaces near Tambo de Mora (Chincha valley) to Tambo Colorado, in a direct line on the map, one finds about 79 km as the distance. As the roads of the coast are more comfortable than those of the mountains, this distance appears to be the length of another (third) journey during the Inca's traveling."